HP3456A Power Supply Parts Help

[Electro-McMech]

Hi!

I am looking for some help to track down a few parts that have me worried on the in-guard power supply of my HP3456A. CR6, CR8, CR9, CR10, as well as CR7 all have the same part number of 1028 with the date code of (I think its the date code) 8452. Using the diode check on my multi meter a forward voltage of 0.600V seems reasonable, and when powered up for a while they're not getting hot (pretty sure the blue ones are replacements). I would rather be safe than sorry about it, and hope to replace all of them, but wanted a second opinion that 1N4001 is a reasonable choice? I've not managed to convince myself the 1N4001 is the part I  need so any help on this would be much appreciated. My thought hope is to replace these diodes and use some of those ceramic spacer rings to get them off of the board and hopefully prevent anymore scorching should something get cooked in the future.

I will also be replacing all of the capacitors, even the monster 5 pin unit will be replaced, planning to make an adapter board up to do that neatly and hold that off of the in-guard main board using more ceramic spacer rings. Also the power entry module will be replaced, I've seen enough horror stories about RIFA caps going nuclear that I would prefer to avoid any such issues. This one does have some filtration, so I would rather be preemptive.   

[wn1fju]

According to the HP 3456A's parts list, CR6,8,9,10 are part number 1920-0028 which are 400V, 0.75A general purpose diodes.  You could replace them with 1N4004 diodes.  The 1N4001 you suggest is only rated to 50V reverse voltage, so why take a chance when the 1N4004 has the correct rating and the parts are so inexpensive and readily available. 

Also, CR7 is an 18V, 5 watt Zener diode, 1N5355B.  It should not be replaced by a 1N4004.  While that won't hurt anything, you would lose the overvoltage protection that the Zener is providing.

But the bigger question is why are you replacing these diodes (and electrolytics)?  Is the power supply malfunctioning?  If the supply is working, I would leave the thing alone.  Unless you see 120 Hz ripple on the power rails, or there is obvious capacitor bulging or leakage, I wouldn't fuss with it.  And diodes heating and discoloring boards is quite commonplace.  You could lift them 1/8" off the board if it upsets you.

A preemptive replacement of the line filter is a toss-up.  These things seem to be more problematic when using 240 VAC, not 120 VAC as the input voltage.  But they do make a mess if and when they explode.  And you are quite correct that if you ever find any of the gold-colored RIFA suppression caps, they should be replaced immediately without question.

[Electro-McMech]

Whoops, I should have started out with that, most of the output voltages are pretty well out of spec. If memory serves the 33V rail was down below 24V. The meter would fire up and work OK, excepting a noisy 1V to 10V range. Going after the power supply first before I start working through the trouble shooting section of the manual for that. I know some of this is likely unnecessary work, but my hope is that replacing some components now will hopefully prevent any kind of future failure mode that could damage something else.

This is the first board that I've really encountered with such discoloration from heating, if its an issue that can be easily prevented then it seems well worth the time to me. The leads on the 1920-0028 components were cut really short, leaving little room to try and lift them off of the board, steering me towards a replacement instead of trying to modify whats there. 

Thankfully I've not found any of the gold colored RIFA caps in this unit, but a couple other pieces of gear I've worked on have had them (TDS 420, HP8165A, HP3478A). The TDS 420 repair has left me feeling a strong urge to replace every electrolytic I see... (so much time cleaning up traces and pads). More likely in this case that the caps have dried out, rather than leak all over for this machine. I think I will hold off on the line filter for the time being then and see how the rest of the repair work goes. At this point, the plan is to get some 1N4004's and all of the caps swapped.

Thank you so much for your help! I'll post back once I can get everything sorted out.

[coromonadalix]

 a trick if you ever replace the diodes,  lift them a few millimeters from the pcb, it will help them dissipate their heat ..

For the rectifier models  yes an 1n4004, 4005  and up is good

And  no you DO not substitute a zenner for an rectifier model,  it was designed that way and should stay that way


For psu problems  normally the culprit is most likely the electrolytic capacitors, the tantalums  and the notorious RIFA caps

Dont change parts for nothing or hoping to fix things, unless your absolutely positive about changing them

[Electro-McMech]

Sorry for the delay in responding further too this, just wanted to say thank you for all of the input/help! The HP3456A is back up and running! Wasn't sure what to suspect for a repair candidate but that ended up being less of a pain than I expected. Not sure if there are any more subtle issues but for the relatively basic stuff I've been using the meter for, its hilariously overkill (Perhaps I'll join in on the home brew voltage standard fun).

[bd139]

Great job and nice to see effort going into quality repairs