I'm building a great little line level preamp with 6922 tubes, and I thought it would be nice to add a headphone output. The trouble is, I don't want anything solidstate, or involving feedback. The output will be a 6922 cathode follower running at 15mA, with a current source below its bias R off the cathode. I'm thinking that maybe a 4:1 transformer could work to drive 32-600ohm headphones. I'm not much of an expert when it comes to transformers, but I'm pretty sure they only work well when sourced and terminated with very close to the proper impedance. 32 - 600 ohms is a pretty wide range. What if I add a SE EL84 buffer instead? Does this need to get complicated? Any thoughts? 
Well,
take a look at the Morgan Jones head-amp. It is also well suited as a line-amp. Chinese rebuilds exist.
HeadWize - Project: The Morgan Jones Mini Tube Headphone Amplifier by Chu Moy
LG Carsten
take a look at the Morgan Jones head-amp. It is also well suited as a line-amp. Chinese rebuilds exist.
HeadWize - Project: The Morgan Jones Mini Tube Headphone Amplifier by Chu Moy
LG Carsten
I'm trying out a ss unity gain buffer for my 6922 preamp. I don't have a cathode follower attached to the single stage tube preamp.
I need to drive down to 32 ohms. The target however is to drive an AKG-701 which is 60 ohms. I'm using MOSFET's . Sim results aren't bad and I will report back after I test it out. The MOSFETS use a separate lower voltage power supply. Currently it is 15 volts because I don't have the 18 V regulator that I had designed it for. Bias was to be at 0.1 Amp. That should easily run the 701 over 115 dBspl . So it will have enough output for clean transients.
I need to drive down to 32 ohms. The target however is to drive an AKG-701 which is 60 ohms. I'm using MOSFET's . Sim results aren't bad and I will report back after I test it out. The MOSFETS use a separate lower voltage power supply. Currently it is 15 volts because I don't have the 18 V regulator that I had designed it for. Bias was to be at 0.1 Amp. That should easily run the 701 over 115 dBspl . So it will have enough output for clean transients.
I'm building a great little line level preamp with 6922 tubes, and I thought it would be nice to add a headphone output. The trouble is, I don't want anything solidstate, or involving feedback. The output will be a 6922 cathode follower running at 15mA, with a current source below its bias R off the cathode. I'm thinking that maybe a 4:1 transformer could work to drive 32-600ohm headphones. I'm not much of an expert when it comes to transformers, but I'm pretty sure they only work well when sourced and terminated with very close to the proper impedance. 32 - 600 ohms is a pretty wide range. What if I add a SE EL84 buffer instead? Does this need to get complicated? Any thoughts?
Hello,
The 6dj8 cathode follower work well!
Try this transformer
EDCOR Electronics Corporation. WSM10K/600
and this capacitor
Dayton Audio DMPC-30 30uF 250V Polypropylene Capacitor
The above parts are inexpensive. Depending on the headphone impedance you can pick a smaller value capacitor.
You can set up the transformer using two impedance outputs (see the center tap) that will work for the range of loads (600 to 32 ohm) that you listed.
DT
All just for fun!
Anybody make any progress ?
I managed to make a pcb and populate the board for a ss unity gain buffer. It runs off a 0-18 V(ac) supply with supply parts on board ( except the trafo).
Will try it out tomorrow. Right now there are no stabilising caps. I'm sure I'd need one ! The scope will tell !
I managed to make a pcb and populate the board for a ss unity gain buffer. It runs off a 0-18 V(ac) supply with supply parts on board ( except the trafo).
Will try it out tomorrow. Right now there are no stabilising caps. I'm sure I'd need one ! The scope will tell !
I got the circuit working after blowing two transistors. Had accidentally shorted out two resistors.
Anyway all's well now. The CFP input stage required a 150pF cap for RF stability.
Now it's noiseless.
Sound is good too. I compared it with a typical ss headphone amp with an OPA2134 driving a pair of TO126 type output transistors.
I think it's slightly better in the HF and bass is more powerful....not just LF extension but 'meaty'! Mid range is similar.
However I feel the unity gain buffer is slightly veiled in some sounds. The 2134 seems a bit cleaner between notes but also seems to lack 'energy'.
Better IMD ?
From memory the JLH headphone amp ( split supply) sounds better. In fact the JLH sounded better than the 2134 when I tried it sometime ago. Right now I can't do that comparison due to lack of time.
I suspect the coupling capacitor to be the cause of the slight veiling .From experience , it's presence had a similar effect in other ss headphone map designs. So a better 470uF/16V cap might make a difference.
I could post the circuit in case anyone is interested. Nothing much really.
Note that the input stage for the unity gain headphone amp was a 6922 preamp. The ss OPA2134 amp ran directly from the DAC output.
Cheers.
Anyway all's well now. The CFP input stage required a 150pF cap for RF stability.
Now it's noiseless.
Sound is good too. I compared it with a typical ss headphone amp with an OPA2134 driving a pair of TO126 type output transistors.
I think it's slightly better in the HF and bass is more powerful....not just LF extension but 'meaty'! Mid range is similar.
However I feel the unity gain buffer is slightly veiled in some sounds. The 2134 seems a bit cleaner between notes but also seems to lack 'energy'.
Better IMD ?
From memory the JLH headphone amp ( split supply) sounds better. In fact the JLH sounded better than the 2134 when I tried it sometime ago. Right now I can't do that comparison due to lack of time.
I suspect the coupling capacitor to be the cause of the slight veiling .From experience , it's presence had a similar effect in other ss headphone map designs. So a better 470uF/16V cap might make a difference.
I could post the circuit in case anyone is interested. Nothing much really.
Note that the input stage for the unity gain headphone amp was a 6922 preamp. The ss OPA2134 amp ran directly from the DAC output.
Cheers.
Hello ashok,
Your discrete SS headphone buffer sounds like a pretty nice project with good results, driven by the 6DJ8 it sounds to be “Moskido” like. All that SS is the opposite end of the DIY from the requested Cathode Follower output version. (last comment added to stay on topic)
DT
All just for fun!
Your discrete SS headphone buffer sounds like a pretty nice project with good results, driven by the 6DJ8 it sounds to be “Moskido” like. All that SS is the opposite end of the DIY from the requested Cathode Follower output version. (last comment added to stay on topic)
DT
All just for fun!
Listening update
After two days of running in I tried listening to the amp again.
I kept playing several disks with headphones connected while I was busy doing household work etc. So last night I listened again. Something has "run in " !
It sounds perfectly clean now with no hint of the slight 'veiling' that I mentioned earlier. Transistors were not on a heat sink and were very hot. Maybe 60 deg C. But it was that way all of two days with the ambient at about 33 deg C.
So a CFP follower does sound very good and is suitable for low impedance headphones. Will try later with my 32 ohm phones. Don't have anything with lower impedance than that.
Cheers.
After two days of running in I tried listening to the amp again.
I kept playing several disks with headphones connected while I was busy doing household work etc. So last night I listened again. Something has "run in " !
It sounds perfectly clean now with no hint of the slight 'veiling' that I mentioned earlier. Transistors were not on a heat sink and were very hot. Maybe 60 deg C. But it was that way all of two days with the ambient at about 33 deg C.
So a CFP follower does sound very good and is suitable for low impedance headphones. Will try later with my 32 ohm phones. Don't have anything with lower impedance than that.
Cheers.
The measured temperature on the transistor tab was 69 deg C with an ambient of 29 deg C . I added a 60x35 mm aluminum heat sink. 2mm thickness. I also painted it matt black. Temp is about 10 deg lower at 60 deg C.
So if I add a finned heat sink it should probably drop to about 45 deg C.
I'll probably make the new pcb to suit the finned heat sink !
So if I add a finned heat sink it should probably drop to about 45 deg C.
I'll probably make the new pcb to suit the finned heat sink !
At this point it looks like I'm going to do a White Fowler output stage using the confusing 6N6pi or 6H6p. Our prototype appears to drive 32 ohms elegantly enough. The Russian tube part number system is different than, and clashes with that of the rest of the world. Our system also has a 6N6, but it's a dual diode tube. And when they write an N, it looks like our H. Or did I get that all backwards? Suffixes vary too, which may not matter. The Little Dot III headphone amp uses these (not sure which topology they are using) and it seems to get nothing but rave reviews, so I don't have a better idea. I really don't want to have to add more tubes such as paralleling output tubes. I'm including this circuit in a Line preamp project that will have 2 6SN7's and a 6922 already.
I may have been wrong about the 6N6pi White Follower being able to drive 32 ohms elegantly. Depending on levels, it may be sort of OK, but if I want quality I'll buy some 300ohm Sennheisers or equiv. I hear the 600's and 650's are hard to beat.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.