Maker Pro
Maker Pro

HP 54510A Power Supply Smelling

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
I am wondering whether there is more dried electrolyte on the circuit board. Is there any way I can clean it without removing the caps again?

Another observation I have just made - the fan "twitches" in time to the clicking noise it doesn't actually turn, but when the power is switched off it starts to run briefly as the clicking stops.
 
Last edited:

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
The clicking may be as a result of gross overload of the power supply.

If the voltage you gave earlier were with the PSU disconnected, check the rectifier diodes for the very low rails (12 and 15v) first. That would be CR24 on the diagram above

It is also possible that C30 or C31 is internally shorted
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
Thanks Steve. The voltages I reported were with no load. They all drop right down under load. I have ordered some fast recovery diodes to replace CR24. I have actually replaced all the electrolytics on the board so C30 and C31 should be ok. Having also replaced the mains filter I hace invested quite a lot in this power supply and will be very pleased to get it working again.
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
Unfortunately the new diode makes no difference. As a matter of interest the old one measures around 900k one way and 2m the other.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
2mΩ, or 2MΩ?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
That's very interesting. Have you measured diodes on your multimeter before? Does it have a diode range?

But that won't help with the problem you're having now.

I'll have to go back to the circuit diagram and look at it again.
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
Yes it does have a diode range and it reads ok on that. I have never measured diodes on it before as my usual interest is more in valve equipment. I will wait for further suggestions as the circuit is way beyond me!

Thanks
Paul
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
OK, so we have eliminated the rectifiers for the 15V supply.

The clicking combined with the low 15V rail and the soft nature of the other rails is consistent with the protection circuit having triggered the SCR used as crowbar protection.

My suggestion is that you disable the protection and see if the clicking stops. Do not have any load on the power supply when you do this.

If the rails are approximately correct, then I would agree that failing capacitors are a distinct possibility.

If you have an ESR meter, or can beg, borrow, or steal one... Without one you can look for suspicious capacitors, but frankly, replacing all the secondary filter caps is maybe the way to go.

Arcing under caps is a bad thing. Is this fully resolved? Isopropyl alcohol is a good cleaning agent.
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
Thanks Steve

There was some bad arcing under one of the caps. I washed underneath it with water and detergent on a cotton bud when I replaced it and scraped away the carbon from the circuit board. That stopped the visible arcing. I washed under the caps in the same way when I replaced them but as the problem of it not working persisted I flooded the whole of the secondary side of the board with Maplin pcb cleaner which seems to be mainly IPA. I have replaced all the caps on the secondary side.

I have a nagging thought that it "sort of" worked until I replaced C19, 100uf 10v, which is the un-numbered cap on the -5.2v rail. I have looked over the back of the board several times since and I cannot see that I have shorted or disturbed anything. I say "sort of" worked as it would run for about a minute then cycle on and off.

Sorry to be thick but how would I disable the protection?
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
I began to doubt my own memory and notes and so decided to replace C30 and 31 again. There was quite a lot of dirt underneath them so I either had not replaced them or had forgotten to clean under them. Anyway I did them again with soapy water and IPA and put new ones in. If anything it is now worse as the fan doesn't move at all and there is very little voltage on the 15.5 (fan) pin with no load. What is odd (to me) however is that I do get a good spark if I short the connections to C30 amd 31 when the power is off. Does that help at all?
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
On the second page of your PDF, you show an SCR across the output of the 15.5V rail.

The gate of this is controlled by the protection circuit. Remove the connection to the gate.

The possibility also exists that this device is faulty. You may have to take it out of circuit.

Once you disable the crowbar protection, the output voltages *MAY* rise to value that are clearly out of spec.

Are your replacement capacitors low ESR types?
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
Thank you Steve. Disconnecting the crowbar at the A2 SCR makes no difference to the voltages on that rail unfortunately. The voltage on the 12v , 15.5 v rails is more or less zero.

The capacitors were not sold as specifically low ESR types. I was thinking of getting an ESR meter but they are quite expensive.

PS I have just seen that with the crowbar still disconnected there is 21v between the cathode of CR24 and pin 19
 
Last edited:

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
I have been looking at the photographs I took before I started changing caps and I see that I had already changed C30 and 31. Putting different ones there changed the outcome so I am wondering whether specifically low ESR ones might help. I have therefore ordered some Panasonic ones and will post the results. Any further suggestions in the meantime would be gratefully received.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Yeah, low ESR caps will reduce ripple. However, excessive ripple might trigger the crowbar, but you've disconnected that...

Did you try removing the SCR and checking it to be sure it's not shorted?
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
There is a 5meg resistance between grid and cathode. All others show open circuit.

I cannot understand how I can get a voltage between pin 19 and the positive leads of the caps when they are charged as it seems to me they are directly connected
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
Are you saying you see a significant voltage between 12 and 19?

There is an inductor been those points. Does the voltage appear across it?

You might want to fix one lead to 12 and then read voltages along the circuit until you get to 19. The may be a crack on the board, a dry joint, or some other sort of damage. If the voltage suddenly rises you've found something.
 

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
I don't think the reference to 12 on the transformer refers to pin 12. There is the same voltage drop between the positive of the caps and either side of the inductor. I am doing these measurements just with the caps charged and no power supplied. I assume that any break must be between the caps and L4. Should hard wire the two?

Edit. I am getting a resistance of 109 ohms between pins 19 and 29 which presumably is R63.
 
Last edited:

Paul Ramsbottom

Aug 12, 2016
46
Joined
Aug 12, 2016
Messages
46
I soldered a piece of wire from C30 and 31 to L4 and am back to where I was before I replaced the caps for the second time. It is a 2 layer board so it is difficult to see where the break is. Presumably I disturbed something as I was manipulating it.

It was better before I replaced C19 and I am wondering whether the same happened then but I am not sure where to start. I don't think anything shorted. I have ordered a low ESR replacement for that but it is on the -5.2 rail. This is also a little low at 3.5v.

Would it be a plan to disconnect the crowbar again? If the voltage recovers I could dig out my ancient valve scope and check for ripple.
 

(*steve*)

¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd
Moderator
Jan 21, 2010
25,510
Joined
Jan 21, 2010
Messages
25,510
It's not making clicking noises still is it? That is a reasonable indication of a short somewhere.

My concern is that something is damaged due to that. The crowbar protection is there to protect the rest of the oscilloscope and it may damage the power supply when it activates.
 
Top