Old Philips PE 1649 DC (150V / 7A) Powersupply "bumping" noise

[kblue]

Greetings!

I rarely use this DC Power Supply for anything "above" 0,5 Amps but recently i discovered some loud "bumping noises" if i cross the 0,5A +/- mark. In the video its "not that loud" but in real life it is Is this "normal" for such a huge beast? Amps seems to be stable if i let it run for about 5 mins, but this "bumping" or whatever is kinda creepy. Even the desk this power supply rests on is vibrating and "bumping".

Test was done with set 50 VDC and ramping up the amps with coarse.



Best wishes!

[ch_scr]

If there are still original "blue axial Philips" capacitors in there, they might have degraded to the point of no longer providing adequate decoupling, and it might break into self-oscilliation (and god knows what else) 

[kblue]

Ye, its all original, except the self-destruct-rifa caps are already changed. I could change the smaller caps but no way the huuuuuge filter caps. New ones would cost hundreds of euro, the filtercaps are massive like beer-bottle-massive.

Well ... guess its time to take it apart.

[ch_scr]

I'd change all the PCB mounted blue (probably axial) Philips. Generally: the lower the volume, the faster they go - the big cans should have some life left.

[kblue]

Thats how it looks like ... below the (single) PCB are the ginormous caps, indeed. I do hope they have some life left As said, replacement are extremely expensive.

Thanks for any hints and tips!

/edit: All the caps test fine in-circuit (DerEE with 120 Hz), except for the very, very tiny ones.

[garrettm]

Looking at the date codes on those ICs... Yeah, I'd say those axial caps are near death, if not already dead. Though I have had some Sprauge electrolytics from 1967 that still tested good. But they were pretty big and had epoxy encapsulation over the rubber seal. In my experience, "small" electrolytic caps this old have been bad (low capacitance or high ESR). The bigger caps usually have better seals and the larger volume of electrolyte keeps them working for longer. But I have seen very large axial caps still fail due to being placed near high power dissipation devices (resistors and transistors). Valhalla and HP are guilty of doing this in some equipment I have (notice the "brown" cap vs the orange one in the second pic...).

[garrettm]

If this PSU uses SCR pre-regulation then the "bump" noise could be caused by them pulsing the transformer with a non-sinusoidal current.

My HP 6110A uses SCR pre-regulation for the HV output and it will start mechanically vibrating at certain output voltages from the transformer being pulsed.

On paper SCR pre-regulation makes sense, but it does cause audible noise and can cause additional AC mains noise if not filtered (HP uses a massive line reactor and filter network to minimize this issue).

[kblue]

Multiple "BTW40"'s are mounted on the massiv heatsink and BTW40 are "Silicon thyristors in metal envelopes" regarding to some documentation. I'm not familiar with these type of elements.

The unit no longer "bumps" after changing EVERY cap on the single board. Surprisingly every single one of them measured good or even "better" (ESR tested with DE5000) compared to new Nippon-Chemicon or Panasonic-Caps i had in my stock. Guess that's a classic case of "they may measure good, but once in actual use the caps are bad". Even the very little ones measured good or "better" than new ones to my surprise regarding capacitance and ESR. But the DE5000 does not run them on max. rated Voltage.

Just did a quick test, ramped amps slowly up to 2A (of 7,5) and stopped. No bumps, but all my lights in my workshop hate this unit, once the amps go high my lights are ... faintly flickering its like a horror-show. Oh and the Amps were stable, very stable in a shorter amount of time than before cap-change.

I didn't test the unit any further, because having it "open" with all the mechanical and electrical noise is just kinda scary and everything screams "one wrong touch and you're gone". The bigger tests like slowly ramping everything up, will be performed outside Once winter is over.

Thanks everyone for the hints, input, information and more! I guess the unit is running fine again without these "bumps".

[ch_scr]

Multiple "BTW40"'s are mounted on the massiv heatsink and BTW40 are "Silicon thyristors in metal envelopes" regarding to some documentation. I'm not familiar with these type of elements.

A "Thyristor" is the semiconductor equivalent to the "Thyratron" tube. They also have a more descriptive name "Silicon controlled rectifier" or SCR for short. And they do just as the name implies, conduct in one direction like a diode, but only when you tell them to... with the caveat that they can be triggered on, but only will turn off when the flow of current through the main path subsides. A slightly later invention that is still somewhat common today is the "Triac", which can also only be triggered on, but will allow current to flow in both directions.

Congrats to the repair 

[kblue]

Weeeellll ... it's back again! Mayhaps it's something else, but i don't know what. I'll check some more caps.

I did a test with more load (4 Amps) (not maxed out yet) and the "bump" just increases in frequency. Amps are stable.

Best wishes!

[Roehrenonkel]

Hi kblue,
 
sounds to me like the thyrisors (SCRs) chargeing the filter-caps
and the transformer-core (or case-sheets or even the wires)
vibrating due to high peak-currents.
Try fastening the bolts of the transformer (core and trafo-->case).
Feel around what's shaking......
Not unusual for Output-transformers of tube-amps to "sing"
with the signal at higher output-levels.

So IMHO nothing to worry about.
You even have a "Audiable feedback" of the current-level. ;-))

Ciao4now

[kblue]

Noticed something "new" or "different" ...

It seems, the bumps are gone, if nothing is "on top". I parked some light (not heavy) stuff on top of it like a stash of small caps, once the things are removed, it seems "silent". Mayhaps the flimsy metal sheet topcover is "too close" to the transformerwindings and due to some magnetic fields, this "thin" metal makes these noises? Especially if "bend" lightly due to something on top?

Guess in the end it was my fault. This unit wants all the space for itself, well ... gonna re-organize the lab.

Ramped up over 5 Amps this time and except from lights flickering, nothing happend and everything is stable.

Best wishes!