A tale of two Fishies - X-101-C integrated amp restorations

The Fisher's low end was pleasantly lumpy before. To a certain extent, it's a very listenable and forgiving of input source frequency curve, such as one would select with an equalizer.
With the work Dave's done on the X-101-C, I've discovered that, with the tone controls out, it very much sounds like:
1. The Fisher X-1000 Dave had worked on before and did some similar, but not the same, work to (different type of tone amp)
2. The Harman Kardon 260A Chorale (HK hot rod) that I restored that has a factory selectable tone control defeat switch.

I believe that in the HK, they made an effort at the design level to have the different stages be closer to flat, rather than to have one stage compensate for the lack of flatness on another stage, as in the X-101-C.

The overall end result? Smooth and non-fatiguing linearity. Very listenable, not tubey, not bumpy, not with any exaggerations, but it also puts out low low bass where it exists and is often obscured by mid-bass (which starts surprisingly low).
 
As a former soundman and therefore toneshaper, I've worked to train myself to let less sound wash over me until my ears adjust to what's there instead of what I want to have there.
It's like that with this amp.
I liked it stock, I liked it the way I restored it, I like it now with Dave's touch. They're all good, but not the same. The amp is now much less of a tone shaping circuit in itself.
I use Spendor mini-monitors with their natural sealed box LF rolloff and their unassuming attention to midrange correctness, plus a powered sub, on just enough to make tympani and bottom mic'd drum heads on floor toms sound like I remember them sounding when doing sound.
 
Yeah it's funny, for years I've been using tannoy super golds with the Manley crossovers and cabs at two studios I work at, one has huge Manley mono blocks, the other we use brystons, then I got a pair to use at other places with a mcintosh, also used these at home for a long time. All sounded just great, and I still use them in the studio, but maybe I like a little color for fun listening to.
I didn't think about it being used as a tone shaping tool, but that's like it is!
 
Nothing wrong with that as long as you recognize it for what it is.

I like amplifiers that inspire passion also.

The one thing I'll say about Dave (dcgillespie)'s touch is that it's not subtractive.
It adds to that midrange quality I crave and that Fisher gear provides so well by providing equivalent subtle but satisfying quality, as opposed to showroom splash, at both the top and the bottom of the audio spectrum.

Having experienced his craftsmanship first hand, I tend to understand his comments and where he's coming from.
 
Last edited:
Well, you wouldn't be doing the kind of work you do if you didn't love sound. I'm a music lover first and a gearhead first too.

I haven't said it yet, but I'm very happy you've posted here and I'd like to say welcome.

Based on your experience, I know that you know good sound . . . probably much better than many of us. Sometimes there are advantages to not being in the trade though, and I can now enjoy those advantages while still using some takeaways from my time inside.

It's difficult to balance full bandwidth and non-fatiguing in a playback system and have it so you just want to keep listening rather than using the playback for analysis. I've pursued that goal for years and what's worked for me works well, yet it sometimes has had other people shaking their heads. I've tried to learn that they're not all wrong while I'm all right. There's some middle ground somewhere but, like you, I like what I like and I accept that about myself. Fortunately, these days I only have to please myself.

Enjoy that X-101-C. I think it's my favorite Fisher. It's very good at getting out of the way of the music.
 
Last edited:
I took this amp to be tested at a Boston Audio Society test clinic in November 2014.

This was just a test of the amplifier portion of the ratrig Fisher X-101-C.
I've got some information I can add to this post now, and some information I'll have to backfill later. That said, I've been holding this information, so posting it will stimulate me to get the rest of the information.

The BAS people don't think much of tube amps, expecting typical tube amp performance and awaiting results that verify their expectations. After all, there haven't been a lot of design advancements in the tube audio arena in the last 50 years, and the art of transformer winding, in general but there are exceptions, is not what it once was for high volume production applications.

Most of what the BAS has seen for tube gear has been either vintage gear, finicky boutique Japanese SET gear, new Chinese gear (think budget gear like Yaqin, not good gear that happens to have been designed and made in China like Line Magnetic) or hobbyist - trying to make a run at a manufacturing career - gear. Yes, I know LM is based on WECO designs, but LM has certainly moved on from simple design reproductions and into original designs based on principles learned and absorbed.

Between the full restoration I'd done and the design upgrades that Dave Gillespie did, including EFB, adding a bias/balance option where there had been cathode bias, actively bypassing the passive tone controls, and building a buffer to support separating the pre from the amp, the ratrig is like that dusty old Buick Riveira your Grandma had in the garage to drive to church on Sundays. I brought a spoiler for testing.

First up was a computer based test. The computer was ancient, but the test was thorough. I have more questions to ask about this testing, and the testing session was not the appropriate time to ask the questions, so I'll come back later and fill in some of this test information. Sorry, these are phone pictures.

Well, since most of us are comfortable with 50+ year old tube gear and 50+ year old test gear, I have to say I'm fine with this 30 year old Olivetti-looking test computer.
attachment.php


It's connected to a bunch of outboard test equipment to drive the channels.
attachment.php


It drives both channels at once, at raises the input until it reads 1% THD on the output.
attachment.php

Having been designed primarily to test solid-state gear, the available impedance taps are 8 ohm, 4 ohm, 2 ohm and 1 ohm. There is no option for testing from the 16 ohm tap.
 

Attachments

  • IMG_5703.jpg
    IMG_5703.jpg
    65.8 KB · Views: 117
  • IMG_0141.jpg
    IMG_0141.jpg
    49.1 KB · Views: 118
  • IMG_1421.jpg
    IMG_1421.jpg
    73.8 KB · Views: 115
Last edited:
So, we tested the 8 ohm taps and the 4 ohm taps of both channels. I've got to take photos of the printouts.

I don't fully understand the output, but with both channels driven at 1% THD, these were the outputs:
8 ohm tap: 37 watts continuous at a 0 degree phase angle. It would be nice to just leap to the 60 degree capacitive phase angle and say 47 watt, but I'm not positive about the test conditions, and 37 watts aligns more closely with Dave Gillespie's testing.
attachment.php

4 ohm tap: 33 watts continuous at a 0 degree phase angle. 42 watts continuous at a 60 degree capacitive phase angle.
attachment.php

There's something about the phase angle in the printouts that I have to ask the BAS about.
Perhaps the output tests depended upon the phase angle of the output. Not sure how that worked.
 

Attachments

  • DSCN7719 copy.jpg
    DSCN7719 copy.jpg
    79.2 KB · Views: 88
  • DSCN7721 copy.jpg
    DSCN7721 copy.jpg
    77.7 KB · Views: 89
Last edited:
BAS member David Hadaway has been seen at AudioKarma Frankenfests, but he is a hard-core test guy. You can see the amp tested two amps after mine on the test jig. The scope shows noise on the power supply. A local dealer had brought that amp, which is new production.

attachment.php


He has a hand-built test jig connected as a routing station between a signal generator, a frequency counter, a scope and other test gear.
Dave tested only the left channel amplifier, but he was amazed at the performance of this amp.

Broadband noise. This was a power supply test.
On the 4 ohm tap, with a 20Mhz bandwidth, there were some spikes at 2mv and 18mv. 8 ohm tap was identical. He called that excellent performance.
With a 100Mhz bandwidth, I can't read his handwriting. The noise measurement turned out to be -97.4dB on the 8 ohm tap using the test gear and plugging it into the equation 20 [log 1/N divided by 200 divided by 2.83].

He had a big metered SAE amp on the bench before this ratrig X-101-C and described all the noise in the power supply and how it showed up on the scope.


The frequency response on the 8 ohm tap with no load was +.4dB at 20khz. He put no other notations other than a straight line, so I assume it was flat from 20hz. He also remarked that he still had output at 2hz. He was laughing about it and telling others.

Next he connected what's called an IHF load, which was made up of two big computer grade capacitors and big coil inductors. This is used to simulate a standard loudspeaker. What a standard loudspeaker is, I'm not quite sure, but a speaker load changes constantly while this is a fixed load, but you've got to start somewhere, and for now I say so be it.
While folks sniff at IHF specs, this thread says a lot about the value of testing and cites some testing examples and how they can be manipulated, whether IHF, RMS, or some other method.
http://www.audiokarma.org/forums/archive/index.php/t-329286.html

Anyway, this IHF test jig puts standard speaker load demands on the amplifier. Dave tested it on the 8 ohm tap.
Results:
-1dB@1hz
+.4dB@20hz
+.9dB@40hz
After that is a straight line, and then:
-.6dB@20khz.

Next, Dave put a 4uf cap in series with a 2 ohm resistor. He had kind of an "Oh yeah, I'll show this amp" look when he did this. In the photo, you can see the two 2uf caps and 2 ohm resistor on the bench in front of the amp at the right. He told me that most speakers are not a purely capacitive load, which is why he added the two ohm resistor. I get the feeling, based on the size of that resistor, that he's tested quite a few high powered speakers.
This test of an almost purely reactive load was done on the 16 ohm tap, the 8 ohm tap, and the 4 ohm tap.
16 ohm tap results into 4uf cap:
+1@1hz. Then straight line up to 7khz. Next markings were:
+.8dB@7khz.
-1dB at 11khz.
-9dB@20khz.

At the 8 ohm tap, the results into a 4uf cap were:
+2dB@2hz.
+.2dB@50hz
-.5dB into 10khz
-4.5dB into 20khz.

At the 4 ohm tap, the results into a 4uf cap were:
I can't interpret what he meant at 1 hz.
+.4dB@10khz
-5dB@20khz.

At this point, he was smiling.

He was very pleased with the performance of this amplifer. For most of the stuff there, he didn't even bother to perform these tests. This one lit up his expression.

I have discussed these test results with Dave Gillespie, who concurs that this is good testing and he was very pleased with the results. DaveG says that it's typical of tube amps to have a slight rise in gain at the low end and the high end of the passband of an audio piece.
 

Attachments

  • IMG_1424.jpg
    IMG_1424.jpg
    97.7 KB · Views: 120
  • DSCN7718 copy.jpg
    DSCN7718 copy.jpg
    70.3 KB · Views: 45
Last edited:
How would this compare to the McIntosh Test Rig that they would haul around to different sites? Seems like this one does a lot more indepth testing than the Mac rig which IIRC just tested WPC and distortion under a static load.
 
I think you're right Larry.
That being said, I'm not confident that there's anyone left who fully understands the computer test. I suspect it's a legacy piece, maintained by someone who recognizes it's value, and can keep it running.
 
How would this compare to the McIntosh Test Rig that they would haul around to different sites? Seems like this one does a lot more indepth testing than the Mac rig which IIRC just tested WPC and distortion under a static load.

Mc used a resistive load, cranked up a 1 kHz signal with both channels driven until reaching the lesser of rated power or visible clipping. Frequency was then varied from 20 Hz to 20 kHz at the same signal level and THD was plotted on a graph. You got the original, they kept a copy.

McIntosh guaranteed performance was 0.5% THD for tube and 0.25% THD for transistor across the band. If a Mc unit did not meet spec, they would repair it on the spot. If that could not be done, they were authorized to give you a new equal or greater amplifier on the spot.

We were able to look through a pile of their copies of the graphs sometime around 1967. There were almost no competing units that came even close to meeting their spec, especially at frequency extremes. The Dynaco 120 was good down to 30 Hz, where the regulated power supply gave up and distortion shot off the graph. Since that unit was single supply, capacitor coupled output, I often wondered at the time if it would be possible to invert the phase of one channel, then reinvert at the output. This way, the channels would not be drawing power simultaneously at low frequencies where there is little separation.

There is also a story about Bob Carver taking an experimental ratrig to one of the Mc clinics, using direct line power with no transformer and a coffee can as a heat sink. At the time, 50-60 watts per channel was considered good. He specified something like 350 watts per channel and the room went silent as Davey O'Brien cranked up the signal. At some point, they tripped a breaker and put half the store in the dark. Eventually, they got the full 350 watts, and Carver left with an "official" document that he was able to take to the bank to get a loan which he used for advertising his first amplifiers. His construction/pricing structure was such that he could bootstrap the company into higher production quantities.
 
Additional information from the BAS. Apparently the PowerCube tester was a productized product sometime in the late '80s or so.

I'm going to have to print out and photograph or scan the PowerCube PDF document. AK doesn't accept PDF as an attached file.

This is good stuff.

From the BAS:
The PowerCube and the IHF Speaker Load are both IHF standard measurement devices. The PowerCube is about 25 years old. We have been in contact with the manufacturer. We are convinced that it is working to IHF specifications.

Dave Hadaway added to the test a load simulating an electrostatic speaker. (Audiodon - that would be the 4uf caps and resistor test jig).
To my knowledge, his parts choice is within the range electrostatic speakers. For more info go to the AudioGraph site (http://www.audiograph.se/). We will supply additional info when the BAS article is complete.
_______________________
Bob Carver,

My heroes are Einstein, Hegeman, Gandhi.

When we look at a power cube representation of amplifier performance, it's easy to jump to a false believe that the "squarer" it is, the better the amp must be.

Let's take a look and see if that's really true. But first, a bit of history. In the old days of vacuum tube design, the ability to drive a reactive load was not even on the minds of tube amp designers because a tube amp has no trouble at all driving reactive loads. This is by the their intrinsic nature. Same for switching amplifiers.
The only problem with tube amps is that they are big and heavy and expensive. The problem with switching amplifiers is that their power supplies have essentially zero common mode rejection and they sound funny because their output networks interact with the loudspeaker load. Still, they do not have any trouble at all driving a reactive load, just like tube amps.

Now we come to linear transistor amplifiers: The problem with solid state amps driving a reactive load is that the output devices are required to deliver simultaneous high voltage and high current. Power is the product of those two, and if both are large numbers, the transistors get very hot, perhaps too hot and they blow up. Big problem!

Now, I like lots of power, and so I have always designed my amps to have lots of power. When the Phase 700 came out in the early seventies, nobody knew how to design protection circuits for big amps. Crown had a DC300 amp which had standard VI limiters, but they snapped and popped into a reactive load, so the amp had a switch on the back labeled normal load and reactive load. The instruction manual said to ALWAYS have the switch in the normal position when hooking the amp up, and when it was all safe and sound, then one was to throw the switch to the reactive position and enjoy a powerful amp that could drive a reactive load (a speaker). But you better not accidentally short the amp's output, or you would have fried DC300 on your plate!

I thought that was pretty dumb, so I set about designing a protection circuit that did not require a switch to be able to drive a reactive load that would allow the amp to blow up if you goofed even once. The ENERGY LIMITER protection circuit was my very first patent, and it made possible a practical high power solid state amp. I began by reading an RCA application note titled "Solid State Amplifiers can be Reliable." I read it at least two hundred times, and it taught me all about VI limiters. And they popped and snapped with difficult speakers. Not good.
 
Fun to read about these tests...
and LOOK, there is David Hadaway of DB Systems Audio fame!
I've got one of his preamps... LOVE it!
BAS member David Hadaway has been seen at AudioKarma Frankenfests, but he is a hard-core test guy. You can see the amp tested two amps after mine on the test jig. The scope shows noise on the power supply. A local dealer had brought that amp, which is new production.

attachment.php


He has a hand-built test jig connected as a routing station between a signal generator, a frequency counter, a scope and other test gear.
Dave tested only the left channel amplifier, but he was amazed at the performance of this amp.

Broadband noise. This was a power supply test.
On the 4 ohm tap, with a 20Mhz bandwidth, there were some spikes at 2mv and 18mv. 8 ohm tap was identical. He called that excellent performance.
With a 100Mhz bandwidth, I can't read his handwriting. The noise measurement turned out to be -97.4dB on the 8 ohm tap using the test gear and plugging it into the equation 20 [log 1/N divided by 200 divided by 2.83].

He had a big metered SAE amp on the bench before this ratrig X-101-C and described all the noise in the power supply and how it showed up on the scope.


The frequency response on the 8 ohm tap with no load was +.4dB at 20khz. He put no other notations other than a straight line, so I assume it was flat from 20hz. He also remarked that he still had output at 2hz. He was laughing about it and telling others.

Next he connected what's called an IHF load, which was made up of two big computer grade capacitors and big coil inductors. This is used to simulate a standard loudspeaker. What a standard loudspeaker is, I'm not quite sure, but a speaker load changes constantly while this is a fixed load, but you've got to start somewhere, and for now I say so be it.
While folks sniff at IHF specs, this thread says a lot about the value of testing and cites some testing examples and how they can be manipulated, whether IHF, RMS, or some other method.
http://www.audiokarma.org/forums/archive/index.php/t-329286.html

Anyway, this IHF test jig puts standard speaker load demands on the amplifier. Dave tested it on the 8 ohm tap.
Results:
-1dB@1hz
+.4dB@20hz
+.9dB@40hz
After that is a straight line, and then:
-.6dB@20khz.

Next, Dave put a 4uf cap in series with a 2 ohm resistor. He had kind of an "Oh yeah, I'll show this amp" look when he did this. In the photo, you can see the two 2uf caps and 2 ohm resistor on the bench in front of the amp at the right. He told me that most speakers are not a purely capacitive load, which is why he added the two ohm resistor. I get the feeling, based on the size of that resistor, that he's tested quite a few high powered speakers.
This test of an almost purely reactive load was done on the 16 ohm tap, the 8 ohm tap, and the 4 ohm tap.
16 ohm tap results into 4uf cap:
+1@1hz. Then straight line up to 7khz. Next markings were:
+.8dB@7khz.
-1dB at 11khz.
-9dB@20khz.

At the 8 ohm tap, the results into a 4uf cap were:
+2dB@2hz.
+.2dB@50hz
-.5dB into 10khz
-4.5dB into 20khz.

At the 4 ohm tap, the results into a 4uf cap were:
I can't interpret what he meant at 1 hz.
+.4dB@10khz
-5dB@20khz.

At this point, he was smiling.

He was very pleased with the performance of this amplifer. For most of the stuff there, he didn't even bother to perform these tests. This one lit up his expression.

I have discussed these test results with Dave Gillespie, who concurs that this is good testing and he was very pleased with the results. DaveG says that it's typical of tube amps to have a slight rise in gain at the low end and the high end of the passband of an audio piece.
 
The problem with VI limiting is that it is cutting power as the speaker is reacting and feeding power back into the amplifier. Cutting power at that time only increases the inductive kickback, causing more problems.

This is similar to foldback current limiting on a DC model railroad supply. At low voltage there is low current available but at low voltage, one needs high current to start the motors turning. VI limiting is fine for constant voltage supplies.

For audio, one should have constant current limiting only, then design the output stage to handle enough power such that the constant current is enough to protect the output stage. McIntosh published a white paper on this quite a few years ago (with graphs) and they appeared to be one of the few that actually understood the problem.
 
Wonderful thread, fantastic job on those amps. They really do make beautiful music don't they? I recently came across an x-101-c myself, with a slew of other Fisher pieces. After cleaning all the pots and switches, replacing some fried components in the bias supply, it's already playing wonderfully. I'd love to bend somebody's ear about the psu caps. I'm nowhere near the skilled, talented and experienced tech that many of you folks are, but not exactly a noob, but I am feeling that way at the moment.

Alan
 
Alan, glad you got your amp going. There's several approaches to doing the psu cans. Some like to order new cans from Hayseed Hamfest or JJ/CE ones from AES. Others, will find space under the chassis for new caps. Myself, I advocate a simple solution of cutting them in place with an oscillating saw, cleaning them out and stuff with smaller, discreet caps. The beauty of this method is that you don't need to de-solder the tabs and components to the old cans and then re-solder everything back. You just drill a hole in the old bases, wrap your leads from the smaller, stuffed caps around the old tabs and solder. The old stuffed cans retain the original look if that's important to you. There's lots of threads on replacing the cans. Whatever method you choose I'm glad you're planning on doing it as failure with the psu caps is probably the number one cause of problems I see here. That and bad volume controls/switches.

Here's a thread I did on cutting the cans in place...
http://audiokarma.org/forums/index....eplacing-fisher-can-caps.644152/#post-8561495
 
Last edited:
Yes sir, thank you very much for the rapid response. All help is gratefully appreciated. It's not the procedure that's troubling me, I've done this dozens of times. It's the first can cap that's got me flustered, and I wish I'd taken pictures before I put it back in its cabinet. I'll have to take it apart again tonight or tomorrow. That first cap in the psu, on the schematic and on the amp, I believe, is supposed to be a single 200uf? But there are several connections to it. If I remember, they look like they are grounds, one jumps to another cap, but they don't test as being related to ground. I do have a drawer full of 40-40-40's and 40-20-20's and the like, I was hoping for a simple solution. That first one is cardboard covered, so I would probably use axials underneath and leave it in place, disconnected. For now I'm off to pick up the kid from school so I'll try and get into it later. Thanks again for the help, I've seen some of your work bham, great work.
 
Back
Top Bottom