I'm sure you're doing it because it's fun. Try both. You're going to hear some 2nd order harmonic distortion in the SE config, but it sounds good to a lot of people.
The P-P will, I think, yield a little more power but you may be down to giving it a little NFB from the top of the output secondary or by biasing it far into class A for distortion. Feeding NFB to a balanced input, however, may be, in the words of Elmer Fudd, "vewy, vewy twicky."
Good luck and do let us know how it works out.
The P-P will, I think, yield a little more power but you may be down to giving it a little NFB from the top of the output secondary or by biasing it far into class A for distortion. Feeding NFB to a balanced input, however, may be, in the words of Elmer Fudd, "vewy, vewy twicky."
Good luck and do let us know how it works out.
Right now I do not have any PP transformers, but I like this tube 6N6P a lot...
Recently I replaced original tubes in Musical Fidelity headphone amp, the original first version, with 6N6P tubes, and boy! does it sound better.
Downside is I had to use external power supply for heaters, as these tubes take quite a lot of current.
Recently I replaced original tubes in Musical Fidelity headphone amp, the original first version, with 6N6P tubes, and boy! does it sound better.
Downside is I had to use external power supply for heaters, as these tubes take quite a lot of current.
Might try...
You might also try the 5965 which has similar characteristics but might be a little less of a heater current hog. What makes these tubes so sweet is their relatively high operating plate current which puts the signal right up in a very linear part of the transfer curve. The operating impedance is also relatively low making them a good choice for an otl headphone amp. I built a 5965 preamp and the result very pleasing to the ear.
5687 is a good tube to look into, too, but at 1 amp per tube heater current you also need lots of transformer. Conrad Johnson used both tubes at one time or another.
You might also try the 5965 which has similar characteristics but might be a little less of a heater current hog. What makes these tubes so sweet is their relatively high operating plate current which puts the signal right up in a very linear part of the transfer curve. The operating impedance is also relatively low making them a good choice for an otl headphone amp. I built a 5965 preamp and the result very pleasing to the ear.
5687 is a good tube to look into, too, but at 1 amp per tube heater current you also need lots of transformer. Conrad Johnson used both tubes at one time or another.
Which MF headphone amp you did upgrade? Would like to know as I have 6N6P tubes and might consider upgrade, too.
thanks
If a 5687 with Va = 200V and Ia = 15mA has ra of about 3k5 ohms, wouldn't the primary impedance of the OPT for a single-ended headphone amp need to be about 8k:VC? That would mean a high inductance primary with a pretty big turns ratio.
6N6P would be very similar to 5687, so the situation should be the same.
--
6N6P would be very similar to 5687, so the situation should be the same.
--
Last edited:
Hi artosalo,
any news about this? I would like to build one if I could have information about the OTP.
I have not been able to build anything so far since I have been abroad.
But I have designed and ordered the circuit boards for this project.
The output transformer should be some small 8k or 10k type, for example this:
TGL 20/002 INDEL - Transformer: speaker | TME - Electronic components
Few pictures and schematics:
But I have designed and ordered the circuit boards for this project.
The output transformer should be some small 8k or 10k type, for example this:
TGL 20/002 INDEL - Transformer: speaker | TME - Electronic components
Few pictures and schematics:
Attachments
Very nice! I was wondering what if we focus in low cost rather than performance, could we then use toroidal power transformer for OTP? Something ready of the shelf, but then what should be the secondary voltage and VA to fit the project? Of course, if all this doesn't make any sense, I apologize.
I managed to build and briefly test the prototype.
THD is not as low as simulations showed, but still quite good. At the "desing goal" 7 W the THD is 0.25 %.
Max. power output to 8 ohms load is some 8 W and to 4 ohms as high as 9 W but with higher THD of course.
I adjusted the high frequency limit (-1 dB) to 62 kHz, and the result is very a good transient response.
Attached the plot from my scope with 10 kHz square wave. The GNFB is about 16 dB.
I noticed that the zener bias works, but is not very convenient. Therefore I made an adjustable bias based on TL431 reference and one power PNP.
Now the adjustment is easier with different tubes.
THD is not as low as simulations showed, but still quite good. At the "desing goal" 7 W the THD is 0.25 %.
Max. power output to 8 ohms load is some 8 W and to 4 ohms as high as 9 W but with higher THD of course.
I adjusted the high frequency limit (-1 dB) to 62 kHz, and the result is very a good transient response.
Attached the plot from my scope with 10 kHz square wave. The GNFB is about 16 dB.
I noticed that the zener bias works, but is not very convenient. Therefore I made an adjustable bias based on TL431 reference and one power PNP.
Now the adjustment is easier with different tubes.
Attachments
Last edited:
I am not working on this project so far. I have built only one channel and I think it is quite completed and ready now.
Most of my experiments remain at this level, because my interest is mainly technical.
I don't even listen all prototypes, because I believe that well performing tube amplifier can not sound bad. At least I have never observed such case. Most essential information is above at post #36, but I also have a bit more extensive test results if you are interested.
Most of my experiments remain at this level, because my interest is mainly technical.
I don't even listen all prototypes, because I believe that well performing tube amplifier can not sound bad. At least I have never observed such case. Most essential information is above at post #36, but I also have a bit more extensive test results if you are interested.
Sure. I have adjusted the idle current to maximum power dissipation, which is 4 W/triode halve. With 230 V anode voltage the max. idle current is 17 mA.
With lower +Ub (<200 V) the idle current can be increased to 20 mA and more. This provides a bit better linearity, but lower output power.
For headphone amplifier this construction provides more than sufficient power.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.