diyAudio (
-   Headphone Systems (
-   -   Hybrid tube headphone amp design (

ashok 21st November 2002 01:06 PM

Hybrid tube headphone amp design
2 Attachment(s)
I have been thinking of building a tube headphone amp with the ability of driving a wide range of headphones . Class A most of the time but capable of higher voltages and currents if needed in class AB mode.

I will attach a circuit that I checked out with Spice. It seems OK. Any comments or suggestions from anyone before I build a prototype , is welcome.
The supply has been restricted to 150 volts and the output MOSFET's are working in Class A mode for currents of less than 11mA.


dhaen 21st November 2002 01:34 PM


Are you sure this isn't "valves for valves sake":confused:


fdegrove 21st November 2002 02:26 PM


At 150V B+ that 12AU7A is not going to be very happy.

A more appropriate choice at that B+ would be a ECC88/6DJ8.

Just my:2c:

ashok 21st November 2002 02:32 PM

Tube sake?
Hi John,
I must admit this is for 'tubes' sake ! Actually I have built so many discrete and opamp based headphone amps in the past I am 'hoping' that this may sound better.
It could be more like let me build this and see how good it is - I have had a good time with valve based circuits. I also admit that not all valve circuits are better than solid state circuits.

In my circuit I goofed the feedback part of it. Having a 47K resistor in series with the input must be a really terrible thing to do with respect to noise. But being a line level stage I am currently ignoring it. The feedback component values are not right and I will post another message after I check it out.

Don't you think it is interesting to see how this will sound ?
I can rig this up in a day!

ashok 21st November 2002 02:40 PM

Hi Frank,
I know the 6DJ8 is much better at low voltages. I did this so that I could find some use for the dozens of ECC82's that I have ! I have very few 6922's that I got from Singapore. Those are reserved for better applications. The low voltage was chosen to use less expensive supply components and to cut dissipation on the MOSFET's.

In the Spice simulation the distortion of this circuit is quite low but this need not necessarily tell us how it sounds. Only way to get a conclusive idea is to build one. This is not a best of components or circuit configuration . Just wanted to use the ECC82 for a class A headphone amp. With the MOSFETS loading it the distortion should not go up too much.


halojoy 21st November 2002 03:01 PM

How much current do your headphones need?
That is, what impedance/sensitivity do they have?
I guess it is 600ohms HP.
11 mA can be enough.

ashok 21st November 2002 03:57 PM

I was aiming for about 10mW into 100 ohms (class A ) . That would also be enough for a 92db/300 ohms phones that I have.
I also have a 92db/40 ohms Audio Technica. For normal listening it should be in the class A range. I do not blast my headphones - cannot afford to kill my hearing.
I just found that the April 1999 issue of Glass Audio has a Tube/MOSFET headphone amplifier design. Wonder what it is like. Does anyone know? In any case I will build this one and let you guys know what it is like.
Thanks for all the feedback.

halojoy 21st November 2002 08:25 PM

300 Ohms 10mW can be quite enough (safe)
This kind of hybrid, I imagine, can Sound very good,
either you use BIPOLAR with low distortion
or MOSFET configuration at the output.

I think I had a quick look at the project you mention,
in Glass Audio archives. There are some good articles there.

ashok 27th November 2002 11:30 AM

Circuit update
2 Attachment(s)
I have figured out another scheme for this amp. The tube is a resistor loaded gain stage. The output is a class A/ AB stage to drive the headphone . The unit can also be used as a preamp with a gain of around 5. Into high impedance loads or low voltages into low impedance phones the stage should work in class A mode.
The distortion seems to be less than about 0.1% when I simulated it in Circuitmaker. Now I have to build the unit to test its real performance. Not ethat in place if the anode load resistor we can use a current source which cuts down the distortion by a factor of 4 or so. It will be necessary to build this unit to find out if that makes an audible difference.

Any comments ?

ashok 27th November 2002 11:38 AM

Some details
The input impedance is not obvious from the circuit as it uses voltage shunt feedback and the gain of tube stage is not very high.
It seems to be in the 100K ohm range as I checked in the Circuitmaker simulation. The output stage quiesent current can be changed by varying the two source resistors. I used IRF 640/9640 because that's what we get here. The tube as explained earlier is because I have so many of them.

The output capacitor needs to be higher for lower impedance headphones. It should be rated higher than 100 volts and so could be a problem size wise and quality wise. I will build mine and post the results later.

If you remove R2 the 330k resistor , the circuit will have no negative feedback but will still work fine. Might be interesting to see how it sounds compared to the feedback version. There will be earlier rolloff but it may not be so audible and distortion on paper is higher.

All times are GMT. The time now is 09:33 AM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio