UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Closed Thread
 
Thread Tools
Old 24th Mar 2021, 7:04 pm   #1
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Weir HSS100/4 PSU blowup

I'm putting this under 'vintage computers' because the supply in question comes out of a vintage computer peripheral and becasue I doubt you'd come across such supplies in other types of vintage technology. But anyway...

I have a Weir HSS100/4 power supply that has failed in a big way... The sequence of events :

I power it up, no load, with a lamp limiter. My experience is that SMPSUs are fine on lamp limiters provided they are supplying a lot less power than the rating of the lamp, so the input voltage is not seriously reduced. It comes up fine, but the +5V output is a little high.

Switch out the lamp limiter and start loading the output. The +5V rail RISES as I load it. This is something I've seen before and generally means high ESR/open capacitors on that output. So I replace the 2 appropriate ones.

+5V rail now seems fine off-load and when supplying 1A (no lamp limiter in the circuit) But the thing seems to be oscillating at an audible frequency, I can hear it.

Connect the 'real' load, 4 boards of logic/microprocessor, etc. It can't supply the current (it is the original supply from the unit), the 5V output is pulled down to 2.5V or so. No lamp limiter in the input circuit of course.

While I am thinking what to do, there is a small 'pop' and the outputs drop to zero.

Turns out the 2A fuse in series with the mains input has blown.

Check over the supply, the chopper transistor, a BUW12A, is shorted all ways round.

My first thought is that the mains smoothing capacitors have dried up, the chopper was thus running with a low input voltage and working too hard.

All rectifiers test OK, by the way.

I shall replace said capacitors and the chopper transistor, but is there anything else I should be suspicious of?
TonyDuell is online now  
Old 25th Mar 2021, 12:39 am   #2
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

Could the audible frequency oscillation be a clue that it wasn't running at the right frequency, and that led to the demise of the chopper transistor - presumably due to excess voltage / current and it didn't overheat (or internal peak power dissipation was such that it failed so quickly that it didn't have time to get hot)

It may also be worth checking any diode / capacitor / resistor snubber components (assuming transistor doesn't have built in 'efficiency' diode clamp on output).

Otherwise it's probably a good idea to check all primary-side capacitors for value / ESR (on electrolytics) and any low resistance-value e,itter current sensing resistors. And also that (Opto-coupler?) feedback is all OK (Maybe putting a temporary resistor across primary-side, to limit power / output voltage to start with)

I suppose the other suspect is unfortunately the Chopper transformer has got shorted turns / is arcing-over.

Another thought is that the audible-frequency could be due to an overload (Although may usually expect it to cycle at rather low 'ticking' frequency in that case, with repeated re-tries)

Last edited by ortek_service; 25th Mar 2021 at 12:56 am.
ortek_service is offline  
Old 25th Mar 2021, 4:54 pm   #3
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

Oh bother...(that's the family-friendly version).

Turns out there's a mosfet in series with the chopper transistor emitter circuit. It's failed open-circuit. The markings are MPT5N06, which Towers tells me is an n-channel enhancement-mode device, 5A. 60V.

Now I can't find it at RS or Farnell, I can't even find a near-equivalent (although my bitter experience suggests that SMPSUs don't like 'equivalents). Any sane suggestions as to what to try?

I think it's unlikely the transformer has failed. My experience is that kills the chopper instantly at switch-on, not after half a minute or so. The supply was running in that time, the outputs (with no load) were all present and correct.

The regulation feedback is another transformer, the chopper current sensing yet another. No low-value resistors. The only electrolytics are the mains and output smoothing capacitors, I intend to replace the lot.
TonyDuell is online now  
Old 25th Mar 2021, 5:25 pm   #4
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

5N06 etc. seems to me a very familiar style of voltage / current rating with some Power MOSFET makes. So MPT could just be a manufacturer prefix.

And searching on just that in mouser, find a lot:
https://www.mouser.co.uk/Semiconduct...d=5N06&FS=True

https://www.mouser.co.uk/Semiconduct...g?keyword=5N06

But not sure what package original ones is in - can't find much on that part in Google - to narrow down a bit.

I'm also wondering why there'd be a MOSFET in the chopper transistor's emitter circuit - normally that goes straight to the -V HT DC or sometimes via a current-sense resistor.
ortek_service is offline  
Old 25th Mar 2021, 5:43 pm   #5
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

It's a normal TO220 package. Metal tab, it's clipped to a heatsink with the appropriate insulator.

It's mt in front of me at the moment, but I think it's a Motorola part.

It appears to be part of the regulation circuitry. The chopper may well be self-oscillating, the mosfet enables/disables it.
TonyDuell is online now  
Old 25th Mar 2021, 5:57 pm   #6
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

Well this one has similar part number - but 55 & 50A rather than 5A
- Although still quite cheap: https://www.switchelectronics.co.uk/...-channel-to220

If it really is just used as a switch, then it should be too critical on equivalents - just same voltage / current /power ratings and Vgs threshold / ratings.

But switching the negative side seems a bit odd, as rest of the circuitry is then floating at high +V DC (although, if switching that will always be some parts at quite high -ve DC voltage). And I would expect the voltage rating of it may want to be a bit higher than 60V, as could be exposed to much greater ones when it's off, due to leakage currents. So maybe a 600V 5A one, if available, would be more expected.
ortek_service is offline  
Old 25th Mar 2021, 6:05 pm   #7
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

Quote:
Originally Posted by TonyDuell View Post

It's mt in front of me at the moment, but I think it's a Motorola part.

Is it MT or MP at the start of the part number?
- I just found full info / datasheet on the MTP5N06? part, on this website (once I got past some Russian? pop-ups!):

https://alltransistors.com/mosfet/tr...ansistor=48015

It didn't seem to find an equivalent, but at least gives enough data to do a detailed search in Mouser / the usual distributors etc.
ortek_service is offline  
Old 26th Mar 2021, 6:08 am   #8
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

You're right, it's MTP5N06. I mis-typed the number...

I can find plenty of modern MOSFETs in a TO220 package (to fit the heatsink) with higher voltage/current ratings, lower channel resistance, etc. They should be 'better'. But my experience of SMPSUs is that things don't always work out like that.
TonyDuell is online now  
Old 26th Mar 2021, 4:18 pm   #9
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

Might be able to find the genuine one somewhere - I'd only originally being looking for the incorrect one - although this one was from a less common manufacturer who may not be doing these anymore. And if the prefix letters are just for that manufacturer, it does seem that other prefixed 5N06 ones should be rather similar.

If it is just being used as a switch, they it shouldn't be anywhere near as critical as a chopper switching one might be. And I'm wondering if it can be just temporarily shorted out, to ensure the rest of the circuitry works / fault-find on that (it was a bit suspicious, that it sounded wrong and blew after 30s)
ortek_service is offline  
Old 26th Mar 2021, 5:13 pm   #10
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,483
Default Re: Weir HSS100/4 PSU blowup

I suppose it depends on the circumstances under which that FET gets switched off. If it switches off when the whole thing is about to go to DEFCON 5 then it might be better to have it both fitted and functional, doing the job it is supposed to do.
SiriusHardware is online now  
Old 29th Mar 2021, 6:46 pm   #11
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

I've had another look at this, and it's rather complicated.

It appears that the MOSFET is the real chopper transistor, in that it gets the drive signal. The base of the BUW12A is held at an almost constant voltage. When the fet turns on, the BUW12A base-emitter junction passes current, thus there is current flow in the collector-emitter circuit, the collector going to the chopper transformer and then to the 350V rail.

I say 'almost constant voltage' because there's a current transformer involved. One winding is in series with the BUW12A emitter. The other supplies a signal to the base via a diode. It appears that idea is that once current starts to flow it increases the base drive to make sue the BUW12A turns on properly.

Now the diode friom that current transformer to the base is a BAX12A. Turns out that's a switching diode with a specified reverse avalanche characteristic, it breaks down between 120V and 170V. I can't see how that could happen in this circut but there must be some reason for choosing an obscure diode like that. It still tests OK but I can only test it as a signal diode.

The 'constant voltage' on the base of the BUW12A comes from a BZW06-13. A strange 13V zener thing (at least it's not the bidirectional one). At first glance a normal 13V zener would work, but again why choose something odd. Anyway, it's dead short so I've got to replace that.

Finding the replacement parts is not going to be easy...
TonyDuell is online now  
Old 31st Mar 2021, 10:30 am   #12
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

And another update...

I've attached a reverse-engineered diagram of the chopper circuit, just so you can follow what I am talking about It's somewhat Weir-d (pun intended!)

Anyway, looks like the mosfet has survived. I simply wasn't giving it enough gate voltage to turn it on.

The real problem is that 13V 'zener', I suspect a normal zener would work, but that the transient supressor type as used here, was designed to fail shorted. Had it gone open-circuit I think there would have been a lot more damage, including the mosfet and several small transistors So it's probably best if I try to track down the right part or a close equivalent. RS do something similar but it's out of stock, no idea when they'll be getting more.
Attached Files
File Type: pdf chopper.pdf (331.4 KB, 56 views)
TonyDuell is online now  
Old 31st Mar 2021, 11:14 am   #13
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,483
Default Re: Weir HSS100/4 PSU blowup

In my long experience even ordinary Zeners fail short-circuit or low resistance when they fail, in fact semis in general tend to fail low-resistance or short circuit rather than open-circuit.
SiriusHardware is online now  
Old 31st Mar 2021, 9:49 pm   #14
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

Yes diodes (and most semiconductors) tend to fail short-circuit with too much current.

Farnell do have lots of the BZW06-13 diode in stock, although you do have to buy 5off (probably the same as RS) and find some other stuff to buy to avoid handling charges (also same with RS, bnless account holder?):

https://uk.farnell.com/taiwan-semico...-do/dp/2677548

- And they do list a few suggested alternative to this, which RS may actually have in stock (Although not necessarily in the UK..)

It does seem a very odd circuit, and I'm wondering if R16 is a king of start-up resistor and T1 / D11 provide the base-bias supply when it is running. It seem the more ac current there is through TR4 emitter, the more feedback drive it gets, so could be some kind of adaptive efficiency measure.
ortek_service is offline  
Old 1st Apr 2021, 2:00 am   #15
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,264
Default Re: Weir HSS100/4 PSU blowup

It might be worth checking if there is an RS trade counter nearby. It was one of the nice things about being in Birmingham as it avoided shipping costs.
Mark1960 is online now  
Old 1st Apr 2021, 4:32 am   #16
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

There is an RS trade counter that I can easily get to (I normally do collect my orders as I find it a lot more convenient than waiting around for a parcel) but I am not sure what they are doing during lockdown. But I never have a problem making my order up to over the 'small order charge limit', even if I just buy a few packets of nuts and bolts, reels of wire, etc. I am sure to need things like that in the future.

As for the circuit operation, I think that when the mosfet is off, the base of TR4 is held at 13V-ish due to R16/D9. Obviously under those conditions there is no collector or emitter current.

When the mosfet is turned on by the drive signal it pulls the emitter down. The emitter-base junction is forward-biased, the base drops below the zener voltage of D9 so the latter no longer passes any current. Current through T1 then causes more current to be fed into the base of TR4 via D11 (the current thrpugh R16 would be nowhere near enough, the hfe of TR4 is low according to Towers). This results in significant current through the collector circuit and T2 primary
TonyDuell is online now  
Old 9th Apr 2021, 3:53 pm   #17
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

ARGH!

I got the BZW06-13 and BUW12A from Farnell and fitted them. Also replaced all the electrolytic capacitors in the PSU with new ones from Farnell.

Powered up with a lamp limiter initially, then shorted out the series bulb. The outputs came up, but the thing was still oscillating at an audible frequnency. Then (lamp limiter not in circuit) I tried a very moderate load on the 5V rail, about 1A. The result was the new transistor and zener are shorted, the expensive fuse on the PSU board has blown.

I am at a loss as to what to try next. I will 'ring' the transformer in case it's shorted turns but the fact that it works at all suggests that's not the problem. Everything else I've checked seems fine.
TonyDuell is online now  
Old 9th Apr 2021, 4:04 pm   #18
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

It does look like the audible oscillation is giving a hint that it's not very happy.
And I wonder if this is due to an overload somewhere, that you might be able to detect by measuring the AC input current / power (those energy monitors may be useful, although some aren't too accurate below a few Watts).

Hopefully you bought a few spare transistors & diodes, which it does seem to repeatedly blow the same ones after a short while.
A variac may also come in handy, to save these, if putting a light-bulb in series also stopped it blowing up (although maybe also from running due to too low input voltage?)

It may be worth checking all the non 5V output-side diodes / capacitors for being s/c, in case these could be causing an overload.
ortek_service is offline  
Old 9th Apr 2021, 4:11 pm   #19
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Weir HSS100/4 PSU blowup

My experience is that most SMPSUs will start up on a lamp limiter but you should short out the bulb before applying significant load to the output. So that's what I did. The lamp was not in-circuit when I put the 1A load (actually a 4.7 Ohm high-power resistor) on the 5V output.

A Variac is probably not a good idea here, with a lower input voltage the chopper will be passing more current. (SMPSUs approximate a contant power load to the mains).

I've replaced all the electrolytics on both the mains and output sides with new ones from Farnell. I think they're Panasonic brand, they are certainly 105 degree ones. I would be very surprised if any of them were high ESR.

I do have a couple of spares of the semiconductors that fail but I don't want to keep on fitting and blowing them as (a) they are not cheap, the transistors are over 5quid at time. and (b) I am not sure how well the PCB will stand up to repeated resoldering.
TonyDuell is online now  
Old 9th Apr 2021, 6:42 pm   #20
ortek_service
Octode
 
ortek_service's Avatar
 
Join Date: May 2018
Location: Northampton, Northamptonshire, UK.
Posts: 1,394
Default Re: Weir HSS100/4 PSU blowup

I presume the lamp was lighting a bit (although won't light much / drop very much when still operating cold) on this limiter. So this would have been dropping quite a few volts, even without you putting a load on the SMPSU - especially if the SMPSU's was running overloaded due to a fault in it.

I was only thinking of using the Variac to run the SMPSU without a load, initially.
- So normally there shouldn't be too much current drawn (even at lower voltages), and it probably wouldn't start to run until >90V anyway. And I suppose you could always switch-on above this, if it got too upset at rather low input voltages.


I couldn't remember if it died after < 1min running, with no load on it. As otherwise, if it did survive like this despite the aubile oscillation, then you may abler to run it quite a bit to do various tests.

I was thinking more of an output capacitor going short / breaking down, so causing an overload. But looks like you've ruled that out, by changing them all.
I also wondered if any output rectifier diodes may have gone s/c, as HF AC into the electrolytics is going to upset it a bit.


Maybe there's a relatively-cheap similar spec transistor, that would be good enough for just running off-load, that you could afford to loose a few. I suppose you could always cut the leads above the PCB and rejoin to those to save re-heating the tracks.
With CRT TV's and Monitors, I often tried using BU508A/D's I had a few of, before putting more expensive original type in there.
ortek_service is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 7:22 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.