Greetings,
I'm rebuilding a HF87 to a configuration I put together back in the late 70's with the 6SN7 configured as a differential phase splitter. I'll followup this post with more on the history of the amp, but for now the question is what to use for a constant current source on the tail of the 6SN7 differential pair. I used a zener/NPN pair previously (schematic below), but was wondering if there's a better way to do this.
I'm rebuilding a HF87 to a configuration I put together back in the late 70's with the 6SN7 configured as a differential phase splitter. I'll followup this post with more on the history of the amp, but for now the question is what to use for a constant current source on the tail of the 6SN7 differential pair. I used a zener/NPN pair previously (schematic below), but was wondering if there's a better way to do this.
I've had the amp since about 1970 when I was in junior high school, and got it from my brother in law who bought it when he worked at Allied Electronics back in the fifties when they were a stereo/hobbyist store. Looks like my grandson who's a vinylhead will inherit it next.
In the late 70's, I changed the phase splitter topology to a differential pair of 6SN7 triodes and cathode coupled the outputs (right terminology? - ran the cathode current through the transformer to provide a little local negative feedback.) Here's the hand scratched schematics of the amp circa 1980.
This seemed to work pretty well for a while, and eventually the amp ended up on a shelf, superseded by a variety of newer amps. I decided to take the amp back to the basic as-shipped Eico config with better parts - hey, it's a classic, right? Unfortunately I like the sound of the previous config better, and so we're going back to that.
I'll post updates as the project progresses. The critical path now is deciding what to do for the differential current source so we can get parts on order. Any feedback, ideas, suggestions are always welcome.
Oh, and here's the link to the end of the back-to-stock project in 2010 when the amp had some stability issues:
https://www.diyaudio.com/community/threads/unstable-eico-hf-87.169866/
In the late 70's, I changed the phase splitter topology to a differential pair of 6SN7 triodes and cathode coupled the outputs (right terminology? - ran the cathode current through the transformer to provide a little local negative feedback.) Here's the hand scratched schematics of the amp circa 1980.
This seemed to work pretty well for a while, and eventually the amp ended up on a shelf, superseded by a variety of newer amps. I decided to take the amp back to the basic as-shipped Eico config with better parts - hey, it's a classic, right? Unfortunately I like the sound of the previous config better, and so we're going back to that.
I'll post updates as the project progresses. The critical path now is deciding what to do for the differential current source so we can get parts on order. Any feedback, ideas, suggestions are always welcome.
Oh, and here's the link to the end of the back-to-stock project in 2010 when the amp had some stability issues:
https://www.diyaudio.com/community/threads/unstable-eico-hf-87.169866/
Osvaldo - thank you.
Looking at this again (decades later), I'm not sure this is a good two pin constant current source application. The semi-linear range of the 6SN7 runs from about -2V to -14V grid voltage. I'm not actually sure what the operating range will be. I noted 10V on the schematic, which I assume is what the grid voltage at idle should be. I'm pretty having a low and wide voltage range is why I biased a zener off B+ to provide a reference voltage to the bipolar current sink, though I'm also pretty sure I never measured the real world performance.
Any pointers to more info/circuits I can reference would be appreciated.
Looking at this again (decades later), I'm not sure this is a good two pin constant current source application. The semi-linear range of the 6SN7 runs from about -2V to -14V grid voltage. I'm not actually sure what the operating range will be. I noted 10V on the schematic, which I assume is what the grid voltage at idle should be. I'm pretty having a low and wide voltage range is why I biased a zener off B+ to provide a reference voltage to the bipolar current sink, though I'm also pretty sure I never measured the real world performance.
Any pointers to more info/circuits I can reference would be appreciated.
Idling cathode voltage looks like about +7VDC to maybe +10VDC, easy enough to find a believable number on the curves. Signal swing at the cathodes is exactly half the signal at the driven grid. Your existing circuit has excellent overhead/swing capacity, if not the very best possible stiffness/impedance. But source impedance only has to be reasonably "very large" compared to the cathodes' impedance (reciprocal of transconductance) and reasonably linear, and yours already is.
A "ring-of-two" current source could give a somewhat better stiffness/impedance, and cascoding (maybe with a high Idss FET to make it easy) helps either choice, but your existing circuit isn't likely holding you back any. For a really top-notch discussion of such topics, and many, many more, Bob Cordell's book can't be recommended too highly. Nothing tubey in there, but amplifiers are amplifiers.
All good fortune,
Chris
A "ring-of-two" current source could give a somewhat better stiffness/impedance, and cascoding (maybe with a high Idss FET to make it easy) helps either choice, but your existing circuit isn't likely holding you back any. For a really top-notch discussion of such topics, and many, many more, Bob Cordell's book can't be recommended too highly. Nothing tubey in there, but amplifiers are amplifiers.
All good fortune,
Chris
After spending too much time wandering the intertubes getting smarter on current sources, I went ahead and ordered the parts to put it back to my original design. Mostly because I want to have it up and running before Christmas. Once running, we can take some measurements to see what the actual performance of the circuit is.
Thanks guys for the replies, I picked up some links from your suggestions and think I have a clue as to what the options are. We'll see if a different approach is justified.
Jim
Thanks guys for the replies, I picked up some links from your suggestions and think I have a clue as to what the options are. We'll see if a different approach is justified.
Jim