diyAudio (
-   Solid State (
-   -   Nakamichi PA-1 AMP Time to replace main caps? (

gto127 27th January 2013 02:27 AM

Nakamichi PA-1 AMP Time to replace main caps?
I've noticed my PA1 is not as open sounding as it used to be & confirmed this with another PA1 for comparison. I measured my main power caps which are supposed to be 15Kuf at 12.8Kuf. The other amp was at 14Kuf. I really don't want to buy new ones if I can help it due to very limited availability of high end large power caps. (I'm afraid of the ones from China on ebay). I know I could put a 2200UF of same voltage in parallell to be back at original spec but will this just patch the problem temporarily or would it be OK. I'm not hearing any hum. Just hoping I don't have to pay ridicolous prices for 15Kuf63v pair. Anyone with experience doing this please let me know.

Fast Eddie D 27th January 2013 05:53 AM

Replace the caps with a parallel array of smaller caps. Construct a sisterboard (thanks Goatee :D ) that fits somewhere close to the poweramp. (Or just wire it right where the old caps are.) You could us a 10x2200 uF array on each rail. You get a nice discount from places like Mouser if you order 25 or more. Look it up; it might be cheaper than the big boys and I bet it works better too.

Fast Eddie D 27th January 2013 05:56 AM

Also look at coupling caps, feedback network caps, and any local power supply bypass caps too. Electrolytics go bad all the time in older equipment. You might as well replace them all and be done with it.

indianajo 27th January 2013 11:51 AM

If your amp is 20 years old, the rubber seal probably won't stop leaking. Depends on the service life of the cap as designed, that is what sealant they used. But a decrease of 15% probably doesn't predict a stop right here. And cap slime all over makes for a bad day, not to mention what a short can do to bridge rectifier or the transformer if the rectifier is not a regulatory rated one with internal fuse.
In my 1994 PV-1.3k I had no cap symptoms, but since I was in there for burnout, and it was 18 years since it left the factory, I replaced every electrolytic cap, including the 10kuf main pair. I used 3x 3300 uf 3000 hour service life rated ones, since they were $5 each instead of $28 for the shorter 10kuf caps available at distributors (newark or mouser). The 3300 were made in C**** also, but since they were from a distributor from a mainline company maybe the QA people get audited by people from the home office.
Since PA amps get thrown around, I made a sandwich of polycarbonate plastic held together with 2" #6 screws, and screwed through the old cap board with washers to keep from breaking the PWB. I used elastic stop nuts, which shouldn't come unscrewed. The 3 terminals caps weren't available from distributors, anyway; they were a Peavey only item, which meant a long distance phone call at the pay phone and another freight charge, too. I drilled for the the snap in terminals through the plastic and bussed 18 ga wire across the 3 caps and down to the old PWB. I then put some silicon caulk around the edges of the top of the caps to keep them from sliding around.
No board etch price, the plastic was left over from a car window project. You cut it with a hacksaw and drill it with normel twist drills. The screws were surplus, too. If you have to buy 2" screws, a bag is $5 usually plus freight.
Works fine.

east electronics 27th January 2013 11:59 AM

you wished that this can be done so easily ....

Dont you think that if this aspect was correct other manufacturers will also adopt it ???

Now days that amplifiers are constructed by robots it could be by far too easy for a robot to drop 10 caps 2200uf each rather than a big one of 22.000 that will require other type of handling harder soldering and so on .

well doesnt actually work like that smaller capacitors are faster obviously have lower ESR BUT !!! implementation can be very tricky , ground and rail traces are getting way too longer while close to the rectifier , and probably closer to other dirty things like transformer ...Work like that and you can kiss your real star ground goodby !!!

Farther more real decoupling becomes less effective ...

It could be done only in cases that the all circuit is designed from scratch like that and all the variables i mention above are taken care off ...

Advice is to first change all the small capacitors first to see if the problem originates from there ...from experience i can tell that capacitors that measure style of 25% less than stated will not behave as you say ..When the amp is working at high power distortion an clip will appear faster and high's will loose some brightness in high power ..In low power you shouldn't be able to tell the difference

kind regards

gto127 27th January 2013 06:51 PM

I would like to try the array of caps in parallell but unfortanetly the 2 large caps hold up a small circuit board. The caps are held down by collar mounted to chasis. It would be very difficult to mount cap array. That's why I had considered putting 1 in parallell with already mounted caps. Will this hurt anything putting a new cap in parallel with an old one? @sakis-I have already replaced quite a few of the smaller caps but not all. I may try replacing the rest in front 2 channels(it's a multi chnnel amp) before buying or altering config with big ones.

Fast Eddie D 28th January 2013 02:49 AM

Actually, your cap measurements are within the typical tolerance for big boys like that. Typical tolerance is -20% to +50%.

Don't overlook the smaller electrolytics on the boards.

Fast Eddie D 28th January 2013 03:57 AM

With regard to the cap array boards, consider this.

You can get small "L" brackets at any hardware store for cheap. Or you can make brackets out of a small quantity of scrap 20 gauge sheet metal. Bolt or solder the brackets to the array board. Do not connect the brackets to the ground on the array board in any way. Then bolt the brackets to the chassis.

Use at least 18 gauge hookup wire if possible. 16 is better. Drill the boards if you can to accept heavier gauge leads. Solder 18 gauge wire (or 16 gauge if you like) to the array board traces. Or just use a plain board and hookup wire.

Stuff as many capacitors as you can on the array boards. Remember you can stuff both sides. The more the merrier, up to a point. I think you get a better discount if you buy 50+. You know you'll use them all eventually.

You can get small boards at rat shack and on the web. Trim them to fit, attach and fit the brackets, then stuff.

How much capacitance is too much? There are caveats to increasing capacitance.

1- You increase the duration of the charging current time and maybe the peak charging current too, which can stress and shorten the lifespan of the transformer and rectifiers. If the rectifiers are marginal to begin with, it can drastically shorten their lifespan. ;) Transformers not so much.

2 - You can increase the peak current capacity and duration ("headroom") to the extent that the output devices are driven outside of their safe operating area.

What guidelines can we use? Specified capacitance tolerance is +50%, so no problem there. In practical terms 100% increase is no problem. In marginal designs I have gone +200% with no problems and a very noticeable improvement in bass and overall sound. And if it blows up so what? It's cheap consumer junk. But for a nice amp +50% is fine and +100% is OK too.

All times are GMT. The time now is 08:09 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio