The End of an Era: The Fisher 55-A

dcgillespie

Fisher SA-100 Clone
Subscriber
The early 1950's. The hifi bug was well on its way to becoming an epidemic, with the stage being set for a scene that more and more amplifier manufacturers would jump into. And, then as now, the arguments raged on as to what approach produced the best sound. Today, much of the discussion is all about boutique component selection, tube rolling, and burn in time. But back then, such discussions were hardly a thought or consideration. Back then, it was all about the heady discussions of pentode or triode design, or how the Circlotron, Unity Coupled, or Ultra-Linear configurations could best be utilized. Class of operation was always at the top of the "discussion" list, with Class A, and AB1 both having their followers, and the high power/low power camps both chiming in right on queue. During those times, there were no wimpy discussions about whether point to point or tag board construction sounded best. It was all about serious real man discussions, whose outcomes formed the basis of entire companies or product lines. Brands became known for the design approaches they championed, many of whom lived or died by the decisions they made. It was an exciting time in the world of high fidelity, and truly the genesis period of this hobby we all know and love.

With Fisher's company freshly reorganized as Fisher Radio Corporation after the war, they jumped into this fray with their own signature design around 1953. In keeping with Avery's burning desire to be unique if not the first with the products his company produced, he introduced the model 50-A power amplifier (series). And what an amplifier it was. The concept of dollar per watt hadn't been conceived yet, but with this amplifier, one of Fisher's first "firsts" must surely have been realized: the concept of pound per watt. As a single chassis device, this mono basic power amplifier achieves a one-to-one ratio, tipping the scale at 50 pounds, producing an advertised 50 watts of power. This then was no chickified model designed to win over the hearts of the fairer sex, but an all out 1953 execution of Tim Taylor's "mo' power, mo power!" concept. With two massive transformers as an anchor, this amplifier is the antithesis of a chassis in search of some transformers to build an amp on. Rather, this is a couple of transformers that happen to have a chassis strapped to them!! It's a beautiful thing to behold, if not hernia inducing. To be sure however, for those of us with grand kids, this ain't your dad's old hifi amp......... it's YOUR grandfather's!........so that should help to keep things in perspective.

As to its design, it was, and remains truly unique in equipment designed for the home high fidelity setting. The power supply is within the conventional realm for its day, but massive. Dual heavy duty rectifier tubes, a (swinging) choke input filter system, and vacuum tube bias regulator circuit get things powered up. The phase inverter and pre-driver small signal circuits borrow heavily from Williamson, and like that design, there is a global NFB loop that (in part) encompasses these stages. But after that, any resemblance to any other design of the day for high fidelity applications ceases. Following the pre-driver stage, two triode tubes (or triode connected pentodes) operate as push-pull Class A cathode followers, whose output is coupled to the output stage via transformer/capacitive coupling. This driver stage then allows the triode connected output stage to operate in class AB2, which is VERY much different than anything else Fisher -- or anyone else in this product category -- ever produced.

Briefly, early high fidelity design was all about low power triode output stage operation, as that was the only practical way to achieve a low output impedance, and low distortion at the time. Pentode operation produced more power, but had a high output impedance (and higher/odd distortion elements). Negative feedback could correct all of that, but proper application of it was still elusive at the time, so low power triode amplifiers and big high efficiency speakers ruled the day. Think about 7-10 watts RMS and K-Horn speakers.

It is against this backdrop then that the Fisher 50A did its cannonball leap off the high dive. Big power, low output impedance, flat response, and good distortion characteristics made for a package that others just did not have at the time. The fact that it could also serve as a boat anchor in an emergency was just an added plus. Over the course of about 4 years, the amplifier went through various versions, with the biggest changes being in the rectifier and output tubes used. Earlier on, there was also a change in driver tubes used, but for all those changes, the circuit itself changed very little if hardly at all. The change in tubes simply represented the use of bigger (higher dissipation) tubes as they came along. The 55-A -- which is the subject of this thread, is then (to my knowledge) the last of this breed, sporting big 5AW4 rectifiers and 6550 output tubes. That's about as big as it got back in that day.

With that background then, I now have a previously restored 55-A on the bench, with no other goals (at this point) than to answer a few basic questions:

1. How well is it operating?

2. Is it meeting published factory specification? and,

3. Are there any reasonable modifications that can be done to this amplifier today to improve its performance?

As a result, basic tests and baseline tests are now being conducted, and ideas formulated, that will all be presented in upcoming posts. But for now, pics include:

1. This is one heavy beast! Displaying the appearance that all Fisher power amplifiers did in the 50s, it is built into a very thick gauge, high brass content chassis, painted in Fisher Brown. From my past experience, these chassis can display poor ground integrity over time, interacting poorly with the dis-similar metals used in the various T-Strips, tube sockets, and ground connection points used. In this unit however, every ground connection was very much in tact -- a bonus!

2. Under the tube cover, where there is also a ceramic resistor mounted to prevent it's heat from building up under the chassis.

3. Underneath, there's plenty of real estate to house the rather simple circuit. As was common in the day with premium equipment, a tag board is used to mount many of the audio circuit components on.

4. Negative grid bias voltage is given as -42 vdc, which the output stage was carefully brought to while monitoring it's various quiescent operating parameters.

5. This bias voltage produced a total cathode current draw of 180 ma in the output stage, and also caused the meter to rise to the 0 Power Output point on the Power Monitor meter. Clearly, this is the intended quiescent bias set point.

6. At 90 ma per tube (about 84 ma per plate) and 420 vdc on the plate, this equates to 35.3 watts per tube, which is basically the dissipation rating for a 6550 tube (of any American manufacture), under the conservative Design Center rating system that was in effect at that time. However, at this dissipation level, neither output tube displayed any color in the plate, even when viewed through the lens of a time lapse pic.

If you ever wondered about some of the earliest efforts at Fisher when the hifi bug was in its infancy, this is it. If you're interested, you might check in and follow along!

More as it becomes available.

Dave
 

Attachments

  • SAM_1282.jpg
    SAM_1282.jpg
    62.5 KB · Views: 589
  • SAM_1273.jpg
    SAM_1273.jpg
    76.7 KB · Views: 686
  • SAM_1274.jpg
    SAM_1274.jpg
    119.5 KB · Views: 611
  • SAM_1279.jpg
    SAM_1279.jpg
    27.1 KB · Views: 514
  • SAM_1277.jpg
    SAM_1277.jpg
    61.8 KB · Views: 562
  • SAM_1278.jpg
    SAM_1278.jpg
    75.1 KB · Views: 539
Last edited:
Whenever I read these threads, I always feel like Jack Nicholson as the joker in the batman movie "where does he get all of those wonderful toys?"
 
Hiya Dave,

And there it is again ... the 6CL6 ... (I have had this tube on the brain this last few weeks)

I will be definitely be following this.

Frannie
 
thanks for putting in the time with these threads Dave; they always read as excellent monographs

looking at the 3rd pic, it's got the exact interstage trans as some of the later model 50AZ amps (T507-123D).

I noticed that Fisher decided to alter the choke by putting wood block underneath it (never had one on any of the 50A/Z's that I've seen)...maybe he should have switched to an input choke instead! :)

Peaking at the 55A's schematic, it appears that both 6cl6 driver tubes share a resistor in addition to using the IT's secondary winding resistance for cathode biasing (the former was never done on the 50 series amps as far as I'm aware). Was this to stabilize the operation point? My blurry schematic pdf reads "R5"? Please, could you tell me the value of this resistor? Why do you suppose it was incorporated in the 55 series?
 
Never heard a 55a, but my 50a pair are wonderful sounding amps and aren't going away any time soon. :no:
 
Finally! My all time favorite old school amp in the hands of the ultimate vintage tube audio Kung Fu master. :thmbsp:
 
Interesting!
I'm thinking of McIntosh on one end and Ed Laurent/Dynaco on the other as I'm following this.

Sent from my SM-N900V using Tapatalk
 
Stock Performance Results

All of the power supply caps have already been replaced in this unit, and a quick check of the resistors in it put a spotlight on ......nothing. They were all easily within tolerance. The tubes were all tested, and while one driver and one output tube were decidedly weaker than their partner's, they were deemed well enough to at least see what ball park this big guy plays in.

With the exception of the coupling cap at the input jack being removed in the audio section, and a few minor modifications done to the power supply to remove the Auxiliary (preamp) Power plug and AC line bypass cap, add a power switch in place of the Auxiliary AC power plug and a new grounded power cord, and replace the selenium bias rectifier with silicon, this amplifier is box stock. So, it should give us a good read on the performance capabilities of the as designed amplifier.

The available specifications were cobbled together from a number of different places, and are presented as follows:

1. Power: No specific power rating given, but taken as "55 watts" from ad banner headings, etc.

2. Power Response: +/- 1 db 15-60 kHz @ 55 watts.

3. Distortion: <1%@ 55 watts, <.08%@ 10 watts, <.05% 5 watts.

4. IM Distortion (40 & 7kHz mixed 4:1): <2%@ 50 watts, <.8%@ 45 watts, <.40%@ 10 watts.

5. Frequency Response: 20-20 kHz +/- .1 db. 5-100 kHz +/- 1 db.

6. Damping: 1 Ohm internal resistance @ 16 Ohm tap.

7. Noise: >92 db below 55 watts.

Against this published data, this particular amplifier, operating from a 122 vac line, returned:

1. Power: 50.0 watts RMS at mid-band frequencies -- 5 watts shy of the published data, although no definition of the power rating system used is given.

2. Power response: 40.64 watts RMS watts at 20 Hz and 20 kHz which is just outside a 1 db down specification based on 55 watts RMS, but is within it based on 50.0 watts RMS. Power above 20 kHz was not tested.

3. Distortion: 1% @ 50 watts RMS, slightly missing the <1% @ 55 watts specification.

4. IM Distortion: 3% at 50 watts (equivalent power), which is 150% high of specification. Due to the design of this amplifier DC balancing of the output stage is not possible, nor is adjusting the AC balance of drive provided to it. Therefore, the only measure available to reduce distortion is to used match pre-driver, driver, and output tubes, which would help lower the distortion produced in this category.

5. Frequency Response: 20Hz to 20 kHz was within +/- .1 db meeting specification. However, supersonic frequency response was down 1 db at just 40 kHz, which in reality is fine, but well off the published specification.

6. Damping: 1.47 Ohm on the 16 Ohm tap -- pretty good, but short of the 1.0 Ohm specification for this tap.

7. Noise: 83 db below 55 watts, which is short of the >92 db given, but I also note that this amplifier did not include a bottom cover, and the AC line bypass cap had been removed.

So, it's close on some performance parameters, and well off on others, but that is hardly surprising for equipment particularly from this era. There are plenty of other quality manufacturer's pieces that will not meet their specifications from this era as well. In 1953, the Naval Research Center performed similar tests on a 50-A, and concluded it only met just one of its advertised specifications.

When judged on the whole however, the performance returned is still very good performance for mid 50's equipment -- and particularly noteworthy, is a specification they didn't even publish, which -- in my opinion -- has more to do with listening quality than most all of the other specifications given: LF & HF transient response (stability). In these two areas, the 55-A really shines (as does likely its predecessors based on the design), for reasons which will be discussed later. For now however, it is noteworthy to say that the amplifier displayed rapid settling under pulsed conditions, and virtually no ringing on loudspeaker loads when passing square waves. These qualities, coupled with a flat frequency response, and adequate clean power across the entire audio bandwidth make for a wonderful listening experience. In a day when high efficiency speakers ruled the day, this series of Fisher amplifiers could deliver that in spades.

Pics include:

1. Uncovering the secrets of the 55-A amplifier. This series of amplifiers is very different from conventional design, with qualities that led both to its success, and the end of its run.

2. 10 kHz square waves are smooth and void of any ringing on any type of load. This shot was taken when a loudspeaker load was connected.

3. Cap load only tests produce an exaggerated, but still controlled ring, with no tendency towards sustained oscillation. For feedback amplifiers of that day, this was stellar performance indeed!

So more than 60 years have passed since the roots of this amplifier were on the drawing board. And what of that time passed? Is there anything that can be done to these amplifiers to improve their performance today? It turns out that there is, and it's more that just a little significant -- and not too hard to implement. But to understand why its such an improvement, you've got to understand what's so different about these amplifiers in the first place.

Next time.

Dave
 

Attachments

  • SAM_1281.jpg
    SAM_1281.jpg
    77.7 KB · Views: 176
  • SAM_1268.jpg
    SAM_1268.jpg
    46.1 KB · Views: 143
  • SAM_1276.jpg
    SAM_1276.jpg
    53.5 KB · Views: 137
Last edited:
Cgo -- My schematic also shows an R5 in the location you mention, with a hand written value indicating it is a 10K resistor, but there is no evidence that it was ever installed in this particular amplifier. The driver stage will be discussed in some detail in the next post, but suffice to say that a 10K resistor in that location would be detrimental to driver performance. The B+ filter for this amplifier series utilizes a choke input design.

Dave
 
Cgo -- My schematic also shows an R5 in the location you mention, with a hand written value indicating it is a 10K resistor, but there is no evidence that it was ever installed in this particular amplifier. The driver stage will be discussed in some detail in the next post, but suffice to say that a 10K resistor in that location would be detrimental to driver performance. The B+ filter for this amplifier series utilizes a choke input design.

Dave

Dave,

What's the serial # of your amp?
 
Coincidence that one of these showed up on eBay shortly after this thread started?

Informative as ever Dave. Thanks.
 
:yes:Ruflerruf,
Its like ripping a hundred dollar bill in half. The second one is where they will hit you.
 
Back
Top Bottom