UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Reply
 
Thread Tools
Old 28th Nov 2020, 11:57 pm   #1
Takapuna
Triode
 
Join Date: Jul 2009
Location: Derby, Derbyshire, UK.
Posts: 21
Default Philips PM3214 Oscilloscope Electrolytic replacement

I'm servicing my Philips PM3214 as a number of controls and switches are noisy. I think this dates from the early seventies.

I would appreciate the guidance and wisdom of fellow members regarding electrolytic capacitor replacement please.

It seems like a good idea to replace some electrolytics, especially those in and around the switching power supply. Nearly all electrolytics throughout are de-couplers and not surprisingly the originals are all Philips blue ones. All the ones I've checked out of circuit so far indicate values and esr nearly as good as or possibly better than the new Panasonic and Wurth near-equivalent values that I bought as replacements. This does not seem to be right! I'm using a Peak ESR meter (new battery!) as the primary measurement.

New caps are physically smaller than originals and radial-style. Nearly all of these have values that measure below nominal, whereas most originals measure typically their printed value or more. I know this can be a sign of leakage but given the low esr values I think they are OK. Only on the Marconi TF2700 bridge do the new ones look "better" in that the null is better defined and the Loss Balance setting critical. The values still measure a few microfarads low however.

Should I go ahead and replace these or leave well alone?

Thanks

Phil
----
Takapuna is offline   Reply With Quote
Old 29th Nov 2020, 2:11 am   #2
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 16,692
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

The ESR meter won't see leakage resistance. The usual capacitance meters will have their reading disturbed by leakage.

The bridge will measure capacitance well and it will measure dissipation factor (inverse of Q). Dissipation factor can be spoiled either by ESR or by effective parallel resistance, and the dip will olso be widened.

So if the Marconi bridge gives a poor dissipation factor, but the peak ESR meter says ESR is OK, then it looks like you have a leaky capacitor.

AVOs use a 15v battery on their top ohms range, so this is good for a leakage check of low voltage capacitors, for higher voltage parts you need a suitable supply to stress them properly while you measure current. Leakage checks at voltages much lower than the parts will run will miss many faulty parts in general cases. But if you've got issues at low voltage then you'd think they must be bad.

But

Electrolytic capacitors work when there ia appreciable bias on them. Bridges intended to measure them often have a facility to inject DC bias while measuring capacitance/dissipation factor. Some posh bridges have an internal DC source you can programme, good bridges have terminals on the back for you to provide an external source. Very basic bridges have no facilities and you have to provide a DC block capacitor and a bias feed resistor externally to bias-up your capacitor.

This is rather involved, so the current check at realistic bias voltage is an easier test.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 29th Nov 2020, 8:54 am   #3
trh01uk
Octode
 
trh01uk's Avatar
 
Join Date: Nov 2010
Location: Ross-on-Wye, Herefordshire, UK.
Posts: 1,356
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

An interesting answer David! It prompts further questions as to the nature of these "loss components" in real capacitors, and how these two types of instruments are measuring them separately. (Please move this question elsewhere if you think I am taking this thread too far off topic).

A real capacitor as you have described has a) series resistance, which we can conceive of as real resistance in the leads and the metal leaves within its construction, and b) a parallel resistance of leakage through the insulation (electrolyte in the caps we are discussing here). Its easy to see real physical parts of an electrolytic providing these loss components.

Turning to theory its fairly easy mathematically to transform a notional series resistance into an equivalent parallel one. So I had always assumed that "ESR" was a mathematical construction of the real resistance a) above, plus the leakage transformed to an "equivalent series resistance". With the ESR measured being the sum of these two parts.

However you are stating above "The ESR meter won't see leakage resistance". Please explain how it does not see the equivalent of the parallel leakage component transformed into a notional series resistor? Is that one type of loss is dominating the other, so the small one can be neglected?

I might add that I am aware that dissipation factor and ESR are related mathematically by the equation:

D.F. = ESR / |Xc|

where Xc = the reactance of the capacitor in ohms


Richard
trh01uk is offline   Reply With Quote
Old 29th Nov 2020, 8:59 am   #4
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 5,808
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

An ESR meter is only one of several ways to check caps,caps can have good ESR but still be leaky. Electrolytics are usually specified as being -10% & up to + 50% within specs, they may just need reforming, though shouldn't as they're new, but it's possible.
Quote:
New caps are physically smaller than originals and radial-style.
It's the same for everything, Mars bars & Cadbury's Cream Eggs,though in the case of caps improved manufacturing, not stingy big corp.

I wouldn't re-cap unless a fault indicates there's something wrong, it's easy to introduce more faults if your not careful, if your going to re-cap, change one at a time then test. A lot of vintage gear works fine even after all these years. Check the power rails first for excessive ripple,either with a scope or click your DMM onto AC volts, anything 1v + ish is suspect.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline   Reply With Quote
Old 29th Nov 2020, 9:05 am   #5
trh01uk
Octode
 
trh01uk's Avatar
 
Join Date: Nov 2010
Location: Ross-on-Wye, Herefordshire, UK.
Posts: 1,356
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

I will just add a comment that the Peak ESR meter measures at 100kHz, while the TF2700 measures at 1kHz. Differences in values obtained look very likely with such a difference in test frequency. I don't know at the moment, which frequency is more relevant to the SMPS in the PM3214.


Richard
trh01uk is offline   Reply With Quote
Old 29th Nov 2020, 10:13 am   #6
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 16,692
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

In an aluminium electrolytic capacitor, the eletrolyte is supposed to be conductive, it's usually the main contributor to the ESR. The electrolyte is, as far as the capacitor goes one of its two electrodes. One of the foils is denuded of oxide in the forming process and serves as the connection between the electrolyte and the outside world.

The other foil has a good build up of aluminium oxide, boosted during the forming process. The aluminium of the foil serves as the other electrode, and the oxide coating serves as the dielectric. Leakage in the capacitor relates to thinning of the oxide or to pinprick flaws in it.

The reason for all this complexity is that aluminium oxide is a good dielectric and can be very thin for reasonable voltages, so if the oxidised electrode started upt by being textured on a microscopic scale (matted) then the oxide follows the magnified surface area. The problem becomes one of making a second electrode to intimately follow this erratic surface. The solution was to use a conductive liquid electrode - the electrolyte. Conductive liquids are poor conductors compared to metals, so only a thin film is needed before the current gets collected by a metal backing electrode - the other foil.

In a feat of tremendous convenience, the capacitor as assembled is symmetrical, the etched (textured) foil is used for both foils. The capacitor is then electrically formed building more oxide on one foil, stripping the other. A once symmetrical structure has turned into a polarised capacitor.

Operation without enough bias voltage, or storage unused for a long time allows the forming operation of building and stripping oxides to degrade and revert. Breakdown voltages diminish and leakage currents rise.

Apply current limited bias to the right voltage and the forming process can be redone, though there is no guarantee that it goes as evenly as the first time.

So high ESR points to a capacitor that has lost electrolyte - mostly the loss of water vapour and the shrinkage-drying out of the chemicals once dissolved in that water.

Low leakage resistance points to problems with the oxide thickness.

The Q or Dissipation Factor of a capacitor gives a real term of impedance (resistance) which can either be considered as a series resistance or a parallel resistance to model it. Conversely, eaither of these two resistances can contribute to the Q/D factor. BUT specific circuits may be more sensitive to one or the other forms of loss. Specific measurements may measure one or other dominantly and relate it to the apparent Q/D. Consequently modelling from just one measurement method can cause confusion.

There really are two different physical processes ar work and an ESR/leakage model is really needed to tell the whole story of a capacitor.. The Q/D model only really works reliably in resonant circuits.

So that's the long way round of why I suggested a capacitance measurement, an ESR measurement, and a leakage measurement. (If you want completeness, ESL can be important for SMPS etc). Leakage is likely to be non-linear as well, so needs measuring at a high enough voltage to make sense given either the rating of the part or the voltage it will be used at.

If you think these capacitors are complicated, you ought to see inductors!

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline   Reply With Quote
Old 29th Nov 2020, 12:40 pm   #7
turretslug
Dekatron
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 3,628
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

Philips "blues" were among the more dependable electrolytics of the era, arguably not really bettered until the good-name Japanese radial types appeared en-masse. That's not to say that some won't have deteriorated by now. Andy (Diabolical Artificer) makes a good point in another thread in that old 'scopes can be a bit crotchety about being disturbed too much and Philips stuff is often a bit of a mare's nest to extensively dismantle. It's your property, your decision but I'd leave capacitor replacement until I was happy with the switch/control operation and general functionality, and then do it in cautious stages. "If it ain't broken...." is a wise saying to keep in mind, and there's also the danger of test gear maintenance becoming a recursive diversion in itself, instead of that gear being used to maintain other things!

It used to be the case that electrolytic capacitor tolerance was often quoted as -20/+50% or even -20/+80% but I've noticed a current trend for it to be quoted as -/+20%, probably reflecting general refinement and tightening of production techniques- that could explain some of the tendency for new electrolytics to appear of lesser capacity. I've certainly noticed the tendency of new electrolytics to measure as surprisingly close to their labelled value, compared to previous wisdom that they were components whose values were of notoriously "woolly" definition,

Colin
turretslug is offline   Reply With Quote
Old 29th Nov 2020, 12:52 pm   #8
chriswood1900
Octode
 
Join Date: Jan 2007
Location: Bristol, UK.
Posts: 1,040
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

Phil
My experience of two Philips scopes of this era was they needed widespread replacement of the blue capacitors particularly around switch mode PSUs. Whilst the Philips capacitors seem to be good in transistor radios etc. less good in SWMPUs. The Peak ESR meter found most of the duds so I ended up re-capping all the PSU and selectively replacing others.
__________________
Chris Wood
BVWS Member
chriswood1900 is online now   Reply With Quote
Old 29th Nov 2020, 4:30 pm   #9
Takapuna
Triode
 
Join Date: Jul 2009
Location: Derby, Derbyshire, UK.
Posts: 21
Default Re: Philips PM3214 Oscilloscope Electrolytic replacement

Thanks to everyone for their replies and the interesting discussion.

So, there's nothing wrong with my replacement components, other than their shape and size. I take Andy's point that unlike chocolate bars they've just got smaller due to manfacturing advances. The missing few microfarads on some of them perhaps is down to cost though, as these things are only a few pence each and and sell in large quantities. A better selection of axial components would be nice even if still undersized. I note that the new capacitors have leads that are attracted to my soldering iron (ferrous) whereas the originals aren't. A couple of the more expensive Vishay branded ones still have copper leads too, so you do get what you pay for and I'm stil a little worried that these new capacitors are not "good enough"!

I'll try a simple leakage test and see what resuts I get.
Unfortunately my Avo does not have its 15V battery fitted and the TF2700 which will allow an external bias to be applied has suddenly taken ill (almost no meter deflection away from zero on any ranges and the PP9 battery, which is 9.6V at its terminals with the unit turned on, measuring only 3V or so when carrying out the battery test between front panel Bias + and chassis)!

With the 'scope PSU card removed I can leak check most capacitors in circuit being mindful of the smoothing chokes effectively connecting capacitors either side in parallel. I will replace the two that I've removed from the card as it seems silly not to but will leave the others alone unless suspicious.

As the front switch assembly is removed this is my best opportunity to replace any of the 18 or so 33uF/16V decouplers on the top board. I don't want to disassemble it again and they don't lend themselves to testing in circuit. The PCBs are fibreglass which is good, but Philips in their infinite wisdom have drilled holes that are barely a clearance for the component leads making them difficult to remove. I'm going to reassemble and test and will have to use the "cut the old component out and solder to the remaining leads" trick for any that need replacement. Ill try to be neater than the guy who does radios in/on "The Repair Shop"...(sorry!).

Stop Press...I've just made a temporary 15V supply for the AVO and can see at least one rail showing <50k leakage and another around 400k so I'll investigate. The 6.3V rated capacitors weren't too happy at this voltage either! This and chriswood1900's input is shifting my thoughts to replacing at least all the ones on the PSU card.

I might come back for ideas on the two 1k Y-shift potentiometers that are erratic beyond repair. They are 17m dia with 4mm shafts, one with a push-pull switch!

Colin's post hit the nail on the head; The scope was being used to help me sort the RIAA curve for a phono pre-amp. Now I'm making AVO batteries and trying to mend a Marconi bridge to fix a scope to...

Thanks again,

Phil
----
Takapuna is offline   Reply With Quote
Reply

Thread Tools



All times are GMT. The time now is 11:21 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2021, vBulletin Solutions, Inc.
Copyright ©2002 - 2021, Paul Stenning.