The Fisher 125-AX – Restoring Fisher’s 1957 beast!

The Fisher 125-AX, Restoring Fisher’s 1957 beast!

These are so rare I have never seen even one before never mind a pair, and were being sold by “tubetycoon”, whom I had never heard of. Well, this was “The Fisher Doctor”, Al Pugliese in disguise. He’s still the same old larcenist!

The picture shows the mismatched cages, if you look closely, but he never mentioned it in text. The meters didn’t work, and he says they never do, but it is obvious that they have recently been swapped in. Assume the good ones were swapped out!

While he so generously included (readily available) extra meter lamps, NE16 bulbs, and meter covers, one of the two 12AX7 tubes and all eight (!) EL34s were worthless! While Al clearly intended to send bad tubes and parts where indicated, he knows these are transient problems and unlikely to result in a return. He’s done this a LOT. He knows what he will get away with.

Internally, both amps were fairly identical as shown, all original parts. Totally unusable/untestable in this condition. That’s expected though. I am happy that no one (Al!) had attempted to repair or restore before me. This is clearly over his head. One amp had been worked on where the driver section meets the output tubes, and parts remained disconnected. I learned why later, and will explain during the restoration section.

The Plan:

  • Test the transformers as quickly as possible to determine if they are restoration worthy. They were fine.
  • Replace ALL of the capacitors, electrolytic and paper, with modern parts while maintaining cosmetic integrity.
  • Replace resistors where necessary.
  • Replace tubes as necessary to get proper performance.
  • Repair the meters, if possible.

Let’s have a look at the original design!

  • The power supply is a dual-rectifier 5R4GY with a choke-input filter. The 5R4GY is the only consumer grade tube that can handle the PIV that would be in a circuit generating 625 VDC. The massive choke (5.5Hy, 160mA) is the center potted item on the chassis top. This is for real! The following filter capacitors are in series, 2 x 120uF at 450 VDC which nets 60uF.
  • The screen voltage is regulated to an adjustable nominal 285 volts using a 6Y6GT series pass tube, 6AU6 error amp, and an NE16 bulb for reference. The low level audio stages are also powered from this source through additional RC filtering.
  • The bias voltage is regulated by a shunt connected NE16 bulb. This would be a bad idea if the plate voltage varied, but a regulated screen voltage makes this acceptable. Together they will hold plate current constant over a varying line voltage.
  • The low level audio section is Williamson style with a 12AU7 concertina or split load phase inverter, with a 12AX7 driver stage after. An interesting choice by the designer to use the 12AU7 first and 12AX7 after! We can assume this was to put the phase inverter (where a 12AU7 may be a better choice) earlier in the signal path. Note that all cathodes are not bypassed, which lowers gain but increases linearity. Negative feedback is applied at the input tube’s cathode and plate.
  • The 12AX7 driver stage is coupled though high-pass filters (to improve LF stability) to the output stage. This is directly connected to one pair of EL34s, then through 100 ohms to the second pair.
  • The output stage is 4 EL34s in push-pull parallel. The operating plate voltage is 625 VDC, screen voltage 285 VDC, and the plate current adjusted to 45 mA per tube. That’s 28.25 watts per plate, a little over the limit as usual for Fisher!

And the (heroic!) specifications?

  • 90 Watts continuous, 125 Watts peak. The unit will deliver 132 Watts continuously with fresh tubes, but this will exceed the EL34 ratings worse than the 28.25 Watts of idle dissipation!
  • Frequency response within .1dB 20Hz to 20000Hz. Power response at 90 Watts within .5dB 20Hz to 20000Hz. These are Freed output transformers!
  • Distortion below .5% at 90 Watts (they probably only mean harmonic, since distortion type is not specified).
  • Hum and noise better than 92dB below rated output.

I have completed the electronic restoration and have been using the amplifiers a bit since. They are:

  • Loud as hell. There’s still knob left when you can’t stand any more!
  • Amazing bass due to excellent damping & excellent frequency response!

The Restoration:

Before any testing can be done, these part upgrades need to be done. The original electrolytics were bad by every measure, as these units had probably not been used/played since the early 1960’s. The only things checked out before this were the transformers themselves. A variable transformer was used to apply voltage to the power and output transformer primaries, and all windings are checked for proportional voltage outputs. All OK!

Overview of the upgrades:

  • Replace each 120uF filter capacitor C1, C2 with an F&T 100uF/100uF @ 500V with both sections in parallel, each also having a .1uF polypropylene and a 100K/3Watt resistor in parallel. These assemblies, on the left and right sides in the picture above, are in series. The peak power supply voltage in these units, when unloaded, can go slightly over 1000V!
  • Replace the rear capacitor can C3A/C3B with an F&T 50uF/50uF @ 500V. Section A is the screen supply filter, and section B is the low level stages filter.
  • Replace all small paper capacitors with Vishay/Sprague 715P Polypropylene, rated at 630V.
  • Three pF-range capacitors replaced with silver-mica units: C8, C9, C14.
  • There are several small electrolytics, replace each with 105-degree units of equal value. The exception, C4, was 1uF, can be replaced with 10uF. This improves the decoupling to the first gain stage.
  • Replace the 3.3M resistors at R24 and R29 with matched parts. High value resistors from that era are drift-prone and at this circuit position any drift will make the output tube currents hard to balance.
  • Replace R53 and R54 with matched 82R resistors, as these are where the metering circuit measures the currents on each push-pull side.
  • Check all other resistors and replace if off by more than 10%, regardless of marked tolerance.
  • Replace selenium rectifier M1 with a fast-recovery diode. Because the bias voltage is regulated by the NE-16 Neon lamp following it, no additional series resistance needs to be added.
  • Link to service manual PDF: Fisher-125-AX-Service-Manual

Underside view of the upgrades:

It’s Tube Time:

Tube selection:

  • For EL34 I selected the new Tung-Sol reissues in matched quads, as these continue to gain good reviews and don’t break the bank.
  • For 5R4G, I selected 5R4GBs, as they have a lower voltage drop and are MUCH easier to find in semi-matching groups. They are also straight-sided like the EL34.
  • For 12AX7, genuine NOS Telefunken.
  • For 12AU7, Telefunken was used again.
  • For 6Y6GT, An original Fisher marked tube was used, but the other had developed a tendency to arc-over, so a GE 6Y6GT was used.
  • For 6AU6, the original Fisher marked tubes remain.
  • NE16 bulbs: Any should work. A failure mode that can occur is a drift upwards in the firing (and thus regulated) voltage across these. 53 to 65V is correct.

Now for Testing…

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: