It works... see this post
Interesting hybrid circuit. If I understand correctly, L3 and L4 in your schematics are in fact a transformer.
What transformer did you use when you built the circuit ?
I'm interested in this kind of opinions since it's the "valve sound" I want from this headphone amplifier.
Can you recommend headphone amplifier circuits that sound good and have the "valve sound" ?
Yes, L3/L4 simulates the transformer.
I used the edcor pcw10k-7k/300-32. You can see them in the pics, it gives you an idea of their size compared to the caps/tubes.
A member here posted more data about it btw : see this post.
I used the edcor pcw10k-7k/300-32. You can see them in the pics, it gives you an idea of their size compared to the caps/tubes.
A member here posted more data about it btw : see this post.
00940's 6E5P triode strapped CCS plate loaded parafeed circuit is NOT going to have "a sound". It's going to be very very fast and transparent.
I've done a few similar circuits with 6E5P, only difference being I use a MOSFET gyrator plate load instead of CCS, but it's very similar in transparency.
I would recommend using only one parafeed cap, no parallelling.
I've done a few similar circuits with 6E5P, only difference being I use a MOSFET gyrator plate load instead of CCS, but it's very similar in transparency.
I would recommend using only one parafeed cap, no parallelling.
I hesitate to recommend an amplifier. Although I understand that some people like these amplifiers, it's purely an intellectual appreciation, they are actually not my thing. I'm sure somebody else will not be so reluctant. You could ask on head-fi, but suggestions are probably going to be (very) expensive and probably not DIY.
Which "valve sound" ?
a) rolled off top end frequency response and flabby bass ?
b)Absence of harmonics above 5th in the distortion spectrum?
c) no cross over distortion nor program related noise floor ?
d) soft, well behaved clipping?
e) lots of second harmonic distortion for "lush" sound?
All of the above can be achieved with valves and each has been called "the valve sound" by various people.
ps - another artifact of running DC in your trannies is that your magnatisation curve is very different to PP/parafeed
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
This should fit the bill, with the top end rolled off a bit... only 5mA in the final stage.
May I present this as food for thought. Performance is good out to 50khz and power toroidals can be used as both input and output transformers. A parafeed arrangement also works well.
A full compliment of transformers can be had for the cost of a single headphone OT. TRIAD make good candidates.
Shoog
A full compliment of transformers can be had for the cost of a single headphone OT. TRIAD make good candidates.
Shoog
Attachments
This is what I would do for a simple and cheap, yet extremely high performing headphone amp.
No need to make compromise for power efficiency for headphones. Linearity and dynamics (the Powerdrive really makes this - any circuit really - shine) with no global feedback. Cheap, simple and relatively easy.
This will not have a sound. It will be fast, musical and dynamic with lots of air and all that.
MOSFETs can be IRF820 or similar.
Tubes can be 6922 or variants, 2C51, 6SN7 or variants with no circuit changes. I would use 6N3P.
Or maybe use 6N16B and change the CCS tails for a bit smaller current and use maybe a bit lower B+. I've been thinking about making this amp for some time; would be very cute and pretty!
Next step: go directly heated.
No need to make compromise for power efficiency for headphones. Linearity and dynamics (the Powerdrive really makes this - any circuit really - shine) with no global feedback. Cheap, simple and relatively easy.
This will not have a sound. It will be fast, musical and dynamic with lots of air and all that.
MOSFETs can be IRF820 or similar.
Tubes can be 6922 or variants, 2C51, 6SN7 or variants with no circuit changes. I would use 6N3P.
Or maybe use 6N16B and change the CCS tails for a bit smaller current and use maybe a bit lower B+. I've been thinking about making this amp for some time; would be very cute and pretty!
Next step: go directly heated.
Attachments
Last edited:
As everyone seems to be pitching in with suggestions, here's mine:
http://www.diyaudio.com/forums/tubes-valves/191632-improved-tube-headphones-amp.html
Cheers
Ian
http://www.diyaudio.com/forums/tubes-valves/191632-improved-tube-headphones-amp.html
Cheers
Ian
Darned close...I'm working on a sim as per sketch but using a directly coupled starvation circuit reminiscent (per modified Mullard 3 watt amp) with generous nfb. As my hearing uppers are still fairly good, the bogey issue is noise. For my ears on classicals, a min noise floor <-80dBu down is the min requirement.
(more later)
richy
(more later)
richy
300 ohm Sennheisers are the only headphones I use, which should reflect 10k. It seems like it should be suitable for a few different kinds of tubes popular in "spud" parafeed like 5687.
The 10K cited is the secondary load, not the load reflected to the primary which is more like 336K with the recommended minimum load. This transformer is really not suitable for driving headphones, honest.
The 10K cited is the secondary load, not the load reflected to the primary which is more like 336K with the recommended minimum load. This transformer is really not suitable for driving headphones, honest.
Where does it say the secondary is 10k?
The winding ratio is 5.8:1
If you look at the lundahl data sheet you will see that the performance is specified into a 10K load on the secondary side. (The secondary is not 10K, it is the source impedance divided by the square of the turns ratio ignoring the effects of the shunting inductive reactance of the primary at low frequencies) These transformers are tiny and can't handle an appreciable amount of power.. Note the application. They do not specify the load impedance for the distortion measurements which probably means solely into the input impedance of the measurement setup which would typically be at least 100K ohms based on my experience.
http://www.lundahl.se/pdf/1930.pdf
http://www.lundahl.se/pdf/1930.pdf
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.