I don't know where the circuit design is going but I have a good idea for the chassis:
An externally hosted image should be here but it was not working when we last tested it.
I think things will become clear when a few of us actually build working models of an all-DHT headphone amp. Yes - they will be different, but all should work! That's what we found with the 26 preamp thread. It started off as theory and by now I wouldn't be surprised if there as many as a hundred working examples. They fall into distinct categories - active load, plate choke, transformer coupled - and people build what they can afford and are comfortable with.
I'm sure Sowter can provide a transformer solution, but I'm expecting that some of us will be looking for cheaper alternatives as well. We haven't talked about the ratio of headphone to speaker listening either. In my case I never use headphones, though I do have them.
I'm sure Sowter can provide a transformer solution, but I'm expecting that some of us will be looking for cheaper alternatives as well. We haven't talked about the ratio of headphone to speaker listening either. In my case I never use headphones, though I do have them.
I am leaning in this direction. Dsvavitz has some real good point on his website in support of Parafeed for headphones. The ability to use a permalloy core really is a benefit for headphones, first its better with low level signal integrity/loss, and second most importnt you don't end up with a secondary 32 ohm winding with 12 ohms DCR and too much induction.
I asked Magnequest to quote their 2A3 nickel parafeed with a 32 ohm secondary months ago, with nothing, they just aren't interested.
I think this little transformer parafeed could be the way to go: Sowter Type 8650 Headphone Output transformer
It only handles 0.5W at 20 hz, but is good for 2W at 50 Hz, if it was the next core size up it would be perfect for a 2A3 or parallel 4P1L's.
Very well said. I respect all input on this as we are treading new ground.
What is your opinion on parafeed?
Well, with just single stage 26 alone I am able to drive my own headphones beyond my comfortable listening level but they are 100dB+/mW efficient (Sony 7506s). I was using a parafeed MQ EXO-45 (1uF motor run cap) in the test circuit and now I'm waiting for Edcor to ship EDCOR - PCW10K-7K/300-32 which would allow for better impedance matching. It certainly wouldn't drive Orthos well but it is dead quiet and would work very well for nearly every other headphone from high 90's sensitivity and above. I think one aspect of the goal of this discussion that has been missing is that can a DIY'er buy the parts for this design with off the shelf components that are not 'the most exotic' in the world? So far, Rod's and Iko's circuits seem to fit that particular concern.
