Hi,
a few questions- could a single 12B4 (per channel) in anode follower connection, drive 300ohm headphones?
how low the output resistance of the headphone amplifier needs to be to be able to drive the headphones?
i am thinking of using 2 pieces of 12b4 and connect them in a classical anode follower connection (one per channel) and maybe also use a ccs in the plate of the tube.
i think that 12b4 will have arround 150-250 ohm output resistance when used that way... is this enough (with the plate current of arround 10-15mA) to drive the headphones like Sennheiser's?
thanks guys
a few questions- could a single 12B4 (per channel) in anode follower connection, drive 300ohm headphones?
how low the output resistance of the headphone amplifier needs to be to be able to drive the headphones?
i am thinking of using 2 pieces of 12b4 and connect them in a classical anode follower connection (one per channel) and maybe also use a ccs in the plate of the tube.
i think that 12b4 will have arround 150-250 ohm output resistance when used that way... is this enough (with the plate current of arround 10-15mA) to drive the headphones like Sennheiser's?
thanks guys
Just had a quick look at the datasheet curves which suggest that the internal anode resistance will be too high; in the region of 700 to 800 Ohms at best.
You could (parallel feed) transformer couple it through something like a Hammond 119DA, but then you might not have enough gain because the mu of a 12b4 is only x6.5.
I suggest that you try the 6H30 (mu about 14), ccs load and parrallel feed transformer couple it.
: )
You could (parallel feed) transformer couple it through something like a Hammond 119DA, but then you might not have enough gain because the mu of a 12b4 is only x6.5.
I suggest that you try the 6H30 (mu about 14), ccs load and parrallel feed transformer couple it.
: )
Gordy - thanks for the reply,
transformer is not an option. I would have to order it from outside the "delta quadrant" and it is, than, to expensive.
...
i tried srpp calculation with tubeCad software and it turned out that Zout is in between the 140 and 200 ohms. so if i use ccs in the anode and classic anode follower - the output impedance should be a bit lower than srpp i think - right
the qustion is if 150ohm is low enough to drive 300ohm headphones??
any opinions about that??!
transformer is not an option. I would have to order it from outside the "delta quadrant" and it is, than, to expensive.
...
i tried srpp calculation with tubeCad software and it turned out that Zout is in between the 140 and 200 ohms. so if i use ccs in the anode and classic anode follower - the output impedance should be a bit lower than srpp i think - right
the qustion is if 150ohm is low enough to drive 300ohm headphones??
any opinions about that??!
i do not have 6s45p. only 6n6p, 12b4, 5687 that are something like usable (regarding low output impedance).....so i have to do it with using them.
but i think if i use line level signal of arround 0db and using 12B4A with ccs in the anode - i might get something like 3-6 times gain of the complete stage. maybe i am wrong - but this could be enough together with bellow 200ohms of output impedance?!
i am not sure...
can somebody tell me - how much gain do i really need for driving the 300ohm headphones from line level voltage as input of the headphone amplifier?! and also - what is the output impedance that i should try to get at least (do not tell me - the lower, the better - i allready know that
). is it 10 times lower than impedance of the headphones or more or less... what to look here ???!
Hi Sunrise
The 6N6P will indeed give to much gain. I think that with a gain of 2 you have more than enough power to drive the headphones to ear splitting levels.
So you need to reduce the gain from about 20 to 2... how much heresy is it to introduce some local NFB (plate to grid?) in the circuit? I do not know the exact equations to calculate the feedback, but a good start may be to use a 100k from grid to ground (gridleak resistor) and a 100k from output (after the 'output coupling capacitor') to the grid. Have a look at the first tube in this schematic by Pete Millet.
http://www.pmillett.com/307a_pushpull_amp.htm
Another source talking about this arrangement can be found at tubecad. It is quite technical and you need some time to go through it.
http://www.tubecad.com/march2001/2001_03.pdf
Hope this helps to bring some nice music to the delta quadrant.
The 6N6P will indeed give to much gain. I think that with a gain of 2 you have more than enough power to drive the headphones to ear splitting levels.
So you need to reduce the gain from about 20 to 2... how much heresy is it to introduce some local NFB (plate to grid?) in the circuit? I do not know the exact equations to calculate the feedback, but a good start may be to use a 100k from grid to ground (gridleak resistor) and a 100k from output (after the 'output coupling capacitor') to the grid. Have a look at the first tube in this schematic by Pete Millet.
http://www.pmillett.com/307a_pushpull_amp.htm
Another source talking about this arrangement can be found at tubecad. It is quite technical and you need some time to go through it.
http://www.tubecad.com/march2001/2001_03.pdf
Hope this helps to bring some nice music to the delta quadrant.
i was just talking with a friend from Delta Quadrant...
it seems that 12b4a in the input and mosfet follower at the output might be the best option to go for.
this way i will have gain of the circuit of arround 3-6 times and also low output impedance (with arround 150mA of bias current) in the second stage.
this sounds good to me.
it seems that 12b4a in the input and mosfet follower at the output might be the best option to go for.
this way i will have gain of the circuit of arround 3-6 times and also low output impedance (with arround 150mA of bias current) in the second stage.
this sounds good to me.
i do not know any more.... there are numerous things that have to fit into place in order to also have a good amplifier. i know what i should do - a 30kg's tube headphone amplifier. but this i can't do because i want to have it small in oreder to fit it inside a small box. but also i want to have a tube. and to be able to drive head's that are 300R. so i guess that i will have to try. i will try 6n6p in parallel (one stage amplifier) or 12B4 as input (or 6n6p) and mosfet voltage follower as output (two stages) - first. than we will see what to do next if these two do not work as i want them to work.
thanks
regards
sunny
thanks
regards
sunny
sunrise said:
i tried srpp calculation with tubeCad software and it turned out that Zout is in between the 140 and 200 ohms. so if i use ccs in the anode and classic anode follower - the output impedance should be a bit lower than srpp i think - right
No. CCS as the anode load will give an output impedance (at the anode) equal to the internal Rp of the tube. Hence not lower, but about 700 - 800 Ohms (at the operating point that I estimate for your application).
However, the output impedance of a single triode will drop lower if you configure it as a 'mu follower'. Perhaps that is what you meant.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.