Depends. Schematic? The CCS I used was a cascode bipolar.
The schematic is attached below, the current is a little less than your measurement subject I think, TP1 is 140V so current is (250-140)/270k = 0.41mA
Not sure what a cascode bipolar looks like TBH, every time I look on the internet I get a different toplogy, some with diodes, some with leds here, leds there, the odd potentiometer etc. Very very confusing. I've attached a picture of what I think a Bipolar cascode current source is (from Walt Jung's paper), but like I say, it could be anything 😕
I think I understand why people use a resistor 😉
Attachments
Probably not suitable for that design.
The bipolar CCS I used was a diyAudio group project; I think Greg the Geek was selling boards and documentation for it; it's pretty similar to the designs shown in Morgan Jones's book (though without the misprint, harrumph). The one you showed is actually a ring of two rather than a cascode.
The bipolar CCS I used was a diyAudio group project; I think Greg the Geek was selling boards and documentation for it; it's pretty similar to the designs shown in Morgan Jones's book (though without the misprint, harrumph). The one you showed is actually a ring of two rather than a cascode.
Hehehe, I knew it would be something sneaky right under our noses 😀
It could be one of many tubes, that's why I've asked about V/A curves.
The schematic is attached below, the current is a little less than your measurement subject I think, TP1 is 140V so current is (250-140)/270k = 0.41mA
You can use LED and CCS with V2 since it is loaded on a high resistance bootstrapped input.
Last edited:
I've previously looked at 12AX7 curves and thought it should do well at a constant 1mA or so, but then thought the output z might be too high to be usefull. I'll look again.
SY, I'm curious... what R and C did you load it with for that test?
(Thank you)
SY, I'm curious... what R and C did you load it with for that test?
(Thank you)
What kind of 12AX7 did you test? A variety of different manufacturers or several of one manufacturer?
John
John
Not sure what a cascode bipolar looks like TBH, every time I look on the internet I get a different toplogy
There was a good document somewhere on this forum called
diyAudio-CCS-beta2.pdf
Or an alternative way from Geek: http://www.classicvalve.ca/docs/CVD_CCS_preliminary_docs.pdf
The particular spectrum I showed was a smooth plate Telefunken. I also tested some small plate eastern European tubes (several brands, all clearly from the same factory). 5 tubes in all (2 Telefunken, 3 small plate), 10 sections, pretty similar results. Haven't tested a JJ, but I have one so I should.
The particular spectrum I showed was a smooth plate Telefunken. I also tested some small plate eastern European tubes (several brands, all clearly from the same factory). 5 tubes in all (2 Telefunken, 3 small plate), 10 sections, pretty similar results. Haven't tested a JJ, but I have one so I should.
Thanks, though I thought I was done hearing about smooth plate Telefunken.🙂
John
Not sure what a cascode bipolar looks like TBH, every time I look on the internet I get a different toplogy,
Glubulator (...and a thousand lurkers...), for a practical example see post #5 from Mr. tubelab in this thread: http://www.diyaudio.com/forums/tubes-valves/61175-parafeed-resistive-load.html
: )
Not sure what a cascode bipolar looks like TBH, every time I look on the internet I get a different toplogy,
(OK, so I'm bored, so...)
Post #25 in this thread: http://www.diyaudio.com/forums/tubes-valves/63652-mosfet-current-source-triode-load-3.html
... time for breakfast.
Ok, guess what is it, if I don't tell about Volt / Ampere characteristics (in this particular case, size of the flower)?
Tip: it's from my garden.
Tip: it's from my garden.
An externally hosted image should be here but it was not working when we last tested it.
Very high. VERY high. 10M, 10pF.
The particular spectrum I showed was a smooth plate Telefunken.
I'd hope the $100 tube into a near infinite load measured well. The real test is the JJ into a 250K load.
If I had gotten to this thread earlier, I would've guessed a 1L6. But I guess there probably aren't too many Zenith TransOceanic owners on here, are there? 😀
I might be making a phono preamp soon...I'm definitely considering that nifty 12AX7.
Kyle
I might be making a phono preamp soon...I'm definitely considering that nifty 12AX7.
Kyle
Funny thing about 12AX7, it's got nice and linear curves; funny thing about 6SL7, it's even more linear, to the point where almost ANY loadline will have very little THD. Amazing.
Tim
Tim
....funny thing about 6SL7...
Distortion spectra or it didn't happen. 😀
I'd hope the $100 tube into a near infinite load measured well. The real test is the JJ into a 250K load.
A 250k load is much too low for good performance from this tube- that was one of the points I was trying to make. Most 12AX7 circuits are sub-optimal.
FWIW, the spectra from the cheap eastern European tubes (some bought at Radio Shack!) were pretty much the same as the Telefunken; I just happened to have the Telefunken curves handy. Based on what I saw from the JJ ECC81, I wouldn't be surprised if their ECC83 performed even better.
If I had gotten to this thread earlier, I would've guessed a 1L6. But I guess there probably aren't too many Zenith TransOceanic owners on here, are there?
I have two, both work fine.
Ok, guess what is it, if I don't tell about Volt / Ampere characteristics Tip: it's from my garden.
OK, I have a power supply, a current meter and some clip leads. I have some doubts about the linearity though.
Ok, guess what is it, if I don't tell about Volt / Ampere characteristics (in this particular case, size of the flower)?
I sense a duel; pistils at dawn. 😀
- Status
- Not open for further replies.
- Home
- Amplifiers
- Tubes / Valves
- Guess That Tube