Maybe about 200 different ones!But i have some brand new CF3,wich i also could use,but i dont have sockets,i made some tests in the past ,and they seem quite linear! and i accept all sugestions for this !I am not a specialist in tubes, i worked in tube tv in the past,but only replacing tubes/or associated circuits,and never went deep in their working characteristics!
Thanks
Why -3V bias on the grid?
I think this is a false target - there is no point or benefit in it, it makes no sense - can you explain?
OK! I want -3V bias because i want to connect a cd player to the input,an the output of a CD player is at least 2V rms, wich could make the tube to cuttof,and
therefore have Cross distortion,i want it to run in class A all the time! Am i wrong in this thought?
Thanks again
Hmmm. You have a conundrum here.
With 2Vrms in you will have a 5.6Vp-p signal at the grid of the first stage, followed by a concertina splitter and what output tube(s)?
Good chance the 2Vrms signal is going to overdrive the amp. I would suggest a volume control followed by the preceeding recomendations for biasing the input tube.
It is not uncommon to see an input sensitivity of 1Vrms or less to drive an amp into clipping, unless a volume control of some sort is used to reduce the input drive sensitivity.
With 2Vrms in you will have a 5.6Vp-p signal at the grid of the first stage, followed by a concertina splitter and what output tube(s)?
Good chance the 2Vrms signal is going to overdrive the amp. I would suggest a volume control followed by the preceeding recomendations for biasing the input tube.
It is not uncommon to see an input sensitivity of 1Vrms or less to drive an amp into clipping, unless a volume control of some sort is used to reduce the input drive sensitivity.
In fact, the slope is not a constant, it changes with anode current and voltage and can become ridiculously low at low anode current.
But at the same time, the Rp changes in the opposite direction so that the Mu (amplification factor) wich is their product remains more or less constant.
However, the limited value of the load resistance reduces this factor unless you use a current source (infinite impedance) as load.
In this case, beware of the input impedance of the next stage that appears in parallel and may ruin your expectations.
Yves.
Oh I see - you are worried about the input tube clipping!
Hmmm.
So at -1V grid you'll swing to -3V to +1V, whereas you only want to go down to 0V if possible. You won't be able to run at 3mA then! Actually +1V on the grid may be OK - it depends how linear it is around the whole point.
The input stage of my Sweet Peach does a good job with 6N3p tubes, a 100k anode resistor from 350V, a 2.2K cathode resistor giving about -4.1V grid voltage with a 14.3x gain (i.e. 2V -> 28.6V).
At 90V anode (or even 120V) I think you may need a lower input. I suppose the way to calculate things is to start from the output and work backwards 😉
My own view is that to pump 2V RMS into a grid you need a bit more anode voltage - and that's the limitation.
Thank you all for the replies i had!
I just finished another test with ECC88 with RP 68K, RK 1K and 90V plate voltage and i have about -2,5 V in grid,and the sound is just FINE !
I am now very Happy😀.
THANKS
Hi again!
I am very happy with this circuit, but it would be perfect with a tube with more gain,the ECC88 has amplification of 33 i think and the ECC81 is about 62 ,But i only can achieve this values with the ECC88(Va 90V-Vg 2,5V) .There must be a tube that can work OK with this values,and having more gain(amplfication).THAT IS WHAT I AM LOOKING FOR ! I dont want to add another tube tu this circuit!
Thanks all
I am very happy with this circuit, but it would be perfect with a tube with more gain,the ECC88 has amplification of 33 i think and the ECC81 is about 62 ,But i only can achieve this values with the ECC88(Va 90V-Vg 2,5V) .There must be a tube that can work OK with this values,and having more gain(amplfication).THAT IS WHAT I AM LOOKING FOR ! I dont want to add another tube tu this circuit!
Thanks all
I'm pretty sure that a pentode could solve all your problems. I don't have any candidate tubes myself but I'm sure someone with more pentode experience can help out here.
/Olof
/Olof
Yes, something like the ECF80 can be used to provide pentode gain stage and triode splitter in one bottle. I think the KEL84 uses this idea.
I have a bunch of those tubes, so might give it a try on my little EL84PP. It has similar B+ to the original poster's amp.
I have a bunch of those tubes, so might give it a try on my little EL84PP. It has similar B+ to the original poster's amp.
SY's step network suggestion is a good one. It solves the problem of voltages very nicely with no loss of gain and can free you to select almost any input tube that you like, with a sensible operating point.
Yves' question is relevant. If the cathode R is unbypassed it will give you more leeway on the amplitude of the IP signal, because the cathode will follow the grid (within limits). Of course, you lose some gain through current feedback.
Hemgjord's idea to use a pentode sounds OK too. EF86 can work quite happily with only 90v on its plate and 60-70v on its screen. If the gain is too high, you can always omit the Rk bypass cap.
Yves' question is relevant. If the cathode R is unbypassed it will give you more leeway on the amplitude of the IP signal, because the cathode will follow the grid (within limits). Of course, you lose some gain through current feedback.
Hemgjord's idea to use a pentode sounds OK too. EF86 can work quite happily with only 90v on its plate and 60-70v on its screen. If the gain is too high, you can always omit the Rk bypass cap.
Yes! I used a 1000mf ! I always do it ,because i dont like NFB! and have more gain!
I am not going to do more tests ,because i am very happy with the sound i have with this one. (ECC88 with Va 90V Rk 1K and -2.5 V on grid And RP 68K with B+250V)! It works just perfect.I think it is very important to have the grid bias at least 2,5V in this input stage!
THANKS FOR YOUR ADVISE
- Status
- Not open for further replies.