Bias set point for 8417s

[Prairie Dawg]

Folks, I have just resurrected a strange amp. It's a SG 410, which is a 100w hybrid production (ss front end, tube rear end) of CMI that was made back in the seventies in beautiful downtown metropolitan El Monte, California.. After a significant amount of screwing around I've got it up and running for the first time in thirty years, and it occurs to me that I have no idea what the bias set point on the power tubes should be. They're strange ones, 8417s, and they cost a fistful of cash to get a NOS RCA pair that the seller alludes are matched-about $200.

I got to find out why it had ended up in a storage unit for thirty years, and it was a torched grid resistor and fried tube socket. So that's fixed. A general freshening up of the power supply and bias network electrolytics has put it into working order, at least far enough that I can now proceed to see what else is kaput with it. Seems the owner gigged with it back in the day, life got in the way, into the storage locker it went and there it stayed until he had the same thought I had: "Better get this guitar playing sheeyit worked out because you aren't getting any younger, fool!".

The plate voltage measures about 624v with the bias at about 30 ma. It was a lot higher and I knocked it back for survival's sake while I consult the brain trust-that's you folks.

Where do I start? A number or a set of rules that lets me figure out the number would be great. I know it has something to do with Ohms' Law, too.

The 8417 data sheet mentions -15v when running at about 550volts on the plates but this thing's a monster.

Thanks.

[Gtr_tech]

Well according to the datsheet, they have a 35w plate. So 21w static diss is a conservative 60% setting. So whatever bias voltage gets you there...

But, I'd do some research first on how to insure these tubes remain stable. They had a bad rep for be'n prone to oscillation due to the higher than normal amplification factor (see datasheet). I'm sure there are some tips out there to help keep these from running away and melting down.

http://frank.pocnet.net/sheets/168/8/8417.pdf

I think of these as bigass 7591s...

[Prairie Dawg]

Thanks for the tip-that is one thing I'm going to address. These tubes do not enjoy a reputation for rock solid reliability. Running them at 624 v seems like pressing the envelope. It was suggested early on to rebuild the bias network and go to 6550s but as this is a hybrid amp with all the bias stuff on a pc board it would be a major task. I did the next best thing which was to get a pair of matched NOS RCA tubes from a reputable vendor.

I see you are from Owosso. Do you know Jim Clark? I used to live down near K'zoo.

[Gtr_tech]

Bias supply, yea you'd need way more for 6550s. Maybe consider KT88s for that kinda plate supply. You'd also need to change the PI to provide more drive since a high transconductance tube like the 8417 didn't require a ton of grid drive signal.

Hmm....can't say that I know a Jim Clark.

[Prairie Dawg]

I wouldn't want to try adapting this to any other tube unless it was a complete gut and rebuild. I myself would look upon it as a source of iron-nice big transformers-and maybe speakers. But the yguy's attached to it.

I knew Jim when I was in the Michigan Antique Radio Club several lifetimes ago when I was actively collecting and screwing around with radio gear.

[Enzo]

I don't recall that name either, but I am in Lansing.

I always thought those old SGs were cool enough amps. You can get a hundred watts out of a pair of those tubes. And in my limited experience with them, they seem to last a lifetime.


I actually have a half dozen of those tubes in my stock. No I am not interested in selling them. Saving them for these SG amps.

The book says 600v B+ and their spec says use a variac to set teh B+ right at 600v on pin 3, then set bias to -16.5v on pin 5..\ If mains control is not available, by the book, the following chart is offered.

Plate volts - grid volts
680v -19.5v
630v -18v
600v -16.5v
570v -16v

Hey, I'm just reporting what they said...

Their test is to bring the amp up and the mains draw ought to be up to 700ma in operate (no signal), and 400ma in standby (no signal). At full output, just at clipping, 1.54A.

The bass amp is similar.

[Prairie Dawg]

Thanks for the update Enzo. I had to bite the bullet for the owner to get a set of decent tubes. They're not matched as well as I might like but they are within 10 ma so they are usable RCA stuff made by Sylvania.

I went at it from the other end by starting with the maximum plate dissipation from the Sylvania data sheet and worked from there to come up with a maximum plate and screen current together and a 60 per cent figure which works out to about 34 ma. As I have a Kill-a-watt watching this thing it ought to be easy enough to see what kind of current it draws at no signal.

Did I say the plate voltage is ferocious? 624 volts measured.

[Enzo]

Well, their chart goes to 680, so they were ready for you.

[Prairie Dawg]

Interestingly enough the factory put some lacquer on the bias pot and at 34 ma for the highest tube the lacquer on the pot lines up to a T.

[Prairie Dawg]

Well, their chart goes to 680, so they were ready for you.

Enzo, do you have any service information on these amps you can copy me on or point me to? It's working well but the phaser quit on me and the reverb barely works at all. I need to troubleshoot them. Thanks again.

[Enzo]

The schematic you have, right? The phaser is described as a complex circuit, no more detail than that. The schematic shows it as just a box. The troubleshooting guide is the sort of information that might come in the owner manual.

What kind of pan is the reverb? Shake the amp to crash the springs, is it loud or real weak? or pull the return plug from the pan and youch it, is the hum loud? The return is just an amp channel, you can feed a signal into it for test. Likewise, see if the signal leaving the reverb drive is weak of OK. The pan itself could be bad, especially if it is one of those godawful piezo things.

As far as I can see, all the amp circuits run on single sided +30vDC, but the phase needs +35 and -35. Got both? The reverb drive needs +12 and -12. I'd wager that if its -12 was missing, it might still function, but would be weak.