FRED Diodes For A Kustom 200B?

[J M Fahey]

Hi earache.
What you posted is very interesting, because it narrows the area problem *a lot*, to begin with that hiss does not come from the front end, even the active tone control itself, but later on.
For all I know now, you might even have a leaky transistor or capacitor or poor solder joint/bad contact in the power amp itself.
Now it's 2AM in Buenos Aires and my neurones are going on strike; tomorrow I'll open and take a long hard look at that preamp and power amp schematics.
Just by heart: the power amp had a couple 33k or 47k resistors at the input of the power amp. Do the preamp boards outputs go straight there or there is another intermediate stage? I don't want to bypass any suspect. We'll talk tomorrow night.

[Eagle]

The Kustoms were rather noisy right from the factory. Hiss was pretty much a part of the sound. The sputter/crackle, however, is caused by aged components.

Replace the output transistors and the drivers. That will get rid of your sputtering and most of the hiss. If you want to go a step further, replace the tantalums also. IIRC, the outputs are 2N3055 in a TO3 package.

I've done quite a few of these and the output transistors are almost always the main culprit of the noise you get.

/

[J M Fahey]

Hi earache.
From what you tested I have almost ruled out the preamps themselves as prime suspects.
I am working with the PC105 preamp posted here and the PC703 Power amp + regulator.
I also think that the blue wire on the right of PC105 (preamp output) and the blue wire(s) on the left of PC103 (Power Amp) are the same, no other elements between them. Do we agree?
In that case, the main suspect is channel mixer Q700.
Tests:
1) with a piece of wire short (soldering to pads) R707. hiss dissapears?
2) undo (1) and lift one end of R705. hiss dissapears?
3) If not , we're up the creek without a paddle.
4) if yes, replace Q700; also check or touch up soldering around it, including all pins of R703/4/6
Following another line of thought:
lift one by one the blue wires arriving to R700 and R701.
If hiss stops, the big suspects are C113 on preamp boards. Replace them. Regular good electrolytics will do, no Solens or Mica or Paper Oil or Orange Drops or N.O.S. or whatever magic components are needed, just regular good commercial quality will do.
Another line, as a last resort only:
solder a couple 100uFx16V from +8 and -8 lines to ground, observing proper polarity.
Either (or both) your + and - 8V supplies mught be oscillating or very dirty.
NOTE: ALL tests follow this sequence: turn amp off, wait for discharge , mod/change whatever was asked, turn amp on and hear/measure/see/whatever, turn amp off for any soldering/desoldering/etc..
Good luck.

[earache]

Will Continue Research/Repair Work Tomorrow

Hi J M Fahey
I won't be able to do any further work on that amp until tomorrow (Sunday the 16th), but am anxious to try the tests that you have specified.

Thanks again for the time and attention. You say you are in Buenos Aires? The internet has sure made the world a more interesting place!

[earache]

MUCH Quieter Now!

J M Fahey - I replaced Q700 per your suggestions and the sputtering hiss disappeared. What a difference! I felt like I was on a roll so I also replaced the(2) 2N3055 transistors on the board and reduced the hiss even further. Great directions and insight, thanks so much for your help!

[uvacom]

I was under the assumption that FREDs were used to simulate tube rectifier "sag", not as a means to improve noise performance. Is this incorrect?

[J M Fahey]

Hi earache, glad to see you found it.
Keep those original 2N3055, I trust them much more than any modern made one.
No doubt noise improved, but probably because of resoldering and/or contact auto-cleaning when you replaced them.
Good luck.

[Steve Conner]

Hi earache, glad to see you found it.
Keep those original 2N3055, I trust them much more than any modern made one.

Do they even make them any more?

I once heard a story about one of these old 2N3055 amps that survived a shorted speaker for 24 hours in spite of having no protection circuitry at all.

As far as I know, the "secret" was that the beta of the 2N3055 fell off so quickly with increasing collector current, that they practically had a current limiter built in. And I agree that modern ones will probably have beta that holds up better, which will make them less rugged.

[Enzo]

On the other hand, today, I am more likely to reach in the drawer for a handful of MJ15024, instead of hunting down more 2N3055.

[Steve Conner]

Well, I know who I'd be betting on in Celebrity Transistor Deathmatch. The MJ15024 would stomp all over the poor 2N3055.

But the NJL4281 could hang them both out to dry.

[earache]

Source Of FRED Idea

The initial reason I asked about FRED diodes is that I read the following post from Stevem on Welcome to VintageKustom.com!!!
"Making a bridge out of fast recovery diodes(freds) will take some switching hash ( noise) out of the amps idle noise floor that is mistaken for hiss."

Found at: VintageKustom.com: Replacement Part Suggestions => Metal Film Replacement Resistors

I am by no means an electronics engineer but just another guitar player and gadget tweaker who looks for info on the internet. I am grateful to anyone who takes the time to share their info or opinions.

J M Fahey - I misspoke in my post when I said I replaced the 2N3055 transistors. I actually used 2N3055's to replace old RCA 36892 transistors. The noise when the amp is turned up full was reduced with this replacement.

Steve Connor - those 2N3055 trannys are still available at local Radio Shack stores here in the States.

Thanks again to everyone for their input.

[Steve Conner]

Trying to stay on topic:

I've built hi-fi and guitar amps with both FREDs and regular diodes, and never noticed much of a difference.

I don't believe claims of hiss reduction. Reason: The diodes produce a burst of hash each time they turn off. Therefore, any noise they generate is modulated at twice the AC line frequency, so if it were getting into the audio, it should sound like a 120Hz buzz or sizzle, not a hiss.

If you're interested, you should be able to pick up the diode switching noise from regular diodes in a power amp, by using an old-school medium wave portable radio. When it's held near the rectifier or the power cord, it should pick up a 120Hz buzz, especially when the amp is loaded and the diodes passing lots of current. Installing FREDs should make the noise all but vanish.

Because FREDs are more expensive, commercial designs just use regular diodes and connect a R-C snubber across each one to absorb the noise. Except in switching power supplies, where FREDs are mandatory, because regular diodes would melt from the switching losses.

The true audiophile might like to check out the new silicon carbide and gallium arsenide schottky rectifiers :-) Nelson Pass already has an amp on the market made from SiC JFETs.

[earache]

Now Amp Has Less Volume

Hi Since I have fixed the hiss on this amp I have only recently tried it out at a high volume. I found that it has much less volume than before.

The transistor that I used to replace Q700 (which was a SE4002) was a 2N2222. Could this be the reason that the output is lower? The substitution list I found on "www.vintagekustoms.com" just called for an "NPN transistor" so I used what I had on hand.

[Jazz P Bass]

SE2002

SE2002
Specification
Si NPN Lo-Pwr BJT
V(BR)CEO (V)=25
V(BR)CBO (V)=30
Absolute Max. Power Diss. (W)=200m
I(CBO) Max. (A)=.20u
f(T) Min. (Hz) Transition Freq=60M
C(obo) (Max) (F)=4p
Package=TO-222AB
MHz: <1 MHz
Volts: 25V
2N2222 is probably overkill. It should work fine.
Are you sure that you got the leads in right.

[J M Fahey]

Hi.
I agree with Jazz P Bass and add: It should work, the 2N2222 is as good or better than the original one, but maybe the pinout is off.
Anyway, there's a sure way to know, without unsoldering: just measure that those 0.6V and 4.4V (+/- 20%) are there, that's to say: on the node (junction) of R700/701/703/704 (to avoid referring to transistor pins which can be misinterpreted) and on the node C701/R 703/705/706
If all measures well, that stage is fine.