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 Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 4th Nov 2022, 5:10 pm   #41
Dekatron
Octode
 
Dekatron's Avatar
 
Join Date: May 2009
Location: Linkoping, Sweden
Posts: 1,463
Default Re: HP 5300B Overcurrent

Is the power supply loaded enough during the test with the mains unit connected?

Some switching power supplies don't work properly if they aren't loaded enough, some even break down. To prevent this individually resistance loading modules are attached to the output voltages while the PSUs are being repaired/tested without the whole unit plugged in.
__________________
Martin, Sweden
Dekatron is offline  
Old 4th Nov 2022, 7:15 pm   #42
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 766
Default Re: HP 5300B Overcurrent

Looking back through this whole thread, I think the answer could lie in post 30. There, you measured a voltage of 21.5V AC on the mains transformer output. That is going to give a peak voltage of around 29V which is way above the 22V+/-2V that this power supply needs. There will be a large ripple voltage so you will measure a lower voltage across A1C6, but the peak voltage is there. That could why you are having so many problems. I wonder if there is a shorted turn or two on one of the primary side windings of the mains transformer?

Try comparing the resistance of the winding between the black and black/yellow wires with that between the black/green and black/brown wires. If they are both the same, then it's another red herring. Do check that the wiring to the switch is correct and that the windings are wired in series, not parallel.

If another member has a 5300B, perhaps they could measure the AC output voltage of the mains transformer on their unit, to give some confidence as to what should be measured.

Martin's (Dekatron) comments above are perfectly valid but your earlier tests had the supply loaded with the 5302 module.

Paula
frsimen is offline  
Old 4th Nov 2022, 10:15 pm   #43
Chris55000
Nonode
 
Join Date: Aug 2007
Location: Walsall Wood, Aldridge, Walsall, UK.
Posts: 2,853
Default Re: HP 5300B Overcurrent

Paula's remarks about a possible mains transformer defect are definitely worth noting!

If you have a reasonable quality a.f. generator with a 50 Ω output, you should be able to disconnect the secondary windings from the power supply circuits, and then run the transformer in reverse using a suitable combination of secondary windings/tappings, you probably won't be able to get anywhere near the rated voltage due to loading on the generator's output, but you should be able to calculate the turns/voltage ratio between your test secondary and the two primary windings!

If the voltage across one of the primary windings is substantially different from the other, this would be suspicious of a defective transformer as mentioned by Paula!

If the two primaries provide equal voltages across their outer ends, and the transformer doesn't load the output of your test generator too much, it is probably OK, and I would get one of those metal–clad wire–wound resistors of about 220 Ω or 330 Ω @ 15 W rating and bolt one to the rear chassis, wiring it in series with the mains input live or neutral lead, whichever is more convenient!

Chris Williams
__________________
It's an enigma, that's what it is! This thing's not fixed because it doesn't want to be fixed!
Chris55000 is offline  
Old 5th Nov 2022, 6:43 pm   #44
Dekatron
Octode
 
Dekatron's Avatar
 
Join Date: May 2009
Location: Linkoping, Sweden
Posts: 1,463
Default Re: HP 5300B Overcurrent

These dual primary voltage transformer sometimes gets damaged if they are accidentally set to 115V and plugged into 230V. I've had several instruments from different manufacturers that have suffered from this type of overload even though the fuse blew (I guess some people just tried with a new fuse one or more times before giving up or realizing that the voltage selector was in the wrong position).
__________________
Martin, Sweden
Dekatron is offline  
Old 5th Nov 2022, 7:23 pm   #45
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: HP 5300B Overcurrent

My random thoughts, having come in late...

It's an HP instrument. HP will have designed it to cope with 240V mains when set to 'primaries in series'. So if it can't handle that, there is a fault, and any modifications to the mains input circuit to reduce the voltage on the input of the mains transformer will just be bodging. The real fault is still there and will bite you.

With no jumpers to connect the output of the mains supply to the switching regulator circuit, what do you measure across A1C5? And what does it look like on a 'scope?

I wonder if A1C5 has dried up and has high ESR or low value now. Ripple on the unregulated input could well confuse the switching regulator.
TonyDuell is offline  
Old 5th Nov 2022, 10:37 pm   #46
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 766
Default Re: HP 5300B Overcurrent

A1C5 is a 4700pF capacitor with a 1 megohm resistor in parallel, so the waveform measured across it will be the raw full wave rectified output from the bridge rectifier. The reservoir capacitor is A1C6 which is only in circuit when a module is installed, or pins 25 and 50 on A1J1 are linked.

Paula
frsimen is offline  
Old 6th Nov 2022, 6:18 am   #47
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,188
Default Re: HP 5300B Overcurrent

Oops.. Yes, you're right, I misread A1C5 as 4700uF.

It might still be worth looking at the waveform across it to make sure it really is full-wave rectified, and there there isn't a problem with the bridge.

I would check A1C6 if you've not already done so.

Another idea would be to disconnect the mains transformer from the input of the bridge and feed in AC from another transformer. If the zener still burns out when doing that it's unliely the problem is in the HP mains transformer.
TonyDuell is offline  
Old 6th Nov 2022, 9:02 pm   #48
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP 5300B Overcurrent

Quote:
Originally Posted by frsimen View Post
Looking back through this whole thread, I think the answer could lie in post 30. There, you measured a voltage of 21.5V AC on the mains transformer output. That is going to give a peak voltage of around 29V which is way above the 22V+/-2V that this power supply needs. There will be a large ripple voltage so you will measure a lower voltage across A1C6, but the peak voltage is there. That could why you are having so many problems. I wonder if there is a shorted turn or two on one of the primary side windings of the mains transformer?

Try comparing the resistance of the winding between the black and black/yellow wires with that between the black/green and black/brown wires. If they are both the same, then it's another red herring. Do check that the wiring to the switch is correct and that the windings are wired in series, not parallel.

If another member has a 5300B, perhaps they could measure the AC output voltage of the mains transformer on their unit, to give some confidence as to what should be measured.

Martin's (Dekatron) comments above are perfectly valid but your earlier tests had the supply loaded with the 5302 module.

Paula
Here are some measurements of the 5300B I imported a month or two back, forgot to check AC at the green wires.

Module removed, rectifier output with scope AC coupled, input power 0.7W.
Click image for larger version

Name:	DSC_9365a.jpg
Views:	48
Size:	104.8 KB
ID:	267548 Click image for larger version

Name:	DSC_9366a.jpg
Views:	58
Size:	71.0 KB
ID:	267549

Forgot to check with scope DC coupled, but would expect a higher DC voltage with no load.

Module removed, power switch bypassed, measured across C6 (950uF 40V) with scope AC coupled, input power 5.1W.
Click image for larger version

Name:	DSC_9368a.jpg
Views:	44
Size:	142.7 KB
ID:	267550 Click image for larger version

Name:	DSC_9369a.jpg
Views:	55
Size:	61.9 KB
ID:	267551

Same as above with scope DC coupled.
Click image for larger version

Name:	DSC_9370a.jpg
Views:	50
Size:	56.9 KB
ID:	267552

David
factory is offline  
Old 6th Nov 2022, 9:05 pm   #49
factory
Octode
 
Join Date: Jul 2009
Location: Cheshire, UK.
Posts: 1,706
Default Re: HP 5300B Overcurrent

Module fitted (5307A), switched on, measured across C6 (950uF 40V) with scope AC coupled, input power 10.9W.
Click image for larger version

Name:	DSC_9377a.jpg
Views:	47
Size:	91.1 KB
ID:	267553 Click image for larger version

Name:	DSC_9379a.jpg
Views:	50
Size:	73.3 KB
ID:	267554

Same as above with scope DC coupled
Click image for larger version

Name:	DSC_9378a.jpg
Views:	44
Size:	96.2 KB
ID:	267555

Mains voltage was 243V.

David
factory is offline  
Old 7th Nov 2022, 7:02 pm   #50
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 766
Default Re: HP 5300B Overcurrent

Thank you David (Factory) for a useful set of measurements. The transformer output voltage would have iced the cake but what you have provided is sufficient, I think, to confirm that the mains transformer is outputting the correct voltage.

The measurement across A1C5 isn't very useful as the tolerance on the capacitor is very wide at +100/-20%, so I've not made use of that result. The ripple voltage there can't be used with any confidence.

Using LTSpice, I was able to reproduce the basic waveforms across A1C6 that you posted using a mains transformer output voltage of 21.2V. That's very close to what has been measured on the faulty unit. The rectified output is higher than was predicted in the manual as the mains transformer is intended for 220V operation.

Based on the ripple voltage measurements, the total current consumption from the rectified DC supply can be calculated, the accuracy rather depending on the actual value of A1C6.

For the case with no module installed, it is an average of 200mA, with an uncertainty of +/-10-15% (due to capacitor tolerance and test gear calibration).

With the 5307 module, the consumption increases to 375mA, again with +/-10-15% uncertainty.

Looking more closely at the no module case, the DC voltage was 26.5V, giving a power consumption of about 5.4W.

I have made an assumption that the efficiency of the switch mode power supply isn't vastly different at 22V to when it is running from 26.5V. That being the case, the current that will be drawn at 22V will be about 245mA+/-15% which is between 210mA and 280mA.

For the faulty unit, it would be useful to measure the current drawn, with no module installed, when running on 22V.

Paula
frsimen is offline  
Old 7th Nov 2022, 11:11 pm   #51
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP 5300B Overcurrent

Has the switching transistor Q17 been changed at any time?
Maybe the replacement CR10 have a lower capacitance or are more sensitive to over voltage?
Is it worth checking the -5V supply without any load?

Last edited by PJL; 7th Nov 2022 at 11:20 pm.
PJL is offline  
Old 8th Nov 2022, 2:10 pm   #52
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP 5300B Overcurrent

Just my thoughts... the PSU does seem to be working at lower supply voltages and it is possible that the design relies on the behaviour of the original parts. I will leave it to others to decide if this could be the answer.
PJL is offline  
Old 8th Nov 2022, 8:12 pm   #53
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

All previous tests, except at the last when I thought the supply might be working, were done without any extra unit connected, and without the multi-pin connector's contacts bridged, so voltage was only being applied to C6.

Historically, mains applied on the variac reached about 160V before CR10 blew.

I have the supply on now with 22V applied across C6, board A2 in place. Voltages on the rails are as before, with slightly low -5V. Measured current draw is 208mA.

I can measure the mains transformer if you think it worthwhile, but it'll be a fiddle as the wires are only accessible on top of the voltage selector (which incidentally is on 230V), and the pins are hidden by the fuse holder!

Q17 has not been changed in my ownership. CR10 is specified as a 68.1V 500mW zener, so I don't think my using an actually obtainable 500mW 68V one is going to have an effect. If it is, then HP are greatly annoying.

Thank you for all the considered remarks. I've had this on the side for a while as I've been fed up with it since my last post.
Uncle Bulgaria is offline  
Old 8th Nov 2022, 8:57 pm   #54
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 766
Default Re: HP 5300B Overcurrent

I can understand you being fed up with this one!

The measured current is towards the bottom end of what I was expecting, but that's not a bad thing.

There isn't anything obviously wrong with what you have measured and you have checked just about everything that can be checked. The unit that David (Factory) measured had more like 26V on the supply rail. You could try increasing the supply voltage (slowly) to see what happens to the current. I would expect it to drop as the voltage rises, then the current will increase once the Zener diode starts to conduct. If the current just rises, with increased voltage that would indicate a problem somewhere. The big question is what could cause that.

Paula
frsimen is offline  
Old 8th Nov 2022, 10:42 pm   #55
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP 5300B Overcurrent

Post #23 shows the waveform of Q17 collector. I have not looked at the circuit closely but assume it is a flyback converter? It shows an overshoot to 60V, presumably limited by the zener, followed by a flat section of around 26V as the stored energy in the transformer feeds the various output reservoir caps.

The 26V flat level includes the the supply voltage and a voltage determined by the transformer turns ratio. The 26V was when the power supply was 11V so the 26V should increase to roughly 40V with 22V supply voltage still well below the Zener voltage. If you check this theory with your scope it would imply that it must be the initial overshoot that is damaging the Zener and that must be timing related.

It would be useful to see a Q17 collector trace of David (factory)'s unit. Does it have the overshoot? Does it run at the same frequency?

Last edited by PJL; 8th Nov 2022 at 10:55 pm.
PJL is offline  
Old 9th Nov 2022, 1:31 pm   #56
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP 5300B Overcurrent

If the transformer behaved as a perfect inductor the current into the Zener at flyback would be the peak current during the Q17 on phase. Zener datasheets don't mention a maximum current so it is possible HP chose a specific type tolerant to high peak currents. One of my first jobs was to test just how quickly a Zener fused s/c when hit with a lightning strike...

The -5V is not regulated so the voltage will be determined by the +5V, the turns ratio between the 5V and -5V winding, the matching of the rectifiers, and the difference in current between the +5V and -5V.

Last edited by PJL; 9th Nov 2022 at 1:45 pm.
PJL is offline  
Old 9th Nov 2022, 8:52 pm   #57
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP 5300B Overcurrent

I suggest you put your scope across CR10 to monitor the Q17 drive. Start at a low power supply voltage and gradually increase it but I suggest there is no point in killing another zener. As the supply voltage increases I would expect the on-time (at zero) and off-time to decrease in width and the off-pulse to increase in height but only by the same amount as the increase in supply voltage. I can see no circumstance (fault or otherwise) where the voltage exceeds 68V other than the initial flyback peak.

If this is seen in practice then it is the flyback that is killing CR10. As well as the possibility that the zener you are using is not up to the job, it could be because the switch off of Q17 is too fast, or the rectifier diodes on the secondaries are too slow. If these parts are original then I would try a higher power Zener.
PJL is offline  
Old 11th Nov 2022, 8:46 pm   #58
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

At 22V on C6, the waveform across CR10 looks as the attached. The flyback peak is around 60V, but not approaching 68V (x10, 1V/div & 10µs/div). It stays constant ramping up to 22V across C6 (it doesn't get absolutely larger in amplitude as the voltage increases).

With mains on the variac, the waveform was very dirty and wouldn't trigger properly so I don't have a 'scope picture. The power supply was chirping as I ran up the supply, and I guessed the picture would clean up once it had enough drive across the supply starter. However, the zener burned out before 9V was reached across C6, and it wasn't possible to determine a waveform peak on the 'scope before this happened.

When the unit came to me, there was a much higher dissipation zener in place of CR10 as part of mole42uk's previous repair efforts, but that hadn't fixed the underlying problem. From memory that was a +5V rail issue, and I think that might have been sorted when I discovered two misplaced components on the A2 board.
Attached Thumbnails
Click image for larger version

Name:	IMG_20221111_192926.jpg
Views:	45
Size:	68.8 KB
ID:	267894  
Uncle Bulgaria is offline  
Old 11th Nov 2022, 9:25 pm   #59
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP 5300B Overcurrent

The scope is triggering on both the rising and falling edges so you are seeing two overlaid traces. I would expect the discharge slope to rise up the screen with an increase in the supply voltage as the transformer primary is connected to the rail.

Have you ever had the Zener fail when running on your PSU? If it has failed what voltage did it fail at?

I have PM'd David (factory) to see if he would mind taking a scope image of Q17 collector.

Last edited by PJL; 11th Nov 2022 at 9:35 pm.
PJL is offline  
Old 11th Nov 2022, 9:52 pm   #60
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 5,997
Default Re: HP 5300B Overcurrent

The service manual has an expected Q17 waveform and shows the discharge slope at around 50V and yours is only 35V.

Can you try again to get the scope to trigger correctly so we have something to compare with David's image.

It looks like the on-time is roughly 15uS and discharge time 18uS, total 33uS=30KHz

What is the bandwidth of your scope?

Last edited by PJL; 11th Nov 2022 at 9:59 pm.
PJL is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 2:35 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.