Fisher 70A design vs Fisher 30A design

[AnyDayNow]

Sam Cogley: Yes, I plan to review your series of posts on the 70A amplifiers before I tear into them. Right now I’m just trying to understand the circuits before I do anything. I didn’t realize that available myself of some kits that include circuit boards for replacing the power capacitors, as well as coupling caps and a thermistor to get me started.

My amps did not include the Z-Magic control itself. From the reviews I’ve read, those that have tried them day the amp sound better without them. So I plan on following your suggestion.

ZacktheDog: I checked the schematics on the 80AZ, and they are similar but different, especially at the output section. The 80AZ is rated at 30 Watts; the 70AZ is rated for 25 Watts. The 80AZ uses EL37 tubes at the output section while the 70AZ uses 5881 (or 6L6). The 80AZ includes a 4 Ohm speaker tap, which the 70AZ lacks.
Finally, the output transformer is very different, it lacks the ultralinear taps noted by DirtyErnie; it has the cathode winding like the 70AZ, but the upper half of the primary side is a simple center-tapped portion like the Fisher 30A I’m getting my “Elmer-ducation” from. The 80AZ also has a higher component count, and not all of it has to do with the ‘peak power’ indicator. Although I notice that it includes a fuse, which I may add to my 70AZ. So, although they may look similar, there is good reason to say they may sound different. I guess I’m trusting in the Fisher brand to deliver the sound my Fisher 80T tuner did back when I was in high school.

Thanks again to everyone who has responded to this thread. I’ll keep you posted on developments, especially if I have anything to add to Sam Cogley’s journal.

 

[gadget73]

The 80AZ output stage is a pentode design with cathode feedback. Screens are fed DC from the power supply rather than from a tap on the transformer.

Driver circuitry is pretty much the same, though I didn't compare all the values. Same topology at least.

 

[GordonW]

yeah UL and cathode feedback are an unusual combo for sure. Honestly don't see a ton of cathode feedback amps but I don't think I'm aware of any others that do both.

currently listening to a pentode amp with cathode feedback. Completely different front end arrangement though.

Yes, and it makes sense that it needs the additional gain of 12AT7s (vs. the usual 12AU7/6SN7 type tubes used in most WIlliamson designs) in the Fisher 70 amps, for enough open-loop gain, having both UL and cathode feedback. That takes a substantial drive voltage and gain in the driver stage, to drive output tubes with that little gain once UL and cathode feedback are both applied!

Regards,
Gordon.

 

[Sam Cogley]

The 70A-2 and 70AZ are the "boxcar" form factor?

Yep. The 70A was the only one made on the more-square aluminum chassis with the tubes in a line in front of the transformers. The 70-A-2, 70-AZ, 80AZ, and 100 all used the same plated steel “boxcar” chassis. The 70-AZ had the Z-Matic control on a plug-in remote cable, the 80AZ and 100 had a faceplate bolted on with the control knob and a cage wrapped over at least part of it. The apparent intent was for the 70AZ (or a 70A/70-A-2 with the retrofit kit) to be installed inside of a cabinet, the Z-Matic control had a little brass faceplate that could be mounted next to the tuner / preamp. I guess Fisher realized that some people just wanted to put their amp on a shelf, or didn’t need to tinker with the variable damping, and put the faceplate on later amps so the control was integrated. The only post 70-AZ Fisher amps to retain such an industrial look were the ones designed for their integrated console systems.

 

[gadget73]

For a long time hifi gear lived in a cabinet, largely out of sight so I guess that isn't too surprising. I guess they got the message when that started to change and made the later amps a little nicer looking. That is one down side of many console pulls, they just aren't very attractive in stock form.

 

[zackthedog]

Sam Cogley: Yes, I plan to review your series of posts on the 70A amplifiers before I tear into them. Right now I’m just trying to understand the circuits before I do anything. I didn’t realize that available myself of some kits that include circuit boards for replacing the power capacitors, as well as coupling caps and a thermistor to get me started.

My amps did not include the Z-Magic control itself. From the reviews I’ve read, those that have tried them day the amp sound better without them. So I plan on following your suggestion.

ZacktheDog: I checked the schematics on the 80AZ, and they are similar but different, especially at the output section. The 80AZ is rated at 30 Watts; the 70AZ is rated for 25 Watts. The 80AZ uses EL37 tubes at the output section while the 70AZ uses 5881 (or 6L6). The 80AZ includes a 4 Ohm speaker tap, which the 70AZ lacks.
Finally, the output transformer is very different, it lacks the ultralinear taps noted by DirtyErnie; it has the cathode winding like the 70AZ, but the upper half of the primary side is a simple center-tapped portion like the Fisher 30A I’m getting my “Elmer-ducation” from. The 80AZ also has a higher component count, and not all of it has to do with the ‘peak power’ indicator. Although I notice that it includes a fuse, which I may add to my 70AZ. So, although they may look similar, there is good reason to say they may sound different. I guess I’m trusting in the Fisher brand to deliver the sound my Fisher 80T tuner did back when I was in high school.

Thanks again to everyone who has responded to this thread. I’ll keep you posted on developments, especially if I have anything to add to Sam Cogley’s journal.

You're right, the 80AZ had pentode outputs, I forgot about that.

 

[Sam Cogley]

Looking forward to seeing your restoration progress! It’s been a long time since I worked on those, but I’ll be happy to help if I can.

 

[AnyDayNow]

Yes, and it makes sense that it needs the additional gain of 12AT7s (vs. the usual 12AU7/6SN7 type tubes used in most WIlliamson designs) in the Fisher 70 amps, for enough open-loop gain, having both UL and cathode feedback. That takes a substantial drive voltage and gain in the driver stage, to drive output tubes with that little gain once UL and cathode feedback are both applied!

Regards,
Gordon.

Thanks for the reply, I’m still working out the gains of each stage as I learn how my amp works. I found an analysis by a fellow who specializes in Electronics for Guitarists (his YouTube channel name) who walked me through how a 6V6 operates in ultralinear mode in his video #90 (there is a playlist). I still have to ‘translate’ his analysis to the 5881 tube. I just googled a formula for calculating cathode feedback from a transformer. And work through the feedback circuit itself.

Meanwhile, I opened up both amplifiers and was surprised to learn that, even though I purchased them both from the same vendor on EBay they had been ‘updated’ very differently. In the first one (SN 51267) the first power can cap had been replaced with two sets of two capacitors placed in series: 100 mF at 250 volts in series with 100 mF also at 250 volts to replace the factory 40 mF at 500 volts. The second can capacitor was original. That seems to have been the only modification. All the other capacitors appeared to be original.

On the second unit (SN 56938) both power can caps appear to be original, but all the other capacitors had been replaced, with the exception of the output common cathode electrolytic resistor. Some were Sprague orange drops, others were poly-pro [I know there are two types that look the same, I can’t tell from their appearance]. My kit provides all the coupling caps in poly-pro and a circuit board to mount the electrolytics in the same space taken up by the cans.

Since I had recently rebuilt a couple of KLH Model 17s, I got the bug to hook’em up for a listen. I also wanted to try out my Douk Audio output tube analog current meters. A pair of analog meters equipped with ceramic socket series taps to capture the current flow in action. I had also purchased a matched pair of cryogenically treated 5881s to replace the mismatched 5881 + 6L6GB that I found in the second one (56938).

The amps sounded great; after it stabilized, the 56938 with new tubes were each reading right at 50 mA; the older 51267’s tube sockets were looser than the ones with the new tubes, and while one of them hit 50 mA after half an hour or so, the other tube never got over 48 mA.

I was listening at about 60dB A-weighted, but kicked it up to 80dB at the end of the session and judging from the voltage at the 8 Ohm taps, and using the current readings of the ammeter, I came up with a grand total of 100 milliWatts (2.0 volts x .05 A). The sound seemed very ‘liquid’ to me. I enjoyed listening to it.

Interested in anyone’s thoughts regarding how to proceed with the restoration. I’m not set on replacing everything, but I am interested in best quality performance and heading off any potential, and known failure points.

 

[gadget73]

"patches" are pretty common on old gear. It strongly depends when it was done, and who is doing it. The "fix only what is broken today" method gets you a random smattering of replaced components. If they also subscribe to the "use only what is on hand right now" method, then you get multiple caps assembled to make up a replacement part.

were these mine I'd probably do both cans and get rid of those wax dipped caps. Wax paper caps are notoriously unreliable, and old electrolytic caps aren't much better. Both amps would get the same parts for symmetry and neatness. Spot check the resistors in circuit, change any that are off-value. Most of them can be measured without unhooking anything.

 

[AnyDayNow]

"patches" are pretty common on old gear. It strongly depends when it was done, and who is doing it. The "fix only what is broken today" method gets you a random smattering of replaced components. If they also subscribe to the "use only what is on hand right now" method, then you get multiple caps assembled to make up a replacement part.

were these mine I'd probably do both cans and get rid of those wax dipped caps. Wax paper caps are notoriously unreliable, and old electrolytic caps aren't much better. Both amps would get the same parts for symmetry and neatness. Spot check the resistors in circuit, change any that are off-value. Most of them can be measured without unhooking anything.

Thank you Gadget73: sounds like a plan to me.

 

[AnyDayNow]

I finished working my way through the 70A/Z schematics with the help of Blue Glow electronics (for the 30A comparison), Electronics for Guitarists (for doping out the Ultralinear configuration), Uncle Doug (for how tube amps work generally, and phase splitters in particular, and especially for teaching me how to work out the impedance ratios empirically by injecting reduced household AC power from the output secondaries back out to the primary leads going into the output tube sockets), and Google AI for help understanding and calculating the Cathode Feedback Circuit. I was heartened to learn from Sam Cogley’s ‘70A Questions’ thread here on AudioKarma that I wasn’t the only one to have an issue identifying the primary windings at the output tube sockets. It doesn’t help that I’m blue-green colorblind.

Anyway, I wasn’t able to figure out that (1) while both the Fisher 30A and the Fisher 70A can claim a Williamson heritage, they are not different versions of the same amplifier! They are very different all because of the reasons brought out by responders to my original post; namely, the ultralinear and cathode windings on the primary side of the output transformer.

Other things I learned along the way were that the ultralinear windings are not tapped at the David Hafler (and 6881 tube spec datasheet’s) recommended 43% from the center tap; but much closer to the plate, at approximately 67%. I’m not sure what effect this has, but it seems to me that the tube would be working more as a triode than a pentode, offering less power and less distortion.

The cathode feedback factor works out to be around 12% (11.7). And the power dissipation come out a little above the transformer’s rated max power of 25 Watts (I came up with 32 Watts), but when (following Denton Dailey’s lead on Electronics for Guitarists), that figure is halved for continuous (or RMS) power at 16 Watts, it’s only one Watt higher than the transformer spec of 15 Watts.

Although I worked out the voltage gains for each stage, including the output stage, I don’t have much confidence in them. I noticed, while reading a 31 page ‘appreciation’ of the ‘new’ Williamson amplifier (that was first presented by Williamson in 1944) written by an Englishman, that Google AI turned up, that someone had gone through the Williamson design and noted the voltage gains and current increases at each section. It was very clear what each stage was continuing to the overall performance of the amplifier. At this point in my journey (this is the first tube amp I’ve ever opened up), I think it would be instructive to perform the same exercise on my amps, one at a time, as I open them up and install my kits and verify each circuit’s performance. Kind of a before and after snapshot and record for future reference when needed.

Aside to Sam Cooley: Man those two amps gave you fits, didn’t they? It seemed like something else broke each time you addressed an item. Kind of a ‘two steps forward, three steps back’ kind of affair. I’m thinking I’ll go ahead and replace the tube sockets on the older of my two amps (S/N 51267), they just aren’t tight like sockets in the newer one (S/N 56938).

When I come back, I’ll likely start a new thread on the 70AZ, since I won’t be looking at the 30A anymore.

 

[gadget73]

The 43% tap placement was derived from testing to get the best performance, using whatever parameters were selected with no cathode feedback. Its entirely possible that adding cathode feedback into the mix changes where that optimal point is.

 

[DirtyErnie]

Been looking at Toroidy's stuff. Their normal ultralinear taps are at 43%. When they do cathode feedback, they move the taps to 33% and list the cathode windings as 10%.

 

[gadget73]

I guess the idea there is that they add?

There were also plenty of other tap points used in that era, some of which worked better than others. 50% seemed to be used on certain things which appears to not be optimal for anything, but when its done on a transformer that has split primaries it does have the benefit of allowing the two primaries to be run in parallel to create a transformer at 1/4 the primary impedance, just in case you might want to do that for some reason. I have a Chicago BO-13 that way, 50% tap points, 10K plate to plate, but it could be re-arranged for 2.5K plate to plate with no screen taps. Those 50% points become the B+ feed.

 

[Ken Boyd]

I deleted my post's the amps I have are not 70A's.

 

[bandtuber]

I just finished restoring a pair of these. On the amp with the original paper capacitors, I would suggest replacing the two .1uf grid coupling caps and the two .022 that connect the 12AU7/12AT7 ASAP before running these again. Those Good-All caps are prone to failure and a short in any of them could be very bad. I thought that I could just get away with replacing the .1uf's, but one of my .022uf failed after about 30min of playing and killed the 68K resistor on the AC balance pot. The amp made a horrible noise! Anyway- all good now.
I would also suggest replacing all of the filter capacitors with 39uF or 47uF Nichicon, Rubycon, Chemi-Con, or JCCON under the chassis. You can mount a terminal strip and mount them underneath the old cans. I was able to get a bit more power output and reduce distortion by changing the cathode bypass filter cap to a 100uf/100v instead of 50uf.
Once restored, these amps sound incredible. Super rich and detailed. I was only able to squeeze about 22.5w out of each before clipping on a 1k sine wave. However, their square wave performance at 1k is phenomenal. Incredible flat response with superb stability. PM if I can be of service!