It’s still not fixed, but it's getting late here. I would try to simply short out the switch, but it’s very difficult to identify and reach its microscopic solder pads and I’m no good at micro soldering. I think my best bet will be to mechanically remove the switch, crushing and cutting
it as necessary, hoping that it will yield access to solder points that
I can bridge with a short. If that cures it, fine. And I’ll probably buy that other light anyway.
I'm not an electronics guy. But speaking of problems, I think I
mentioned a few days ago that on a recent night ride, my headlight (B&M
Eyc N plus), powered by my good old sidewall generator, was randomly
and sporadically blinking off. I put the bike on the workstand, checked
out the wiring (it’s original and pristine), and spent some time
spinning the front wheel. Sometimes the headlight worked, sometimes not,
and it changed state at random without having been touched. I was
suspicious of the tiny on/off switch hidden behind a rubber plug, in
part because I felt no satisfying click.
I substituted a different (lower quality) LED dynamo light, which seemed
to work perfectly and probably absolved the generator. I eventually
talked by phone to Peter White, who sold me the light far longer ago
than I remembered - ten years! I asked if he had clues on diagnosing the problem. He did not, but suggested that the identical light without the possible failure point of an on/off switch is just $35. (A sidewall
dyno doesn't need a switch.) Still, I opted to open the light to see if
I could spot anything because I have a compulsion about fixing things -
but a headlight has to be really, really reliable.
Here are photos of the electronic guts, with a vernier caliper for size
(and to educate Tom!).
Bottom view. The downward facing LED is at the top of the photo, the suspicious switch is the tiny white box at the photo’s bottom. https:// www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-public/
Top of the circuit board. The large copper pad is part of the heat sink system, contacting a metal casting that forms the top of the headlamp
body. I’ll probably add some heat conducting goop when I reassemble: https://www.flickr.com/photos/16972296@N08/54435279589/in/dateposted-
public/
Although a bike generator or dynamo puts out AC, I had hoped to diagnose
it using DC, figuring a 6 volt battery would essentially supply half of
the sine wave so half of the input circuit. That naturally works with incandescents, and it worked with one Avenir LED dyno headlight that I repaired. Not so in this case. So I ran another generator using a roller chucked in my drill press, attached a digital voltmeter in parallel, and poked around best I could. I saw nothing loose, no scorched components, nothing obviously wrong.
With no power attached, I tried checking for continuity across the
switch. It always showed close to zero ohms, never infinity - but the resistance readings with pinpoint probes were jumpy, and since the
switch wasn’t isolated from the circuit, that may not mean much.
I used contact cleaner to blast out the switch and repeatedly clicked
it, including while power was applied. Except for a few occasions, the
light did not respond to the switch, so I still suspect the switch. When
the light was off, I had about 12 VDC. When on, about 4 VDC.
It’s still not fixed, but it's getting late here. I would try to simply short out the switch, but it’s very difficult to identify and reach its microscopic solder pads and I’m no good at micro soldering. I think my best bet will be to mechanically remove the switch, crushing and cutting
it as necessary, hoping that it will yield access to solder points that
I can bridge with a short. If that cures it, fine. And I’ll probably buy that other light anyway.
BTW, Peter White is not pleased about the tariff situation. He says he won’t raise prices on current stock, but with the next shipment he’ll probably have to.
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
I'm not an electronics guy. But speaking of problems, I
think I mentioned a few days ago that on a recent night
ride, my headlight (B&M Eyc N plus), powered by my good
old sidewall generator, was randomly and sporadically
blinking off. I put the bike on the workstand, checked out
the wiring (it’s original and pristine), and spent some
time spinning the front wheel. Sometimes the headlight
worked, sometimes not, and it changed state at random
without having been touched. I was suspicious of the tiny
on/off switch hidden behind a rubber plug, in part because
I felt no satisfying click.
I substituted a different (lower quality) LED dynamo
light, which seemed to work perfectly and probably
absolved the generator. I eventually talked by phone to
Peter White, who sold me the light far longer ago than I
remembered - ten years! I asked if he had clues on
diagnosing the problem. He did not, but suggested that the
identical light without the possible failure point of an
on/off switch is just $35. (A sidewall dyno doesn't need a
switch.) Still, I opted to open the light to see if I
could spot anything because I have a compulsion about
fixing things - but a headlight has to be really, really
reliable.
Here are photos of the electronic guts, with a vernier
caliper for size (and to educate Tom!).
Bottom view. The downward facing LED is at the top of the
photo, the suspicious switch is the tiny white box at the
photo’s bottom. https:// www.flickr.com/
photos/16972296@N08/54435279894/in/dateposted-public/
Top of the circuit board. The large copper pad is part of
the heat sink system, contacting a metal casting that
forms the top of the headlamp body. I’ll probably add some
heat conducting goop when I reassemble:
https://www.flickr.com/photos/16972296@N08/54435279589/in/
dateposted- public/
Although a bike generator or dynamo puts out AC, I had
hoped to diagnose it using DC, figuring a 6 volt battery
would essentially supply half of the sine wave so half of
the input circuit. That naturally works with
incandescents, and it worked with one Avenir LED dyno
headlight that I repaired. Not so in this case. So I ran
another generator using a roller chucked in my drill
press, attached a digital voltmeter in parallel, and poked
around best I could. I saw nothing loose, no scorched
components, nothing obviously wrong.
With no power attached, I tried checking for continuity
across the switch. It always showed close to zero ohms,
never infinity - but the resistance readings with pinpoint
probes were jumpy, and since the switch wasn’t isolated
from the circuit, that may not mean much.
I used contact cleaner to blast out the switch and
repeatedly clicked it, including while power was applied.
Except for a few occasions, the light did not respond to
the switch, so I still suspect the switch. When the light
was off, I had about 12 VDC. When on, about 4 VDC.
It’s still not fixed, but it's getting late here. I would
try to simply short out the switch, but it’s very
difficult to identify and reach its microscopic solder
pads and I’m no good at micro soldering. I think my best
bet will be to mechanically remove the switch, crushing
and cutting it as necessary, hoping that it will yield
access to solder points that I can bridge with a short. If
that cures it, fine. And I’ll probably buy that other
light anyway.
BTW, Peter White is not pleased about the tariff
situation. He says he won’t raise prices on current stock,
but with the next shipment he’ll probably have to.
One possibility is open/"cold" solder connection somewhere.
These assemblies are built with what is known as a 'reflow'
soldering process - essentially, the PCB has a solder paste
applied through a process similar to silk-screen, then the
parts a put on the board using a "pick and place" machine,
then it's run through a hot oven which melts the solder
paste and hopefully wets the solder to both the component
and the pad under it.
Sometimes, it doesn't quite flow right and ends up with a
"cold" connection. The connection may work for some time,
over time will oxidize such that it opens the connection.
Normal cleaning techniques seldom work to fix this, it
usually takes reheating the offending connection to reflow
the solder, and some type of flux is highly recommended.
One way to diagnose this is to apply power to the circuit
and press on the components, though this sometimes doesn't
work either. If it does, a light touch with a soldering iron
and on the component connections _usually_ works.
Someone with good understanding of soldering techniques
might be able to see a cold solder joint under a microscope
(10X usually suffices).
I recently bought a new blinky light for my commuter and the
switch was intermittent out of the box. A cold solder joint
on the switch contacts was the culprit.
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
I did exactly that (reflow poor solder joints) for a control
on the board of our shop truck's radio last autumn. For Mr
Krygowski's problem, the scale is dauntingly smaller. I am
no expert but it looks difficult to me.
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there are
ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as
it would only be lit for half the cycle.
For an inexpensive LED light I wouldn't spend too much time trying to
repair it. If it were a high-end, LED light then that would be a
different story.
I also have a cheap dynamo light on one bike, the Planet Bike Blaze
dynamo, the only dynamo light that I'm aware of that has a flash mode.
It has a switch that I never use, it's always on flash mode, and it's
powered by a hub dynamo. At night I have a more powerful light of course.
AMuzi <am@yellowjersey.org> wrote:
I did exactly that (reflow poor solder joints) for a control
on the board of our shop truck's radio last autumn. For Mr
Krygowski's problem, the scale is dauntingly smaller. I am
no expert but it looks difficult to me.
Indeed, for us old folks 8-)
There's an instructive YouTube channel called "northridgefix" which illustrates modern methods.
But, with a pointy iron and good flux it's worth a try. Especially
when, as in this case, there's nothing to lose.
bob prohaska
On 4/7/2025 11:56 AM, sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there
are ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as
it would only be lit for half the cycle.
For an inexpensive LED light I wouldn't spend too much time trying to
repair it. If it were a high-end, LED light then that would be a
different story.
I also have a cheap dynamo light on one bike, the Planet Bike Blaze
dynamo, the only dynamo light that I'm aware of that has a flash mode.
It has a switch that I never use, it's always on flash mode, and it's
powered by a hub dynamo. At night I have a more powerful light of course.
A rectified AC wave form would work as long as the voltage is within
spec of the LED. Pulsatile waveforms are exceeding common in LED driver systems. It's usually driven by a PWM for the intensity control for
lighting and through a shift register for a matrix intended for
information.
It isn't wasteful, it's actually more efficient to drive with a lower
duty cycle than straight DC.
On Sun Apr 6 22:22:33 2025 Frank Krygowski wrote:
I'm not an electronics guy. But speaking of problems, I think I
mentioned a few days ago that on a recent night ride, my headlight (B&M
Eyc N plus), powered by my good old sidewall generator, was randomly
and sporadically blinking off. I put the bike on the workstand, checked
out the wiring (it?s original and pristine), and spent some time
spinning the front wheel. Sometimes the headlight worked, sometimes not,
and it changed state at random without having been touched. I was
suspicious of the tiny on/off switch hidden behind a rubber plug, in
part because I felt no satisfying click.
I substituted a different (lower quality) LED dynamo light, which seemed
to work perfectly and probably absolved the generator. I eventually
talked by phone to Peter White, who sold me the light far longer ago
than I remembered - ten years! I asked if he had clues on diagnosing the
problem. He did not, but suggested that the identical light without the
possible failure point of an on/off switch is just $35. (A sidewall
dyno doesn't need a switch.) Still, I opted to open the light to see if
I could spot anything because I have a compulsion about fixing things -
but a headlight has to be really, really reliable.
Here are photos of the electronic guts, with a vernier caliper for size
(and to educate Tom!).
Bottom view. The downward facing LED is at the top of the photo, the
suspicious switch is the tiny white box at the photo?s bottom.
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-public/ >>
Top of the circuit board. The large copper pad is part of the heat sink
system, contacting a metal casting that forms the top of the headlamp
body. I?ll probably add some heat conducting goop when I reassemble:
https://www.flickr.com/photos/16972296@N08/54435279589/in/dateposted-public/ >>
Although a bike generator or dynamo puts out AC, I had hoped to diagnose
it using DC, figuring a 6 volt battery would essentially supply half of
the sine wave so half of the input circuit. That naturally works with
incandescents, and it worked with one Avenir LED dyno headlight that I
repaired. Not so in this case. So I ran another generator using a roller
chucked in my drill press, attached a digital voltmeter in parallel, and
poked around best I could. I saw nothing loose, no scorched components,
nothing obviously wrong.
With no power attached, I tried checking for continuity across the
switch. It always showed close to zero ohms, never infinity - but the
resistance readings with pinpoint probes were jumpy, and since the
switch wasn?t isolated from the circuit, that may not mean much.
I used contact cleaner to blast out the switch and repeatedly clicked
it, including while power was applied. Except for a few occasions, the
light did not respond to the switch, so I still suspect the switch. When
the light was off, I had about 12 VDC. When on, about 4 VDC.
It?s still not fixed, but it's getting late here. I would try to simply
short out the switch, but it?s very difficult to identify and reach its
microscopic solder pads and I?m no good at micro soldering. I think my
best bet will be to mechanically remove the switch, crushing and cutting
it as necessary, hoping that it will yield access to solder points that
I can bridge with a short. If that cures it, fine. And I?ll probably buy
that other light anyway.
BTW, Peter White is not pleased about the tariff situation. He says he
won?t raise prices on current stock, but with the next shipment he?ll
probably have to.
What is the purpose of holding that PC board in a caliper?
Was the size of the board in question?
Tell us the function of the PC board
and the likely source of your blinking problem?
Frank, you were an instructor, you do not have any practical knowledge of anything.
On 4/7/2025 9:08 AM, Zen Cycle wrote:
On 4/7/2025 11:56 AM, sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there
are ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful
as it would only be lit for half the cycle.
For an inexpensive LED light I wouldn't spend too much time trying to
repair it. If it were a high-end, LED light then that would be a
different story.
I also have a cheap dynamo light on one bike, the Planet Bike Blaze
dynamo, the only dynamo light that I'm aware of that has a flash
mode. It has a switch that I never use, it's always on flash mode,
and it's powered by a hub dynamo. At night I have a more powerful
light of course.
A rectified AC wave form would work as long as the voltage is within
spec of the LED. Pulsatile waveforms are exceeding common in LED
driver systems. It's usually driven by a PWM for the intensity control
for lighting and through a shift register for a matrix intended for
information.
It isn't wasteful, it's actually more efficient to drive with a lower
duty cycle than straight DC.
Correct that rectified AC would work but Frank was not using rectified
AC, he was using DC.
PWM DC is very different than AC.
With pure AC, the LED would be
conducting during only half the cycle.
With AC, it's done like this withYup, that's one way to do it.
two LEDs <https://www.heathcote-electronics.co.uk/images-leds/two-leds- wired-to-AC-power-supply.jpg>.
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
It?s still not fixed, but it's getting late here. I would try to simply short out the switch, but it?s very difficult to identify and reach its microscopic solder pads and I?m no good at micro soldering. I think my best bet will be to mechanically remove the switch, crushing and cutting
it as necessary, hoping that it will yield access to solder points that
I can bridge with a short. If that cures it, fine. And I?ll probably buy that other light anyway.
If the switch used to click and now doesn't, it's likely bad. I'd
be somewhat surprised if the illustrated switches were able to click
at all, they seem mighty small. They also look well-sealed.
With ohmmeter probes across the switch contacts it should either be
zero (closed) or something else (rest of circuit) when open. If it's
not zero, I'd agree either the switch is faulty or you've got a bad connection in your test leads. Jumpy readings are always suspect.
A close inspection with a magnifier might reveal unsound solder joints.
It might be possible to reflow at least those that look dewetted or
lean. Long shot, but I've fixed one or two gadgets that way. In the
second photo the solder joints either side of SW2 look less than ideal.
On Mon Apr 7 12:17:17 2025 Zen Cycle wrote:
On 4/7/2025 12:10 PM, cyclintom wrote:
On Sun Apr 6 22:22:33 2025 Frank Krygowski wrote:
I'm not an electronics guy. But speaking of problems, I think I
mentioned a few days ago that on a recent night ride, my headlight (B&M >>>> Eyc N plus), powered by my good old sidewall generator, was randomly
and sporadically blinking off. I put the bike on the workstand, checked >>>> out the wiring (it?s original and pristine), and spent some time
spinning the front wheel. Sometimes the headlight worked, sometimes not, >>>> and it changed state at random without having been touched. I was
suspicious of the tiny on/off switch hidden behind a rubber plug, in
part because I felt no satisfying click.
I substituted a different (lower quality) LED dynamo light, which seemed >>>> to work perfectly and probably absolved the generator. I eventually
talked by phone to Peter White, who sold me the light far longer ago
than I remembered - ten years! I asked if he had clues on diagnosing the >>>> problem. He did not, but suggested that the identical light without the >>>> possible failure point of an on/off switch is just $35. (A sidewall
dyno doesn't need a switch.) Still, I opted to open the light to see if >>>> I could spot anything because I have a compulsion about fixing things - >>>> but a headlight has to be really, really reliable.
Here are photos of the electronic guts, with a vernier caliper for size >>>> (and to educate Tom!).
Bottom view. The downward facing LED is at the top of the photo, the
suspicious switch is the tiny white box at the photo?s bottom.
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-public/
Top of the circuit board. The large copper pad is part of the heat sink >>>> system, contacting a metal casting that forms the top of the headlamp
body. I?ll probably add some heat conducting goop when I reassemble:
https://www.flickr.com/photos/16972296@N08/54435279589/in/dateposted-public/
Although a bike generator or dynamo puts out AC, I had hoped to diagnose >>>> it using DC, figuring a 6 volt battery would essentially supply half of >>>> the sine wave so half of the input circuit. That naturally works with
incandescents, and it worked with one Avenir LED dyno headlight that I >>>> repaired. Not so in this case. So I ran another generator using a roller >>>> chucked in my drill press, attached a digital voltmeter in parallel, and >>>> poked around best I could. I saw nothing loose, no scorched components, >>>> nothing obviously wrong.
With no power attached, I tried checking for continuity across the
switch. It always showed close to zero ohms, never infinity - but the
resistance readings with pinpoint probes were jumpy, and since the
switch wasn?t isolated from the circuit, that may not mean much.
I used contact cleaner to blast out the switch and repeatedly clicked
it, including while power was applied. Except for a few occasions, the >>>> light did not respond to the switch, so I still suspect the switch. When >>>> the light was off, I had about 12 VDC. When on, about 4 VDC.
It?s still not fixed, but it's getting late here. I would try to simply >>>> short out the switch, but it?s very difficult to identify and reach its >>>> microscopic solder pads and I?m no good at micro soldering. I think my >>>> best bet will be to mechanically remove the switch, crushing and cutting >>>> it as necessary, hoping that it will yield access to solder points that >>>> I can bridge with a short. If that cures it, fine. And I?ll probably buy >>>> that other light anyway.
BTW, Peter White is not pleased about the tariff situation. He says he >>>> won?t raise prices on current stock, but with the next shipment he?ll
probably have to.
What is the purpose of holding that PC board in a caliper?
He stated it in his message.
Was the size of the board in question?
Wow...You don't know how to read a caliper?
Tell us the function of the PC board
He did
and the likely source of your blinking problem?
He speculated, not being an EE or Elec Tech.
Frank, you were an instructor, you do not have any practical knowledge of anything.
did you actually bother to read any of this before posting?
Why do you have to show you're a blithering idiot all of the time? Gemeral knowledge gained by someone else telling him doesn't mean he knows what he ius looking at or understands possible problem areas. And it is plain that neither do you.
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there are
ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as
it would only be lit for half the cycle.
For an inexpensive LED light I wouldn't spend too much time trying to
repair it. If it were a high-end, LED light then that would be a
different story.
I also have a cheap dynamo light on one bike, the Planet Bike Blaze
dynamo, the only dynamo light that I'm aware of that has a flash mode.
It has a switch that I never use, it's always on flash mode, and it's
powered by a hub dynamo. At night I have a more powerful light of course.
On Mon Apr 7 08:56:02 2025 sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there are
ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as
it would only be lit for half the cycle.
For an inexpensive LED light I wouldn't spend too much time trying to
repair it. If it were a high-end, LED light then that would be a
different story.
I also have a cheap dynamo light on one bike, the Planet Bike Blaze
dynamo, the only dynamo light that I'm aware of that has a flash mode.
It has a switch that I never use, it's always on flash mode, and it's
powered by a hub dynamo. At night I have a more powerful light of course.
Stop trying to explain things to someone who doesn't have the concepts. The Problem wikth simply using two reversed LED's is that dynamos increase their ACV with speed. if you simply cut the tops off of the waves if they havemore than the forwaed biasmost of the time. But that throws away a lot of energy. And the chopper would have to be able to get rid of a lot of heat. I can think of some nice ways to increase performance but Frank is in no position to design PC boards. And Flunky thinks that wave
One of the reasons I was worth so much money was because I knew every step of manufacturing and could not only design and progam but analyze every procedure from assembly to end product and even packaging and shipping.
On 4/7/2025 9:08 AM, Zen Cycle wrote:
On 4/7/2025 11:56 AM, sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there
are ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as
it would only be lit for half the cycle.
For an inexpensive LED light I wouldn't spend too much time trying to
repair it. If it were a high-end, LED light then that would be a
different story.
I also have a cheap dynamo light on one bike, the Planet Bike Blaze
dynamo, the only dynamo light that I'm aware of that has a flash mode.
It has a switch that I never use, it's always on flash mode, and it's
powered by a hub dynamo. At night I have a more powerful light of course. >>
A rectified AC wave form would work as long as the voltage is within
spec of the LED. Pulsatile waveforms are exceeding common in LED driver systems. It's usually driven by a PWM for the intensity control for lighting and through a shift register for a matrix intended for information.
It isn't wasteful, it's actually more efficient to drive with a lower
duty cycle than straight DC.
Correct that rectified AC would work but Frank was not using rectified
AC, he was using DC.
PWM DC is very different than AC. With pure AC, the LED would be
conducting during only half the cycle. With AC, it's done like this with
two LEDs <https://www.heathcote-electronics.co.uk/images-leds/two-leds-wired-to-AC-power-supply.jpg>.
On Mon Apr 7 13:09:45 2025 Zen Cycle wrote:bias most of the time. But that throws away a lot of energy. And the chopper would have to be able to get rid of a lot of heat. I can think of some nice ways to increase performance but Frank is in no position to design PC boards. And Flunky thinks that
On 4/7/2025 1:01 PM, cyclintom wrote:
On Mon Apr 7 08:56:02 2025 sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there are >>>> ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as >>>> it would only be lit for half the cycle.
For an inexpensive LED light I wouldn't spend too much time trying to
repair it. If it were a high-end, LED light then that would be a
different story.
I also have a cheap dynamo light on one bike, the Planet Bike Blaze
dynamo, the only dynamo light that I'm aware of that has a flash mode. >>>> It has a switch that I never use, it's always on flash mode, and it's
powered by a hub dynamo. At night I have a more powerful light of course. >>>
Stop trying to explain things to someone who doesn't have the concepts. The Problem wikth simply using two reversed LED's is that dynamos increase their ACV with speed. if you simply cut the tops off of the waves if they havemore than the forwaed
And, As usual....
https://www.mclpcb.com/blog/solder-joint/
And yes, that article is about voids in surface mount reflow systems.
Besides that, you dumbass, I didn't mention voids. I mentioned cold
solder - different than a void. That you obviously don't know the
difference speaks even more to your overall ineptitude.
One of the reasons I was worth so much money was because I knew every step of manufacturing and could not only design and progam but analyze every procedure from assembly to end product and even packaging and shipping.
And yet you never had to deal with the FDA when designing medical
equipment....lol
I never had to deal with the FDA because it wasn't necesary for a machine, not being a food or a drug.
Chapter I Food and Drug Administration, Department of Health andHuman Services 1 – 1299
Subchapter H Medical Devices 800 – 898
While heart regulators are required to meet FDA standards - lawyers delt with tyhat.
But explain why a rectal thermometer - your favorite toy -
would need FDA approval beyond the initial calibration of the production equipment?
I'm not an electronics guy. But speaking of problems, I think I
mentioned a few days ago that on a recent night ride, my headlight (B&M
Eyc N plus), powered by my good old sidewall generator, was randomly
and sporadically blinking off. I put the bike on the workstand, checked
out the wiring (it’s original and pristine), and spent some time
spinning the front wheel. Sometimes the headlight worked, sometimes not,
and it changed state at random without having been touched. I was
suspicious of the tiny on/off switch hidden behind a rubber plug, in
part because I felt no satisfying click.
I substituted a different (lower quality) LED dynamo light, which seemed
to work perfectly and probably absolved the generator. I eventually
talked by phone to Peter White, who sold me the light far longer ago
than I remembered - ten years! I asked if he had clues on diagnosing the problem. He did not, but suggested that the identical light without the possible failure point of an on/off switch is just $35. (A sidewall
dyno doesn't need a switch.) Still, I opted to open the light to see if
I could spot anything because I have a compulsion about fixing things -
but a headlight has to be really, really reliable.
Here are photos of the electronic guts, with a vernier caliper for size
(and to educate Tom!).
Bottom view. The downward facing LED is at the top of the photo, the suspicious switch is the tiny white box at the photo’s bottom. https:// www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-public/
Top of the circuit board. The large copper pad is part of the heat sink system, contacting a metal casting that forms the top of the headlamp
body. I’ll probably add some heat conducting goop when I reassemble: https://www.flickr.com/photos/16972296@N08/54435279589/in/dateposted-
public/
Although a bike generator or dynamo puts out AC, I had hoped to diagnose
it using DC, figuring a 6 volt battery would essentially supply half of
the sine wave so half of the input circuit. That naturally works with incandescents, and it worked with one Avenir LED dyno headlight that I repaired. Not so in this case. So I ran another generator using a roller chucked in my drill press, attached a digital voltmeter in parallel, and poked around best I could. I saw nothing loose, no scorched components, nothing obviously wrong.
With no power attached, I tried checking for continuity across the
switch. It always showed close to zero ohms, never infinity - but the resistance readings with pinpoint probes were jumpy, and since the
switch wasn’t isolated from the circuit, that may not mean much.
I used contact cleaner to blast out the switch and repeatedly clicked
it, including while power was applied. Except for a few occasions, the
light did not respond to the switch, so I still suspect the switch. When
the light was off, I had about 12 VDC. When on, about 4 VDC.
It’s still not fixed, but it's getting late here. I would try to simply short out the switch, but it’s very difficult to identify and reach its microscopic solder pads and I’m no good at micro soldering. I think my best bet will be to mechanically remove the switch, crushing and cutting
it as necessary, hoping that it will yield access to solder points that
I can bridge with a short. If that cures it, fine. And I’ll probably buy that other light anyway.
BTW, Peter White is not pleased about the tariff situation. He says he won’t raise prices on current stock, but with the next shipment he’ll probably have to.
On Sun Apr 6 22:22:33 2025 Frank Krygowski wrote:
I'm not an electronics guy. But speaking of problems, I think I
mentioned a few days ago that on a recent night ride, my headlight (B&M
Eyc N plus), powered by my good old sidewall generator, was randomly
and sporadically blinking off. I put the bike on the workstand, checked
out the wiring (it?s original and pristine), and spent some time
spinning the front wheel. Sometimes the headlight worked, sometimes not,
and it changed state at random without having been touched. I was
suspicious of the tiny on/off switch hidden behind a rubber plug, in
part because I felt no satisfying click.
I substituted a different (lower quality) LED dynamo light, which seemed
to work perfectly and probably absolved the generator. I eventually
talked by phone to Peter White, who sold me the light far longer ago
than I remembered - ten years! I asked if he had clues on diagnosing the
problem. He did not, but suggested that the identical light without the
possible failure point of an on/off switch is just $35. (A sidewall
dyno doesn't need a switch.) Still, I opted to open the light to see if
I could spot anything because I have a compulsion about fixing things -
but a headlight has to be really, really reliable.
Here are photos of the electronic guts, with a vernier caliper for size
(and to educate Tom!).
Bottom view. The downward facing LED is at the top of the photo, the
suspicious switch is the tiny white box at the photo?s bottom.
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-public/ >>
Top of the circuit board. The large copper pad is part of the heat sink
system, contacting a metal casting that forms the top of the headlamp
body. I?ll probably add some heat conducting goop when I reassemble:
https://www.flickr.com/photos/16972296@N08/54435279589/in/dateposted-public/ >>
Although a bike generator or dynamo puts out AC, I had hoped to diagnose
it using DC, figuring a 6 volt battery would essentially supply half of
the sine wave so half of the input circuit. That naturally works with
incandescents, and it worked with one Avenir LED dyno headlight that I
repaired. Not so in this case. So I ran another generator using a roller
chucked in my drill press, attached a digital voltmeter in parallel, and
poked around best I could. I saw nothing loose, no scorched components,
nothing obviously wrong.
With no power attached, I tried checking for continuity across the
switch. It always showed close to zero ohms, never infinity - but the
resistance readings with pinpoint probes were jumpy, and since the
switch wasn?t isolated from the circuit, that may not mean much.
I used contact cleaner to blast out the switch and repeatedly clicked
it, including while power was applied. Except for a few occasions, the
light did not respond to the switch, so I still suspect the switch. When
the light was off, I had about 12 VDC. When on, about 4 VDC.
It?s still not fixed, but it's getting late here. I would try to simply
short out the switch, but it?s very difficult to identify and reach its
microscopic solder pads and I?m no good at micro soldering. I think my
best bet will be to mechanically remove the switch, crushing and cutting
it as necessary, hoping that it will yield access to solder points that
I can bridge with a short. If that cures it, fine. And I?ll probably buy
that other light anyway.
BTW, Peter White is not pleased about the tariff situation. He says he
won?t raise prices on current stock, but with the next shipment he?ll
probably have to.
Frank, I have to appologize concerning the caliper measurements of bolt I was sure was 4 mm. The digital caliper was giving me intermittent readings and the manual caliper has a mm scale too small for me to accurately read (with my recent stroke myeyesight is affected) but what looks like 4 mm. But it is a dial caliper and with your insistance as well as Andrew's in mind I zerod the dial gauge and read the inch scale with was exactly 0.19 inches which is a little more than 4.8 mm So, indeed, the
"PWM DC"? PWM by definition isn't DC. PWM generators even at maximum
drive aren't 100% duty cycle (yes, effectively DC considering the
response of the LED, but still not exactly DC).
On 4/7/2025 9:28 AM, Zen Cycle wrote:
<snip>
"PWM DC"? PWM by definition isn't DC. PWM generators even at maximum
drive aren't 100% duty cycle (yes, effectively DC considering the
response of the LED, but still not exactly DC).
Yes. PWM is a DC voltage with varying pulse widths.
It never goes below 0V like AC.
BTW, Frank could have just used a transformer like ><https://www.amazon.com/dp/B094NRRKC1> to simulate a 6VAC dynamo, or--
even a 16-24VAC doorbell transformer with some resistors to divide the >voltage and limit the current, but it's really not worth spending much
money to try to fix an inexpensive bicycle headlight.
On 4/7/2025 9:28 AM, Zen Cycle wrote:
<snip>
"PWM DC"? PWM by definition isn't DC. PWM generators even at maximum
drive aren't 100% duty cycle (yes, effectively DC considering the
response of the LED, but still not exactly DC).
Yes. PWM is a DC voltage with varying pulse widths.
It never goes below
0V like AC.
BTW, Frank could have just used a transformer like <https:// www.amazon.com/dp/B094NRRKC1> to simulate a 6VAC dynamo, or even a
16-24VAC doorbell transformer with some resistors to divide the voltage
and limit the current, but it's really not worth spending much money to
try to fix an inexpensive bicycle headlight.
On 4/7/2025 2:58 PM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:It has no numbers or other codes at all! It's solid black with a sort of
I'm not an electronics guy. But speaking of problems, I think I
mentioned a few days ago that on a recent night ride, my headlight
(B&M Eyc N plus), powered by my good old sidewall generator, was
randomly and sporadically blinking off. I put the bike on the
workstand, checked out the wiring (it’s original and pristine), and
spent some time spinning the front wheel. Sometimes the headlight
worked, sometimes not, and it changed state at random without having
been touched. I was suspicious of the tiny on/off switch hidden
behind a rubber plug, in part because I felt no satisfying click.
I substituted a different (lower quality) LED dynamo light, which
seemed to work perfectly and probably absolved the generator. I
eventually talked by phone to Peter White, who sold me the light far
longer ago than I remembered - ten years! I asked if he had clues on
diagnosing the problem. He did not, but suggested that the identical
light without the possible failure point of an on/off switch is just
$35. (A sidewall dyno doesn't need a switch.) Still, I opted to open
the light to see if I could spot anything because I have a compulsion
about fixing things - but a headlight has to be really, really reliable. >>>
Here are photos of the electronic guts, with a vernier caliper for
size (and to educate Tom!).
Bottom view. The downward facing LED is at the top of the photo, the
suspicious switch is the tiny white box at the photo’s bottom.
https:// www.flickr.com/photos/16972296@N08/54435279894/in/
dateposted- public/
Top of the circuit board. The large copper pad is part of the heat
sink system, contacting a metal casting that forms the top of the
headlamp body. I’ll probably add some heat conducting goop when I
reassemble:
https://www.flickr.com/photos/16972296@N08/54435279589/in/dateposted-
public/
Just as a matter of personal interest, can you read the numbers off
the 16 pin IC?
fine stippled finish.
Not going below 0V is the most common way PWM is used. However, it'sYeah, that is true, it's possible for PWM to go below zero, but that's
not mandatory or the only way: <https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
On 4/8/2025 1:40 PM, Zen Cycle wrote:
On 4/8/2025 12:44 PM, Frank Krygowski wrote:
On 4/7/2025 2:58 PM, Zen Cycle wrote:
Just as a matter of personal interest, can you read the numbers offIt has no numbers or other codes at all! It's solid black with a sort
the 16 pin IC?
of fine stippled finish.
Take another look. I can see 4 rows of text but the resolution on the
picture isn't good enough to make it out. FWIW, it's the square device
in your second picture with 4 solder joints on each side, next to the
octagonal device with "220" written on it (that's a 220uH inductor,
likely part of a PWM or PS switcher circuit).
Sorry, Zen, it absolutely has no writing on it. I don't know if I can
manage a micro photograph to convince you, but I'm using a high powered
loupe in sunlight. It's just a stippled black finish.
4 lines of text. I think I can make out 4x35x on the first line, but
that's it.
On 4/8/2025 1:00 PM, Zen Cycle wrote:
<snip>
4 lines of text. I think I can make out 4x35x on the first line, but
that's it.
I would guess that it's a PWM Controller in a PLCC-20 package, i.e. TI/ Unitrode UC18xx.
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
<snip>
Not going below 0V is the most common way PWM is used. However, it'sYeah, that is true, it's possible for PWM to go below zero, but that's
not mandatory or the only way:
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
not the way it's normally used for LED light dimming or PWM fan speed control.
With LED bicycle lights, it's surprising that no bicycle light company
has a dynamo input into their battery powered lights. Even at only 3W
from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when
all you have on is the daytime running light.
On 4/8/2025 3:04 PM, Frank Krygowski wrote:
On 4/8/2025 1:40 PM, Zen Cycle wrote:
On 4/8/2025 12:44 PM, Frank Krygowski wrote:
On 4/7/2025 2:58 PM, Zen Cycle wrote:
Just as a matter of personal interest, can you read the numbers offIt has no numbers or other codes at all! It's solid black with a sort
the 16 pin IC?
of fine stippled finish.
Take another look. I can see 4 rows of text but the resolution on the
picture isn't good enough to make it out. FWIW, it's the square device
in your second picture with 4 solder joints on each side, next to the
octagonal device with "220" written on it (that's a 220uH inductor,
likely part of a PWM or PS switcher circuit).
Sorry, Zen, it absolutely has no writing on it. I don't know if I can
manage a micro photograph to convince you, but I'm using a high powered
loupe in sunlight. It's just a stippled black finish.
Try again, using this as a reference:
https://1drv.ms/i/c/5c034a6b86a172e9/EX6snfFdTeJMl1tfB4xqOUwBXw_4YXVPsbTNjw04Bpgcdw?e=lFRSWg
4 lines of text. I think I can make out 4x35x on the first line, but
that's it.
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
Not going below 0V is the most common way PWM is used. However, it'sYeah, that is true, it's possible for PWM to go below zero, but that's
not mandatory or the only way:
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
not the way it's normally used for LED light dimming or PWM fan speed control.
With LED bicycle lights, it's surprising that no bicycle light company
has a dynamo input into their battery powered lights. Even at only 3W
from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when
all you have on is the daytime running light.
On 4/7/2025 11:56 AM, sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there
are ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as
it would only be lit for half the cycle.
As I mentioned in discussions here years ago, one LED dyno headlight I
own has a very simple circuit: IIRC just a bridge rectifier, a voltage regulator, a resistor and a capacitor feeding one LED.
It's obvious from my photos of this B&M lamp that the electronics are
much more complicated. I'm disappointed that the electronics experts
here have never commented on what the likely functions of all that complexity.
On 4/8/2025 6:29 PM, Frank Krygowski wrote:
On 4/8/2025 5:15 PM, Frank Krygowski wrote:Here's a larger version of the better one. I think I clicked wrong
On 4/8/2025 4:00 PM, Zen Cycle wrote:
On 4/8/2025 3:04 PM, Frank Krygowski wrote:
On 4/8/2025 1:40 PM, Zen Cycle wrote:
On 4/8/2025 12:44 PM, Frank Krygowski wrote:
On 4/7/2025 2:58 PM, Zen Cycle wrote:
Just as a matter of personal interest, can you read the numbers >>>>>>>> off the 16 pin IC?It has no numbers or other codes at all! It's solid black with a >>>>>>> sort of fine stippled finish.
Take another look. I can see 4 rows of text but the resolution on
the picture isn't good enough to make it out. FWIW, it's the
square device in your second picture with 4 solder joints on each
side, next to the octagonal device with "220" written on it
(that's a 220uH inductor, likely part of a PWM or PS switcher
circuit).
Sorry, Zen, it absolutely has no writing on it. I don't know if I
can manage a micro photograph to convince you, but I'm using a high
powered loupe in sunlight. It's just a stippled black finish.
Try again, using this as a reference:
https://1drv.ms/i/c/5c034a6b86a172e9/
EX6snfFdTeJMl1tfB4xqOUwBXw_4YXVPsbTNjw04Bpgcdw?e=lFRSWg
4 lines of text. I think I can make out 4x35x on the first line, but
that's it.
Those are reflections. There is no text, no letters, no numbers. I'll
try to get a super-closeup photo later, but this is silly.
Here are two more shots, through a loupe:
https://live.staticflickr.com/65535/54440036155_e9437c6d58_m.jpg
https://live.staticflickr.com/65535/54438807352_4bb9fb0d30_m.jpg
Honest, no text, no letters, no numbers.
before generating that link.
https://www.flickr.com/photos/16972296@N08/54438807352/in/dateposted-
public/
It does seem odd to have an unlabeled 16 pin device. I can check to see
if that's some sort of coating applied over some text.
... Nope. Some fine black stuff scraped away, and the surface now looks smoother, but still no visible characters.
On Tue, 8 Apr 2025 16:00:01 -0400, Zen Cycle <funkmaster@hotmail.com>
wrote:
On 4/8/2025 3:04 PM, Frank Krygowski wrote:
On 4/8/2025 1:40 PM, Zen Cycle wrote:
On 4/8/2025 12:44 PM, Frank Krygowski wrote:
On 4/7/2025 2:58 PM, Zen Cycle wrote:
Just as a matter of personal interest, can you read the numbers off >>>>>> the 16 pin IC?It has no numbers or other codes at all! It's solid black with a sort >>>>> of fine stippled finish.
Take another look. I can see 4 rows of text but the resolution on the
picture isn't good enough to make it out. FWIW, it's the square device >>>> in your second picture with 4 solder joints on each side, next to the
octagonal device with "220" written on it (that's a 220uH inductor,
likely part of a PWM or PS switcher circuit).
Sorry, Zen, it absolutely has no writing on it. I don't know if I can
manage a micro photograph to convince you, but I'm using a high powered
loupe in sunlight. It's just a stippled black finish.
Try again, using this as a reference:
https://1drv.ms/i/c/5c034a6b86a172e9/EX6snfFdTeJMl1tfB4xqOUwBXw_4YXVPsbTNjw04Bpgcdw?e=lFRSWg
4 lines of text. I think I can make out 4x35x on the first line, but
that's it.
I tried feeding the original photo to an online AI photo enhancer: <https://yce.perfectcorp.com/photo-enhance>
The photo is too far out of focus to produce more than a few
characters.
Frank. If you photograph something at an angle as in the photo of the
PCB, the depth of field will limit the surface area that will be in
focus. I suggest you take the photo again and position the camera
more perpendicular to the PCB.
perpendicular or you might see a reflection. If your camera has an
autofocus area mode feature, that allows you focus on a specific area
of interest, it will autofocus on this area of interest and blur
everything outside this area. The area of interest is usually the
center of the screen, but on some cameras, can be resized and moved
anywhere on the image.
Also, brighter light will improve the depth of field. (Smaller
apertures have a greater depth of field). However, don't use a point
source of light or your photos will have shadows.
"What are Autofocus Area Modes? | Camera Focusing Basics Pt.3" <https://www.youtube.com/watch?v=nlzlKZWUI38>
On 4/7/2025 11:56 AM, sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a sine wave."
An LED dynamo light would be rectifying the AC into DC, though there
are ways around this if the light has multiple LEDs (two LEDs each
conducting for half the cycle). If there is only a single LED it can
still be powered by the AC from the dynamo but it would be wasteful as
it would only be lit for half the cycle.
First, here's what I said: "I had hoped to diagnose it using DC,
figuring a 6 volt battery would essentially supply half of the sine wave
so half of the input circuit. That naturally works with incandescents,
and it worked with one Avenir LED dyno headlight that I repaired."
My reference to "essentially half the sine wave" meant that electron
flow from a DC battery would be in the same direction as the flow during
half the sine wave, and might serve to determine what part of the
circuit was open.
And as I said, the DC source has worked with other LED dyno headlights.
It works with the Avenir I own, it works with a B&M Lyt that I just
tested. Both are single LED headlamps. It didn't work with the B&M Eyc
I'm trying to repair. But maybe that's related to its inconsistent and intermittent fault.
As I mentioned in discussions here years ago, one LED dyno headlight I
own has a very simple circuit: IIRC just a bridge rectifier, a voltage regulator, a resistor and a capacitor feeding one LED.
It's obvious from my photos of this B&M lamp that the electronics are
much more complicated. I'm disappointed that the electronics experts
here have never commented on what the likely functions of all that complexity.
On 4/8/2025 6:44 PM, Frank Krygowski wrote:
On 4/8/2025 6:29 PM, Frank Krygowski wrote:
On 4/8/2025 5:15 PM, Frank Krygowski wrote:Here's a larger version of the better one. I think I
On 4/8/2025 4:00 PM, Zen Cycle wrote:
On 4/8/2025 3:04 PM, Frank Krygowski wrote:
On 4/8/2025 1:40 PM, Zen Cycle wrote:
On 4/8/2025 12:44 PM, Frank Krygowski wrote:
On 4/7/2025 2:58 PM, Zen Cycle wrote:
Just as a matter of personal interest, can you readIt has no numbers or other codes at all! It's solid
the numbers off the 16 pin IC?
black with a sort of fine stippled finish.
Take another look. I can see 4 rows of text but the
resolution on the picture isn't good enough to make
it out. FWIW, it's the square device in your second
picture with 4 solder joints on each side, next to
the octagonal device with "220" written on it (that's
a 220uH inductor, likely part of a PWM or PS switcher
circuit).
Sorry, Zen, it absolutely has no writing on it. I
don't know if I can manage a micro photograph to
convince you, but I'm using a high powered loupe in
sunlight. It's just a stippled black finish.
Try again, using this as a reference:
https://1drv.ms/i/c/5c034a6b86a172e9/
EX6snfFdTeJMl1tfB4xqOUwBXw_4YXVPsbTNjw04Bpgcdw?e=lFRSWg
4 lines of text. I think I can make out 4x35x on the
first line, but that's it.
Those are reflections. There is no text, no letters, no
numbers. I'll try to get a super-closeup photo later,
but this is silly.
Here are two more shots, through a loupe:
https://
live.staticflickr.com/65535/54440036155_e9437c6d58_m.jpg
https://
live.staticflickr.com/65535/54438807352_4bb9fb0d30_m.jpg
Honest, no text, no letters, no numbers.
clicked wrong before generating that link.
https://www.flickr.com/photos/16972296@N08/54438807352/in/
dateposted- public/
It does seem odd to have an unlabeled 16 pin device. I can
check to see if that's some sort of coating applied over
some text.
... Nope. Some fine black stuff scraped away, and the
surface now looks smoother, but still no visible characters.
I don't know what to tell you except maybe get someone with
better eyesight to look at it. The First of the new photos
is a good angle, I can see the lines of text but still
blurry. The second and third of the new photos are a bad
angle such that the light doesn't catch the text, but better
resolution.
I know Jeff has already commented about seeing the text, I'd
be interested if anyone else has seen the photos and can
comment.
_All_ devices of this type are laser etched with the part
number and some sort of factory/lot/date code, or at the
very least with some type of abbreviated code that you can
decipher with the data sheet. Some companies _do_ have
devices custom built with no markings to prevent reverse
engineering, but those are rare and generally restricted to
higher dollar items.
No big deal. It was just a matter of personal interest,
don't bother putting anymore effort into it if you don't
want to.
On 4/9/2025 6:53 AM, zen cycle wrote:
On 4/8/2025 6:44 PM, Frank Krygowski wrote:
On 4/8/2025 6:29 PM, Frank Krygowski wrote:
On 4/8/2025 5:15 PM, Frank Krygowski wrote:Here's a larger version of the better one. I think I clicked wrong
On 4/8/2025 4:00 PM, Zen Cycle wrote:
On 4/8/2025 3:04 PM, Frank Krygowski wrote:
On 4/8/2025 1:40 PM, Zen Cycle wrote:
On 4/8/2025 12:44 PM, Frank Krygowski wrote:
On 4/7/2025 2:58 PM, Zen Cycle wrote:
Just as a matter of personal interest, can you read theIt has no numbers or other codes at all! It's solid black with >>>>>>>>> a sort of fine stippled finish.
numbers off the 16 pin IC?
Take another look. I can see 4 rows of text but the resolution >>>>>>>> on the picture isn't good enough to make it out. FWIW, it's the >>>>>>>> square device in your second picture with 4 solder joints on
each side, next to the octagonal device with "220" written on it >>>>>>>> (that's a 220uH inductor, likely part of a PWM or PS switcher
circuit).
Sorry, Zen, it absolutely has no writing on it. I don't know if I >>>>>>> can manage a micro photograph to convince you, but I'm using a
high powered loupe in sunlight. It's just a stippled black finish. >>>>>>>
Try again, using this as a reference:
https://1drv.ms/i/c/5c034a6b86a172e9/
EX6snfFdTeJMl1tfB4xqOUwBXw_4YXVPsbTNjw04Bpgcdw?e=lFRSWg
4 lines of text. I think I can make out 4x35x on the first line,
but that's it.
Those are reflections. There is no text, no letters, no numbers.
I'll try to get a super-closeup photo later, but this is silly.
Here are two more shots, through a loupe:
https://live.staticflickr.com/65535/54440036155_e9437c6d58_m.jpg
https://live.staticflickr.com/65535/54438807352_4bb9fb0d30_m.jpg
Honest, no text, no letters, no numbers.
before generating that link.
https://www.flickr.com/photos/16972296@N08/54438807352/in/dateposted-
public/
It does seem odd to have an unlabeled 16 pin device. I can check to
see if that's some sort of coating applied over some text.
... Nope. Some fine black stuff scraped away, and the surface now
looks smoother, but still no visible characters.
I don't know what to tell you except maybe get someone with better
eyesight to look at it. The First of the new photos is a good angle, I
can see the lines of text but still blurry. The second and third of
the new photos are a bad angle such that the light doesn't catch the
text, but better resolution.
I know Jeff has already commented about seeing the text, I'd be
interested if anyone else has seen the photos and can comment.
_All_ devices of this type are laser etched with the part number and
some sort of factory/lot/date code, or at the very least with some
type of abbreviated code that you can decipher with the data sheet.
Some companies _do_ have devices custom built with no markings to
prevent reverse engineering, but those are rare and generally
restricted to higher dollar items.
No big deal. It was just a matter of personal interest, don't bother
putting anymore effort into it if you don't want to.
I'd require a better camera to get a better photo than this: https://www.flickr.com/photos/16972296@N08/54438807352/in/dateposted-
public/
What can I say? Using my eyes with a variety of magnifiers, I see no markings. After some light scraping, I see no markings. In the photos I
see no markings. I could try heavier scraping in case B&M was very
diligent about keeping their chips secret...
But I don't see any more effort in identifying that chip is justified.
It certainly has no bearing on getting the headlight to work.
I'm still curious if electronics people here can address the complexity
of the circuit, at least in general. Any ideas what it's all for?
remind you that I do have some LED dynamo headlights that light up when
I apply DC. This one does not. Is that a clue to what the circuit is doing?
On 4/9/2025 7:44 AM, zen cycle wrote:
On 4/8/2025 12:33 PM, Frank Krygowski wrote:
On 4/7/2025 11:56 AM, sms wrote:
On 4/7/2025 8:16 AM, Zen Cycle wrote:
On 4/6/2025 10:22 PM, Frank Krygowski wrote:
<snip>
A battery puts out DC, it does not provide "half of a
sine wave."
An LED dynamo light would be rectifying the AC into DC,
though there are ways around this if the light has
multiple LEDs (two LEDs each conducting for half the
cycle). If there is only a single LED it can still be
powered by the AC from the dynamo but it would be
wasteful as it would only be lit for half the cycle.
First, here's what I said: "I had hoped to diagnose it
using DC, figuring a 6 volt battery would essentially
supply half of the sine wave so half of the input
circuit. That naturally works with incandescents, and it
worked with one Avenir LED dyno headlight that I repaired."
My reference to "essentially half the sine wave" meant
that electron flow from a DC battery would be in the same
direction as the flow during half the sine wave, and
might serve to determine what part of the circuit was open.
And as I said, the DC source has worked with other LED
dyno headlights. It works with the Avenir I own, it works
with a B&M Lyt that I just tested. Both are single LED
headlamps. It didn't work with the B&M Eyc I'm trying to
repair. But maybe that's related to its inconsistent and
intermittent fault.
As I mentioned in discussions here years ago, one LED
dyno headlight I own has a very simple circuit: IIRC just
a bridge rectifier, a voltage regulator, a resistor and a
capacitor feeding one LED.
It's obvious from my photos of this B&M lamp that the
electronics are much more complicated. I'm disappointed
that the electronics experts here have never commented on
what the likely functions of all that complexity.
You didn't ask that question previously:
The lamp is significantly more complex in order to control
multiple LEDs with different drive levels by processing
input sensor sensor data.
According to the user manual, it has several modes of
operation:
https://www.bumm.de/files/Produkte/LUMOTEC%20Eyc_Avy_IQ-
XS.PDF
"Functionality of the sensor mode:
During daylight, the headlight operates in “day mode”.
The driving light LED in the headlight shines dimmed onto
the road. The additional DRL LEDs shine with full
luminance. Maximal visibility for oncoming traffic!
(Eyc T, IQXS T: The indicator LED in the rear push button
shines with full intensity.)
During dusk or darkness, the headlight automatically
switches to “night mode”. The driving light emits the full
luminance. The daytime running LEDs stay dark (Eyc T,IQXS
T) or shine with diminished luminance (Avy T). Maximum
vision, additional visibility for oncoming traffic! (Eyc
T, IQXS T: The indicator LED in the rear push button
shines dimly.)
The light/dark sensor switches the headlamp from mode
“Night” to “Day” with a predetermined delay of 8 sec. so
that a temporary brightness (e.g. car headlights turned
up) will not cause immediate switching."
These different modes require processing information from
the sensors to switch the drive control of the different
LEDs. That 16 pin device I've been asking about is likely
a micro controller with a simple program performing a
Boolean function.
This LEDs are likely being driven by a PWM circuit to
control the brightness of more than one LED. The PWM
circuit is being controlled by the sensor circuits.
As noted in the manual excerpt, the lamp will autoswitch
dring daylight hours to a DRL mode, and at night has a
delay of 8 seconds so car headlights don't autodim the
lamp (likely controlled by the uC). Then there's the
'standby mode' where the unit switches from the power
supply to the storage cap while stopped. This may be a
simple analog circuit but it may also be a digitally
controlled switch that senses when power is being
generated from the dynamo.
So, yes, significantly more complex than a basic LED being
driven from a power supply.
I suspect most of that document refers to a later model of
this headlight. (Mine is ten years old.) I say that because
of the remarks about multiple LEDs, etc. This one has only
one LED, it has no "senso" or auto-control function. There
is an ineffective yellow "indicator LED" at the switch end.
It's incredibly tiny - maybe 0.1mm diameter - and so dim I
didn't notice it at first.
I suppose it's possible they designed one circuit board for
all existing and upcoming models of the headlight. There
seem to be several unused solder pads near the one LED in
this photo: https://www.flickr.com/ photos/16972296@N08/54435279894/in/dateposted-public/
The LED is the bright white rounded square toward the "top"
of the photo. That "top" edge faces forward in the assembled
headlight.
If that's the case, then most of the circuitry might be
redundant in the headlight I own.
Incidentally, Rolf mentioned getting extra power out of the
dynamo. My digital meter consistently shows 4 VAC when the
lamp is running off my drill press. (That's at an effective
road speed of about 14 mph.) Given that bike dynamos are
essentially constant current devices pumping about half an
amp, it seems like it's drawing only about two Watts
electrical power. Mechanical input would be more than two
Watts, of course - maybe four to six Watts.
Which reminds me that in my phone call with Peter White, he
says he never bothers to turn his headlamps off. Especially
with a hub dyno, the power draw is negligible. This is a
bottle dyno, so zero watts drawn when I swing it out of
contact.
I'm with Mr White on that. None of my dynamo systems have switches and haven't since I wired past my first failed one in 1972.Switches, especially low-voltage switches are prone to failure.
I think I
mentioned a few days ago that on a recent night ride, my headlight (B&M
Eyc N plus), powered by my good old sidewall generator, was randomly
and sporadically blinking off.
I put the bike on the workstand, checked
out the wiring (its original and pristine), and spent some time
spinning the front wheel. Sometimes the headlight worked, sometimes not,
and it changed state at random without having been touched. I was
suspicious of the tiny on/off switch hidden behind a rubber plug, in
part because I felt no satisfying click.
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-public/ >https://www.flickr.com/photos/16972296@N08/54435279589/in/dateposted-public/
With no power attached, I tried checking for continuity across the
switch. It always showed close to zero ohms, never infinity - but the >resistance readings with pinpoint probes were jumpy...
I used contact cleaner to blast out the switch and repeatedly clicked
it, including while power was applied. Except for a few occasions, the
light did not respond to the switch, so I still suspect the switch. When
the light was off, I had about 12 VDC. When on, about 4 VDC.
Its still not fixed, but it's getting late here. I would try to simply
short out the switch, but its very difficult to identify and reach its >microscopic solder pads and Im no good at micro soldering.
Am 08.04.2025 um 22:12 schrieb sms:
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
Not going below 0V is the most common way PWM is used. However, it'sYeah, that is true, it's possible for PWM to go below zero, but that's
not mandatory or the only way:
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
not the way it's normally used for LED light dimming or PWM fan speed
control.
With LED bicycle lights, it's surprising that no bicycle light company
has a dynamo input into their battery powered lights. Even at only 3W
from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when
all you have on is the daytime running light.
"No" is a not quite correct, "extremely rare" would describe the market >better.
<https://nabendynamo.de/produkte/scheinwerfer/ladelux/>
Slightly more common: a "dynamo driven power supply" produces power to a
USE output, and theoretially you can route that USB power into your bikt >light but with no fine-tuned optimization the output is too low to make
it worthwhile in any sense.
A customer produces an industrial machine based on his
father's design patent for which my customer (an academic
electrical engineer PhD) makes up the controller boards. He
scuffs the top of some chips and coats them with epoxy.
I can see where a manufacturer would request electronic bits
with no number if it would otherwise reveal information to
competitors but I don't think that's likely for a bicycle
lamp. Not unreasonable for a $50K machine.
On Wed, 9 Apr 2025 12:35:15 +0200, Rolf Mantel <news@hartig-mantel.de>
wrote:
Am 08.04.2025 um 22:12 schrieb sms:
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
Not going below 0V is the most common way PWM is used. However, it's >>>> not mandatory or the only way:Yeah, that is true, it's possible for PWM to go below zero, but that's
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
not the way it's normally used for LED light dimming or PWM fan speed
control.
With LED bicycle lights, it's surprising that no bicycle light company
has a dynamo input into their battery powered lights. Even at only 3W
from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when
all you have on is the daytime running light.
"No" is a not quite correct, "extremely rare" would describe the market
better.
<https://nabendynamo.de/produkte/scheinwerfer/ladelux/>
Slightly more common: a "dynamo driven power supply" produces power to a
USE output, and theoretially you can route that USB power into your bikt
light but with no fine-tuned optimization the output is too low to make
it worthwhile in any sense.
This article shows 26 such bicycle USB chargers. That seems to be a
popular use for bicycle dynamos. <https://www.cyclingabout.com/list-of-hub-dynamo-power-supplies-for-usb-devices/>
On Wed, 9 Apr 2025 12:35:15 +0200, Rolf Mantel <news@hartig-mantel.de>
wrote:
Am 08.04.2025 um 22:12 schrieb sms:
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
Not going below 0V is the most common way PWM is used. However, it's >>>> not mandatory or the only way:Yeah, that is true, it's possible for PWM to go below zero, but that's
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
not the way it's normally used for LED light dimming or PWM fan speed
control.
With LED bicycle lights, it's surprising that no bicycle light company
has a dynamo input into their battery powered lights. Even at only 3W
from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when
all you have on is the daytime running light.
"No" is a not quite correct, "extremely rare" would describe the market
better.
<https://nabendynamo.de/produkte/scheinwerfer/ladelux/>
Slightly more common: a "dynamo driven power supply" produces power to a
USE output, and theoretially you can route that USB power into your bikt
light but with no fine-tuned optimization the output is too low to make
it worthwhile in any sense.
This article shows 26 such bicycle USB chargers. That seems to be a
popular use for bicycle dynamos. <https://www.cyclingabout.com/list-of-hub-dynamo-power-supplies-for-usb-devices/>
On 4/8/2025 8:56 PM, Frank Krygowski wrote:
The switch is the tiny white box toward the bottom of this photo:
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-
public/ just below "3R 46 16V" which is probably a capacitor. The black
pushbutton sticks out downward in the photo (or backwards, in the
mounted headlight) and looks odd in the photo, again because of
reflections. That switch is about 4.5mm x 2.5mm x 1.8mm tall. I removed
the switch from the circuit board.
I had assumed it was a toggle, but it seems instead to be a momentary
switch. It connects to five tiny solder pads. The two big ones, at the
"bottom" of the switch in the photo, are just for fastening it to the
board. At the "top" face in the photo are three very tiny pads, call
them A, B and C.
Playing with a jumper wire ending in two sharp pins, I found that
momentary contact from A to C changed the state of the light, on to off
to back on, etc. A to B did the same. I had to be careful to touch
fairly quickly and not bounce.
At last check, the light is in "on" state. I think if I don't disturb
things, it should stay in that state, and I can try reassembling and
testing it on the bike. I've got other stuff to work on tonight, so I'll
get back to it tomorrow.
More bench testing this morning. The headlamp started right up,
perfectly consistently when driven by the dynamo spun with my drill
press. Momentarily jumping A to C (or B) of those switch solder pads
toggled the lamp on or off while running the dynamo. Shutting down the
dynamo put the lamp into "standlight" mode. In that mode, with no
voltage input, jumping A to C turned off the standlight. Applying power turned the lamp back on, no switching necessary. It seems the headlight
is operating as it should.
I'm convinced the problem was that faulty switch, probably with internal parts rattling around enough to occasionally toggle that A to C or A to
B contact. I suspect it will work fine if I successfully reassemble it without the switch.
It's still slightly worrisome, because a headlight really does have to
be reliable. Maybe I'll carry a little battery light as a spare for a
while. I often do that anyway when night riding with others, because of
so many incidents of other folks' headlights failing on rides.
Am 11.04.2025 um 03:56 schrieb Jeff Liebermann:
On Wed, 9 Apr 2025 12:35:15 +0200, Rolf Mantel <news@hartig-mantel.de>
wrote:
Am 08.04.2025 um 22:12 schrieb sms:
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
Not going below 0V is the most common way PWM is used. However, it's >>>>> not mandatory or the only way:Yeah, that is true, it's possible for PWM to go below zero, but that's >>>> not the way it's normally used for LED light dimming or PWM fan speed
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
control.
With LED bicycle lights, it's surprising that no bicycle light company >>>> has a dynamo input into their battery powered lights. Even at only 3W
from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when >>>> all you have on is the daytime running light.
"No" is a not quite correct, "extremely rare" would describe the market
better.
<https://nabendynamo.de/produkte/scheinwerfer/ladelux/>
Slightly more common: a "dynamo driven power supply" produces power to a >>> USE output, and theoretially you can route that USB power into your bikt >>> light but with no fine-tuned optimization the output is too low to make
it worthwhile in any sense.
This article shows 26 such bicycle USB chargers. That seems to be a
popular use for bicycle dynamos.
<https://www.cyclingabout.com/list-of-hub-dynamo-power-supplies-for-usb-devices/>
Sure, it's common to use a bike dynamo to charge low-power devices.
Charging mobile phones barely works because many phones stop charging on
a variable power supply.
I have no direct knowledge of anybody using a dynamo to top up their
battery lights.
On 4/9/2025 10:04 AM, AMuzi wrote:
I'm with Mr White on that. None of my dynamo systems have switches andSwitches, especially low-voltage switches are prone to failure.
haven't since I wired past my first failed one in 1972.
But higher-end dynamo lights will often have multiple modes that require
a switch but it is often handlebar mounted.
For a mid-cost, switchless, dynamo light, I recommend the 100 Lux
H-Black Pro Dynamo <https://herrmans.eu/products/front-lights-4147/dynamo-10038/h-black-pro-dynamo/>
for about $50 <https://www.bike24.com/p2214107.html>.
It's a lot better than the low-cost dynamo lights at 20-40 lux. I bought
one in 2017 for €62,94 so they come down in price.
Not the latest in LED efficiency, but more than adequate, and the optics
are very good.
A lot of people don't realize how vitally important the optics are in
bicycle lighting, especially in dynamo lights where you only have a few
watts of power to work with.
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
On 4/8/2025 8:56 PM, Frank Krygowski wrote:Who on earth to you ride with that have light failures? Ive never had a >light fail on the ride, Ive had in the early days ooh batteries is a bit >tired as the light dims as the battery cant hold the voltage any more, and >had cheap lights mounts and so on start to wear.
The switch is the tiny white box toward the bottom of this photo:
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-
public/ just below "3R 46 16V" which is probably a capacitor. The black
pushbutton sticks out downward in the photo (or backwards, in the
mounted headlight) and looks odd in the photo, again because of
reflections. That switch is about 4.5mm x 2.5mm x 1.8mm tall. I removed
the switch from the circuit board.
I had assumed it was a toggle, but it seems instead to be a momentary
switch. It connects to five tiny solder pads. The two big ones, at the
"bottom" of the switch in the photo, are just for fastening it to the
board. At the "top" face in the photo are three very tiny pads, call
them A, B and C.
Playing with a jumper wire ending in two sharp pins, I found that
momentary contact from A to C changed the state of the light, on to off
to back on, etc. A to B did the same. I had to be careful to touch
fairly quickly and not bounce.
At last check, the light is in "on" state. I think if I don't disturb
things, it should stay in that state, and I can try reassembling and
testing it on the bike. I've got other stuff to work on tonight, so I'll >>> get back to it tomorrow.
More bench testing this morning. The headlamp started right up,
perfectly consistently when driven by the dynamo spun with my drill
press. Momentarily jumping A to C (or B) of those switch solder pads
toggled the lamp on or off while running the dynamo. Shutting down the
dynamo put the lamp into "standlight" mode. In that mode, with no
voltage input, jumping A to C turned off the standlight. Applying power
turned the lamp back on, no switching necessary. It seems the headlight
is operating as it should.
I'm convinced the problem was that faulty switch, probably with internal
parts rattling around enough to occasionally toggle that A to C or A to
B contact. I suspect it will work fine if I successfully reassemble it
without the switch.
It's still slightly worrisome, because a headlight really does have to
be reliable. Maybe I'll carry a little battery light as a spare for a
while. I often do that anyway when night riding with others, because of
so many incidents of other folks' headlights failing on rides.
I still have multiple lights that Ive upgraded from which still work, but >time and technology has moved on.
Likewise I see a lot of bikes, and meet up with folks for a Wednesday night >ride, most weeks its not a thing that is common, its not impossible but >definitely rare.
Roger Merriman
Rolf Mantel <news@hartig-mantel.de> wrote:
Am 11.04.2025 um 03:56 schrieb Jeff Liebermann:
On Wed, 9 Apr 2025 12:35:15 +0200, Rolf Mantel <news@hartig-mantel.de>
wrote:
Am 08.04.2025 um 22:12 schrieb sms:
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
Not going below 0V is the most common way PWM is used. However, it's >>>>>> not mandatory or the only way:Yeah, that is true, it's possible for PWM to go below zero, but that's >>>>> not the way it's normally used for LED light dimming or PWM fan speed >>>>> control.
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
With LED bicycle lights, it's surprising that no bicycle light company >>>>> has a dynamo input into their battery powered lights. Even at only 3W >>>>> from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when >>>>> all you have on is the daytime running light.
"No" is a not quite correct, "extremely rare" would describe the market >>>> better.
<https://nabendynamo.de/produkte/scheinwerfer/ladelux/>
Slightly more common: a "dynamo driven power supply" produces power to a >>>> USE output, and theoretially you can route that USB power into your bikt >>>> light but with no fine-tuned optimization the output is too low to make >>>> it worthwhile in any sense.
This article shows 26 such bicycle USB chargers. That seems to be a
popular use for bicycle dynamos.
<https://www.cyclingabout.com/list-of-hub-dynamo-power-supplies-for-usb-devices/>
Sure, it's common to use a bike dynamo to charge low-power devices.
Charging mobile phones barely works because many phones stop charging on
a variable power supply.
Even so its quite a power hungry device at least smart phones, so would
need to be cycling at sufficient speed for enough time, and phones
throttling of the charge rate as it probably believes its connected to a >faulty power supply, hence folks use of power banks, which also would have >the advantages of not needing to charge and cable up devices on the move.
Considering that the batteries in lights are quite a bit larger capacity
I have no direct knowledge of anybody using a dynamo to top up their
battery lights.
and the device is even more power intensive, and generally can charge back
up relatively quickly, I dont see much use for it, I get that be a few
folks whod like the idea of it but cant see a good use for it.
Even the folks doing a hr or more commute, the lights last multiple days
even in winter and charge it up over night when its getting low, or even
at work!
Roger Merriman
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
On 4/8/2025 8:56 PM, Frank Krygowski wrote:Who on earth to you ride with that have light failures? I’ve never had a light fail on the ride, I’ve had in the early days ooh batteries is a bit tired as the light dims as the battery can’t hold the voltage any more, and had cheap lights mounts and so on start to wear.
The switch is the tiny white box toward the bottom of this photo:
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-
public/ just below "3R 46 16V" which is probably a capacitor. The black
pushbutton sticks out downward in the photo (or backwards, in the
mounted headlight) and looks odd in the photo, again because of
reflections. That switch is about 4.5mm x 2.5mm x 1.8mm tall. I removed
the switch from the circuit board.
I had assumed it was a toggle, but it seems instead to be a momentary
switch. It connects to five tiny solder pads. The two big ones, at the
"bottom" of the switch in the photo, are just for fastening it to the
board. At the "top" face in the photo are three very tiny pads, call
them A, B and C.
Playing with a jumper wire ending in two sharp pins, I found that
momentary contact from A to C changed the state of the light, on to off
to back on, etc. A to B did the same. I had to be careful to touch
fairly quickly and not bounce.
At last check, the light is in "on" state. I think if I don't disturb
things, it should stay in that state, and I can try reassembling and
testing it on the bike. I've got other stuff to work on tonight, so I'll >>> get back to it tomorrow.
More bench testing this morning. The headlamp started right up,
perfectly consistently when driven by the dynamo spun with my drill
press. Momentarily jumping A to C (or B) of those switch solder pads
toggled the lamp on or off while running the dynamo. Shutting down the
dynamo put the lamp into "standlight" mode. In that mode, with no
voltage input, jumping A to C turned off the standlight. Applying power
turned the lamp back on, no switching necessary. It seems the headlight
is operating as it should.
I'm convinced the problem was that faulty switch, probably with internal
parts rattling around enough to occasionally toggle that A to C or A to
B contact. I suspect it will work fine if I successfully reassemble it
without the switch.
It's still slightly worrisome, because a headlight really does have to
be reliable. Maybe I'll carry a little battery light as a spare for a
while. I often do that anyway when night riding with others, because of
so many incidents of other folks' headlights failing on rides.
I still have multiple lights that I’ve upgraded from which still work, but time and technology has moved on.
Likewise I see a lot of bikes, and meet up with folks for a Wednesday night ride, most weeks it’s not a thing that is common, it’s not impossible but definitely rare.
Roger Merriman
Am 11.04.2025 um 03:56 schrieb Jeff Liebermann:
On Wed, 9 Apr 2025 12:35:15 +0200, Rolf Mantel <news@hartig-mantel.de>
wrote:
Am 08.04.2025 um 22:12 schrieb sms:
On 4/7/2025 5:15 PM, Jeff Liebermann wrote:
Not going below 0V is the most common way PWM is used. However, it's >>>>> not mandatory or the only way:Yeah, that is true, it's possible for PWM to go below zero, but that's >>>> not the way it's normally used for LED light dimming or PWM fan speed
<https://en.wikipedia.org/wiki/Pulse-width_modulation>
Note that the waveform shown goes below 0V.
control.
With LED bicycle lights, it's surprising that no bicycle light company >>>> has a dynamo input into their battery powered lights. Even at only 3W
from the dynamo, you could power a high-end bicycle light at lower
power, and you could be charging the battery during daytime rides when >>>> all you have on is the daytime running light.
"No" is a not quite correct, "extremely rare" would describe the market
better.
<https://nabendynamo.de/produkte/scheinwerfer/ladelux/>
Slightly more common: a "dynamo driven power supply" produces power to a >>> USE output, and theoretially you can route that USB power into your bikt >>> light but with no fine-tuned optimization the output is too low to make
it worthwhile in any sense.
This article shows 26 such bicycle USB chargers. That seems to be a
popular use for bicycle dynamos.
<https://www.cyclingabout.com/list-of-hub-dynamo-power-supplies-for-usb-devices/>
Sure, it's common to use a bike dynamo to charge low-power devices.
Charging mobile phones barely works because many phones stop charging on
a variable power supply.
I have no direct knowledge of anybody using a dynamo to top up their
battery lights.
On 4/11/2025 4:15 AM, Roger Merriman wrote:
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
On 4/8/2025 8:56 PM, Frank Krygowski wrote:Who on earth to you ride with that have light failures? I’ve never had a >> light fail on the ride, I’ve had in the early days ooh batteries is a bit >> tired as the light dims as the battery can’t hold the voltage any more, and
The switch is the tiny white box toward the bottom of this photo:
https://www.flickr.com/photos/16972296@N08/54435279894/in/dateposted-
public/ just below "3R 46 16V" which is probably a capacitor. The black >>>> pushbutton sticks out downward in the photo (or backwards, in the
mounted headlight) and looks odd in the photo, again because of
reflections. That switch is about 4.5mm x 2.5mm x 1.8mm tall. I removed >>>> the switch from the circuit board.
I had assumed it was a toggle, but it seems instead to be a momentary
switch. It connects to five tiny solder pads. The two big ones, at the >>>> "bottom" of the switch in the photo, are just for fastening it to the
board. At the "top" face in the photo are three very tiny pads, call
them A, B and C.
Playing with a jumper wire ending in two sharp pins, I found that
momentary contact from A to C changed the state of the light, on to off >>>> to back on, etc. A to B did the same. I had to be careful to touch
fairly quickly and not bounce.
At last check, the light is in "on" state. I think if I don't disturb
things, it should stay in that state, and I can try reassembling and
testing it on the bike. I've got other stuff to work on tonight, so I'll >>>> get back to it tomorrow.
More bench testing this morning. The headlamp started right up,
perfectly consistently when driven by the dynamo spun with my drill
press. Momentarily jumping A to C (or B) of those switch solder pads
toggled the lamp on or off while running the dynamo. Shutting down the
dynamo put the lamp into "standlight" mode. In that mode, with no
voltage input, jumping A to C turned off the standlight. Applying power
turned the lamp back on, no switching necessary. It seems the headlight
is operating as it should.
I'm convinced the problem was that faulty switch, probably with internal >>> parts rattling around enough to occasionally toggle that A to C or A to
B contact. I suspect it will work fine if I successfully reassemble it
without the switch.
It's still slightly worrisome, because a headlight really does have to
be reliable. Maybe I'll carry a little battery light as a spare for a
while. I often do that anyway when night riding with others, because of
so many incidents of other folks' headlights failing on rides.
had cheap lights mounts and so on start to wear.
I still have multiple lights that I’ve upgraded from which still work, but >> time and technology has moved on.
Likewise I see a lot of bikes, and meet up with folks for a Wednesday night >> ride, most weeks it’s not a thing that is common, it’s not impossible but
definitely rare.
Roger Merriman
I have had light failures. Mine run incandescent globes not
LEDs. Those fail at about 8~10 years. Those bikes have been
1/2 year daily rides, that is, a winter bike and a summer bike.
On 4/11/2025 5:52 AM, Roger Merriman wrote:
sms <scharf.steven@geemail.com> wrote:
I’d of thought that the folks buying lights in the above price ranges would
A lot of people don't realize how vitally important the optics are in
bicycle lighting, especially in dynamo lights where you only have a few
watts of power to work with.
do at least to some degree after all could just buy some cheap almost
disposable lights as some folks do! But they are choosing to buy expensive >> lights.
Mr. Scharf's statement is correct. Optics are important, and not only
for dynamo lights. I've been saying that for many, many years. It makes
no sense to waste half of a headlight's output in directions that you
never need to see.
I think only a small percentage of cyclists are aware of the importance
of optics. I think almost all cyclists get their "education" (such as it
is) from product advertisements. And ever since white LEDs began their
rapid rise in lumen output, lumen counts have been the almost exclusive message in advertisements
That's led to nonsense like lights for road bikes that blind oncoming
road users, or otherwise waste lumens shooting them skyward. We've even
seen arguments in this discussion group, long and furious, that one
_must_ have a simplistic perfectly round headlight beam, because
otherwise a road rider would be sure to run into a low hanging branch
and injure his head.
(Yes, really. And apparently those making that argument were not joking.
Go figure.)
On 4/11/2025 12:44 PM, Roger Merriman wrote:
AMuzi <am@yellowjersey.org> wrote:
On 4/11/2025 4:15 AM, Roger Merriman wrote:
I have had light failures. Mine run incandescent globes not
LEDs. Those fail at about 8~10 years. Those bikes have been
1/2 year daily rides, that is, a winter bike and a summer bike.
Which rather proves my point really both you and Frank are outliers in
this
regard! With incandescent being rather rarer than Dynamos, I don’t recall >> seeing any for a very long time!
AA/AAA lights aren’t that common being supplemented by USB or coin
battery
powered lights, though do see them still.
I probably still own some AA powered halogen headlights, mostly because
I seldom throw away anything remotely useful. Yes, it's a personal
fault.
But for actual riding, I switched to LEDs many years ago.
I've used generator powered lights since about 1977, experimenting back
then to improve light output (when others were charging lead acid
batteries to power MR spotlight bulbs) even though good quality stock
units were fine for my uses. But when high output LED dynamo headlights
came on the market, I bought them and considered the problem solved.
Since then, the LEDs have gotten brighter, but I've seen no reason to
chase every incremental improvement. The headlights I use light up
street signs 1/4 mile away, so they must be visible to motorists half a
mile away. They give me plenty of illumination of the roadway, and light
it better than the headlights of my friends. That's good enough for me.
Similarly, the fact one can buy a 16 pound bicycle doesn't mean everyone
with 18 pound bicycles must trade up. Not even those with 26 pound
bicycles. What matter is whether one's equipment meets their needs, not whether it's the latest, greatest technology.
On 4/11/2025 1:37 PM, Roger Merriman wrote:
Frank Krygowski <frkrygow@sbcglobal.net> wrote:I don't think that's true at all. Take a look at some of the beam shots
I think only a small percentage of cyclists are aware of the importance
of optics. I think almost all cyclists get their "education" (such as it >>> is) from product advertisements. And ever since white LEDs began their
rapid rise in lumen output, lumen counts have been the almost exclusive
message in advertisements
I believe most of the bike lights manufacturers have always used lenses to >> greater or lesser degree, was a period when some of the Chinese led
lights/torches that where lumens only and not true figures!
at this site: https://bikelightdatabase.com/beamshots
Or go to https://road.cc/content/feature/roadcc-front-bike-lights-beam-test-310999
and look at the beam shapes of the Ravemen and the Knog.
In the past few years, there's been more attention to optics, but I
think there are still plenty of small manufacturers who use rotationally symmetrical optics. Peter White has an explanation of why those are far
from optimum, at https://www.peterwhitecycles.com/plight.php
On 4/11/2025 12:44 PM, Roger Merriman wrote:
AMuzi <am@yellowjersey.org> wrote:
On 4/11/2025 4:15 AM, Roger Merriman wrote:
I have had light failures. Mine run incandescent globes not
LEDs. Those fail at about 8~10 years. Those bikes have been
1/2 year daily rides, that is, a winter bike and a summer bike.
Which rather proves my point really both you and Frank are outliers in this >> regard! With incandescent being rather rarer than Dynamos, I don’t recall >> seeing any for a very long time!
AA/AAA lights aren’t that common being supplemented by USB or coin battery >> powered lights, though do see them still.
I probably still own some AA powered halogen headlights, mostly because
I seldom throw away anything remotely useful. Yes, it's a personal
fault. But for actual riding, I switched to LEDs many years ago.
I've used generator powered lights since about 1977, experimenting back
then to improve light output (when others were charging lead acid
batteries to power MR spotlight bulbs) even though good quality stock
units were fine for my uses. But when high output LED dynamo headlights
came on the market, I bought them and considered the problem solved.
Since then, the LEDs have gotten brighter, but I've seen no reason to
chase every incremental improvement. The headlights I use light up
street signs 1/4 mile away, so they must be visible to motorists half a
mile away. They give me plenty of illumination of the roadway, and light
it better than the headlights of my friends. That's good enough for me.
Similarly, the fact one can buy a 16 pound bicycle doesn't mean everyone
with 18 pound bicycles must trade up. Not even those with 26 pound
bicycles. What matter is whether one's equipment meets their needs, not whether it's the latest, greatest technology.
On 4/11/2025 3:24 PM, Roger Merriman wrote:
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
[about headlight beam optics, generating "specially shaped beams":]
I don't think that's true at all. Take a look at some of the beam shots
at this site: https://bikelightdatabase.com/beamshots
Or go to
https://road.cc/content/feature/roadcc-front-bike-lights-beam-test-310999 >>> and look at the beam shapes of the Ravemen and the Knog.
Both of which have shaped beams, Raveman and Knog both claim road specific >> beams and reviews tend in Road.cc and others mention this, Raveman make lot >> out of their High and low beams, though they are far from the first to do
that sort of thing!
I get why they do this, but a camera isn’t an eye and it’s rather false if
not misleading to use beam shots.
I think you and I have different definitions of beam shapes, and
different standards. (Andrew will be thinking "Of course!")
But the beams shown for the Raveman and Knog are rotationally symmetric.
That can be called "specially shaped" only in comparison to totally
random shapes, or to a light source with no focus at all.
Here's a good article on beam design that highlights the shortcoming of
a round beam for road use. https://www.renehersecycles.com/myth-14-more-lumens-make-a-better-light/ Essentially, the illumination of the road surface can't be uniform. It's closest to uniform if you're wasting half the light above the road
surface. Otherwise it generates hot spots that are detrimental to your
night vision. A properly designed beam will do a much better job of
giving uniform illumination and will better show road contours,
obstacles, potholes, rocks etc. The information is much the same as the
Peter White article below, but with graphics that may make the ideas
more clear.
In the past few years, there's been more attention to optics, but I
think there are still plenty of small manufacturers who use rotationally >>> symmetrical optics. Peter White has an explanation of why those are far
from optimum, at https://www.peterwhitecycles.com/plight.php
On 4/11/2025 11:29 AM, Frank Krygowski wrote:
The light now works. I tried it on a short test ride last night.
I do like that headlight. Since the non-switch version is now only $35
from Peter White, I plan to buy another to upgrade the headlight on my
folding bike.
Final report: A friend and I did a two hour ride late Sunday night. The generator and the headlight performed very well, no problems.
The headlight's mate arrived by mail last night. It will be going on one
of the folding bikes.
On 4/16/2025 1:39 AM, Roger Merriman wrote:
Frank Krygowski <frkrygow@sbcglobal.net> wrote:BTW, that illustrates a maintenance problem I frequently encounter.
On 4/11/2025 11:29 AM, Frank Krygowski wrote:
Final report: A friend and I did a two hour ride late Sunday night. The
generator and the headlight performed very well, no problems.
The headlight's mate arrived by mail last night. It will be going on one >>> of the folding bikes.
Has spurred me into spending my light or rather the old (ish) one I use for >> the commute to get fixed the mount will not tighten I had a crash with it
years ago and it’s never been right since!
While I'm on the bike, some problem or other will annoy me, as in "I've
got to fix that squeak when I get home."
But when I park the bike, it naturally doesn't squeak (or whatever) so I forget about it until the next time I'm riding the bike. :-/
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
On 4/16/2025 1:39 AM, Roger Merriman wrote:Your post plus as it was winter Ive being using the light on the old
Frank Krygowski <frkrygow@sbcglobal.net> wrote:BTW, that illustrates a maintenance problem I frequently encounter.
On 4/11/2025 11:29 AM, Frank Krygowski wrote:
Final report: A friend and I did a two hour ride late Sunday night. The >>>> generator and the headlight performed very well, no problems.
The headlight's mate arrived by mail last night. It will be going on one >>>> of the folding bikes.
Has spurred me into spending my light or rather the old (ish) one I use for >>> the commute to get fixed the mount will not tighten I had a crash with it >>> years ago and its never been right since!
While I'm on the bike, some problem or other will annoy me, as in "I've
got to fix that squeak when I get home."
But when I park the bike, it naturally doesn't squeak (or whatever) so I
forget about it until the next time I'm riding the bike. :-/
school roadie on early starts, and that the remote doesnt fit naturally to >the bars, and certainly nowhere just to hand like the MTB derived commuter, >so Ive been toggling the beam from High to low using the switch on the >light, and so have noticed it moving slightly.
Clearly still works in that its not going to fall off but its tipped into >this is really quite annoying from Meh! Such as one of the old rear lights >that does occasionally loose connections and go dim, but as rear lights are >such a tedious bits of kit I run two on the commute bikes, and that light >definitely isnt worth repairing.
Roger Merriman
On 16 Apr 2025 15:43:28 GMT, Roger Merriman <roger@sarlet.com> wrote:
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
On 4/16/2025 1:39 AM, Roger Merriman wrote:Your post plus as it was winter Ive being using the light on the old
Frank Krygowski <frkrygow@sbcglobal.net> wrote:BTW, that illustrates a maintenance problem I frequently encounter.
On 4/11/2025 11:29 AM, Frank Krygowski wrote:
Final report: A friend and I did a two hour ride late Sunday night. The >>>>> generator and the headlight performed very well, no problems.
The headlight's mate arrived by mail last night. It will be going on one >>>>> of the folding bikes.
Has spurred me into spending my light or rather the old (ish) one I use for
the commute to get fixed the mount will not tighten I had a crash with it >>>> years ago and its never been right since!
While I'm on the bike, some problem or other will annoy me, as in "I've
got to fix that squeak when I get home."
But when I park the bike, it naturally doesn't squeak (or whatever) so I >>> forget about it until the next time I'm riding the bike. :-/
school roadie on early starts, and that the remote doesnt fit naturally to >> the bars, and certainly nowhere just to hand like the MTB derived commuter, >> so Ive been toggling the beam from High to low using the switch on the
light, and so have noticed it moving slightly.
Clearly still works in that its not going to fall off but its tipped into
this is really quite annoying from Meh! Such as one of the old rear lights >> that does occasionally loose connections and go dim, but as rear lights are >> such a tedious bits of kit I run two on the commute bikes, and that light
definitely isnt worth repairing.
Roger Merriman
I don't need any lights, but for other maintenance issues, I carry a
lot of tools and I'm inclined to stop in some shade and fix or try to
fix whatever is not working right. I also carry a huge amount of tools
in my truck, including a vice, and what I cannot fix while out on a
ride is likely to under the wrench before I drive home. I am the
opposite of a procrastinator. My wife serves that function at our
home.
--
C'est bon
Soloman
Catrike Ryder <Soloman@old.bikers.org> wrote:
On 16 Apr 2025 15:43:28 GMT, Roger Merriman <roger@sarlet.com> wrote:This will require rather a lot of disassembly, as its the cleat and its >bolt and the nut it screws into, which is the mostly likely culprit to be >damaged as the bolt threads looked visually fine.
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
On 4/16/2025 1:39 AM, Roger Merriman wrote:Your post plus as it was winter I?ve being using the light on the old
Frank Krygowski <frkrygow@sbcglobal.net> wrote:BTW, that illustrates a maintenance problem I frequently encounter.
On 4/11/2025 11:29 AM, Frank Krygowski wrote:
Final report: A friend and I did a two hour ride late Sunday night. The >>>>>> generator and the headlight performed very well, no problems.
The headlight's mate arrived by mail last night. It will be going on one >>>>>> of the folding bikes.
Has spurred me into spending my light or rather the old (ish) one I use for
the commute to get fixed the mount will not tighten I had a crash with it >>>>> years ago and it?s never been right since!
While I'm on the bike, some problem or other will annoy me, as in "I've >>>> got to fix that squeak when I get home."
But when I park the bike, it naturally doesn't squeak (or whatever) so I >>>> forget about it until the next time I'm riding the bike. :-/
school roadie on early starts, and that the remote doesn?t fit naturally to >>> the bars, and certainly nowhere just to hand like the MTB derived commuter, >>> so I?ve been toggling the beam from High to low using the switch on the
light, and so have noticed it moving slightly.
Clearly still works in that it?s not going to fall off but it?s tipped into >>> this is really quite annoying from Meh! Such as one of the old rear lights >>> that does occasionally loose connections and go dim, but as rear lights are >>> such a tedious bits of kit I run two on the commute bikes, and that light >>> definitely isn?t worth repairing.
Roger Merriman
I don't need any lights, but for other maintenance issues, I carry a
lot of tools and I'm inclined to stop in some shade and fix or try to
fix whatever is not working right. I also carry a huge amount of tools
in my truck, including a vice, and what I cannot fix while out on a
ride is likely to under the wrench before I drive home. I am the
opposite of a procrastinator. My wife serves that function at our
home.
--
C'est bon
Soloman
So likely to require rather a lot of dissembling! Im assuming they would >need to take the inner electronics (likely to be by some margin more
complex than Franks lamb) plus batteries out, to then get at the nut from >inside the case.
They started off making kit for divers/cavers then MTBers so durable kit
that can be repaired is very much there thing.
And its the new bike that and using it during winter when I needed a light >to see by, that it became apparent that it was getting worse, and the
problem was highlighted.
But it didnt stop the light working, it still worked fine so certainly >during winter when its by far the best tool for the job I was reluctant to >be without it, but since Im now commuting even early in daylight and
Franks thread reminded me seemed the time to do so, Id hope to get it back >last next week or so.
Roger Merriman
On 16 Apr 2025 20:48:28 GMT, Roger Merriman <roger@sarlet.com> wrote:
Catrike Ryder <Soloman@old.bikers.org> wrote:
On 16 Apr 2025 15:43:28 GMT, Roger Merriman <roger@sarlet.com> wrote:This will require rather a lot of disassembly, as its the cleat and its >> bolt and the nut it screws into, which is the mostly likely culprit to be
Frank Krygowski <frkrygow@sbcglobal.net> wrote:
On 4/16/2025 1:39 AM, Roger Merriman wrote:Your post plus as it was winter I?ve being using the light on the old
Frank Krygowski <frkrygow@sbcglobal.net> wrote:BTW, that illustrates a maintenance problem I frequently encounter.
On 4/11/2025 11:29 AM, Frank Krygowski wrote:
Final report: A friend and I did a two hour ride late Sunday night. The >>>>>>> generator and the headlight performed very well, no problems.
The headlight's mate arrived by mail last night. It will be going on one
of the folding bikes.
Has spurred me into spending my light or rather the old (ish) one I use for
the commute to get fixed the mount will not tighten I had a crash with it
years ago and it?s never been right since!
While I'm on the bike, some problem or other will annoy me, as in "I've >>>>> got to fix that squeak when I get home."
But when I park the bike, it naturally doesn't squeak (or whatever) so I >>>>> forget about it until the next time I'm riding the bike. :-/
school roadie on early starts, and that the remote doesn?t fit naturally to
the bars, and certainly nowhere just to hand like the MTB derived commuter,
so I?ve been toggling the beam from High to low using the switch on the >>>> light, and so have noticed it moving slightly.
Clearly still works in that it?s not going to fall off but it?s tipped into
this is really quite annoying from Meh! Such as one of the old rear lights >>>> that does occasionally loose connections and go dim, but as rear lights are
such a tedious bits of kit I run two on the commute bikes, and that light >>>> definitely isn?t worth repairing.
Roger Merriman
I don't need any lights, but for other maintenance issues, I carry a
lot of tools and I'm inclined to stop in some shade and fix or try to
fix whatever is not working right. I also carry a huge amount of tools
in my truck, including a vice, and what I cannot fix while out on a
ride is likely to under the wrench before I drive home. I am the
opposite of a procrastinator. My wife serves that function at our
home.
--
C'est bon
Soloman
damaged as the bolt threads looked visually fine.
So likely to require rather a lot of dissembling! Im assuming they would >> need to take the inner electronics (likely to be by some margin more
complex than Franks lamb) plus batteries out, to then get at the nut from >> inside the case.
They started off making kit for divers/cavers then MTBers so durable kit
that can be repaired is very much there thing.
And its the new bike that and using it during winter when I needed a light >> to see by, that it became apparent that it was getting worse, and the
problem was highlighted.
But it didnt stop the light working, it still worked fine so certainly
during winter when its by far the best tool for the job I was reluctant to >> be without it, but since Im now commuting even early in daylight and
Franks thread reminded me seemed the time to do so, Id hope to get it back >> last next week or so.
Roger Merriman
I love my electronics and electrics but I have no interest in
disassembling them to work on them. They're mostly inexpensive enough
to be easily replaced. I recently shitcanned a bike computer that
didn't work right. I have an older bike computer that works fine but
got replaced by one with more functionality. I prefer the mechanical
aspects of bike maintenence.
--
C'est bon
Soloman
| Sysop: | Keyop |
|---|---|
| Location: | Huddersfield, West Yorkshire, UK |
| Users: | 546 |
| Nodes: | 16 (2 / 14) |
| Uptime: | 34:31:14 |
| Calls: | 10,391 |
| Calls today: | 2 |
| Files: | 14,064 |
| Messages: | 6,417,137 |
