• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Fisher 500B OPT/Feedback Questions

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
A while back , I bought the iron from a Fisher 500B on Ebay (power and output XFMRs). I plan to build a new 30-35W stereo power amp using the iron and a nice quad of EH 7591s. I'm not sure about the primary impedance of the OPTs. I wanted to know this so I could plan out the gain of the amplifier, as I'm going open loop for my first try. Does anyone know the primary impedance off the top of their heads, or will I have to torture the transformers with a sig gen to make them reveal their secrets?

I was looking at the schematic of the 500B for inspiration, and I noticed that the designer returned the cathode side of the phase splitter to the summing node along with the classic feedback network. Can anyone wih more vacuum stte experience than me explain the rationale behind this? Are we talking nested feedback loops here?

What I'm thinking of trying first is a simple open loop setup using a 6CM7 as preamp and splitter, with the high gain triode as a common cathode amp with solid state current source loading, followed by a classic "concertina" style phase splitter using the low gain triode. Next would be a feedback approach using a folded cascode tube/SS hybrid as the first gain stage, and the same splitter.
 
I've been following the other 7591 threads, and have gleaned some ineresting stuff from them. Still looking for answers to my other questions, though. If I'm forced to scope out the primary impedance of my output XFMRs, though, I'll post the information here.
 
You can measure the ratios with an ordinary voltmeter, albeit not without some danger and thrills. Measure your line voltage. Connect a line cord to the primary, then connect a voltmeter to the secondary. CAREFULLY plug it in and note the voltage reading. Unplug, then break out your calculator.
 
I'm not crazy enough to apply line voltage to my precious audio iron, even through a variac. A sig gen at a few kHz and a scope will work just fine, and won't zorch anything. At a few kHz, the loading effect of the transformer magnetizing current shouldn't be an issue. If I was paranoid about this I could also gin up a buffer to go between the sig gen and the transformer, but I'll try the simple approach first.
BTW - thanks for the information - that was pretty much what I seeking. Now about the 500B phase splitter feedback to the summing node...
 
I don't have a print handy, and I also have intermittent Rumzheimers, but it seems that Fisher would often stand a concertina splitter on an asymmetrically large cathode resistor, put a higher positive voltage on the grid with a voltage divider from B+ to grid to cathode resistor and balance out the AC signal with a variable pot/resistor combo at the plate.

So they could operate the first stage at a higher plate voltage and more linear operating point and still direct couple it to the concertina.

I think I remember that..... don't I?
 
Wait, my brain is still on phone modem....loading...

Maybe the concertina had a self bias resistor and the voltage divider went to the junction of that and a larger value resistor from there to ground.

Whatever they did, it was to operate the first stage at a higher than usual plate voltage and use direct coupling to the concertina.

Maybe stage one would be a good place for mu-follower with that trick. :idea:
 
Maybe we're talking about something else. I've got a 400 and I've had Fishers, here and there, in for repair, including 500s. And I don't remember anything that brought global feedback back to the splitter. That would probably have caught my attention. Or, maybe not.

I think the coolest thing that could happen here, is a schematic graphics program, so people could draw it and put it in with the post. Instead of draw, scan, file and send to image host, etc.

7591 is actually even easier to drive than 6V6. An EL-84 front end with some headroom would do it.
 
Here is a copy of the input schematic for the 500B power amp section, so you can see what I was talking about with the phase splitter being returned to the summing node of the input stage.
 

Attachments

  • 500b.gif
    500b.gif
    10.2 KB · Views: 175
Well, I couldn't find a print, so I looked inside my 400. Similar to the 500. And see what I found!

An externally hosted image should be here but it was not working when we last tested it.


The negative feedback loop is, sure enough, applied to both the gain stage and the splitter. And since both cathodes are drawing through the same 220 ohm resistor, there is some positive feedback as well. This acts to increase the gain of the first stage.

That's a little more complex than I remembered it. It is a dangerous critter but it sounds good.
 
Seemed to have misplaced the edit button. It was there a minute ago.

I forgot to note that the 400 uses a 12ax7 in this circuit. Also, I left out an 18 pf cap to ground after the .047 output cap in the cathode leg of the splitter. For hf balance.

And there is a 330 pf cap to ground at the tx secondary side of the 2.7k feedback resistor. Hf comp shaping? Rf trap on speaker wire?
 
I can't place that venue, but I remember being there a long time ago. Pirates World?

Actually The Strat is a Squire model, cheap but effective, and the Les Paul is an Epiphone. I can't play well enough to justify expensive guitars any more. Have an old Guild and a Fender Coronado that needs some serious help though.
 
To answer Eli's earlier question, I plan to run a preamp with a fairly mild amout of line input gain (~5) in front of the tube amp. I'll be wanting a power amp sensitivity of 1-2V input for full output.
I've been researching sand state open loop preamps using JFETs for about a year now. Anyone interested can find the threads by searching under my user name. I eventually plan to do a vacuum state preamp for the heck of it, but for now, any tube power amp I build will be driven by my current solid state setup (whichever mutation is sitting in my living room at the time).

Since I already have the models, I've been thinking of simulating the Fisher amplifier/splitter feedback setup using JFETS to understand a bit better what is going on. The insights gained should still be applicable to tubes with a bit of grunting and straining.

It would not be too far a stretch to take the JFET circuit from that exercise and use it to drive a pair of MOSFETs in push-pull using an output transformer. The operating voltages and primary impedance required would be very different than what is required for the 7591s, but it still might be possible to find some suitable stock iron. It would be interesting to see how much juice one could squeeze out of a pair of IRF820s, and what the whole mess would sound like. I can't recall anyone who has tried something like this, but maybe some of you all out there may know something about a similar effort. One thing at a time,, though...
 
tubelab.com said:
I can't place that venue, but I remember being there a long time ago. Pirates World?

Actually The Strat is a Squire model, cheap but effective, and the Les Paul is an Epiphone. I can't play well enough to justify expensive guitars any more. Have an old Guild and a Fender Coronado that needs some serious help though.

Daytona. Bandshell on the beach. Miller Beer, Hawaiian Tropic and a radio station used to throw big bashes and we'd furnish the square dance tunes.

I was in a bunch of FL bands even though I'm from NC. (Don't remember how that happened, all I wanted was some mushrooms.) ;)
 
wrenchone said:
To answer Eli's earlier question, I plan to run a preamp with a fairly mild amout of line input gain (~5) in front of the tube amp. I'll be wanting a power amp sensitivity of 1-2V input for full output.
I've been researching sand state open loop preamps using JFETs for about a year now. Anyone interested can find the threads by searching under my user name. I eventually plan to do a vacuum state preamp for the heck of it, but for now, any tube power amp I build will be driven by my current solid state setup (whichever mutation is sitting in my living room at the time).

Since I already have the models, I've been thinking of simulating the Fisher amplifier/splitter feedback setup using JFETS to understand a bit better what is going on. The insights gained should still be applicable to tubes with a bit of grunting and straining.

It would not be too far a stretch to take the JFET circuit from that exercise and use it to drive a pair of MOSFETs in push-pull using an output transformer. The operating voltages and primary impedance required would be very different than what is required for the 7591s, but it still might be possible to find some suitable stock iron. It would be interesting to see how much juice one could squeeze out of a pair of IRF820s, and what the whole mess would sound like. I can't recall anyone who has tried something like this, but maybe some of you all out there may know something about a similar effort. One thing at a time,, though...

You are ambitious. Good luck.

(With my solid state chops, I'd watch many a cloud of smoke roll up to the ceiling!) :bawling:
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.