Ansmann Energy 16 PSU

[T i m]

Hi All,

I fired my Ansmann Energy 16, cell battery charger up earlier to
charge 4 x D NiMh's and initially it looked like it was behaving
normally but when I looked at it again a few minutes later, there were
no LEDs illuminated? ;-(

Checked the mains (3A) fuse and it was blown, so I removed the cells,
replaced the fuse, tried it again and it blew again.

I took it to bits, blew a bit of dust out of it and running the DMM
over it on low Ohms, it seems there is a fairly low resistance across
the AC input (~ 2 ohms).

<https://www.dropbox.com/s/r93ghlg58qgcxhv/Ansmann%20PSU%20top.jpg?dl=0>

<https://www.dropbox.com/s/yx4vg5td3mqkiu2/Ansmann%20PSU%20bottom.jpg?dl=0>

I then disconnected the PSU and (carefully) ran it up on the bench and
again, the fuse blew instantly (however, I'm not sure if it's ok to
run the PSU like that with no load etc)?

I de-soldered one end of both the input filter caps, and the first
transformer and the very low resistance went away on the input side
(so it wasn't a shorted cap) but was still across the board where the
txfmr output was. FWIW, of the 4 bridge rectifier diodes D1-4 in the
middle of the board, D2-3 test out ok in cct, D1,4 test as short cct.
All the other diodes and caps around that part of the circuit seem to
test out ok on the DMM (diode test, cap continuity charge 'beep' upon
probe connection) so I was wondering if someone could point me in the
right direction re checking it out further etc please?

Given there isn't much else on that board, is it likely it's the first
txfmr, one (or both) of the TOP223Y's (PWM switch) and how could I
test them if it was please? Or could it be the (PC817) opto-isolators
as I assume they would be fairly easy to test out of the board?

Thanks for your help in advance ... ;-)

Cheers, T i m

p.s. I might have got some of the above wrong so please assume that to
be the case as I don't want to mislead anyone. ;-(

[Robert Roland]

On Sat, 29 Jul 2017 17:27:21 +0100, T i m <ne...@spaced.me.uk> wrote:

>D1,4 test as short cct.

I would take those out and check outside the circuit.

If the short is at the output of the bridge rectifier, and all four
diodes are good, you should measure a diode drop (0.6 V or so) across
the diodes in both directions.
--
RoRo

[Dave M]

Replace the shorted recctifiers!!!! As a precaution, replace all 4 of them;
they're cheap. Even in circuit, they should not test as a short corcuit.

Cheers,
Dave M

[tabb...@gmail.com]

if you test individually: 0.6v one way only
if testing at bridge input: 1.2v both ways

An easy way to find faults is to run the psu via a filament light bulb, then it's easy to probe the [dangerously live] circuit for voltages with due care. Since I don't know if you can do this safely, don't do it.


NT

[T i m]

On Sat, 29 Jul 2017 13:17:05 -0500, "Dave M" <dgmi...@mediacombb.net>
wrote:

>Replace the shorted recctifiers!!!! As a precaution, replace all 4 of them;
>they're cheap. Even in circuit, they should not test as a short corcuit.
>

<snip>

Thanks Robert / Dave,

I haven't played much with SMPS (for obvious (safety) reasons) so
assumed there might be some strange 'appears short but isn't in use'
type thing going on. I should have pulled those diodes before asking
here. ;-(

However, I just desoldered one end of both and one appears to be dead
short but the other was ok (but I've pulled them all).

Problem is, while I probably have 1,000 IN4004's, I don't have any
1N4007's and I'm guessing we need the 1000V peak reverse
/ DC blocking voltage (rather than just 400) and 700 over 280V RMS?
;-(

So, I can pick some of those up next week (eBay) or it looks like
Maplin have 1N4007S's in stock but are they an acceptable equivalent?
(Peak forward surge current, 30 not 50A)?

http://pdf1.alldatasheet.com/datasheet-pdf/view/329384/CHENG-YI/IN4004.html

http://www.mouser.com/ds/2/395/1N4001S%20SERIES_F15-1082941.pdf

Cheers, T i m

p.s. How long is a piece of string question but ... Is it 'likely'
that diode just went (the charger is a good few years old now and has
been used quite a bit), or is it more likely something further down
the line took it out?

[T i m]

On Sat, 29 Jul 2017 11:49:28 -0700 (PDT), tabb...@gmail.com wrote:

>On Saturday, 29 July 2017 19:03:01 UTC+1, Robert Roland wrote:
>> On Sat, 29 Jul 2017 17:27:21 +0100, T i m <ne...@spaced.me.uk> wrote:
>>
>> >D1,4 test as short cct.
>>
>> I would take those out and check outside the circuit.
>>
>> If the short is at the output of the bridge rectifier, and all four
>> diodes are good, you should measure a diode drop (0.6 V or so) across
>> the diodes in both directions.
>
>if you test individually: 0.6v one way only

Yeah, my diode test function was showing around 520mV on the good
ones.

>if testing at bridge input: 1.2v both ways

Understood.

>
>An easy way to find faults is to run the psu via a filament light bulb,

I did remember that earlier and was going to rig up a plug - lamp
-socket jig to make it easier. I thought I'd better ask here first.
;-)

>then it's easy to probe the [dangerously live] circuit for voltages with due care.

Understood.

>Since I don't know if you can do this safely, don't do it.

Understood. Whilst I have co-designed and scratch built plenty of
mains powered kit (including disco sound-to-light, sequencers and
strobes) I prefer not to be playing with live kit unless it's the only
way (or the kit is very well mounted and otherwise insulated /
protected, not just loose on the bench etc).

Cheers, T i m

[tom]

"T i m" <ne...@spaced.me.uk> wrote in message
news:h9npnct8gjd7pmoh3...@4ax.com...

It is likely that the off line switch regulator is also failed. Grab a copy
of the TOP223Y datasheet to see the reference design for your supply.

Here is a source for those chips:

http://www.ebay.com/itm/5PCS-TOP223Y-Encapsulation-TO-220-Three-terminal-Off-line-PWM-Switch-/371080857393?hash=item5666248731:g:h~8AAOSwnLdWrEJR

[Dave M]

Yes, the 1N4007 recrifiers should be good replacements. Likely to be the
only problem in your unit.

As a precaution, you should check the large transistors on the board. They
are likely to be high voltage MOSFETs, and should NOT be shorted or low
resistance from drain to source or from gate to any other terminal. Best to
take them off the board to test, since they're certain to be connected to
the low resistance transformer windings, which could lead you astray.

Cheers,
Dave M

T i m wrote:

[tom]

Dave,

The "large transistors" are TOP223Y off line switching regulator ICs. Google
the part number for the datasheet and reference design.

Regards



"Dave M" <dgmi...@mediacombb.net> wrote in message
news:1KmdnWTYV6-adOHE...@giganews.com...

[T i m]

On Sat, 29 Jul 2017 15:54:33 -0400, "tom" <tmille...@verizon.net>
wrote:

<snip>

>> p.s. How long is a piece of string question but ... Is it 'likely'
>> that diode just went (the charger is a good few years old now and has
>> been used quite a bit), or is it more likely something further down
>> the line took it out?
>
>It is likely that the off line switch regulator is also failed.

There are two in this PSU so could that apply to one or both?

> Grab a copy
>of the TOP223Y datasheet to see the reference design for your supply.

I have looked at this:

https://www.digchip.com/datasheets/parts/datasheet/373/TOP223Y-pdf.php

... and see roughly how it would work but not sure if that means I
should be able to test it somehow? FWIW, I have one of the little
component testers but I don't know if it will know what that is or
even damage it?

>
>Here is a source for those chips:
>
>http://www.ebay.com/itm/5PCS-TOP223Y-Encapsulation-TO-220-Three-terminal-Off-line-PWM-Switch-/371080857393?hash=item5666248731:g:h~8AAOSwnLdWrEJR
>

Thanks for that. However I am in the UK and but get them here also.

[T i m]

On Sat, 29 Jul 2017 16:02:34 -0400, "tom" <tmille...@verizon.net>
wrote:

>Dave,
>
>The "large transistors" are TOP223Y off line switching regulator ICs. Google
>the part number for the datasheet and reference design.
>

Would you think one of those component testers would be able to
identify the PWM reg, at least with a go/no-go result?

From testing around there were no shorts around the devices but that
doesn't stop them being OC or just non-functional?

Cheers, T i m

[tom]

"T i m" <ne...@spaced.me.uk> wrote in message

news:2nqpnct2pd9g7dna6...@4ax.com...

A quick ohmmeter check across the three pins should not show a short
circuit.

If it does so in circuit, you may need to remove the device to check.

[tom]

"T i m" <ne...@spaced.me.uk> wrote in message

news:2nqpnct2pd9g7dna6...@4ax.com...

Look at fig 8 in the datasheet for how your unit works. If I read your post
right, this charges LiIon batteries, right? So the outputs should be about
4.0 to 4.2 volts. There is some feedback coming from the battery that
probably looks at the battery temperature for a safety shutdown.

You could probably reverse engineer the whole schematic (or half of it) to
get a better view of what is going on.

BTW, the transformer on the right next to the capacitor is not a
transformer. It is a common mode choke for RFI filtering.

[T i m]

On Sat, 29 Jul 2017 16:32:22 -0400, "tom" <tmille...@verizon.net>
wrote:

<snip>

>> I have looked at this:
>>
>> https://www.digchip.com/datasheets/parts/datasheet/373/TOP223Y-pdf.php
>>
>> ... and see roughly how it would work but not sure if that means I
>> should be able to test it somehow? FWIW, I have one of the little
>> component testers but I don't know if it will know what that is or
>> even damage it?
>>>

<snip>

>Look at fig 8 in the datasheet for how your unit works.

OK. Again, I get the rough idea but I was more into digital
electronics than SMPSU etc.

>If I read your post
>right, this charges LiIon batteries, right?

Sorry, no, it's a general purpose multi-cell NiCad / NiMh charger Tom.

http://www.conrad.com/medias/global/ce/2000_2999/2500/2500/2503/250363_RB_00_FB.EPS_1000.jpg
http://www.ansmann.de/en/service/downloads/send/6-bedienungsanleitungen-aelterer-produkte/66-5207123-energy-16

<snip>

The PSU board feeds a fairly complex logic / output board (4 PSU
outputs to the logic board) and in turn the 16 charger outputs.

>
>You could probably reverse engineer the whole schematic (or half of it) to
>get a better view of what is going on.

Ok.

>
>BTW, the transformer on the right next to the capacitor is not a
>transformer. It is a common mode choke for RFI filtering.

Ah, ok, thanks ... that (now) makes more sense as even with that
(choke) in and D1-4 out, I've lost the short I was seeing across the
mains input. ;-)

Cheers, T i m

[tom]

"T i m" <ne...@spaced.me.uk> wrote in message

news:respncdm39h95fhbh...@4ax.com...

Ok, good. That makes more sense for why so many outputs.

Doing any measurements on this device (scope, etc) will require an isolation
transformer to look at the primary (line) side. The outputs all appear to be
isolated.

Hope it's just the shorted diodes.

Good luck.

Regards

[tabb...@gmail.com]

On Saturday, 29 July 2017 20:59:12 UTC+1, Dave M wrote:
> Yes, the 1N4007 recrifiers should be good replacements. Likely to be the
> only problem in your unit.
>
> As a precaution, you should check the large transistors on the board. They
> are likely to be high voltage MOSFETs, and should NOT be shorted or low
> resistance from drain to source or from gate to any other terminal. Best to
> take them off the board to test, since they're certain to be connected to
> the low resistance transformer windings, which could lead you astray.
>
> Cheers,
> Dave M

I've been led astray by small amounts of capacitance before


NT

[T i m]

On Sat, 29 Jul 2017 17:17:26 -0400, "tom" <tmille...@verizon.net>
wrote:

<snip>

>> Ah, ok, thanks ... that (now) makes more sense as even with that
>> (choke) in and D1-4 out, I've lost the short I was seeing across the
>> mains input. ;-)
>>
>

>Ok, good. That makes more sense for why so many outputs.

It's quite an expensive unit (I think it cost me ~100 UKP some years
ago) so I'd like to get it going if I can.

>
>Doing any measurements on this device (scope, etc) will require an isolation
>transformer to look at the primary (line) side.

Understood.

>The outputs all appear to be
>isolated.

Ok.

>
>Hope it's just the shorted diodes.

Fingers crossed. ;-)
>
>Good luck.

Thanks for your help and I'll let you know how I get on.

Cheers, T i m

[tabb...@gmail.com]

On Saturday, 29 July 2017 22:38:41 UTC+1, T i m wrote:
> On Sat, 29 Jul 2017 17:17:26 -0400, "tom" <tmille...@verizon.net>
> wrote:
>
> <snip>
>
> >> Ah, ok, thanks ... that (now) makes more sense as even with that
> >> (choke) in and D1-4 out, I've lost the short I was seeing across the
> >> mains input. ;-)

you would, 99% of the time

> >Ok, good. That makes more sense for why so many outputs.
>
> It's quite an expensive unit (I think it cost me ~100 UKP some years
> ago) so I'd like to get it going if I can.
> >
> >Doing any measurements on this device (scope, etc) will require an isolation
> >transformer to look at the primary (line) side.
>
> Understood.
>
> >The outputs all appear to be
> >isolated.
>
> Ok.
> >
> >Hope it's just the shorted diodes.
>
> Fingers crossed. ;-)
> >
> >Good luck.
>
> Thanks for your help and I'll let you know how I get on.
>
> Cheers, T i m

if it's only 1 bad diode try running it without it, on half load or less it should be ok. 2x400v diodes in series should do as replacements temporarily.


NT

[T i m]

On Sat, 29 Jul 2017 14:45:34 -0700 (PDT), tabb...@gmail.com wrote:

<snip>

>if it's only 1 bad diode try running it without it, on half load or less it should be ok.

Now I've got them all out it seems to be two bad diodes but one was
reading strange things so maybe died fully as I got it out?

>2x400v diodes in series should do as replacements temporarily.

Ok, I can give that a go tomorrow ... especially as I have so many
1N4004's. ;-)

Cheers, T i m

[Robert Roland]

On Sat, 29 Jul 2017 11:49:28 -0700 (PDT), tabb...@gmail.com wrote:

>if you test individually: 0.6v one way only

Unless the output is shorted.
--
RoRo

[Robert Roland]

On Sat, 29 Jul 2017 21:22:53 +0100, T i m <ne...@spaced.me.uk> wrote:

>Would you think one of those component testers would be able to
>identify the PWM reg, at least with a go/no-go result?

I do not know which component tester you are talking about, but I
would be very impressed if it could.
--
RoRo

[T i m]

On Sun, 30 Jul 2017 12:21:53 +0200, Robert Roland <fa...@ddress.no>
wrote:

Well, whilst I think they are 'quite clever' I agree that it probably
couldn't test the TOP223Y's.

http://www.eetimes.com/author.asp?section_id=30&doc_id=1329331

I bought one (ironically) for measuring the ESR on SMPS caps (Topfield
STB) but have used it (mostly successfully) for many other things
since. ;-)

On that ... do you think it might be worth testing (or even just
replacing) the two main caps in case they are getting weak? I note
they have an extra heat shrink 'skin', probably because they are
pressed up against the tiny heatsink on the TOP223Y's?

Cheers, T i m

[Robert Roland]

On Sun, 30 Jul 2017 11:54:52 +0100, T i m <ne...@spaced.me.uk> wrote:

>Well, whilst I think they are 'quite clever' I agree that it probably
>couldn't test the TOP223Y's.
>
>http://www.eetimes.com/author.asp?section_id=30&doc_id=1329331

I have one similar. Yes, they are impressively clever for the money.
By they do not do ICs, only generic components. The only component
that I found will fool it, was a germanium power transistor. The low
Vbe, combined with the low DC gain probably threw it off.

>On that ... do you think it might be worth testing (or even just
>replacing) the two main caps in case they are getting weak?

Hard to say. There are so many factors that affect the aging of caps.
It does not take many seconds to take them out and test them, though.

>I note
>they have an extra heat shrink 'skin', probably because they are
>pressed up against the tiny heatsink on the TOP223Y's?

Yes, I noticed that, too. Hopefully, the TOP does not get all that
hot.
--
RoRo

[T i m]

On Sun, 30 Jul 2017 17:18:11 +0200, Robert Roland <fa...@ddress.no>
wrote:

>On Sun, 30 Jul 2017 11:54:52 +0100, T i m <ne...@spaced.me.uk> wrote:
>
>>Well, whilst I think they are 'quite clever' I agree that it probably
>>couldn't test the TOP223Y's.
>>
>>http://www.eetimes.com/author.asp?section_id=30&doc_id=1329331
>
>I have one similar. Yes, they are impressively clever for the money.

Aren't they. ;-)

>By they do not do ICs, only generic components.

My only thought was that whist the TOP is actually a 'clever' 3 pin
device, there may be some basic logic that it could respond to that
might make the tester think it was a working FET or Triac or some
such? eg. You might be able to use it as a go / no-go test, *if* it
detected something positive (however misdirected) in comparison to a
good one?

>The only component
>that I found will fool it, was a germanium power transistor. The low
>Vbe, combined with the low DC gain probably threw it off.

Ok ... and that's not even a 'clever' component at such. eh. ;-)

>
>>On that ... do you think it might be worth testing (or even just
>>replacing) the two main caps in case they are getting weak?
>
>Hard to say. There are so many factors that affect the aging of caps.
>It does not take many seconds to take them out and test them, though.

True, as whatever 'glue' they used to hold stuff down seems to have
gone off a bit in any case. Maybe I'll replace the diodes first and if
the unit then works, treat it to a fair of quality (low ESR?) caps at
that point.

>
>>I note
>>they have an extra heat shrink 'skin', probably because they are
>>pressed up against the tiny heatsink on the TOP223Y's?
>
>Yes, I noticed that, too. Hopefully, the TOP does not get all that
>hot.

I doubt it would (for that heatsink) as it's got next to no radiative
qualities (being partially masked by the cap) and only a small thermal
mass. It might just be there to rest the cap on. ;-)

Cheers, T i m