Schottky diode in SM PSU o/p?...

T

T i m

Guest
Hi all,

I was given a \'dead\' cheapo 20V Lithium drill battery, charger and
wall wart to look at by a mate tonight.

Plug PSU (wart) into wall socket, clip \'remote\' charger interface onto
battery, switch socket on ... nothing.

I took the covers of the charger clip module and connected it onto the
battery, measured 19 or so volts on the battery pins but nothing
coming from the PSU.

However, after a few minutes I felt the PSU was slightly warm so
assumed it was doing something?

I disconnected the PSU output leads from the PSU at the charger module
and connected my bench PSU at 20 and with the current limited to the
same 400mA as the PSU and the green \'Charged\' LED came on constantly.

I rigged up a load (a couple of 12V 60W headlights in series) onto the
battery and let it run for a while and then hooked it back up to the
charger, charging LED now flashing, indicating it was charging.

DMM on the unloaded o/p of the PSU, nothing.

After leaving it disconnected for a while ... I cracked the PSU open
with my toffee hammer and gave the board a look over, nothing
obviously burned and no dry joints etc. I checked the bridge rec and a
couple diode over and they seemed ok, but an axial series diode (SJ220
from memory) on the output that was driving a cap that went to the
remote charger module seemed to be short?

I found the spec somewhere that suggested it was a 2A 1000V (?) device
and I have a couple of questions please?

I know Schottky diodes have fast switching times (and a low forward
voltage drop but not sure if that would be relevant here?) so is a
Schottky required in what I think might be part of a \'charge pump\'
circuit because of the frequency of the output through the SMPSU and
assuming I decoded the markings correctly, why would it need a 1000V
device there?

Cheers, T i m
 
On Friday, October 9, 2020 at 4:51:25 PM UTC-4, T i m wrote:
Hi all,

I was given a \'dead\' cheapo 20V Lithium drill battery, charger and
wall wart to look at by a mate tonight.

Plug PSU (wart) into wall socket, clip \'remote\' charger interface onto
battery, switch socket on ... nothing.

I took the covers of the charger clip module and connected it onto the
battery, measured 19 or so volts on the battery pins but nothing
coming from the PSU.

However, after a few minutes I felt the PSU was slightly warm so
assumed it was doing something?

I disconnected the PSU output leads from the PSU at the charger module
and connected my bench PSU at 20 and with the current limited to the
same 400mA as the PSU and the green \'Charged\' LED came on constantly.

I rigged up a load (a couple of 12V 60W headlights in series) onto the
battery and let it run for a while and then hooked it back up to the
charger, charging LED now flashing, indicating it was charging.

DMM on the unloaded o/p of the PSU, nothing.

After leaving it disconnected for a while ... I cracked the PSU open
with my toffee hammer and gave the board a look over, nothing
obviously burned and no dry joints etc. I checked the bridge rec and a
couple diode over and they seemed ok, but an axial series diode (SJ220
from memory) on the output that was driving a cap that went to the
remote charger module seemed to be short?

I found the spec somewhere that suggested it was a 2A 1000V (?) device
and I have a couple of questions please?

I know Schottky diodes have fast switching times (and a low forward
voltage drop but not sure if that would be relevant here?) so is a
Schottky required in what I think might be part of a \'charge pump\'
circuit because of the frequency of the output through the SMPSU and
assuming I decoded the markings correctly, why would it need a 1000V
device there?

Cheers, T i m

Use the Schottky diode. Sometimes a fast/soft recovery diode will work, but I\'ve run into problems where the drive IC or transistor will run hot or fail or the secondary voltage will be lower if the Schottky diode is subbed from original spec. Also, sometimes those diodes check very leaky even when they\'re not - depending on your meter and the polarity of the leads, even when not using the diode scale. You can put in a fast recovery to see if the circuit works, but I wouldn\'t run it more than a few seconds.
 
On Fri, 09 Oct 2020 21:51:22 +0100, T i m <news@spaced.me.uk> wrote:

Hi all,

I was given a \'dead\' cheapo 20V Lithium drill battery, charger and
wall wart to look at by a mate tonight.

Plug PSU (wart) into wall socket, clip \'remote\' charger interface onto
battery, switch socket on ... nothing.

I took the covers of the charger clip module and connected it onto the
battery, measured 19 or so volts on the battery pins but nothing
coming from the PSU.

However, after a few minutes I felt the PSU was slightly warm so
assumed it was doing something?

I disconnected the PSU output leads from the PSU at the charger module
and connected my bench PSU at 20 and with the current limited to the
same 400mA as the PSU and the green \'Charged\' LED came on constantly.

I rigged up a load (a couple of 12V 60W headlights in series) onto the
battery and let it run for a while and then hooked it back up to the
charger, charging LED now flashing, indicating it was charging.

DMM on the unloaded o/p of the PSU, nothing.

After leaving it disconnected for a while ... I cracked the PSU open
with my toffee hammer and gave the board a look over, nothing
obviously burned and no dry joints etc. I checked the bridge rec and a
couple diode over and they seemed ok, but an axial series diode (SJ220
from memory) on the output that was driving a cap that went to the
remote charger module seemed to be short?

I found the spec somewhere that suggested it was a 2A 1000V (?) device
and I have a couple of questions please?

I know Schottky diodes have fast switching times (and a low forward
voltage drop but not sure if that would be relevant here?) so is a
Schottky required in what I think might be part of a \'charge pump\'
circuit because of the frequency of the output through the SMPSU and
assuming I decoded the markings correctly, why would it need a 1000V
device there?

Cheers, T i m

Output rectifiers of simple wide-range input flybacks
will tend to see voltages stresses unrelated to the
output voltage.

At the same time, HV diodes will be relatively inefficient,
due to slow speed and high Vf.

If you can find real info on the part, it\'ll be easier to make
a safe replacement. Seeing as the original part failed, it
would be unwise to skimp on any of it\'s intended ratings.

RL
 
T i m wrote:

=============

I found the spec somewhere that suggested it was a 2A 1000V (?) device
and I have a couple of questions please?

** That cannot be a Schottky.

The highest voltage available is about 200V.

Most of your facts are WRONG !


...... Phil
 
On Fri, 9 Oct 2020 16:21:38 -0700 (PDT), \"ohg...@gmail.com\"
<ohger1s@gmail.com> wrote:

<snip>

>Use the Schottky diode.

Yeah, I will, I was just wondering why / if they were used in that
role? ;-)

>Sometimes a fast/soft recovery diode will work, but I\'ve run into problems where the drive IC or transistor will run hot or fail or the secondary voltage will be lower if the Schottky diode is subbed from original spec.

Ok.

>Also, sometimes those diodes check very leaky even when they\'re not - depending on your meter and the polarity of the leads, even when not using the diode scale.

I\'ll check it again but I believe I measured the resistance both ways
and it looked short ... and the PSU wasn\'t working (but warming) and
everything else I could check tested out ok.

>You can put in a fast recovery to see if the circuit works, but I wouldn\'t run it more than a few seconds.

Understood. I\'ll wait for the right things to arrive and keep my
fingers crossed it was that. ;-)

Assuming the spec I looked up was correct, and it was a 2A device, and
the PSU itself was rated as 400mA, I wouldn\'t have thought straight
current would have taken it out so would there me a recommended \'safe\'
startup procedure that might put the least shock load on the diode?

eg, Should you connected charger clip to the battery first, then
switch on at the wall (and likely what it says in the instructions)
.... or switch on first (allowing the PSU to get the output cap charged
up) and then connect it to the battery?

Wall wart / PSU <> lv Lead <> charging clip / LED status <> Battery.

Cheers, T i m
 
On Fri, 09 Oct 2020 21:08:30 -0400, legg <legg@nospam.magma.ca> wrote:

<snip>

I know Schottky diodes have fast switching times (and a low forward
voltage drop but not sure if that would be relevant here?) so is a
Schottky required in what I think might be part of a \'charge pump\'
circuit because of the frequency of the output through the SMPSU and
assuming I decoded the markings correctly, why would it need a 1000V
device there?


Output rectifiers of simple wide-range input flybacks
will tend to see voltages stresses unrelated to the
output voltage.

Ah, ok.
At the same time, HV diodes will be relatively inefficient,
due to slow speed and high Vf.

Check.
If you can find real info on the part, it\'ll be easier to make
a safe replacement.

<checks diode again with better light and loupe> Ah, it\'s an SR2200 JF
(not SJ220, that was a result of my dyslexia <g>) that looks to be a
200V 2A Schottky Barrier Diode.

https://www.4project.co.il/product/210423

Seeing as the original part failed, it
would be unwise to skimp on any of it\'s intended ratings.

Understood, I\'ll get the right thing ordered now.

Cheers T i m
 
On Fri, 9 Oct 2020 19:19:34 -0700 (PDT), Phil Allison
<pallison49@gmail.com> wrote:

T i m wrote:

=============


I found the spec somewhere that suggested it was a 2A 1000V (?) device
and I have a couple of questions please?

** That cannot be a Schottky.

Oh?
The highest voltage available is about 200V.

The eBay listing suggested it was a \'Axial Lead Schottky Diode HER208
Ultra-fast Recovery Diode MIC 2A1000V DO-15\' so maybe it was wrong /
misleading?
Most of your facts are WRONG !

Most of my facts were questions but as I have mentioned since and with
a better / closer look, it\'s an SR2200 (not an SJ220) and that is
indeed a 200V, 2A device so will order one now.

Cheers, T i m
 
T i m wrote:

----------------

I found the spec somewhere that suggested it was a 2A 1000V (?) device
and I have a couple of questions please?

** That cannot be a Schottky.

Oh?

The highest voltage available is about 200V.

The eBay listing suggested it was a \'Axial Lead Schottky Diode HER208
Ultra-fast Recovery Diode MIC 2A1000V DO-15\' so maybe it was wrong /
misleading?

** The HER208 is a fast recovery silicon diode, not a Schottky.


.... Phil
 
On Saturday, October 10, 2020 at 5:03:02 AM UTC-4, T i m wrote:
On Fri, 9 Oct 2020 16:21:38 -0700 (PDT), \"ohg...@gmail.com\"
ohg...@gmail.com> wrote:

snip

Use the Schottky diode.

Yeah, I will, I was just wondering why / if they were used in that
role? ;-)
Sometimes a fast/soft recovery diode will work, but I\'ve run into problems where the drive IC or transistor will run hot or fail or the secondary voltage will be lower if the Schottky diode is subbed from original spec.
Ok.
Also, sometimes those diodes check very leaky even when they\'re not - depending on your meter and the polarity of the leads, even when not using the diode scale.
I\'ll check it again but I believe I measured the resistance both ways
and it looked short ... and the PSU wasn\'t working (but warming) and
everything else I could check tested out ok.
You can put in a fast recovery to see if the circuit works, but I wouldn\'t run it more than a few seconds.
Understood. I\'ll wait for the right things to arrive and keep my
fingers crossed it was that. ;-)

Assuming the spec I looked up was correct, and it was a 2A device, and
the PSU itself was rated as 400mA, I wouldn\'t have thought straight
current would have taken it out so would there me a recommended \'safe\'
startup procedure that might put the least shock load on the diode?

400ma is the maximum continuous duty design output, but during the first few cycles, it\'s essentially feeding a short circuit until the filter caps charge. Most modern smps circuits ramp up the drive in a controlled manner using a slow or soft start strategy to mitigate that. Still, sometimes diodes just short, and I\'ve seen more random Schottky failures than other types..



eg, Should you connected charger clip to the battery first, then
switch on at the wall (and likely what it says in the instructions)
... or switch on first (allowing the PSU to get the output cap charged
up) and then connect it to the battery?

Wall wart / PSU <> lv Lead <> charging clip / LED status <> Battery.

I wouldn\'t worry about the start up procedure unless there are specific instructions to the contrary. Consumer electronics are generally designed to be plug and play regardless of the sequence.
 
On Sat, 10 Oct 2020 03:49:20 -0700 (PDT), \"ohg...@gmail.com\"
<ohger1s@gmail.com> wrote:
<snip>

Assuming the spec I looked up was correct, and it was a 2A device, and
the PSU itself was rated as 400mA, I wouldn\'t have thought straight
current would have taken it out so would there me a recommended \'safe\'
startup procedure that might put the least shock load on the diode?


400ma is the maximum continuous duty design output, but during the first few cycles, it\'s essentially feeding a short circuit until the filter caps charge.

Understood.

>Most modern smps circuits ramp up the drive in a controlled manner using a slow or soft start strategy to mitigate that.

This is the PSU OOI: https://ibb.co/VtBY4Rt

The IC is a CR6338T.

>Still, sometimes diodes just short, and I\'ve seen more random Schottky failures than other types.

Oh, ok.
snip

I wouldn\'t worry about the start up procedure unless there are specific instructions to the contrary. Consumer electronics are generally designed to be plug and play regardless of the sequence.
Agreed, I was just thinking that this being a \'cheap and cheerful\'
example of this sort of thing, if they may have cut more corners than
more expensive brands might, making them slightly less \'durable\' under
some circumstances ... how they handle the \'what if\' situation?

But as you say, it could have just been one of those things and if it
works it could be fine till it\'s worn out. ;-)

Cheers, T i m
 
On Sat, 10 Oct 2020 03:31:22 -0700 (PDT), Phil Allison
<pallison49@gmail.com> wrote:

T i m wrote:

----------------


I found the spec somewhere that suggested it was a 2A 1000V (?) device
and I have a couple of questions please?

** That cannot be a Schottky.

Oh?

The highest voltage available is about 200V.

The eBay listing suggested it was a \'Axial Lead Schottky Diode HER208
Ultra-fast Recovery Diode MIC 2A1000V DO-15\' so maybe it was wrong /
misleading?


** The HER208 is a fast recovery silicon diode, not a Schottky.

As I was looking back into it that\'s what I surmised Phil.

I\'ve ordered the right thing now that should be here soon and I\'ll let
the group know how I get on.

If it\'s not that there isn\'t much else outside the CR6338T. ;-)

Initially I was just going to look for a generic 20V, 400mA PSU but
thought I\'d open this one up just to see if there was anything
obvious.

Cheers, T i m
 
On Saturday, October 10, 2020 at 8:17:15 AM UTC-4, T i m wrote:
On Sat, 10 Oct 2020 03:49:20 -0700 (PDT), \"ohg...@gmail.com\"
ohger1s@gmail.com> wrote:
snip

Assuming the spec I looked up was correct, and it was a 2A device, and
the PSU itself was rated as 400mA, I wouldn\'t have thought straight
current would have taken it out so would there me a recommended \'safe\'
startup procedure that might put the least shock load on the diode?


400ma is the maximum continuous duty design output, but during the first few cycles, it\'s essentially feeding a short circuit until the filter caps charge.

Understood.

Most modern smps circuits ramp up the drive in a controlled manner using a slow or soft start strategy to mitigate that.

This is the PSU OOI: https://ibb.co/VtBY4Rt

The IC is a CR6338T.

Still, sometimes diodes just short, and I\'ve seen more random Schottky failures than other types.

Oh, ok.

snip

I wouldn\'t worry about the start up procedure unless there are specific instructions to the contrary. Consumer electronics are generally designed to be plug and play regardless of the sequence.

Agreed, I was just thinking that this being a \'cheap and cheerful\'
example of this sort of thing, if they may have cut more corners than
more expensive brands might, making them slightly less \'durable\' under
some circumstances ... how they handle the \'what if\' situation?

But as you say, it could have just been one of those things and if it
works it could be fine till it\'s worn out. ;-)

Cheers, T i m

Years ago when everything was discrete, we saw plenty of power supplies so badly designed that an excessive load or even a brown out would cause a catastrophic failure (Sony). Today, an properly designed IC SMPS controller is literally pennies each, and they have all that solved internally. Your IC does have a soft start internal to it and overcurrent protection, so the shorted Schottky diode did not seem to destroy the IC (check pin 5/6 to 4, should be no short on the internal mosfet).
 
On Sat, 10 Oct 2020 09:34:11 -0700 (PDT), John-Del <ohger1s@gmail.com>
wrote:

<snip>

This is the PSU OOI: https://ibb.co/VtBY4Rt

The IC is a CR6338T.

snip

>Years ago when everything was discrete, we saw plenty of power supplies so badly designed that an excessive load or even a brown out would cause a catastrophic failure (Sony).

;-(

>Today, an properly designed IC SMPS controller is literally pennies each, and they have all that solved internally.

Ok.

>Your IC does have a soft start internal to it and overcurrent protection, so the shorted Schottky diode did not seem to destroy the IC

Phew.

>(check pin 5/6 to 4, should be no short on the internal mosfet).

5/6 to 4 test out like a pair of diodes but with a low resistance
(either way) between 5 and 6?

Cheers, T i m
 
On Sunday, October 11, 2020 at 3:01:40 PM UTC-4, T i m wrote:
On Sat, 10 Oct 2020 09:34:11 -0700 (PDT), John-Del <ohger1s@gmail.com
wrote:

snip

This is the PSU OOI: https://ibb.co/VtBY4Rt

The IC is a CR6338T.

snip

Years ago when everything was discrete, we saw plenty of power supplies so badly designed that an excessive load or even a brown out would cause a catastrophic failure (Sony).

;-(

Today, an properly designed IC SMPS controller is literally pennies each, and they have all that solved internally.

Ok.

Your IC does have a soft start internal to it and overcurrent protection, so the shorted Schottky diode did not seem to destroy the IC

Phew.

(check pin 5/6 to 4, should be no short on the internal mosfet).

5/6 to 4 test out like a pair of diodes but with a low resistance
(either way) between 5 and 6?

Cheers, T i m

5 and 6 should read short as they are the common drain pin and internally tied together inside the IC. Don\'t use the diode scale when checking for leakage between those pins. Pins 5 and 6 are the drain, and pin 4 is the source. Out of circuit, it should read virtually open between pins 5/6 to pin 4.
 
On Fri, 09 Oct 2020 21:51:22 +0100, T i m <news@spaced.me.uk> wrote:

<snip>

After leaving it disconnected for a while ... I cracked the PSU open
with my toffee hammer and gave the board a look over, nothing
obviously burned and no dry joints etc. I checked the bridge rec and a
couple diode over and they seemed ok, but an axial series diode (SJ220
from memory) on the output that was driving a cap that went to the
remote charger module seemed to be short?
Just a quick update for the records ... I replaced the diode with the
right / same one, taped the PSU together, stuck my DMM on 200V DV on
the output and gingerly turned it on. There was a slight pause and
then the DMM showed the voltage quickly (but not instantly) ramping up
to 20 volts. ;-)

I switched it off, unplugged it and watched the output drop back down
to 0V and then soldered the wires back onto the charger clip, screwed
it back together and slid it onto the battery.

Plugged in, turned on, a second or so pause then the green \'Charging\'
LED started flashing <wahoo>! ;-)

I\'ll leave it till it indicates it\'s fully charged then seal the PSU
back up again [1] and should have a happy mate. ;-)

Thanks to all those who responded ... another load of stuff (charger,
battery, drill) that has been saved from the waste ... for the sake of
a few pence worth of diode (and my free time of course, which is the
big issue for those not like us <g> [2]).

Cheers, T i m

[1] I will leave him with the instruction that he\'s not to pass the
thing onto anyone else.

[2] Now to try and find out why our soup maker was working fine one
minute and then completely dead the next. ;-(
 
On Wednesday, October 14, 2020 at 5:59:33 AM UTC-4, T i m wrote:

[2] Now to try and find out why our soup maker was working fine one
minute and then completely dead the next. ;-(

MOM?????
 
On 15.10.20 17:35, John-Del wrote:
On Wednesday, October 14, 2020 at 5:59:33 AM UTC-4, T i m wrote:


[2] Now to try and find out why our soup maker was working fine one
minute and then completely dead the next. ;-(

MOM?????
It tried to feed on your own soup?
 
On Thu, 15 Oct 2020 08:35:12 -0700 (PDT), John-Del <ohger1s@gmail.com>
wrote:

On Wednesday, October 14, 2020 at 5:59:33 AM UTC-4, T i m wrote:


[2] Now to try and find out why our soup maker was working fine one
minute and then completely dead the next. ;-(

MOM?????

At 90+ she still seems to be plodding along <g> ... no, one of those
electric jobbies where you just chuck the ingredients and some water
in, press \'Chunky\' or \'Smooth\' and it beeps to tell you when it\'s done
....

Actually, since I mentioned it here I took it to bits, tested it in
bits (carefully, another SMPSU) it seemed to come to life and it still
seems alive now it\'s back together?

I couldn\'t see any dry / bad solder joints so will see how it goes and
if it dies again, check all the crimps / plugs / sockets out.

(I bought a disability scooter that was my Dads, back of a mate with a
fault and that turned out to be a connector within a multipole plug
had pushed though and was not making contact).

Cheers, T i m
 

Welcome to EDABoard.com

Sponsor

Back
Top