Chinese commodity power supplies have tended to use recognizable configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist, particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Case in point is a 5V 40A unit advertised 'for use in LED sign',
commonly used in Onbon product. In the application where a repair
or replace decision was made, actual consumption was in the 35W
range, though a test sequence could draw much higher power.
replacement with an identically rated unit was Cdn$22.00.
The replacement was physically and schematically identical, but
relaid as a mirror image for component placement. Different
brand name.
Anyways - a basic self-oscillating bipolar transistor half bridge
with forced beta, synchronized/steered and pwm'd by opening and
shorting the resistor-limited, center-tapped 'drive' winding.
Open collector drive out of a KA7500.
What's a KA7500 ? Turns out to be pin compatible to TL494, but
mfrd by Samsung/Fairchild/ONS.
http://ve3ute.ca/query/TL494_vs_KA7500.pdf
Oodles of data and apps for the 494, not so much for the 7500.
If anyone's got app info published for the KA7900, in any
language, I'd be interested to see it.
legg wrote:
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist,
particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Case in point is a 5V 40A unit advertised 'for use in LED sign',
commonly used in Onbon product. In the application where a repair
or replace decision was made, actual consumption was in the 35W
range, though a test sequence could draw much higher power.
replacement with an identically rated unit was Cdn$22.00.
The replacement was physically and schematically identical, but
relaid as a mirror image for component placement. Different
brand name.
Anyways - a basic self-oscillating bipolar transistor half bridge
with forced beta, synchronized/steered and pwm'd by opening and
shorting the resistor-limited, center-tapped 'drive' winding.
Open collector drive out of a KA7500.
What's a KA7500 ? Turns out to be pin compatible to TL494, but
mfrd by Samsung/Fairchild/ONS.
http://ve3ute.ca/query/TL494_vs_KA7500.pdf
Oodles of data and apps for the 494, not so much for the 7500.
If anyone's got app info published for the KA7900, in any
language, I'd be interested to see it.
As you say, KA7500 specific notes are nearly non-existent; other than
this circuit schematic of an application available at DiodeGoneWild [1]:
<https://danyk.cz/s_atx01h.png>
Don wrote:
legg wrote:<snip>
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist,
particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Case in point is a 5V 40A unit advertised 'for use in LED sign',
commonly used in Onbon product. In the application where a repair
or replace decision was made, actual consumption was in the 35W
range, though a test sequence could draw much higher power.
replacement with an identically rated unit was Cdn$22.00.
The replacement was physically and schematically identical, but
relaid as a mirror image for component placement. Different
brand name.
Anyways - a basic self-oscillating bipolar transistor half bridge
with forced beta, synchronized/steered and pwm'd by opening and
shorting the resistor-limited, center-tapped 'drive' winding.
Open collector drive out of a KA7500.
What's a KA7500 ? Turns out to be pin compatible to TL494, but
mfrd by Samsung/Fairchild/ONS.
http://ve3ute.ca/query/TL494_vs_KA7500.pdf
Oodles of data and apps for the 494, not so much for the 7500.
If anyone's got app info published for the KA7900, in any
language, I'd be interested to see it.
As you say, KA7500 specific notes are nearly non-existent; other than
this circuit schematic of an application available at DiodeGoneWild [1]:
<https://danyk.cz/s_atx01h.png>
That's basically the controller drive and regulation section found in commodity single output jobs, but they use output copper track and
links to develop a bulk current limit, within the input compliance
range of the error amp inputs (one would hope).
I note that in the Sunny Tech schematic, the slow start and aux
overvoltage line is disconnected from either regulation path.
The 12V 40A unit used MBRF30100 output rectifiers (fully insulated
TO220). I'll believe the part msrkings/ratings when I see it.
legg wrote:
Don wrote:
legg wrote:<snip>
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist,
particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Case in point is a 5V 40A unit advertised 'for use in LED sign',
commonly used in Onbon product. In the application where a repair
or replace decision was made, actual consumption was in the 35W
range, though a test sequence could draw much higher power.
replacement with an identically rated unit was Cdn$22.00.
The replacement was physically and schematically identical, but
relaid as a mirror image for component placement. Different
brand name.
Anyways - a basic self-oscillating bipolar transistor half bridge
with forced beta, synchronized/steered and pwm'd by opening and
shorting the resistor-limited, center-tapped 'drive' winding.
Open collector drive out of a KA7500.
What's a KA7500 ? Turns out to be pin compatible to TL494, but
mfrd by Samsung/Fairchild/ONS.
http://ve3ute.ca/query/TL494_vs_KA7500.pdf
Oodles of data and apps for the 494, not so much for the 7500.
If anyone's got app info published for the KA7900, in any
language, I'd be interested to see it.
As you say, KA7500 specific notes are nearly non-existent; other than >>>this circuit schematic of an application available at DiodeGoneWild [1]:
<https://danyk.cz/s_atx01h.png>
That's basically the controller drive and regulation section found in
commodity single output jobs, but they use output copper track and
links to develop a bulk current limit, within the input compliance
range of the error amp inputs (one would hope).
I note that in the Sunny Tech schematic, the slow start and aux
overvoltage line is disconnected from either regulation path.
The 12V 40A unit used MBRF30100 output rectifiers (fully insulated
TO220). I'll believe the part msrkings/ratings when I see it.
My knowledge of SMPS is limited. Where are the slow start and aux
overvoltage lines?
It seems like the thermister, tau set from R34 and C16, and the output >control (pin 13) ought to contribute something to a slow start.
Does the circuit around pin 15 sense overvoltage?
Danke,
this circuit schematic of an application available at DiodeGoneWild [1]:
<https://danyk.cz/s_atx01h.png>
On 03/04/2025 16:11, legg wrote:
this circuit schematic of an application available at DiodeGoneWild [1]: >>>>
<https://danyk.cz/s_atx01h.png>
The circuitry around Q5 and Q6, far from being some kind
of OVP protection, is actually an undervoltage latch.
When either transistor turns off, the slow-start capacitor
is discharged fairly quickly.
If the missing connection to pin 4 is present, the PW is
inhibited and conversion is latched off, until the
collapsing housekeeping supply turns the chip off.
Combined with a current limit, this can produce a hiccoughing
response to overload or short circuit on the lower-powered
outputs.Sort of a Hail Mary approach.
Q5 disables output undervoltage effects while the housekeeping
supply is rising (at start-up), or has gross ripple.
Some of the commodity supplies of the type don't have this
added circuitry, hence their sensitivity to output
shorts and overload.
Single output units may monitor the one output or simply
count on current limiting to reduce PW sufficiently to
collapse the housekeeping supply.
RL
Slightly relevant to the above discussion - I recently
wanted to get an old Dell notebook PC running and found
that many of the (supposedly) genuine Dell power supplies
I had to hand did not work. One was completely broken,
but several of the others allowed the notebook to run for
a few seconds before shutting down. All were adequately rated.
I checked them all with a variable load resistor and
found that they would deliver significantly more current
than their rated output. However, once they current limited
they latched off and could only be restarted by power cycling
the mains input.
The two supplies I found that would run the notebook were
very different. One, labeled as a genuine Dell unit kept
delivering more and more current until I stopped as the
output dropped from 19.5V to 12V at about 8.5A.
The other one fold-back current limited at a sensible degree of
overload and restarted when the overload was removed. This
was made by Lite-am. This is the one I am now using.
The notebook PC may of course be drawing far too much current
at startup. I will check this later. The battery is dead.
However, I was quite surprised by these results.
John
All these power supplies are the three-contact barrel type. They
all deliver a fixed 19.5V, even into a resistive load. They can
also all deliver well over their rated output currents. The issue
seems to be that their failure mode is an inability to retry when
the current limit has been triggered.
I don't think they have any mechanism for negotiating like a USB-C
PD supply. Instead they just have a memory accessed through a 1-wire interface which tells the computer what their rating is.
John
Chinese commodity power supplies have tended to use recognizable configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist, particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Case in point is a 5V 40A unit advertised 'for use in LED sign',
commonly used in Onbon product. In the application where a repair
or replace decision was made, actual consumption was in the 35W
range, though a test sequence could draw much higher power.
replacement with an identically rated unit was Cdn$22.00.
The replacement was physically and schematically identical, but
relaid as a mirror image for component placement. Different
brand name.
Anyways - a basic self-oscillating bipolar transistor half bridge
with forced beta, synchronized/steered and pwm'd by opening and
shorting the resistor-limited, center-tapped 'drive' winding.
Open collector drive out of a KA7500.
What's a KA7500 ? Turns out to be pin compatible to TL494, but
mfrd by Samsung/Fairchild/ONS.
http://ve3ute.ca/query/TL494_vs_KA7500.pdf
Oodles of data and apps for the 494, not so much for the 7500.
If anyone's got app info published for the KA7900, in any
language, I'd be interested to see it.
On 3/25/25 6:25 AM, legg wrote:
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist,
particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Burn-in? Doesn't that happen at the customer? :-)
Case in point is a 5V 40A unit advertised 'for use in LED sign',
commonly used in Onbon product. In the application where a repair
or replace decision was made, actual consumption was in the 35W
range, though a test sequence could draw much higher power.
replacement with an identically rated unit was Cdn$22.00.
The replacement was physically and schematically identical, but
relaid as a mirror image for component placement. Different
brand name.
Anyways - a basic self-oscillating bipolar transistor half bridge
with forced beta, synchronized/steered and pwm'd by opening and
shorting the resistor-limited, center-tapped 'drive' winding.
Open collector drive out of a KA7500.
What's a KA7500 ? Turns out to be pin compatible to TL494, but
mfrd by Samsung/Fairchild/ONS.
http://ve3ute.ca/query/TL494_vs_KA7500.pdf
Oodles of data and apps for the 494, not so much for the 7500.
If anyone's got app info published for the KA7900, in any
language, I'd be interested to see it.
You mean a datasheet? It's here:
https://www.mouser.com/datasheet/2/149/KA7500C-89501.pdf?srsltid=AfmBOoqgRNYOTol-O6pVSUwQ0v07-Tqdgmcnph3f_Uq5xj18Tvq5uAYY
On Tue, 8 Apr 2025 21:22:37 -0700, Joerg <news@analogconsultants.com>
wrote:
On 3/25/25 6:25 AM, legg wrote:
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist,
particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Burn-in? Doesn't that happen at the customer? :-)
No, burn-in is a well-defined process control step used in the
manufacturing of equipment to achieve and maintain low failure
rates (ppm).
On Wed, 09 Apr 2025 09:42:22 -0400, legg <legg@nospam.magma.ca> wrote:
On Tue, 8 Apr 2025 21:22:37 -0700, Joerg <news@analogconsultants.com> >>wrote:
On 3/25/25 6:25 AM, legg wrote:
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold
- why even bother with burn-in, in that case? If no burn-in or field
return failure analysis is ever consudered, the small errors persist,
particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Burn-in? Doesn't that happen at the customer? :-)
No, burn-in is a well-defined process control step used in the >>manufacturing of equipment to achieve and maintain low failure
rates (ppm).
What fraction of the parts and equipment that you buy has been
burned-in? And how do you know?
I'd expect 0%, and that you don't know.
On Wed, 09 Apr 2025 08:51:52 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Wed, 09 Apr 2025 09:42:22 -0400, legg <legg@nospam.magma.ca> wrote:
On Tue, 8 Apr 2025 21:22:37 -0700, Joerg <news@analogconsultants.com> >>>wrote:
On 3/25/25 6:25 AM, legg wrote:
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold >>>>> - why even bother with burn-in, in that case? If no burn-in or field >>>>> return failure analysis is ever consudered, the small errors persist, >>>>> particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Burn-in? Doesn't that happen at the customer? :-)
No, burn-in is a well-defined process control step used in the >>>manufacturing of equipment to achieve and maintain low failure
rates (ppm).
What fraction of the parts and equipment that you buy has been
burned-in? And how do you know?
I'd expect 0%, and that you don't know.
Purchases of assembled hardware, here, are generally consumer
grade, with no obvious indication that infant mortality has
been addressed.
Some incoming aql levels are specified in the data sheets at
the component level. Things like pumps, motors and power supply
units are included in this category.
Designs or products that go out the door can only achieve ppm
failure rates if a burn-in strategy is included after final assy.
Nuts and bolts can fail just as often at this stage.
RL
On Sat, 12 Apr 2025 15:08:18 -0400, legg <legg@nospam.magma.ca> wrote:
On Wed, 09 Apr 2025 08:51:52 -0700, john larkin <jl@glen--canyon.com> >>wrote:
On Wed, 09 Apr 2025 09:42:22 -0400, legg <legg@nospam.magma.ca> wrote:
On Tue, 8 Apr 2025 21:22:37 -0700, Joerg <news@analogconsultants.com> >>>>wrote:
On 3/25/25 6:25 AM, legg wrote:
Chinese commodity power supplies have tended to use recognizable
configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable
product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold >>>>>> - why even bother with burn-in, in that case? If no burn-in or field >>>>>> return failure analysis is ever consudered, the small errors persist, >>>>>> particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Burn-in? Doesn't that happen at the customer? :-)
No, burn-in is a well-defined process control step used in the >>>>manufacturing of equipment to achieve and maintain low failure
rates (ppm).
What fraction of the parts and equipment that you buy has been
burned-in? And how do you know?
I'd expect 0%, and that you don't know.
Purchases of assembled hardware, here, are generally consumer
grade, with no obvious indication that infant mortality has
been addressed.
Some incoming aql levels are specified in the data sheets at
the component level. Things like pumps, motors and power supply
units are included in this category.
Designs or products that go out the door can only achieve ppm
failure rates if a burn-in strategy is included after final assy.
Nuts and bolts can fail just as often at this stage.
RL
With modern electronics, burnin isn't necessary or feasible.
Temperature cycling and vibration would improve reliability a bit, but
that's not practical either.
The biggest failure cause is bad engineering.
Modern electronics, except for the obvious cheap junk, is remarkably >reliable.
On Sat, 12 Apr 2025 13:07:19 -0700, john larkin <jl@glen--canyon.com>
wrote:
On Sat, 12 Apr 2025 15:08:18 -0400, legg <legg@nospam.magma.ca> wrote:
On Wed, 09 Apr 2025 08:51:52 -0700, john larkin <jl@glen--canyon.com> >>>wrote:
On Wed, 09 Apr 2025 09:42:22 -0400, legg <legg@nospam.magma.ca> wrote:
On Tue, 8 Apr 2025 21:22:37 -0700, Joerg <news@analogconsultants.com> >>>>>wrote:
On 3/25/25 6:25 AM, legg wrote:
Chinese commodity power supplies have tended to use recognizable >>>>>>> configurations from times gone by. In doing so, it's easy to
miss some of the 'small stuff' that actually produced a reliable >>>>>>> product, in the day.
Even more so, when pricing reaches the 'replace vs repair' threshold >>>>>>> - why even bother with burn-in, in that case? If no burn-in or field >>>>>>> return failure analysis is ever consudered, the small errors persist, >>>>>>> particularly if vendors play wack-a-mole with the same hardware
offered under different brand names and paperwork.
Burn-in? Doesn't that happen at the customer? :-)
No, burn-in is a well-defined process control step used in the >>>>>manufacturing of equipment to achieve and maintain low failure
rates (ppm).
What fraction of the parts and equipment that you buy has been >>>>burned-in? And how do you know?
I'd expect 0%, and that you don't know.
Purchases of assembled hardware, here, are generally consumer
grade, with no obvious indication that infant mortality has
been addressed.
Some incoming aql levels are specified in the data sheets at
the component level. Things like pumps, motors and power supply
units are included in this category.
Designs or products that go out the door can only achieve ppm
failure rates if a burn-in strategy is included after final assy.
Nuts and bolts can fail just as often at this stage.
RL
With modern electronics, burnin isn't necessary or feasible.
Temperature cycling and vibration would improve reliability a bit, but >>that's not practical either.
The biggest failure cause is bad engineering.
Modern electronics, except for the obvious cheap junk, is remarkably >>reliable.
Apart from the price, there's nothing 'obvious' about modern
electronic reliability. Modern's got nothing to do with it.
Off-shore hardware can suffer 200% tarrifs and still be
competative at the retail level. There's plenty of room
for quality control.
Those margins are, instead, being absorbed by shareholders;
hence the stock market sensitivity at that end.
RL
| Sysop: | Keyop |
|---|---|
| Location: | Huddersfield, West Yorkshire, UK |
| Users: | 489 |
| Nodes: | 16 (2 / 14) |
| Uptime: | 46:07:46 |
| Calls: | 9,670 |
| Calls today: | 1 |
| Files: | 13,719 |
| Messages: | 6,170,073 |
