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 16th Jul 2022, 2:08 pm   #1
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default HP 5300B Overcurrent

This might look familiar to some of you. I'm starting a new thread on the same problem to make things clearer now I've done further testing.

Problem: zener diode CR10 (0.5W) in SMPS is heavily overloaded and burns out on unit switch-on. An AVO measuring current between winding point 1 on T2 and cathode of CR10 was almost immediately up to 5A before I switched off.

With board A2 removed, the +22V rail is OK when mains is applied, so it fails somewhere in the switcher.

TR2 has infinite resistance between windings, measured with a Megger. I have it connected with flying leads for ease of measurement.

Where's this current flowing?

I've combined the circuit diagram pages from the only online PDF available and it's legible. The PSU appears consistent across the revisions. PSU operation explanation is culled from the manual as well.
Attached Thumbnails
Click image for larger version

Name:	PSU Block Diagram.jpg
Views:	137
Size:	55.5 KB
ID:	261043   Click image for larger version

Name:	PSU Theory 1.jpg
Views:	96
Size:	79.7 KB
ID:	261044   Click image for larger version

Name:	PSU Theory 2.PNG
Views:	81
Size:	55.3 KB
ID:	261045   Click image for larger version

Name:	PSU Theory 3.jpg
Views:	60
Size:	67.6 KB
ID:	261046  
Attached Files
File Type: pdf Circuit Diagram HP3500B optimised.pdf (2.65 MB, 122 views)
Uncle Bulgaria is offline  
Old 16th Jul 2022, 2:11 pm   #2
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

Main Board.
Attached Files
File Type: pdf HP 5300B Main Board.pdf (3.04 MB, 124 views)
Uncle Bulgaria is offline  
Old 16th Jul 2022, 3:01 pm   #3
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Dorridge, West Midlands, UK.
Posts: 1,475
Default Re: HP 5300B Overcurrent

You might find this useful someone else with a similar problem. https://www.eevblog.com/forum/repair...-supply-whine/ and here https://www.barbouri.com/2017/12/03/...suring-system/
Unfortunately without a solution!
__________________
Chris Wood
BVWS Member

Last edited by chriswood1900; 16th Jul 2022 at 3:08 pm. Reason: improve clarity
chriswood1900 is offline  
Old 16th Jul 2022, 5:50 pm   #4
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 767
Default Re: HP 5300B Overcurrent

Looking at the manual extracts, the voltage across CR10 should be 50V, so it should not be burning out as it is a 68V Zener.

The current you are measuring is presumably flowing into the collector of A1Q17. The start up circuit is working, otherwise there would be no current into A1Q17. According to the manual, A1Q17 is turned off by the circuit on A2.

The collector current drawn by A1Q17 rises linearly until it is turned off, the manual shows the current as rising up to 2-5A, similar to what you are measuring. The voltage which will be generated when A1Q17 shuts off will be related to the rate of change of current. The higher the rate of change, the larger the voltage. It’s probable that if the control circuit is shutting the transistor off later than it should the voltage across the Zener diode, A1CR10 will be high enough to damage it.

There are some waveforms given in the manual which should be checked. I would check Q1C29 as that is responsible for feeding the turn off waveform to A1Q17. Low value electrolytic capacitors don’t have the best reputation for reliability.

Paula
frsimen is offline  
Old 18th Jul 2022, 12:23 am   #5
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

Thanks. Chris - alas the problems aren't similar. The EEVBlog chap's wondering whether his PSU whine is excessive, but it seems to be working. The other fellow's is working as is after fixing some mechanical damage.

I did find one online example with an overcurrent fault which turned out to be shorts in T2, which was why I took it out. The author managed to fix it by squeezing T2 in a vice, having noticed the leak resistance increased to infinity when doing so. Mine doesn't seem to have this leakage problem.

frsimen - thank you for the analysis. C29 is a tantalum electrolytic, and measures <3µA leakage at 35V so I think it's OK, worse luck. My Chinesium component tester suggests an ESR of 0.54R. I'll try substitution with a new electrolytic in case.
Uncle Bulgaria is offline  
Old 25th Jul 2022, 12:12 pm   #6
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

I'm awaiting some eyelets to repair some through-holes on the board, so haven't been able to test the board yet with C29 substituted.

In the meantime I've been checking PSU board A2 against the circuit diagram to be sure the connections and components are all correct.
Uncle Bulgaria is offline  
Old 25th Jul 2022, 1:00 pm   #7
dave cox
Nonode
 
dave cox's Avatar
 
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,059
Default Re: HP 5300B Overcurrent

To me, it looks like this circuit expects the collapsing field in T2 (when Q17 turns off) to be dumped into the secondary. if this DOESN'T happen, I can see CR10 getting whacked pretty hard as the voltage on the primary of T2 will rise very quickly until CR10 conducts, and with only 0.5W of capability ...

Something not loading the secondary of T2 enough ??

dc
dave cox is offline  
Old 4th Aug 2022, 3:22 pm   #8
Alistair D
Nonode
 
Alistair D's Avatar
 
Join Date: Jan 2010
Location: Edinburgh, UK.
Posts: 2,008
Default Re: HP 5300B Overcurrent

Quote:
Originally Posted by Uncle Bulgaria View Post
I'm awaiting some eyelets to repair some through-holes on the board, so haven't been able to test the board yet with C29 substituted.

In the meantime I've been checking PSU board A2 against the circuit diagram to be sure the connections and components are all correct.
Can I ask where you get your eyelets from?

Al
__________________
I won't tell you how I discovered that.
Alistair D is offline  
Old 4th Aug 2022, 9:27 pm   #9
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

The eyelets have come, and I've got a few odd-value resistors to replace some that close inspection has shown are cracked. I'll solder them in and report.

Alistair D: I got a little box on eBay, where there are lots. I can't recommend it as despite my best due diligence it turned out they came from China. I have managed to repair the relevant holes and T1 can now be reconnected.
Uncle Bulgaria is offline  
Old 4th Aug 2022, 11:09 pm   #10
Alistair D
Nonode
 
Alistair D's Avatar
 
Join Date: Jan 2010
Location: Edinburgh, UK.
Posts: 2,008
Default Re: HP 5300B Overcurrent

Thanks for that. I can think of a number of PCB repairs where these would have made the repair much stronger. In particular I am thinking of those coaxial power connecters that seem to cause all sorts of track damage.

Sorry for taking this thread off topic. Lets get back on track. Pun intended

Al
__________________
I won't tell you how I discovered that.
Alistair D is offline  
Old 7th Aug 2022, 11:36 pm   #11
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

Track or trace, Al?

I think I saw the technique mentioned on here, and as the access on top of this board is sometimes poor, components are soldered from underneath. On some pads, the trial of testing and repair (and my no doubt poor technique) has lifted some of the small lower pads that are through-holes to the track above. Some solder on the track, an eyelet pushed through and flared (I used some centre punches) and then soldered to the track seems to do the trick.

It turns out the 5k9 resistor is on back-order, so unless I can make up a temporary one I'll have to wait until it arrives to continue testing.
Uncle Bulgaria is offline  
Old 7th Oct 2022, 9:44 pm   #12
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

New resistors are in place. I'm following the troubleshooting flowchart again (attached), with +6V attached across A1C6.

The waveform at the collector of A1Q17 is shown in the photograph with a x10 probe, 1V/div. and 0.1ms/div. It's supposed to be 10V pk-pk, but I read that as 20V, with a pronounced overshoot at the start.

+5V rail is OK.
-5V is 0.7V
+17V is 18.1V
-17V is -18.3V
+3.5V is 3.9V
Attached Thumbnails
Click image for larger version

Name:	IMG_20221007_213333.jpg
Views:	69
Size:	59.3 KB
ID:	265788  
Attached Files
File Type: pdf HP3500B Maintenance Flowchart.pdf (301.4 KB, 39 views)
Uncle Bulgaria is offline  
Old 8th Oct 2022, 11:04 am   #13
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 767
Default Re: HP 5300B Overcurrent

The photo does seem to be showing 20V peak to peak with a large overshoot, but the output voltages are, with one exception, close to what they should be. I would check your 'scope calibration against a known voltage to be sure that the reading is genuine.

Have you checked A1C29, as that is on the list of parts to check if the waveform isn't 10V peak to peak?

The -5V rail clearly has a fault, so it would be a good idea to sort that fault out before going too much further. The fault chart suggests A1CR4 but a low resistance on the -5V line would also cause low output voltage.

Paula
frsimen is offline  
Old 13th Oct 2022, 11:11 pm   #14
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

That's a good idea. The 'scope does slightly overread on its own calibration signal, but not to double its value. Monitoring the calibration signal with a multimeter suggests it's OK for voltage (accounting for square wave duty cycle).

A1C29 is replaced with two new, paralleled electrolytics.

A1CR4 measures OK out of circuit. In circuit it has the 0.7V of its output, the -5V rail, at the anode, and 0.2V at the cathode.

A1Q16 and Q17 read OK on a component tester.

A1CR9 and A1C23 are new.

A1CR10 blows when mains is applied.

Checking the -5V line, it only goes to pin 2 of the main multipin connector that links the measuring unit to its accessory and battery packs, or pin 8 of the A2 connector.

With the A2 board removed, measuring resistance between the -5V rail and ground shows 400k ohms dropping to about 4k before climbing again. Could it be C22 or C25 passing DC to ground?
Uncle Bulgaria is offline  
Old 14th Oct 2022, 6:13 pm   #15
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 767
Default Re: HP 5300B Overcurrent

Ideally, you should check the operation of your oscilloscope/probe combination on a known voltage on the range that you are using to measure the 20V p-p. The input attenuator range is likely to be different for the calibrator measurement to the one used for measuring the 20V p-p signal.

A 9V battery, for example will allow you to confirm that the oscilloscope is telling the truth.

Are you sure that you are applying 6V to A1C6 and that it still measures 6V when you are testing? If your 'scope is telling the truth, it suggests that the supply voltage could be higher than 6V.

The overshoot is certainly the cause of the demise of A1CR10 when driven from the mains supply.

I'm not sure quite what you mean about the voltages around A1CR4. I think you are saying that the output at the cathode is 0.2V and you're measuring 0.7V on the anode side. Clearly, that isn't correct. The behaviour of the capacitors is a little odd, but you're not measuring a short circuit, changing them won't do any harm but I don't think they are the cause of the fault.

Board A2 has a Zener diode, CR6 across the -5V supply. I suspect that is short circuit and is the reason for the -5V being so low.

This supply uses the +5V rail which is compared with a reference on board A2. Your measurement has shown that the +5V rail is correct, when driven from 6V. With a short on the -5V line, the supply is going to be working harder than it should to maintain the other supplies. That could well be the cause of the overshoot and the damage to A1CR10.

Paula
frsimen is offline  
Old 14th Oct 2022, 7:08 pm   #16
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

Thank you for your considerations. I need to be clearer in writing up what I'm doing!

I should have said I measured the output from my bench PSU with the 'scope, RS Fluke 77 and an AVO 8. The PSU is a Thurlby Thandar PL310 with digital display which agrees with the measured voltage on the multimeters. The 'scope reads within the error of the thickness of the beam when measuring 20VDC (four divisions at x10 and 0.5V/div).

I have to use the K1 module for powering the 5300B as the PL310 part goes into current limiting when starting up. This is peculiar as it has a limit of about 1.1A, and powering the 5300B from the K1 module (max. 3.5A at 6V), which displays current drawn when under load, shows a slow climb to a constant 530mA. The manual does say a 2A supply, so maybe it's an initial surge.

When it's running at a constant current (yes, 6V is supplied under load) I made the measurements in post #14 at the test points of the power rails with the meters, using the 'scope for the waveform at Q17 only.

I changed the A1CR4 diode. Yes, the figures I gave are peculiar and were the voltages measured to ground at those points. 0.2V at the A1CR4 cathode - T2 pin 4 junction.

The -5V rail also goes to the TCXO. Removing this shows no change. It seems more and more likely that your surmise that the fault is on the A2 board is correct.

I removed and checked C22 (electrolytic, OK at rated voltage for leakage) and C25 (orange 10V tantalum). I replaced C25 with an aluminium electrolytic. A couple of blue 6V tantalums are on the 5V rails so I changed them for tested electrolytics based on previous experience.

I have had most of the components on A2 off and checked, but will have another hard look around CR6. Thanks again, I hope to have more to report this evening...
Uncle Bulgaria is offline  
Old 14th Oct 2022, 7:31 pm   #17
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 767
Default Re: HP 5300B Overcurrent

Check that you put CR6 back in the right way around. It will try to clamp the voltage to 0.6-0.7V if you get it wrong, it may well fail if the current is excessive if you did that.

Once the -5V supply has been restored, things may start to look more sensible. Your oscilloscope is fine as is the power supply, but it was worth checking.

Paula
frsimen is offline  
Old 14th Oct 2022, 7:37 pm   #18
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

5.9V axial zeners appear to be unusual. I replaced CR6 with a 5.6V zener, and still have +0.7V on the -5V rail. It's lucky I have thick hair.

Edit: crossed post. Yes, CR6 was the right way round (original Motorola part - checks out as a diode on the component tester, but haven't checked the zener voltage). Continuity checked between it and neighbouring components.

Indeed, a very good idea to check the test equipment! Quis custodiet ipsos custodes?
Uncle Bulgaria is offline  
Old 14th Oct 2022, 8:03 pm   #19
frsimen
Heptode
 
Join Date: Jan 2011
Location: Croydon, London, UK.
Posts: 767
Default Re: HP 5300B Overcurrent

This is a strange one. Is there continuity to chassis on the feed side of CR4? Did you use a suitable diode to replace the old A1CR4? A 1N400x type isn't suitable at the frequencies involved here and will tend to behave more like a resistor than a diode.

Beyond that there isn't very much else which could cause this fault.

One last thought is to monitor the voltage on the feed to CR4 and compare it with that on the input to A1CR2. The amplitude should be similar.

Paula
frsimen is offline  
Old 15th Oct 2022, 3:23 pm   #20
Uncle Bulgaria
Nonode
 
Uncle Bulgaria's Avatar
 
Join Date: May 2016
Location: Cornwall, UK.
Posts: 2,315
Default Re: HP 5300B Overcurrent

Newsflash. I had a reply written out but have now had to delete it as there's progress of a sort. The original A1CR4 is back in (I had used a 1N5404 as there was nothing in the parts list about high frequency).

A2CR6 remains as a new 5.6V zener. With +6V applied at A1C6, I now have 4.95V on the +5V rail, and -4.52V on the -5V rail!

However, the attached is the waveform at the collector of A1Q17, at x10, 0.5V/div. Still high by my reckoning.
Attached Thumbnails
Click image for larger version

Name:	IMG_20221015_151957.jpg
Views:	51
Size:	51.4 KB
ID:	266169  
Uncle Bulgaria is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 5:41 am.


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.