A Dedicated Headphone Amplifier.

If you would rather go small and low power here is a solid state headphone driver you could build an amplifier around.

Photo of finished amplifier.

If you like to listen to tube amplifiers you can buy one on eBay and listen to it through speakers to your hearts content. If you like listening to headphones you can plug them into a solid state amplifier and listen to your hearts content. If you would like to listen to tube amplifiers through headphones you are out of luck because the old tube amplifiers don't have headphone jacks.


It seems I was wrong. I received an email informing me that tube amplifiers with headphone jacks were made by Scott, Fisher, and others. They didn't find it necessary to fully load the output when the speakers were switched off. Some designs connected a 25 ohm resistor to the 16 ohm tap and others used values as high as 100 ohms. This was apparently sufficient to damp the inductive kick and prevent oscillation. If you are fortunate enough to own one of these amplifiers you really don't need to build this project. Of coarse you may want to just for the fun of it.

Potential Problems with Headphone Jacks on Tube Amplifiers?

That speakers on/off switch could be a problem. Every solid state amp has one but I have never seen one, although I now know they do exist, on a tube amp. An S S amp acts more like a regulated power supply than a tube amp. The amp output supplies what ever amount of current is demanded by the combination of load resistance and output voltage. The voltage remains uneffected by load resistance. There is no such thing as impedance matching. That's why there are not separate terminals for 4, 8, and 16 ohm speakers on a solid state amplifier. If the load is an open circuit there is no current but the voltage remains the same; it does not rise to component damaging values.

A tube amp on the other hand is matched to the speaker impedance. If you have 4 ohm speakers you have to connect them to the 4 ohm tap on the output transformer not the 8 or 16 ohm tap. Moreover tube amps do not like operating into an open circuit. The design engineers were smart enough to keep them from going into oscillation but there is a distinct possibility of damaging the output transformer. If you turned on your tube amp and left the speaker switch off you might turn the volume up all the way in an attempt to hear something. The output tubes would be driven between saturation and cutoff and this would cause the inductance inherent in the output transformer winding to generate high voltage spikes, possibly as much as 2000 volts, which could cause the transformer windings to arc over and ruin the transformer. When a load is on the secondary this effect is damped and these dangerous voltages are not generated. This load does not have to be equal to the normal speaker impedance but can be somewhat larger depending on the design. And so I no longer believe that you will never see a tube amp with a headphone jack.

Never say never.

Another way it could be done. The speakers off switch could totally shut down the power output stage and the headphones could be driven by a circuit similar to the output stage in this project. I don't think this was ever done in a commercially available amplifier.


Modern headphones have an impedance of 100 ohms. They only require about 50 millivolts to drive them to a normal listening level. (As opposed to the hearing destroying level some people listen at in their cars.) That's 2.2 milliwatts; believe it or not. Even if somebody wanted to listen 20 dB higher than that it's still only 22 milliwatts. It doesn't take much power to drive headphones.

The 100 ohm impedance and the small amount of power means that an output transformer is not necessary. (The headphone amp for the original 4 ohm Koss phones did need one.) Some engineers might come up with other ways but my approach was to use a cathode follower. The tube I selected was a 6CG7. When you look up this tube in the Sylvania tube manual it refers you to the data for a 6SN7. So it is a 9 pin mini version of that venerable tube.

I set up the circuit on the breadboard and gave it a listening and distortion test. It sounded really good and measured less than 0.1% distortion. All without any negative feedback applied except that inherent in the cathode follower itself.


Noise below normal listening level.

Magnetic Phono Input,

58 dB

Auxiliary Input,

more than 80 dB.

Frequency response plus or minus 1 dB,

8 to 26000 cycles.

Distortion at 10 dB above normal listening level,


If I were to put the whole article on one page it would take a long time to download over a dialup connection. I have broken it up into smaller parts. This way you can read what ever you want to and skip the rest.

Schematic diagram and circuit description.
Parts list and where to buy them.
Construction details.

A heart in love with the beauty
of glowing tubes never grows old.


This page last updated July 31, 2004.