|
Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders. |
|
Thread Tools |
23rd Sep 2017, 1:16 pm | #1 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Philips PM5519 Pattern Generator runs out of water.
I'm sure this is all too familiar. I was working on a project with a video monitor and to confuse issues the workshop equipment starts to malfunction. A PM5519 video pattern generator has 100Hz ripple in its video signal. So I drop what I'm doing and disassemble it. The ripple is on the 5V rail and I notice in the past I had replaced the "in line bridge" rectifier assembly. This time the 2200uF 16V filter electro, on the input to the 5V regulator had failed and the voltage was dipping low enough to put 100Hz ripple in the 5V output.
On testing this capacitor (Valvo brand not Philips but still blue) it read 1700uF, and ESR of 3.5 Ohms (way too high). No sign of bulging at the end, no sign of liquid electrolyte leaking out of it either. So I opened it up, as I always do, for a forensic inspection. No sign of corrosion inside and perfect insulation & foil, only thing being the paper insulation was as dry as parchment paper. Despite the good looking rubber seal, water had abandoned the capacitor. Whether the H2O molecules left directly over time, or some in the form of hydrolysed H2 and O2 gas, I do not know. Then it occurred to me that if this was the case and just water alone had left and no electrolyte solutes, if I added the water back in, the capacitor should return to normal. So I placed it in a glass of de-ionized water for a few minutes to thoroughly wet it and slipped it back into the metal can and left it for 6 hours. On testing the uF value read 2300 and the ESR dropped back to a respectable 0.06 Ohms. So the capacitor in this case simply "ran out of water" This means of course it would have been possible to repair it easily with a small hole drilled in the metal end and introducing a little water with a syringe & needle and sealing that with some epoxy resin. It just goes to show that longer lasting electros, among other things, must have better water/gas seals. |
23rd Sep 2017, 7:26 pm | #2 |
Retired Dormant Member
Join Date: Dec 2007
Location: Shropshire, UK.
Posts: 3,051
|
Re: Philips PM5519 Pattern Generator runs out of water.
Restuffing electrolytics is one thing, but repairing them takes it to another level!
|
23rd Sep 2017, 9:19 pm | #3 |
Moderator
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
|
Re: Philips PM5519 Pattern Generator runs out of water.
Sticking in some more distilled water sounds good, but you'd also want to find and repair the leak via which the original water did its vanishing act.
Dunk it in hot soapy water and look for bubbles? Tiny puncture repair outfit, finished off with French chalk? David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done |
23rd Sep 2017, 9:31 pm | #4 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Philips PM5519 Pattern Generator runs out of water.
Quote:
One thing I note with caps like this, they feel light in the hand compared to a new one and they have definitely lost mass. Probably if the end with the rubber seal was varnished over it might help seal the micro-leak. |
|
24th Sep 2017, 12:53 am | #5 |
Dekatron
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
|
Re: Philips PM5519 Pattern Generator runs out of water.
I wonder if the "Valvo" caps were actually Philips made- after all, Philips made Valvo valves. I've recently been working on a piece of kit with Roederstein electrolytics that appear to be identical to classical Philips axial elkos- lead welds, can crimping, translucent spiral-wrap sleeve, font style- except that that sleeve is a fetching Lucozade orange in colour, rather than blue.
I wouldn't be surprised if there is a slow loss of water through even apparently effective seals long-term, after all the volume is low and water is a volatile substance- in open air it vanishes very quickly when it's even slightly warm! I'm wary of the whole business of trying to reform/otherwise recover elderly electrolytics, after all they're comparatively cheap and ones in PSU positions could destroy a tricky-to-replace transformer. I know the originality debate is a perennial and intractable one, but new capacitors neatly fitted in venerable kit give me a feeling of confidence and if the original designers could have used the types available now, they'd probably be delighted at the size/performance/price revelation. Talking of "light" electrolytic capacitors, does anyone else recall the light purple sleeve Dubiliers of the late '70s/ early '80s? (I think a lot of "RS" badged ones in that colour were Dubiliers).They just felt instinctively wrong to me, they were so light, but I'm sure that if they were particularly troublesome, a lore would have developed around them. I think we develop an instinct for how heavy something "ought" to be approximately based around water's SG- moving away from electrolytics, PCB-filled capacitors can feel strangely heavy. |
24th Sep 2017, 1:53 pm | #6 | |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,669
|
Re: Philips PM5519 Pattern Generator runs out of water.
Quote:
The purple Dubiliers have gone back into the spares drawer. I think they're in excellent condition considering their age. Chris
__________________
What's going on in the workshop? http://martin-jones.com/ |
|
24th Sep 2017, 2:25 pm | #7 |
Dekatron
Join Date: Nov 2003
Location: Heckmondwike, West Yorkshire, UK.
Posts: 9,637
|
Re: Philips PM5519 Pattern Generator runs out of water.
It's about immortality in the end. Everything has a life cycle which we can cheat only so far, however we try. Even the best components eventually succumb to the inevitable.
|
24th Sep 2017, 3:53 pm | #8 |
Dekatron
Join Date: Sep 2003
Location: Near Swindon, North Wiltshire, UK.
Posts: 3,595
|
Re: Philips PM5519 Pattern Generator runs out of water.
"I wonder if the "Valvo" caps were actually Philips made- after all, Philips made Valvo valves. recently been working on a piece of kit with Roederstein electrolytics that appear to be identical to classical Philips axial elkos- lead welds, can crimping, translucent spiral-wrap, etc."
I think Valvo is to Philips (in Germany) as Mullard is to Philips (in the UK). |
24th Sep 2017, 11:17 pm | #9 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Philips PM5519 Pattern Generator runs out of water.
Quote:
On the other hand it appears that electrolytic caps are grossly out of step with the life cycle of practically every other electronic component. At least 9/10 faults or close in aged (30 year old) semiconductor gear are failed electrolytic caps and they can take out other parts when they leak. Currently I'm working on a 30 year old video monitor; each high capacity capacitor is an axial tantalum. These have been sealed at the ends with a type of glassy hard resin, very similar to that used to seal lead wires in TO-3 transistors, rather than the folded over aluminium lip and rubber disc seal of the electrolytic cap. The Tants all test perfectly with no sign of degradation. I wonder if the electrolytic cap was sealed this way we might have a lot less trouble with them, but I guess that would have also increased the initial cost. (Also I noticed recently that if a typical radial electrolytic capacitor has leaked in the past, but its not obvious looking at it, it renders the rubber base seal plug conductive. So if the meter probes are placed on the rubber it conducts. This is then a clue that there is electrolyte on nearby pcb tracks that can cause faults. Also this invisible surface electrolyte, especially on boards with a matt conformal coating, won't clean off with contact cleaner, but will with hot water). Last edited by Argus25; 24th Sep 2017 at 11:27 pm. |
|