When to replace filter caps?

[Restless Rocks]

While contemplating all this even more I just had a realization. I also own some old tube hifi gear and just remembered one of the factors in evaluating the quality of a power amp is the Slew Rate of it's power supply. This is the speed from which it can recover or recharge after transient peaks. As caps get old the slew rate becomes slower and therefore we notice the flubby bass and lack of punchiness in the amps (guitar or hifi) response. I know a lot of guys who, when recapping old Dynaco's and Eico's and Scott's etc. will replace the filter caps with much larger values. This seriously improves the bass response and even gives a slight increase in output power. You are limited in these capacitance increases by the abilities of your power transformer. I have a Dynaco Stereo 70 which had a 30uf,20uf,20uf,20uf can type filter in it. I replaced it with 80uf, 40uf, 30uf, 30uf and you betcha the bass and tightness were improved many fold.

 

[Monsta-Tone]

I think this is why Mesa uses 2-220mf caps in the 1st stage of their power supplies, where most Marshalls use 2-50 or 2-100mf caps in theirs, and Fender uses 2-80mf caps for the 1st filter stage in their amps.


Randall probably did this at 1st because he was adding gobs of gain and wanted the amps to sound more legible at higher gain stages.

 

[Restless Rocks]

There are four 220uf caps in the simulclass amps.

 

[stokes]

Restless,I have always been a proponent of using larger value caps for the main and screen supply to tighten up the overall response,particularly the bass.We are supposedly limited to how large these caps can go by the rectifier,solid state rectifiers having virtually no limit to the mf value.Tube rectifiers have recomended values that should not be exceded,but I have done tests with a 5R4 which has a recomended input capacitance of 20uf.I was told by a well known amp manufacturer and author that a reliable test is to hook up the cap in question and look for arcing in the tube,no arcing indicates the tube can take the capacitance.I went as high as 220uf with the 5R4 and it was fine.I used a 100uf and so far after almost a year of very regular use,the tube is fine and still tests like it did when it was new.So it is safe to exceed the recomendations in the manuals,they are likely very conservative.
Monsta,the 220uf's in the Mesa are wired in series for a total of 110uf,most Marshalls have 2x50uf in parallel to give you 100uf.Not being an engineer myself,but I suspect that 100uf is about the highest useful value in the main filter.
Shep,you can measure the ac volts at the caps in the B+ rail and it will indicate a faulty cap,but my point is that after 10 years you dont know when they will fail,they tend to fail suddenly and explode.
PRS-Ian,you sound like you could use a cap job.
Stuart,to bring the volts up slow with a variac,you should remove all the tubes,if you have a tube rectifier you need to wire in a couple of diodes in place of the tube.You then plug the amp into the variac and turn the amp on,not in standby but in play mode.I like to start with the variac at 40 and raise it ten every hour till you get the voltage at the B+ to the caps rating.You have to be careful because with no tubes in you will reach full dc voltage before you get the variac to the full wall voltage.

 

[Monsta-Tone]

Monsta,the 220uf's in the Mesa are wired in series for a total of 110uf,most Marshalls have 2x50uf in parallel to give you 100uf.Not being an engineer myself,but I suspect that 100uf is about the highest useful value in the main filter.

Yeah, I should have worded that differently.

How about this....

I wonder if Randall Smith placed the 2 220mf caps in series for the tonal differences (if there are any) between series & parallel wiring, or if it was just because he was replacing parts on a Fender board and this is how they were situated?


That makes more sense to me.

 

[stokes]

Wiring the caps in series or totempoled is to up the voltage rating,2x220uf/300v caps wired as such gives you 110uf/600volt total rating.Wont effect the tone,it is purely a profit margin issue.They use Illinois caps here,2 of these caps are cheaper than one Sprague 100/600v cap,if they are even available.I know they do make some 600v caps but not sure if they have the uf value needed.I am sure Mesa could have them made special by Sprague,but that would cut further into the profit.

 

[jvk]

Some cool cap info in this thread, learnt some interesting stuff.

Does anyone have any formulas for ideal cap values for a given DC voltage to remove most of the AC, or does it not work like that? I hardly remember my power supply design part of my course in uni, too much beer...

 

[jvk]

Ok, well here are some pics of the internals of my Mk III green. Please note that there seems to be one area of the circuit board that is burned (maybe a resistor burnt or something arc'd out???) Can someone please shed some light on what may have happened here or what problem(s) this may cause? The amp seems to be working flawlessly. thx.

Here's what that area should look like:

 

[Russ]

OUCH! :shock:

 

[Monsta-Tone]

:lol: :lol: :lol: :lol: :lol: :lol: :lol: :lol:

 

[stokes]

here's an example of an old Marshall (1970's) with original electrolytic caps:

http://www.vintageamps.com/plexiboard/viewtopic.php?t=61919

I read that the liquid like substance in the filter caps solidifies over time and a sudden bump in voltage/current from the AC may cause the filter cap to fail. So they may last long like this example, but it becomes a time bomb plus worse tone and/or noises, hum, etc.

Exactly! This is what I was told years ago when learning about amps,and still live by it.

 

[stokes]

The guy in that video just needs a new stabilizer in his drinking shoes.

 

[STUART]

Stokes thanks big time for your info much appreciated here..

I just replaced my Mark IV caps with 2 X F&T 220uF 300V and
3 X 30uF 500V
WAHOOOO!!!
Muchos greater improved bottom end and generally more dynamic sounding overall.
Although this amp was manufactured in 2001 it definitely needed the caps doing probably through lack of use with the original owner.
Visually the original caps looked perfectly fine with no leakage etc etc but hey looks can be deceiving...
Ghost notes now gone too !!

All the best !!

Stu.

 

[stokes]

6 years and the caps were showing signs of degrading already?Could be that when they were installed by Mesa,they were old already old,or they werent properly formed when they were installed.That is exactly why I say waiting more than 10 years to do a cap job is foolish,and to those who swear there is no reason to change 20 year old caps,you are just asking for a problem.

 

[tony777]

Are filter caps smaller in size nowadays? I bought some replacemnts from tubeampdoctor, for my simulclass, and the caps are smaller than the original ones. They sound better though, so it was worth doing.

 

[kdove]

My 30uf/500v were about the same size, but the 220uf/285v were smaller. Luckily this made them easier to replace. Mine sounded better too.

 

[6L6C]

Hey
To all of you who did cap‘s

Am I going insane? Or do you find it easier to control the volume of the amp at lower volumes.
At “bedroom levels” I always found myself adjusting the volume ever since the cap job just set it and forget it.

 

[kdove]

Seemed more stable to me ... in fact at low volumes is when I detected my problem. So I'd say yes.

 

[Octavarius]

Hi,

I received my '85 IIC+ a couple of days ago, and I want to replace the filter caps. However, I really know very little about electronics, so could someone here tell me which ones I'll have to replace?

 

[Daevid]

Although the Caps on my (1992) Studio 22+ look fine i do think it could benefit from changing them (i only looked at the Blue Caps)... what other caps do you think i should change?