Capacitor tolerance - UK Vintage Radio Repair and Restoration Discussion Forum

Linnovice · 25th May 2016, 10:54 am

Hi all, before I start please accept my apologies for ignorance.
I am in the process of refurbishing my Revox A77 MkIII. Being at the stage of buying capacitors to recap the machine I find myself in a very confusing situation. When replacing capacitors do I have to stick rigidly to what's specified in the service manual? For example; on the input board there are two 125uF 3V electrolytically specified. For the life of me I can't find them. I've tried RS, Mouser, Digitech they have umpteen caps in stock but none of that spec. So, thinking sideways, I've plumped for 120uF 35V on the assumption that the tolerance quoted is 20% +/- which covers the 125 figure and the 35V gives a wide safety margin. Am I right in thinking this?
Mike

60 oldjohn · 25th May 2016, 11:13 am

That cap will be ok, given the +50% -10% of a lot of capacitors used I have wondered myself why bother with 120uf when there are 125uf available. In theory worst case scenario they could both measure around 180uf. The voltage is above the original so ok, If the voltage was many X what was needed say 2000wv the cap would suffer in time with the lack of volts. I am sure others can explain better than me.

John.

'LIVEWIRE?' · 25th May 2016, 11:14 am

Two things occur tome, Mike. The first is that Electrolytic capacitors have a wide tolernace +/- their nominal value, so replacing a 125uf with 120 or even 220uf will not normally cause any problems, neither should the use of a 35v cap. to replace a 3v. one, although a 6.3 or 10v type would be better. The second thing is that capacitors (of any type) and resistors have been made in 'preferred' values for many years now. The most common range is 1, 2.2, 3.3, 4.7, 6.8. 8.2 and their multiples. Because of this, a 50uf capacitor is replaced with a 47uf, 2.5uf with a 2.2uf, etc., a 500 ohm resistor with a 470 ohm, and so on.

peter_sol · 25th May 2016, 11:31 am

This depends on where the caps are in the circuit.
If they are decoupling, then nearest and yes the next available higher voltage is OK.
If coupling then same.
If they are in a frequency dependant circuit such as the EQ section, as it is a tape recorder, then you may need to be more exact in your value selection.

turretslug · 25th May 2016, 11:32 am

Some European electrolytic manufacturers (in particular) stuck to an odd old sequential series for a long time with values like 80uF, 125uF, 250uF, 640uF etc. As said, they're not precision components, so 120uF will be fine. Similarly, 150 or 220uF wouldn't harm the circuit operation.

Linnovice · 25th May 2016, 11:52 am

Thanks Gents, your advice is much appreciated and noted. I'm sure I'll be back . . . as the humanoid said (before running California!).

GMB · 25th May 2016, 12:54 pm

One point worth noting:

The tolerance (and indeed life rating) of electrolytics is more about aging than initial manufacture. So don't make the mistake of thinking that because you just bought one that measures in the expected tolerance range of the one being replaced, that doesn't mean it will necessarily stay in that range. I.e. assume the worst by the quoted tolerances and so decide if it will remain OK.

60 oldjohn · 25th May 2016, 2:48 pm

I understand that capacitance value can be viewed as a piece of elastic, capacitance can get stretched in time but will never return to it's original value.

John.

Diabolical Artificer · 26th May 2016, 5:40 am

I was faced with the same dillema and used a 100u and 22u in parallel. Your 120u's sound a better bet.

I've been doing a lot of capacitor measurement recently on a Marconi TF2700 bridge. All new caps measured were pretty spot on in value. Older used electrolytic caps had all drifted higher in capacitance, but the circuit's in which they were in still worked. This is in audio gear, I couldn't hear any difference.

Andy.

Radio Wrangler · 26th May 2016, 9:48 am

Higher capacitance in an old electrolytic capacitor is due to the dielectric layer thinning a little. It's fairly certain that the area hasn't been growing

This is just part of the de-forming process and goes with a reduction in the voltage it will stand. Careful re-forming to (eventually) a high enough voltage will put the capacitance back.

What capacitance tolerance a circuit can stand before something goes out of spec depends on a circuit. If anything needs 125uF and not 120uF then it shouldn't be using an electrolytic part.

David

stevehertz · 26th May 2016, 9:37 pm

As was the subject of a recent thread started by myself, you can, in general, use electrolytic caps of a higher working voltage than the one being replaced. However, to be pedantic, electrolytic caps 'like' to be operated or 'worked' at a voltage getting up towards their rated voltage. So, for example, an electrolytic cap rated at 30V will 'prefer' to be operated at 20V or so than it will 2V. Don't shoot me, I'm only the messenger, their are phycists and scientists on here who can explain the theory. But in general, it doesn't matter that much so long as the voltage applied does not exceed the rated voltage. In other words, I've just told you about a bit of theory that you need to worry about too much.

ricard · 27th May 2016, 8:59 am

I've seen at least one case where a circuit has been redesigned seemingly because the electrolytic capacitor saw a very low polarizing voltage (like half a volt or 0.1 volt or so), and consequently would fail prematurely. The circuit was then redesigned so that the capactor saw something more like its rated voltage.

25th May 2016, 10:54 am	#1
Linnovice Hexode Join Date: Jun 2014 Location: Chelmsford, Essex, UK Posts: 386	Capacitor tolerance Hi all, before I start please accept my apologies for ignorance. I am in the process of refurbishing my Revox A77 MkIII. Being at the stage of buying capacitors to recap the machine I find myself in a very confusing situation. When replacing capacitors do I have to stick rigidly to what's specified in the service manual? For example; on the input board there are two 125uF 3V electrolytically specified. For the life of me I can't find them. I've tried RS, Mouser, Digitech they have umpteen caps in stock but none of that spec. So, thinking sideways, I've plumped for 120uF 35V on the assumption that the tolerance quoted is 20% +/- which covers the 125 figure and the 35V gives a wide safety margin. Am I right in thinking this? Mike __________________ Small Scotch, New Aids and Happy Ears

25th May 2016, 11:13 am	#2
60 oldjohn Dekatron Join Date: Dec 2004 Location: East Yorkshire, UK. Posts: 3,988	Re: Capacitor tolerance That cap will be ok, given the +50% -10% of a lot of capacitors used I have wondered myself why bother with 120uf when there are 125uf available. In theory worst case scenario they could both measure around 180uf. The voltage is above the original so ok, If the voltage was many X what was needed say 2000wv the cap would suffer in time with the lack of volts. I am sure others can explain better than me. John.

25th May 2016, 11:14 am	#3
'LIVEWIRE?' Rest in Peace Join Date: Mar 2006 Location: N.W. Oxfordshire(Chipping Norton) Posts: 7,306	Re: Capacitor tolerance Two things occur tome, Mike. The first is that Electrolytic capacitors have a wide tolernace +/- their nominal value, so replacing a 125uf with 120 or even 220uf will not normally cause any problems, neither should the use of a 35v cap. to replace a 3v. one, although a 6.3 or 10v type would be better. The second thing is that capacitors (of any type) and resistors have been made in 'preferred' values for many years now. The most common range is 1, 2.2, 3.3, 4.7, 6.8. 8.2 and their multiples. Because of this, a 50uf capacitor is replaced with a 47uf, 2.5uf with a 2.2uf, etc., a 500 ohm resistor with a 470 ohm, and so on.

25th May 2016, 11:31 am	#4
peter_sol Octode Join Date: Aug 2003 Location: Welwyn Garden City, Herts. UK. Posts: 1,906	Re: Capacitor tolerance This depends on where the caps are in the circuit. If they are decoupling, then nearest and yes the next available higher voltage is OK. If coupling then same. If they are in a frequency dependant circuit such as the EQ section, as it is a tape recorder, then you may need to be more exact in your value selection.

25th May 2016, 11:32 am	#5
turretslug Dekatron Join Date: Nov 2011 Location: Surrey, UK. Posts: 4,397	Re: Capacitor tolerance Some European electrolytic manufacturers (in particular) stuck to an odd old sequential series for a long time with values like 80uF, 125uF, 250uF, 640uF etc. As said, they're not precision components, so 120uF will be fine. Similarly, 150 or 220uF wouldn't harm the circuit operation.

25th May 2016, 11:52 am	#6
Linnovice Hexode Join Date: Jun 2014 Location: Chelmsford, Essex, UK Posts: 386	Re: Capacitor tolerance Thanks Gents, your advice is much appreciated and noted. I'm sure I'll be back . . . as the humanoid said (before running California!). __________________ Small Scotch, New Aids and Happy Ears

25th May 2016, 12:54 pm	#7
GMB Dekatron Join Date: Aug 2003 Location: near Reading (and sometimes Torquay) Posts: 3,098	Re: Capacitor tolerance One point worth noting: The tolerance (and indeed life rating) of electrolytics is more about aging than initial manufacture. So don't make the mistake of thinking that because you just bought one that measures in the expected tolerance range of the one being replaced, that doesn't mean it will necessarily stay in that range. I.e. assume the worst by the quoted tolerances and so decide if it will remain OK.

25th May 2016, 2:48 pm	#8
60 oldjohn Dekatron Join Date: Dec 2004 Location: East Yorkshire, UK. Posts: 3,988	Re: Capacitor tolerance I understand that capacitance value can be viewed as a piece of elastic, capacitance can get stretched in time but will never return to it's original value. John.

26th May 2016, 5:40 am	#9
Diabolical Artificer Dekatron Join Date: Jul 2009 Location: Sleaford, Lincs. UK. Posts: 7,667	Re: Capacitor tolerance I was faced with the same dillema and used a 100u and 22u in parallel. Your 120u's sound a better bet. I've been doing a lot of capacitor measurement recently on a Marconi TF2700 bridge. All new caps measured were pretty spot on in value. Older used electrolytic caps had all drifted higher in capacitance, but the circuit's in which they were in still worked. This is in audio gear, I couldn't hear any difference. Andy. __________________ Curiosity hasn't killed this cat...so far. Last edited by Diabolical Artificer; 26th May 2016 at 5:49 am.

26th May 2016, 9:48 am	#10
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,896	Re: Capacitor tolerance Higher capacitance in an old electrolytic capacitor is due to the dielectric layer thinning a little. It's fairly certain that the area hasn't been growing This is just part of the de-forming process and goes with a reduction in the voltage it will stand. Careful re-forming to (eventually) a high enough voltage will put the capacitance back. What capacitance tolerance a circuit can stand before something goes out of spec depends on a circuit. If anything needs 125uF and not 120uF then it shouldn't be using an electrolytic part. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

26th May 2016, 9:37 pm	#11
stevehertz Dekatron Join Date: Sep 2011 Location: Rugeley, Staffordshire, UK. Posts: 8,832	Re: Capacitor tolerance As was the subject of a recent thread started by myself, you can, in general, use electrolytic caps of a higher working voltage than the one being replaced. However, to be pedantic, electrolytic caps 'like' to be operated or 'worked' at a voltage getting up towards their rated voltage. So, for example, an electrolytic cap rated at 30V will 'prefer' to be operated at 20V or so than it will 2V. Don't shoot me, I'm only the messenger, their are phycists and scientists on here who can explain the theory. But in general, it doesn't matter that much so long as the voltage applied does not exceed the rated voltage. In other words, I've just told you about a bit of theory that you need to worry about too much. __________________ A digital radio is the latest thing, but a vintage wireless is forever..

27th May 2016, 8:59 am	#12
ricard Octode Join Date: Mar 2006 Location: Lund, Sweden Posts: 1,632	Re: Capacitor tolerance I've seen at least one case where a circuit has been redesigned seemingly because the electrolytic capacitor saw a very low polarizing voltage (like half a volt or 0.1 volt or so), and consequently would fail prematurely. The circuit was then redesigned so that the capactor saw something more like its rated voltage.