Recapping old solid state amp

xerox · Post by **xerox** » 08 Feb 2009, 14:54

I brought back an old Hagström GA85 reverb from a friend of mine a few days ago. It sounds fairly ok, at least with clean tones. It doesn't seem to like high gain pedals though. To much nasty treble. Pots are fairly scrathy as well. But I am getting some unpleasant noise that seems to increase with the volume of the signal. Almost like a noisy signal with a soft noise gate on it. It could possibly be some kind of high frequency distortion too.

It's currently the only amp I have at home so I am prepared to spend some time and money on it if there are things that I can do that will make a change with a moderate amount of effort. I am thinking of recapping it, or at least one channel. That's only 20 something caps or so. How much of a difference would that do to the sound? It's something like 35-40 years old so the electrolytes are probably not in their best condition. On a few of them the the plastic cover is degraded and they seem to run quite hot.

I have fixed it one time before, about 5 years back or so. At that point I replaced the power transistors. The originals were ASZ15, I swapped them for NTE121. Don't really remember why but I think I consulted some equivalence chart. Yesterday I took a look at the datasheets however and found that asz15 had a hfe of 20-55 while nte121 had 60-165. And I also saw that the nte121s were available as a matched pair. Naturally I didn't bother about that back then, but am I right in thinking that matching them might be quite important? They are wired in a class AB fashion if I'm reading the schematic correctly.

TubeDude22 · Post by **TubeDude22** » 08 Feb 2009, 15:05

It would be a good idea to replace all the electrolytics in a SS amp that old. I have an Acoustic 150 from the 70's and replacing them with new caps made a big improvement in the sound. Capacitors should not run hot. Rplace the hot ones and be sure to replace all the electrolytic capacitors in the power supply circuit, too.

Regarding the output transistors, using a matched set couldn't hurt however, proper bias is a must. Like a tube, if they are biased cold, the amp will sound harsh and stiff.

Best of luck on the upgrade. Restoring that amp is the way to go.

xerox · Post by **xerox** » 08 Feb 2009, 15:19

Thanks for the answer, I'll definitely try replacing the caps.

Regarding bias, you don't have any handy link that explains how to adjust that?
I'm guessing that I would need to tweak the two 1,2k resistors, r12 and r13 but I have no idea in which direction and what to listen for.

Schematics are here btw, if anyone is interested, I found a pdf of them somewhere else but I can't seem to remember where.
http://www.hagstrom.org.uk/Schematics/H ... matics.htm (bottom of page)

xerox · Post by **xerox** » 08 Feb 2009, 17:36

Did some measurements and I'm getting a base-emitter voltage of 0.1 volt on one transistor and 0.04 volts on the other. That doesn't seem right, it should be more like 0.3 volts or so. Should I try to reduce the values of R12 and 13 to raise the b-e voltages?

TubeDude22 · Post by **TubeDude22** » 08 Feb 2009, 20:57

You can measure the voltage across the output cathode resistors to calculate the current. The voltage at the junction of the outputs and output coupling capacitor should be 50% of the power supply voltage. You can balance and bias the circuit by changing R12 and R13's values. Both resistors should be matched.

Good luck!

xerox · Post by **xerox** » 08 Feb 2009, 23:26

Thanks

The voltage at the junction is around 28 V while the supply voltage is 52 volts. Not exactly 50 % but fairly close.

I found this through google, http://www.geofex.com/Article_Folders/v ... oxprot.htm
http://www.schematicheaven.com/voxamps/ ... _v1081.pdf The Ga85s output stage is virtually identical to the vox's except it wired "upside down". However the vox uses two additional small resistors (r34 and 33) while the ga85 only seems to be using the dc resistance of the secondary windings (5,5 ohm measured) together with R12 and 13 to bias the transistors. Maybe I should add a really small resistor here 2-3 ohms and see what happens. How much is too much when it comes to biasing the transistors?
R. G. writes around 80 millivolts at the base for germanium. I realise that increasing the biasing will push the output stage further into class AB operation but how far can I safely go before things go south?

xerox · Post by **xerox** » 10 Feb 2009, 11:52

I am such an idiot. When I fixed this thing many years ago I had apparently connected the biasing resistors wrong, which is why the biasing current to one of the transistors was essentially nonexistent.

I soldered them the right way and it sounds much better now. It really sounded cold and lifeless before compared to now. Especially since I got the tremolo circuit working as well. Unfortunately it's a bit noisier now, but I'm hoping that can be fixed.

xerox · Post by **xerox** » 20 Apr 2009, 20:59

So I finally got round to recap this amp, spent some time on it yesterday and today, surprisingly unproblematic except for the fact that modern electrolytes are way, way smaller than their 40 year old counterparts.

I recapped one channel, PSU, reverb and the output stage. The amp has an additional channel for microphones and/or line inputs that I left as it was since I don't use it. And what about the sound, well I can't say that it was a dramatic improvement but clearly noticeable. I have a/b:ed the recapped channel against the one with the old caps as well and the recapped channel has a cleaner sound and has a nice sparkle in it while the other is murkier and seems to lack the highest treble frequencies completely. The occasional 50 Hz hum I was hearing has decreased significantly as well.

The amp is actually surprisingly quiet except for the reverb which is noisier than a freaking waterfall. The reverb does sound fairly good once I dug out the tank and removed the cloth that was rubbing on the springs. It has a pretty long double spring tank in the bottom of the cabinet. But the noiselevel renders the reverb useless.

Is there something that can be done about the reverb? It is not a very complicated circuit (schematic below) but I don't really see what is causing all the noise. Could one of the transistors be in bad shape? Or is it just the gain of all of them adding up?

http://www.hagstrom.org.uk/Schematics/S ... -44008.jpg

JiM · Post by **JiM** » 20 Apr 2009, 22:03

xerox wrote:Could one of the transistors be in bad shape?

That's the first time i see such schematic symbols for transistors !
It must really be vintage, for the symbols not being standardized yet !

xerox · Post by **xerox** » 20 Apr 2009, 22:27

JiM wrote:
xerox wrote:Could one of the transistors be in bad shape?
That's the first time i see such schematic symbols for transistors !
It must really be vintage, for the symbols not being standardized yet !

Or just some crazy Swede who drew it

No, but I guess it could be, it's from somewhere in the late sixties probably, but I don't know the exact year it was made.

xerox · Post by **xerox** » 11 Jul 2009, 23:52

Just a little bumb, I finally got round to uploading the pictures I took when working on it in case any one is interested in looking at it. http://www.flickr.com/photos/10853352@N ... 176851955/

royal jelly · Post by **royal jelly** » 22 Oct 2011, 01:14

Great stuff,I am also restoring an older ss amp a Tel-Ray supernova to be exact. Mine was buzzing really bad with ac ripple,and just had no volume or musical voice. I love ss amps,I build tube amps but I am now getting some favorite sounds with a Sunn Concert Lead and my diy dirt pedals Rat,Honey Bee clone,Bad Monkey.etc.I believe that ss amps take pedals better cause they dont color the sound as much. My Supernova Has really nice parts too!! Tropical fish poly caps. Ero-Vishay electrolytic signal caps and I couldnt believe that they would put 2 Altec Lansing Alnico speakers in a ss amp but they did and they sound nice.I will also post some pics here soon.Lets hear it for solid state mojo gut shots.