Converting A Fisher 660A to 6L6 tubes !

It will certainly benefit from the original OPT. Awesome on getting the 6L6's working.Do you have a scope and sig gen?
 
They are obsolete surplus korean war tubes with industrial numbers with one choice new.
I Beg to DIFFER ! Westinghouse designed the 7591 in 1959 with sales starting 1960. It was designed for audio, NOT Industrial applications, but with most tubes you can find different applications. Sweep tubes as audio output tubes. etc. But the 7591 was designed to fit in a smaller envelope than the 6l6's of the time, while maintaining a 19W Maximum dissipation. These ARE NOT Korean War (1950-1953 note the dates Gary) Surplus tubes. And there is more than 1 new stock replacement. EH7591a in the bog bottle, Tung-Sol 7591A re-Issues (original size envelope), and the ever popular on the guys with more $$$ than (add your own expletive or commnet here) JJ's with SIDS. Granted they are all made in Russia, but the QC on the Tung-Sols and the EH's have much improved over the last 5-6 years. JJ's are still a flip of the coin as to whether it will encounter SIDS within the 1st 2 weeks.
 
It will certainly benefit from the original OPT. Awesome on getting the 6L6's working.Do you have a scope and sig gen?
Yes on scope and signal generator.
I Beg to DIFFER ! Westinghouse designed the 7591 in 1959 with sales starting 1960. It was designed for audio, NOT Industrial applications, but with most tubes you can find different applications. Sweep tubes as audio output tubes. etc. But the 7591 was designed to fit in a smaller envelope than the 6l6's of the time, while maintaining a 19W Maximum dissipation. These ARE NOT Korean War (1950-1953 note the dates Gary) Surplus tubes. And there is more than 1 new stock replacement. EH7591a in the bog bottle, Tung-Sol 7591A re-Issues (original size envelope), and the ever popular on the guys with more $$$ than (add your own expletive or commnet here) JJ's with SIDS. Granted they are all made in Russia, but the QC on the Tung-Sols and the EH's have much improved over the last 5-6 years. JJ's are still a flip of the coin as to whether it will encounter SIDS within the 1st 2 weeks.
You may be right on the tubes being for audio , I was guessing from the industrial numbers on them that they were most likely military surplus tubes or some other type of applications. The russian and china tubes are better now then they were at first but other than the Groove Tube 6L6 China GE and Svetlana EL34 they still don't sound very good to me. With a mix of NOS and them you can get a pretty good sound sometimes.
 
This set up is out of a '62 Ambassador Console. The Amp is actually a 660A. Why go to a 6L6GC when it's optimized for 7591's?
Thanks for the info about the console and year ! As far as the tubes go it is Just my choice, I like The Sound of The RCA Holy Grail Blackplate 6L6 tubes and There are many other choices with NOS 6L6 tubes. The voltages were in the right place for the RCA's also ! . I do not think that the origional 7591 tubes sound better than RCA blackplate 6L6 tubes !.
 
Last edited:
Thanks for the info about the console and year ! As far as the tubes go it is Just my choice, I like The Sound of The RCA Holy Grail Blackplate 6L6 tubes and There are many other choices with NOS 6L6 tubes. The voltages were in the right place for the RCA's also ! . I do not think that the origional 7591 tubes sound any better than RCA blackplate 6L6 tubes !.
 
Your project is very doable, but doing it correctly depends on how much re-engineering capability you have -- not only in the understanding of theory, but test equipment capability as well. I say the word "correctly" based on either making the amplifier simply "work" (i.e., operate reasonably well) dependably and safely with the alternate tubes, or having it produce all the performance of the original design that Fisher intended for it with the alternate tubes installed. That is two very different things. In either case, basic issues to address include:

1. Mismatched OPTs. The original Fisher OPT is no doubt significantly superior to the replacement piece. It would be best to get an original replacement transformer. Otherwise, understand that there will be performance given up by using the replacement unit installed.

2. 6L6 class tubes have about half the Gm of 7591 tubes, therefore, the 6L6 tubes will require about double the signal drive voltage and double the required grid bias voltage. But these amplifiers do not include a bias supply but rather, used the heater elements in the audio tubes on the adjoining tuner preamp to provide cathode bias to the output stage. This will clearly need to be adjusted to suit. The increased bias voltage required will represent a loss of B+ voltage, so power output will be reduced with such a move. NFB will be reduced as well due to the lower output stage gain, so that if it is restored to its original value, the input sensitivity of the unit will be cut in half -- again, all owing to the fact that the Gm of 6L6 tubes is much lower than that of the 7591.

3. To promote a proper impedance match to (at least) the original OPT, the screen voltage will need to be dropped to about 350 vdc. But doing this by simply upping the value of the screen grid dropping resistor will produce poor results as then the regulation of the screen grid voltage will be very poor.

All these things would need to be addressed to achieve the performance of the original design. If you just want to get it working reasonably well, I would consider installing a separate negative bias supply and an EFB™ screen grid regulator. This will allow the tubes to be biased properly and develop a similar amount of power output (if not more) than that of the original tubes. But the NFB, sensitivity, and stability issues would all have to be sorted out from there as desired. Is there any reason you don't want to use new manufactured tubes as used in the original design in the unit?

Either way, good luck with your project!

Dave
Thanks for that detailed information for me to look into , my plan was to put another Fisher output transformer back on it . I disabled the NFB already and the amp has good clean output , amps are like building a race car, you can over do things sometimes ! A good or winning combination can happen sometimes when the right parts come together, the tone quality of The RCA 6L6 blackplate tubes is more important to me then the Fisher and 7591 specs are to me ! . Certainly voltages are important to be within a certain range but they are not the only thing to consider and they can be over or under spec by 20% with no problems usually, good parts and tighter tolerances help, I was thinking a smaller 3 watt grid resistor would be safer than the 10 watt I used. The caps and resistors are not exact values to begin with and neither are the replacements ! . I am not trying to squeeze every bit of power out of it or to make it the best Fisher 600A on the planet , I just want a reasonably nice tone out of it and to me it is pretty loud already , I do not agree with lowering the voltages, I upped the screen resistor already to 500 ohm from 470 ohm and it sounds very good there !. I am not one to lower voltages in my amps. Output Tube specs are being exceeded all the time with good results. There are two rating systems also and that must be considered when you are comparing tube types. That is valuable information you gave me and I thank you for the deeper thought ! , I used to get that kind of input from James Henderson of James Audio in FT. Worth before he passed on and I learned a lot from it. But he usually in the end told me to do it my way and added a bass cap or something minor to my designs. I would try his ideas out and if I liked the sound I would use it if not I would remove it. It always came back to the sound over the spec in the end !. I am not a full time tech this is just a hobby for me and I am self taught from a WW2 course, I admit that I still have a lot to learn but I have basic electronics training and I have always been pretty good at diagnostics. This amp has two balance pots and a bias adjustment , I do not like new tubes in any of my amps. Thanks for the time Dave !
 
Last edited:
Further update , I cleaned the tuner up and controls and found a bad connection on a 6AU6 tube that was robbing a lot of volume, the amp is about 5 times louder now than it was. My plate voltage is 443 and screen is 433 . This thing sounds pretty good now and I don't think it will sound any better without adding tube rectification, that is the one thing I don't really like about it that it has a solid state rectifier in it , why build something high end without using tube rectification ? It Seems kind of stupid or being cheap to me ! . I checked the replacement output transformer to the fisher one and both have exactly the same 303 ohms each in total . Fisher = 157 ohms and 146 ohms primary from CT, replacement OT = 161 ohms and 142 ohms from CT. The secondary is a single 4 ohm output on the replacement OT and 8 and 16 ohm output on the fisher OT. I don't hear any difference in quality of sound between the two output transformers on 8 ohm speakers. I will still try to get back the fisher OT that came off of it , but I don't think that the fisher OT sounds any better than the replacement that is on it. I am not saying that I do not like the amp because I do like it and I do think it sounds good , I just think a tube rectifier would have made it sound even better. Next after the capacitors break in for awhile then I will roll some pre amp tubes with blackplates , mullards and amperex tubes to see which ones I like the best . Thanks again to everyone !
 
Last edited:
If it were mine, I would check it with 10k square waves, and see what it's doing.I have had amps that sounded "ok", but sounded much better after correcting the square waves.
 
If it were mine, I would check it with 10k square waves, and see what it's doing.I have had amps that sounded "ok", but sounded much better after correcting the square waves.
I do not own a square wave generator and I dId not plan on buying one either ! , The scope check will have to do ! Do You see anything wrong with this scope pattern at 15 watts output ?
 

Attachments

  • IMG_20180612_150403775.jpg
    IMG_20180612_150403775.jpg
    29.7 KB · Views: 41
I don't see sign waves, most basic checking is done at 10khz at 1 watt, it should look like a clean sign wave if that is your input.
 
I don't see sign waves, most basic checking is done at 10khz at 1 watt, it should look like a clean sign wave if that is your input.
Why would I want to check for square waves on a analog tube amplifier that has round waves ? Seems like the wave patterns and slopes would be more important ! The linear output matching the tube charts is what I always thought I was after ? The Fisher analog round wave tuner it came with is my input.
 
Last edited:
I believe a square wave would represent a most difficult type of signal to reproduce accurately, so viewing square wave performance will help to see how well the amp is performing. Ringing instability and high frequency rolloff are easily visualized with a 10KHz square wave and oscilloscope. If your objective is the most accurate, uncolored reproduction of your source material, this might be important.
 
Last edited:
Why would I want to check for square waves on a analog tube amplifier that has round waves ? Seems like the wave patterns and slopes would be more important ! The linear output matching the tube charts is what I always thought I was after ? The Fisher analog round wave tuner it came with is my input.
Need a distortion analyzer, tone generator and load box for a start.
Do I really need all of that if the amp already sounds good to me ?
 
At the very least you need to input a clean sign wave to start, ans compare it to the output. A 10k square wave reveals the shortcomings in an amplifier.
Here is an interesting thread dealing with my previously unknown transformers for this build.
http://www.audiokarma.org/forums/in...ncy-response-of-an-output-transformer.533185/

The build
http://www.audiokarma.org/forums/index.php?threads/i-am-starting-a-conn-mullard-5-20-project.679030/

You said your amp sounds good, but not great, that is the difference. I'm only trying to help you get the most out of your setup.
 
I believe a square wave would represent a most difficult type of signal to reproduce accurately, so viewing square wave performance will help to see how well the amp is performing. Ringing instability and high frequency rolloff are easily visualized with a 10KHz square wave and oscilloscope. If your objective is the most accurate, uncolored reproduction of your source material, this might be important.
Thanks for explaining what the purpose was for , and it is not that important to me. I'm using the Fisher analog tuner and it is solid state rectified so I already understand there a few limitations built into the amps sound, that is why I decided on using the 6L6 tubes.
 
Since the input square wave has sharp corners, the amp will try to reproduce those sharp corners. It will take very high supersonic bandwidth to reproduce them accurately, which of course no output transformer coupled amp has, so what you get is the square wave being a simple way to explore the high frequency limits and characteristics of the amp in a deterministic way, rather than needing to rely solely on your hearing. Because there is global feedback applied, frequency response above human hearing is really important to analyze as it can make the difference between a mediocre sounding amp and a great sounding amp.

Of course, doing the above requires a square wave generator. If you don't have one then you gotta play it by ear like you're doing.

But regardless of the above, additionally with the change in tube types you should test stability of the amp with varying types of loads. Speakers can present vastly varying load depending on frequency. I'd try resistor only, resistor in parallel with cap, and cap only loads. Usually a difficult load for the amp to manage is the cap only load of approx 0.1 uF to .01 uF. You can easily see it behave or misbehave on the scope.
 
Last edited:
Back
Top Bottom