HP 6274B Power Supply Repair

[Luciano2572]

Hello everyone! Long time lurker here

I had this one sitting here for awhile, figured I'd give it a shot.
I bought this power supply a while ago, for about 100USD + 50USD shipping from a guy here in Brazil, in the pictures he had in the ad it wasn't going up to 60v so I knew something was up.
Upon receipt, I tested it and yep, it wouldn't provide even a hundred mA and would slowly crawl to 40V.

Sooo I decided to give it a go last week, here are my findings:

Someone had a go on it before by the flux residues on the PCB;
Using the service manual troubleshooting guide, table 5-2 on page 5-9 here: https://www.avionteq.com/Document/HP-Agilent-6267B-Operating-and-Service-Manual.pdf
I already found a couple problems:

Step 4. Was getting a sinewave on TP 44, right after the diodes CR13 and CR14. Replace those four 1N5059 with four FR207 diodes (seemed to match the avg & peak current ratings) and checked the C12 capacitor, which was shot (measuring 2uF on the fluke 179 - was supposed to be 325uF). I bodged in a 330uF 16V and TP44 was normal again.
From here, the power supply could reach it's max output voltage ( ) and I loaded it up using three Agilent 66332B, which take a maximum of 20V 5A each. It worked if I slowly increased the current from 5A to 12A, if I tried to go 0-12A in one step it would shut off the output and not come back until power cycled. If I did it in 1A steps I got it to 12A.

Step 5. TP45 , should read 4.8V +/- 10%, was near 6V. The only possible causes left was VR6 and VR7, which had flux residue. Turns out they were supposed to be a 2,37V diode from HP. I couldn't identify the code on the zener installed so I measured it to around 2.8V. MFG Part number: SZ 10939-2 / HP Part Number 1902-3002. I found one on ebay for 7USD + 32USD shipping (ouch), however seller only has one. Sent him a message asking if he has more.

Seeing that bad capacitor motivated me on replacing the eletrolytics on the main PCB, and since I will have to order stuff on digikey (40$ shipping), I will try to get a list of everything I need.
I saw there's a bunch of 2.4V zeners, not sure if they would work replacing a 2.37V...

I am taking note of the capacitors on the board and cross-checking with service manual to set up a Digikey order, will do the same for components which had flux residue just in case. Will also get some exact replacement 1N5059 diodes
From my readings, Nichicon VX are a good call, Vishay 118AHT excelent but costy. I'm lost on the 2.37V diode - messaged the ebay seller asking if he has two, but that 32$ shipping to Brazil...

High-res album: https://imgur.com/a/zevp7UM

Any help is appreciated!
Yours,
Luciano

[bd139]

Nice teardown shots. Thanks for posting.

I would stuff a 2.4V 500mW zener in there and see what happens. Those are dirt cheap and plentiful.

It's not going to be the master ref zener as there's a 6.19V one in there which is better suited. SOP in HP kit is 6.2Vish references (at least the units I've seen).

Any old 105oC crap will do for the capacitors to be honest. Literally anything is better than the caps that ship in those. I think I used Vishay units on the last HP unit I had to rebuild.

[capt bullshot]

Regarding the 2.4V zener diodes:
The manual describes them as used for bias voltages, so I'd guess these are not high precision diodes nor have any special properties. I'd replace them by some standard 2V4 zeners, available for some 10cents at Digikey et al.
Alas, the manual doesn't have schematics and I wasn't able to find one by a quick googleing, a schematic would allow some analysis and hints to find your fault. Might be some faulty component within the control loop, if it isn't fixed by replacing the zeners.

[Luciano2572]

Thanks guys! I plan on giving it a more careful clean in the next couple days, might bring my camera to take some more pictures
Checked the 6.2V zeners on this unit as @bd139 mentioned, and they seem untouched, so that's good news as those are on the expensive side

I spent the last hour selecting components on digikey, done capacitors (mostly vishay, one or two Nippon), some transistors (focused on the ones w/ flux residue and in the CURRENT segment of the PCB) aaand 1N5059 diodes.


Now I will check if any friend wants to order something from digikey to help on shipping 


I only found a board layout too, no schematics either.
But on the bright side, the PCB has markings separating each section (Crowbar / mixer & error amp / reference / voltage / current / meter circuit / turn-on control / pre reg control)

[Chris56000]

Hi!

Try this one!

https://www.dropbox.com/s/jdx7njua5tjcnyf/6256_6263_6264_6265_6266_6267_6271_6274b_dc_power_supply_sm.pdf?dl=0

(It seems like bloody keysight are carefully deleting all the component layouts and schematics from all their old manuals before they upload the things!!)

I'm not sure, of course, wether the unit you have is a really early model before the serial no. range quoted on the front cover of this one, but hopefully the differences aren't too major if it is!

Note – one of the parts of the schematic has been put right at the bottom of the file behind the back cover so make sure you print it all! Hope this helps!

Chris Williams

PS!

Quite a few of Agilent's power supplies use this 2.37V device and it's quoted as being "2.37V 5% DO–7" so I suspect HP specified it before low–voltage zeners became much more common, and I can't see why a standard 2.4V one shouldn't do either!

It might be worth buying 10 or 20 from eBay or Ali Express, Banggood etc., and selecting a few for about 2.35–2.37V – you can connect the diode in series with a 2k2 resistor to a PP3/1604 battery to help you select them!

[coromonadalix]

Thks Chriss56000    thats a good service manual

[Luciano2572]

Hi!

Try this one!

https://www.dropbox.com/s/jdx7njua5tjcnyf/6256_6263_6264_6265_6266_6267_6271_6274b_dc_power_supply_sm.pdf?dl=0

(It seems like bloody keysight are carefully deleting all the component layouts and schematics from all their old manuals before they upload the things!!)

I'm not sure, of course, wether the unit you have is a really early model before the serial no. range quoted on the front cover of this one, but hopefully the differences aren't too major if it is!

Note – one of the parts of the schematic has been put right at the bottom of the file behind the back cover so make sure you print it all! Hope this helps!

Chris Williams

PS!

Quite a few of Agilent's power supplies use this 2.37V device and it's quoted as being "2.37V 5% DO–7" so I suspect HP specified it before low–voltage zeners became much more common, and I can't see why a standard 2.4V one shouldn't do either!

It might be worth buying 10 or 20 from eBay or Ali Express, Banggood etc., and selecting a few for about 2.35–2.37V – you can connect the diode in series with a 2k2 resistor to a PP3/1604 battery to help you select them!

Thanks Chris! Thats a nice service manual! Will be sure to print and study it a bit.


I have placed an order on digikey, ordered around 10 2.4V zeners, if I cant get some around 2.35-2.37V from the lot I will go on a local electronics store and buy some, those chinese websites take almost 2 months to deliver.
DHL says it will arrive on tuesday, if customs doesnt delay it a lot. So expect news soon 

Yesterday I cleaned it up a bit, measured the two 22000uF caps, got 25800uF and 25500uF using a resistor + power supply + osciloscope to measure the time constant. Took some more pictures using my digital camera, uploaded them on google drive because they are around 6mb each:
https://drive.google.com/drive/folders/11zuOL5IPSd2c4-krWp3E1L4b8UZJPm1f?usp=sharing
(in these pictures: C10 is removed, C12 is bodged, CR10,12,13 14 replaced with FR203 diodes, VR6 and VR7 removed)
There's quite a bit of rust on some screws
Oh, fan is not present on the pictures because I removed it long ago, sounds like a jet engine taking off. Of course will put it back once it's fixed or replace for a quieter fan and check thermals with a thermal camera... not sure yet

edit: forgot to mention, my serial number is 1029A01276 ; theres some components that have a datecode of 76 & 77


Luciano

[Luciano2572]

Not so great news:

1) Digikey order got stuck in customs due to order missing my CPF (like tax id) number, it was scheduled to return but I asked to cancel that and provided my cpf number... should be here this week hopefully

2) Will have to put this project off for a couple weeks due to having a speaker blow up here at work , I will start a thread for that separately.

[bd139]

Ah that sucks

Hope you find some luck and time

[Luciano2572]

KRK speaker fixed! Back to this power supply

My order from digikey is still held in customs (took 2 days from US to Brazil, but it's been sitting since november 5 on customs here  )
So some minor updates: I left some very rusty screws and locknuts (mainly from the on/off breaker) in vinegar for a week and heres the end result: (attached picture)

And I've got some magic sponge on the way to give a good rub on the chassis

Can't wait for the parts to arrive   

[Luciano2572]

Happy new years everyone!!

Good news,

It's been awhile, Digikey order arrived, replaced all the eletrolytics, all the 1N5059 diodes by new 1N5059 since some were blown and the VR6-VR7 zeners by Vishay 2.4V Zeners.

It was working! I had to tweak R30 a little bit otherwise it would act strange if I switched on a large load (it would "die" and stay on 0V/0A). After a little bit of tweaking... here's the results, loading it at ~20V with 3x HP 66332A power supplies

Also modded in a 12V Noctua Industrial 120mm fan, used a punny 12V transformer with a full bridge rectifier and capacitor, since the original fan ran on 110V. I used the same wiring to power the 12v transformer, then made sure things stayed in place with some 2 part epoxy.


In the next post I'll include PCB pictures due to forum limitation

[Luciano2572]

PCB pictures attached

High res pictures: https://imgur.com/a/mKJa7Ya

It all seems to be in working order! To sum it up: 1N5059 & C12 were shot, VR6-VR7 weren't original parts, replaced those & all electrolytics just in case.

Correction: New installed fan model: Noctua NF-S12B Redux

Thanks everyone for the help! 
Will keep posted if it released the magic smoke (actually it did, but thankfully I only blew the AC power cord... was using a crappy one, wire melted and bang 220VAC short  )

//
Luciano