Gearslutz.com

All Advertisers
Go Back   Gearslutz.com > The Forums > Geekslutz forum


New Reply New Reply Thread Tools Search this Thread
Old 26th January 2009   #1
Lives for gear
 
Join Date: Mar 2007
Location: Pittsboro NC
Posts: 543

Thread Starter
old electrolytic caps LOOK fine but..

If the little electrolytic caps in my 1974 modules look good...no swelling or discoloration at all...no leaks...do they really have to be replaced? How do I check?

Or do I go ahead and replace them all as a matter of course?
__________________
Terry McInturff
President, Terry C. McInturff Guitars, Inc.
www.mcinturffguitars.com
www.thetcmforum.org
tcmzodiac@yahoo.com
919-742-2958
Terry McInturff is offline   Reply With Quote
Old 26th January 2009   #2
Gear maniac
 
Join Date: Oct 2004
Location: winter park
Posts: 278

check by running a signal generator through them. usually lose low end check so watch the roll off of all the channels and replace ones that have a higher Lf roll off. then run a 3kish square wave through and view output with scope, bad caps will have poor flatness in the square. (with drastic leading and trailing edges.)
amorris is offline   Reply With Quote
Old 27th January 2009   #3
Lives for gear
 
Join Date: Feb 2004
Location: beautiful Carlsbad, CA
Posts: 9,364

Replace them. 1974 caps are crap. Even if they work right, they are low quality. Replace them with low impedance el caps like Panasonic FM, Rubycon Z and Nichicon HE. Increase their values, especially on outputs and psu bypasses.

Jim Williams
Audio Upgrades
Jim Williams is offline   Reply With Quote
Old 31st January 2009   #4
Lives for gear
 
Join Date: Jul 2007
Posts: 1,013

The hell with these things. Everything I've got is >20 years old and they're probably all dried up. I'm trying to restore my cassettes and reel to reels now, which I figured would be mostly mechanical. I got the first deck (a decent consumer cassette) back to running like brand new, got the azimuth almost perfect with a nice loud playback at 10k. 6.3k is off by about a quarter of a dB. 315Hz is equally impressive, then I get to 63Hz and I can barely even hear the S.O.B. This of course has no playback EQ adjustment or anything. I go under the board with an ESR meter and everything looks too high. There's like 70 caps and it isn't even a big board. They're mostly the sub-10uF small ones. I find a pair which looks puffy at the top and replace those plus two nearby caps, suddenly 63Hz is only a couple dB below spec. Those particular caps were just barely past the play head in the schematic, but I know damn good and well the rest of the electrolytic population has to be doing something nasty to some signal somewhere.

Where do you guys draw the line, and do you normally have to sit there and feed something to every single cap to determine that? I can't imagine pulling everything in this one deck, and the larger rackmount ones look like a solid box of PCBs inside.

Quote:
Originally Posted by amorris View Post
check by running a signal generator through them. usually lose low end check so watch the roll off of all the channels and replace ones that have a higher Lf roll off. then run a 3kish square wave through and view output with scope, bad caps will have poor flatness in the square. (with drastic leading and trailing edges.)
Thanks for the tips there. So I'm guessing this is all safely done in-circuit? Also, do you have to determine a healthy LF rolloff point for each particular value of cap and/or a specific frequency for the square wave? And what sort of levels are you talking about running through them? Of course I'd just stick with the ESR meter if I can, but like I said, most everything looks high there and I'm not sure what's too high.

Much Thanks!
Jidis is online now   Reply With Quote
Old 31st January 2009   #5
Lives for gear
 
Join Date: Feb 2004
Location: beautiful Carlsbad, CA
Posts: 9,364

Quote:
Originally Posted by Jidis View Post
Where do you guys draw the line, and do you normally have to sit there and feed something to every single cap to determine that? I can't imagine pulling everything in this one deck, and the larger rackmount ones look like a solid box of PCBs inside.
Waste of time. They could all be pulled in 15 minutes with a proper tool. At 8 cents per cap, I would just replace all of them. The one you leave in is the one that will fail next. If you don't want to do this more than once, replace them all.

Jim Williams
Audio Upgrades
Jim Williams is offline   Reply With Quote
Old 31st January 2009   #6
Lives for gear
 
Join Date: Jul 2007
Posts: 1,013

Thanks Jim!

I actually have two desoldering stations thanks to Weller's refusal to stock old tips.

I'd still like to learn more about those checks mentioned though and have a couple more questions if you or anyone else doesn't mind:

- When you say you'd do all of them, does that hold the same for any unit you have (for instance if you were just tuning up a cassette machine like in my case)?

- On a shotgun million cap replacement, do I go with the same electro caps I've bought for more important stuff (Panasonic FM/Nichicon HE)?

Thanks Again,

George
Jidis is online now   Reply With Quote
Old 1st February 2009   #7
Lives for gear
 
ulysses's Avatar
 
Join Date: Mar 2003
Location: Minneapolis and Wiesbaden
Posts: 1,429

Yes, I agree that measuring old caps is a waste of time. Wholesale replacement is the only cost-effective approach. Even mediocre caps today have a lower impedance and higher capacitance per unit volume than anything made back in the 70s. Personally, I think that good caps (Panasonic FM or Nichicon HE) are cheap enough that there's no reason to bother with anything less, regardless of the application. High-temp long-life caps ensure you won't be redoing the job for a while. And by using new caps that have a significantly lower impedance and higher capacitance than the original, your new caps could last even longer since they can go further out of spec before they'll cause any degradation of the equipment's performance.

Of course, there are exceptions. Making caps arbitrarily larger than stock is a good rule of thumb but there's a limit. For most applications, that limit is inrush current - the most likely symptom being nuisance blowing of the mains fuse. But it could manifest itself in other ways too. And of course vintage circuits may have their sound in large part due to the fact that the higher-impedance caps tended to mask the unpleasantness of the amplifier's HF response. It's not unheard of for a recap of a vintage piece exposing some serious sonic weaknesses that were blissfully not present before the recap.

The other thing is that certain pieces of gear just aren't worth the effort. It had better be an awfully nice cassette deck.
__________________
Justin Ulysses Morse
Roll Music Systems
Minneapolis, MN

Put a bottle of juice in your Lunchbox.
ulysses is offline   Reply With Quote
Old 2nd February 2009   #8
Lives for gear
 
Join Date: Jul 2007
Posts: 1,013

Thanks Justin! Yeah, I decided the other night just to order a whole load of them and do all on as much stuff as I can. I had already forgotten my prices from the last order and was probably thinking of the larger cap costs. Stuff 10 cents and under should be no problem. Unfortunately, I have no idea how much of each to get. That first service manual I was looking at (the consumer Teac) seems to have been the only one which actually listed all the individual cap values. For the others, I'd have to go through each layout diagram and play find the caps. I'm open to any pointers on what percentage (how many) of each value I should order on the sub-1000uF stuff.

When you say "larger than stock", you're talking voltage rating right? That shouldn't be a problem either as I'm guessing current stuff will probably be smaller for the same values anyway. BTW- Is 5mm or whatever the standard lead spacing for most small ones?

Quote:
Originally Posted by ulysses View Post
certain pieces of gear just aren't worth the effort. It had better be an awfully nice cassette deck.
Probably not on that first one. It just has some sentimental value to me and it sounds decent when it works. Also has onboard dbx. The others are better (rack 122,122mk2,234,134B, plus a couple open reels).

Ooh ooh.... Forgot- The 1/4" Fostex appears to use surface mounts on the majority of the electrolytics. Do they croak at the same rate or do you guys leave them alone? If you swap them, I may need recommendations on what series to get there too. I'm probably going with Mouser (Nichicon) but may possibly do Digikey depending on what else I need.

This help has been much appreciated,

George
Jidis is online now   Reply With Quote
Old 2nd February 2009   #9
Lives for gear
 
ulysses's Avatar
 
Join Date: Mar 2003
Location: Minneapolis and Wiesbaden
Posts: 1,429

Quote:
Originally Posted by Jidis View Post
When you say "larger than stock", you're talking voltage rating right? That shouldn't be a problem either as I'm guessing current stuff will probably be smaller for the same values anyway. BTW- Is 5mm or whatever the standard lead spacing for most small ones?
No, the voltage rating should be the same. We're talking about increasing the capacitance. The amount of capacitance originally used was probably limited by cost, and by the amount of capacitance you could fit in a given space at that time. So for a unit built in the early 80s, you might have a radial-lead capacitor that has a 0.3" lead spacing, a 0.6" diameter and a length of 1". Maybe that was a 150µF cap. If you look in the catalog for Panasonic FM caps with the same voltage rating and the same lead spacing (which is important for getting them to sit flat on the board), you might find the smallest capacitance available is 470µF, and that's for one that's only 0.6" long. You might even be able to get 1000µF if you buy the one that's 1" long like the original. So you ask yourself "Wow, isn't that way too much capacitance for this circuit?" and the answer - often times, but not always - is "No, that'll give you a nice improvement in the bass response, and the new cap will last a lot longer than the old one did."


Quote:
Ooh ooh.... Forgot- The 1/4" Fostex appears to use surface mounts on the majority of the electrolytics. Do they croak at the same rate or do you guys leave them alone? If you swap them, I may need recommendations on what series to get there too. I'm probably going with Mouser (Nichicon) but may possibly do Digikey depending on what else I need.
I've never had to do a wholesale recap of a board that uses SMT electrolytics, but that day is fast approaching. For a long time, SMT electrolytics had a very bad reputation, not just in audio circles but all around because there were some really bad ones made, and some with really bad reliability problems. So there is surely a whole lot of older stuff out there with SMT caps that really needs to be redone. I don't know which specific varieties were the culprits. In theory, replacing the SMTs could be much easier than the through-hole stuff (since you don't have to suck all the solder out of a hole) but in practice I've found it difficult to reach the solder on an SMT electrolytic with any equipment I have (whether installing or removing them). So good luck with that! As far as replacements, you go for the same varieties as with the through-holes stuff.
ulysses is offline   Reply With Quote
Old 2nd February 2009   #10
Lives for gear
 
JohnRoberts's Avatar
 
Join Date: Nov 2006
Location: Hickory, MS
Posts: 1,946

I am an advocate of first do no harm, and not fixing things that aren't broken.

If the original design was conservative it's very possible the caps will still be adequate after a couple decades. Also if you are inexperienced with soldering, using the wrong tools, and working on single sided circuit boards you could lift traces and do more damage than good.

I like the idea of running sweeps and looking for outliers. If you identify more than a handful of bad capacitors then replace them all, but if all channels measure similarly, and meet original response specifications, I would not consider this automatic.

JR
__________________
John Roberts
www.CircularScience.com
JohnRoberts is offline   Reply With Quote
Old 3rd February 2009   #11
Lives for gear
 
Join Date: Apr 2003
Location: Australia
Posts: 1,315

Does anyone use low ESR electrolytics for audio replacement? I use them when I repair switchmode supplies, I always wondered what effect they would have if used in audio circuits.
triez is offline   Reply With Quote
Old 3rd February 2009   #12
Lives for gear
 
JohnRoberts's Avatar
 
Join Date: Nov 2006
Location: Hickory, MS
Posts: 1,946

Quote:
Originally Posted by triez View Post
Does anyone use low ESR electrolytics for audio replacement? I use them when I repair switchmode supplies, I always wondered what effect they would have if used in audio circuits.
You can pretty much calculate the difference from the data sheets. Add the ESR and ESL in series with cap as if they were independant components.

In the simple example of a DC blocking cap, with say a 22 uf and 100k to ground would nominally be a .072343Hz HPF, but with 10 ohms of ESR that pole would shift to .072335 Hz.. In fact the calculation is nonsense since the ESR is within the tolerance range of most resistors and surely any cap you would use.

IMO the only place in audio where this might make a difference is in power supply decoupling for stiffer rails, and perhaps cheap passive loudspeaker crossovers (ugh).

I generally try not to use electrolytic capacitors to make audible frequency poles. Since changing terminal voltage across electrolytics can cause other nonlinearities to express (voltage coefficient, dielectric absorption, etc). So set your poles adequately low and relax.

JR
JohnRoberts is offline   Reply With Quote
Old 5th February 2009   #13
Gear maniac
 
Join Date: Feb 2008
Location: Huntsville
Posts: 174

Quote:
Originally Posted by Terry McInturff View Post
If the little electrolytic caps in my 1974 modules look good...no swelling or discoloration at all...no leaks...do they really have to be replaced? How do I check?

Or do I go ahead and replace them all as a matter of course?
Terry,

the best way would be to check them with and LCR meter to see if their values have drifted. when I am building an amp I am very meticulous about cap values and types I am also fussy with resistors too!! if you have any questions contact me..


Bill
bibleboy3 is offline   Reply With Quote
Old 5th February 2009   #14
Lives for gear
 
Join Date: Feb 2004
Location: beautiful Carlsbad, CA
Posts: 9,364

A LCR or cap meter is a good way to check a cap, if it's removed first from the circuit. If left in, readings can be skewed by ajacent components, like a resistor across the cap's output as a bias discharge resistor. Impedance loads around the caps will also change readings. Solder pad corrosion will also skew readings unless you bury the probe into the metal first.

1974 caps were crap. I said it, they're CRAP! When they were new they were crap.

Capacitor technology has improved as much as PC's have over say a Commador 64 or 8088's. With the low cost of these caps, typically 8 cents at 100 pieces and the availability of low cost high speed component removal tools like the Hakko 808, dicking around with 30+ year old caps is more than a waste of time, it's futile as the cap you leave in is the next one to fail. Newer cap fomulations will improve performance, that's enough of a reason to replace every one of them.

Jim Williams
Audio Upgrades
Jim Williams is offline   Reply With Quote
New Reply New Reply Submit Thread to Facebook Facebook  Submit Thread to Twitter Twitter  Submit Thread to LinkedIn LinkedIn 



Thread Tools Search this Thread
Search this Thread:

Advanced Search

Similar Threads
Thread Thread starter Forum Replies Last Post
Replace Decoupling Caps as Well as Coupling Caps on MCI538C? dbbubba Geekslutz forum 1 27th July 2008 06:30 PM
Source for Electrolytic Caps dbbubba So much gear, so little time! 2 9th April 2008 11:53 PM
Is there an alternative to electrolytic caps ? gainreduction Geekslutz forum 7 25th November 2006 01:01 AM
Nichicon Muse KZ Audio Electrolytic Caps Westlake HR1 Geekslutz forum 1 22nd November 2005 09:36 PM
Using electrolytic capacitors instead of tantulum? carbon Geekslutz forum 4 20th March 2005 02:39 AM


All times are GMT +1. The time now is 12:55 AM.

 
 
Powered by vBulletin®
Gearslutz.com Limited - UK Company Number 7597610.
Registered Office: 35 Ballards Lane, London, N3 1XW.

SEO by vBSEO ©2010, Crawlability, Inc.