Huh? I can reach it. But that site was down some time.h_a said:
An externally hosted image should be here but it was not working when we last tested it.
Mike,
If you're going to do an apples-to-apples comparison between Curl and Pass, you'll need an early Threshold model from the same time period, not something recent.
The output devices for the JC-3 are still available, but I'm not sure what you'd do for the front end parts. On the other hand, if you update the parts--presumably including the outputs--you're looking at a different animal and comparisons become somewhat more hazy. You're using John's topology, but parts choices will lead to different operating points/resistor values/gain/feedback/etc. How much it would sound like the original thing is anybody's guess.
I could (and still may) do a version of the JC-3 that uses bipolars in one or more positions, but for now it amuses me to do an all FET version in a manner that I imagine might, in some alternate universe, qualify as "how John would do it" in the here and now. I'm faced with the same general problem regarding parts choices, but I'm free of many of the constraints that you will face in trying to reproduce something from that era.
I seem to recall that a former member here tried something along these lines. From what I recall of the schematic, I'm making different choices and will end up with yet a third take on the JC-3. It's an interesting project.
Grey
If you're going to do an apples-to-apples comparison between Curl and Pass, you'll need an early Threshold model from the same time period, not something recent.
The output devices for the JC-3 are still available, but I'm not sure what you'd do for the front end parts. On the other hand, if you update the parts--presumably including the outputs--you're looking at a different animal and comparisons become somewhat more hazy. You're using John's topology, but parts choices will lead to different operating points/resistor values/gain/feedback/etc. How much it would sound like the original thing is anybody's guess.
I could (and still may) do a version of the JC-3 that uses bipolars in one or more positions, but for now it amuses me to do an all FET version in a manner that I imagine might, in some alternate universe, qualify as "how John would do it" in the here and now. I'm faced with the same general problem regarding parts choices, but I'm free of many of the constraints that you will face in trying to reproduce something from that era.
I seem to recall that a former member here tried something along these lines. From what I recall of the schematic, I'm making different choices and will end up with yet a third take on the JC-3. It's an interesting project.
Grey
Folks, it is the topology that matters, not the parts. You can use 2sk170, 2sj74 for the inputs. The second stage can be any reasonable 1W bipolar transistor pair. The RCA devices were slightly better than most. The driver stage can be any 2A minimum high speed complemenary pair, and the output devices were the 2N5884, 2N5886. These were the best parts available in 1970 or so. This amp can do about 100V/us and should be compared to an Electrocompaniet, if you want a real horse race.
The original Electrocompaniet was a pretty nice amplifier, particularly at the time. We used to carry those back when I was in retail. Sold very few, though, given the price/wattage ratio. They sold for something like $1400 if memory serves. People would almost inevitably go for something that provided more watts for the buck. The Conrad Johnson MV-75 sold for roughly the same price and provided half again as much power. Needless to say, we sold Hafler DH-200s by the truck load, MV-75s once every month or two, and an Electrocompaniet...almost never; even though people admired the sound.
For those who think $1400 is really quite reasonable, all I can say is to find an online inflation calculator and look at it in 1980 dollars. Trust me, it was a horrendous amount of money back then.
John,
I agree that the topology is the interesting part, although using MOSFETs for the outputs is forcing me to use higher rails, which in turn means that I can't use the 2SK389/2SJ109 without cascoding them to take some of the voltage (and arguably increase the performance, depending on your point of view), which in turn...oh, well.
Incidentally, the Electrocompaniet schematics are, or were, available on the web at one point. If anyone wants to build one of the classic solid state amps, that would be a good place to start.
Grey
For those who think $1400 is really quite reasonable, all I can say is to find an online inflation calculator and look at it in 1980 dollars. Trust me, it was a horrendous amount of money back then.
John,
I agree that the topology is the interesting part, although using MOSFETs for the outputs is forcing me to use higher rails, which in turn means that I can't use the 2SK389/2SJ109 without cascoding them to take some of the voltage (and arguably increase the performance, depending on your point of view), which in turn...oh, well.
Incidentally, the Electrocompaniet schematics are, or were, available on the web at one point. If anyone wants to build one of the classic solid state amps, that would be a good place to start.
Grey
John,
I'm not aware of a 2SK51/2SJ51 part, and Google doesn't seem to be able to find such, either. Did you mean something like the 2SK215?
Like the Toshiba parts, Hitachi/Renesas are unobtanium as far as I can tell. Digikey carries Renesas, bwess their widdle hearts, but any of the likely TO-220 parts are "non-stock." (Predictably.) Man, I hate this parts-sourcing problem. I had to buy hundreds of the J271s in order to get the Vishay/Siliconix parts. What I'll do with that many, I do not know.
I vaguely remember reading a post somewhere--perhaps from you--saying that the Hitachi MOSFETs are not optimal...something about the N-ch (or P-ch) parts operating in triode region while its complement operates in pentode region. I came away from that with the impression that the Toshiba parts are better choices. Naturally, they're no easier to find.
Yes, I agree about dropping the follower driver stage for MOSFET outputs. My prototype goes straight off the VAS to the outputs. As it stands, I'm running Iq 10-15mA on the VAS (threatening to go to 50mA) and debating the number of output pairs (two vs. three pairs) vs. Iq (assumed to be minimum 1A per output device, but no more than 2A) vs. cumulative Gate capacitance (vertical MOSFETs, so it's comparatively high).
Add to that my quest for more open loop bandwidth, and suffice it to say that I've still got a bit of work to do before I'm satisfied.
Grey
P.S.: I've had requests for the schematic for my "JC-3." I don't feel that it's up to snuff yet and since I can't afford to hire John to improve my design (sorry, twins, no diapers for you this week...Poppa wants to pay John Curl to help him with his hobby circuit), I'm going to have to keep slogging through this on my own. Anyone who remembers the Aleph-X saga will know that I didn't start the actual thread until I had played with the thing for several months.
P.P.S.: If anyone thinks I'm going to water-cool this thing, you may be right. I've got a few moderate-sized heat sinks on hand. If the Pd passes what I can handle with those, I'll just say screw it and do the extra plumbing. I may have to resurrect my old water-cooling thread for an update.
I'm not aware of a 2SK51/2SJ51 part, and Google doesn't seem to be able to find such, either. Did you mean something like the 2SK215?
Like the Toshiba parts, Hitachi/Renesas are unobtanium as far as I can tell. Digikey carries Renesas, bwess their widdle hearts, but any of the likely TO-220 parts are "non-stock." (Predictably.) Man, I hate this parts-sourcing problem. I had to buy hundreds of the J271s in order to get the Vishay/Siliconix parts. What I'll do with that many, I do not know.
I vaguely remember reading a post somewhere--perhaps from you--saying that the Hitachi MOSFETs are not optimal...something about the N-ch (or P-ch) parts operating in triode region while its complement operates in pentode region. I came away from that with the impression that the Toshiba parts are better choices. Naturally, they're no easier to find.
Yes, I agree about dropping the follower driver stage for MOSFET outputs. My prototype goes straight off the VAS to the outputs. As it stands, I'm running Iq 10-15mA on the VAS (threatening to go to 50mA) and debating the number of output pairs (two vs. three pairs) vs. Iq (assumed to be minimum 1A per output device, but no more than 2A) vs. cumulative Gate capacitance (vertical MOSFETs, so it's comparatively high).
Add to that my quest for more open loop bandwidth, and suffice it to say that I've still got a bit of work to do before I'm satisfied.
Grey
P.S.: I've had requests for the schematic for my "JC-3." I don't feel that it's up to snuff yet and since I can't afford to hire John to improve my design (sorry, twins, no diapers for you this week...Poppa wants to pay John Curl to help him with his hobby circuit), I'm going to have to keep slogging through this on my own. Anyone who remembers the Aleph-X saga will know that I didn't start the actual thread until I had played with the thing for several months.
P.P.S.: If anyone thinks I'm going to water-cool this thing, you may be right. I've got a few moderate-sized heat sinks on hand. If the Pd passes what I can handle with those, I'll just say screw it and do the extra plumbing. I may have to resurrect my old water-cooling thread for an update.
john curl said:
I agree... I ran a business for a number of years that updated classic designs leaving the topology intact while improving the basic circuit environment. A bit ahead of it's time but I learned alot... More fun than being a textbook nerd and some people got it.
Regards, as always,
Mike.
Mike,
You don't need to worry about PC boards for me. I've already got a functional layout for a single-ended version etched and tested. Do whatever suits you. I'll be beating on this version for a little while longer, then I'll probably try another version of the circuit with a slightly different VAS. That will require a slightly different layout. So trying to do something to suit me is a waste of time--I'm a moving target.
What parts are you planning on using?
John,
If you've got a moment, I'd appreciate it if you could clarify your suggestion for the K51/J51.
Grey
You don't need to worry about PC boards for me. I've already got a functional layout for a single-ended version etched and tested. Do whatever suits you. I'll be beating on this version for a little while longer, then I'll probably try another version of the circuit with a slightly different VAS. That will require a slightly different layout. So trying to do something to suit me is a waste of time--I'm a moving target.
What parts are you planning on using?
John,
If you've got a moment, I'd appreciate it if you could clarify your suggestion for the K51/J51.
Grey
