A lot of interest in vintage integrated amps these days and for good reason, they have high voltage rails and have no problem with deliverying fairly low distortion at high power levels into headphones. Fact is many of the famous old school headphones were designed specifically for a receiver output, not an opamp.
Digression,
I know the headphone community at this time is divided into the O's and the S's, but headphone design is still a black art and not engineered by objectivists, but rather folks more along the lines of violin builders. A headphone is a musical instrument At least I hope we continue to look at other design philosophies, 'cause until a 140dB high BW nano-tech microphone is implanted in my ear, I can acknowledge that there are still a lot of unknowns while I do appreciate the works of the dScope and designing an opamp based amplifier around one, I hate to see a stagmentation with the "wire-with-gain and a whole lot of NFB" philosophy being the only option..
Modelling Mosfet emitter followers in Spice, and one can fairly easily put together a buffer that has less than .005% THD with no NFB, should be pretty transperant.
The issue with the old receivers, is there is a lot of stuff we don't need for headphones, and with their age replacing a bunch of caps restoring one is not so much fun when something else goes wrong. The other issue is much of their sound characteristics are from the input gain stage which aren't necessarilly a good choice for especially for headphones, then there are the dropper restors that take away the noise and some detail with them thats the drawbacks to old vintage receivers.
There are a million debates on the subject but according to LTspice the old audio spec's lateral Mosfets used in vintage gear show less distortion, many folks prefer these old mosfets to the new Hexfets. And it may be another reason folks are driving up the prices of old japanese receivers on ebay.
So what if we take the approach of a tube stage that is very linear like the work being done in the #26 & 4P1L preamp threads and instead of an output transformer we use a vintage SS amp output stage?
I have an problems with tube output transformers that are currently on the market, none are really ideal for the low signals. Every dang one on the market is oversized. Output transformers for headphones are a unicorn, they exist on paper but very few have really made a proper one.
So the concept is a single 4P1L DHT (mu=8) driving a Tumos. Its an idea heavily stolen from the beau2317's Moskido speaker amp project and mogliaa's 4P1L DHT preamp project.
The Tumos could be called a vintage receiver output stage with a more reliable/modern biasing and dc servo.
Of course one would load the anode follower 4P1L stage with a CCS powered by a SSHV2, and the Rodecoleman DHT regs would probably be a requirement for headphones.
So the headphone amp would look like this:
Digression,
I know the headphone community at this time is divided into the O's and the S's, but headphone design is still a black art and not engineered by objectivists, but rather folks more along the lines of violin builders. A headphone is a musical instrument At least I hope we continue to look at other design philosophies, 'cause until a 140dB high BW nano-tech microphone is implanted in my ear, I can acknowledge that there are still a lot of unknowns while I do appreciate the works of the dScope and designing an opamp based amplifier around one, I hate to see a stagmentation with the "wire-with-gain and a whole lot of NFB" philosophy being the only option..
Modelling Mosfet emitter followers in Spice, and one can fairly easily put together a buffer that has less than .005% THD with no NFB, should be pretty transperant.
The issue with the old receivers, is there is a lot of stuff we don't need for headphones, and with their age replacing a bunch of caps restoring one is not so much fun when something else goes wrong. The other issue is much of their sound characteristics are from the input gain stage which aren't necessarilly a good choice for especially for headphones, then there are the dropper restors that take away the noise and some detail with them thats the drawbacks to old vintage receivers.
There are a million debates on the subject but according to LTspice the old audio spec's lateral Mosfets used in vintage gear show less distortion, many folks prefer these old mosfets to the new Hexfets. And it may be another reason folks are driving up the prices of old japanese receivers on ebay.
So what if we take the approach of a tube stage that is very linear like the work being done in the #26 & 4P1L preamp threads and instead of an output transformer we use a vintage SS amp output stage?
I have an problems with tube output transformers that are currently on the market, none are really ideal for the low signals. Every dang one on the market is oversized. Output transformers for headphones are a unicorn, they exist on paper but very few have really made a proper one.
So the concept is a single 4P1L DHT (mu=8) driving a Tumos. Its an idea heavily stolen from the beau2317's Moskido speaker amp project and mogliaa's 4P1L DHT preamp project.
The Tumos could be called a vintage receiver output stage with a more reliable/modern biasing and dc servo.
Of course one would load the anode follower 4P1L stage with a CCS powered by a SSHV2, and the Rodecoleman DHT regs would probably be a requirement for headphones.
So the headphone amp would look like this:
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.