DIY headphone amp - checking my thoughts/calculations

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
The headphones are AKG 240mkII. 55ohm, 200mW.

Q1. I’ve read that a multiplier is good, is 10x but 2W seems too high?

Q2. Reduction of impedance, using push pull and multiple devices, 110mA max each in an OTL push pull class AB1 configuration.
If I use 4 devices that’s 1/32 = 4* 1/R therefore each tube is 1/4 so 8ohm.

Then using Negative feedback to get from 8 to 4ohms.

Calculate 2W 32ohms sqr(2/32)= 250mA, 32*.25 = 8V peak.
sqr(2/4)= 707mA, 4*.707= 2.829V peak.

Gain 8V/447mV = 17.8971x

But if the impedance dropped to Vout/Vin, or more accurately iirc G/(G+bG) so the gain we actually have into 4ohms is 8/2.829 = x2.82, so we have a ratio we can use to work out the gain we actually need with the NFB in place to be?

I’ve been staring at numbers too much and crap sleep but I think I’m onto the right track?
 
200mw is the MAXIMUM input power. Not the recommended input power.

Here is how I would do it. I would first find your sensitivity for the headphones. In this case it is 104db/V input. That means that if you input 1V to the headphones, you get 104 db out. 104db will be pretty darn loud, but I usually try to shoot for 120db just because

120db would be a 16 db more than 104. If we convert the decibel gain to voltage gain we get about 6.31v or about 6.4v. That means that you need an amplifier that can provide 6.4v peak to the headphones or 12.8v peak to peak.

Now if we want to find the power output of the device, we use the formula V^2/R and we get ~0.75 watts peak or about 0.525watts rms. Obviously these figures are above your maximum power input for the headphones, so please don't go shoving 0.75 watts through your headphones. This is just an arbitrary example.

Continuing on with the example, 0.75 watts peak is the minimum power requirement. You would obviously want to increase the power of your amp to be able to drive your headphones with some amount of headroom to spare. If I had to make a recommendation, I would build a amp capable of driving a 32 ohm load to 1 watt. With that sort of power, you can drive darn near any headphone as loud as you want.
 
Last edited:
Good points. That makes sense and a good sanity check.

A 6.4V peak with a 110mA max per 6AS7 triode means 704mW which would be enough as you say.

Into 32ohm, that P=RI^2 gives sqr(.704/32) needs 148mA, so you're over one triode so the option here is either run a second 6AS7 (remember the other 1/2 is running on the other side of the push-pull) or run a solid state cathode follower.
So with two 110mA devices in parallel at 32*(110*2)^2 = 1.55W, or more interesting is R=1.55/.220 = 7 ohms flat out.

The AKGs frequency impedance bottom end of 55ohm, the average is 84 and the top is 200ohm. The idea is to support a wide range - it could be possible to support higher voltages for higher impedances with this too with some changes to the design.

So in short with 55ohm, sqr(.704/55) only needs 113mA so a single tube would drive it but perhaps a little ragged on the bottom end but the 6AS7 is quoted at a max of 120mA but the guidance is a max 110mA in class A. So with a Class AB1 there's a little saving which is likely to cover the 3mA (just being retentive :) on that).

I assume with AB1 is the load sees two devices for a portion of the overlap until the cutoff, at which point the load then sees one so at the roll off the impedances/current change. I could set the bias so that up to a certain dB/V is running on the two tubes as classA but then for the transients/deep bass/loud runs into class B.

It's sounding like a single 6AS7 in PP AB1 for each channel will be more than enough for normal volumes. If I want to drive speakers I could add SS CF and a second 6AS7 per channel and use a second external PS if needed.

Thank you :)
 
Last edited:
unfortunately you need +/- 110mA = 220mA max.

Push pull configured with two 6AS7 has two triodes per tube:
Left Channel Triode 1A B+ 110mA, Triode 1B B- 110mA = 220mA
Right Channel Triode 1A B+ 110mA, Triode 1B B- 110mA = 220mA

As the tubes are running in AB1 the bias overlap steals some of the current capacity as the tube is not running just 50% of the sine wave but 60%.

I hope that works :)
 

PRR

Member
Joined 2003
Paid Member
Put a 'scope across any handy amplifier and these headphones with your ears in them. How loud is good? What is the peak Volts of that?

I think even touching 200mW will blow your ears out. (127dB SPL peaks).

2mW (107dB SPL) ought to be enough, unless you are going deaf (many of us are). So 8mA-10mA ought to be plenty.
 
The plate is 280ohm, with two tubes in push-pull 140ohm before feedback.

One option was/is to use a NPN/PNP to follow each output tube and possibly additional tubes - the net effect would be to lower the impedance further.

The simulations are coming along nicely. The current working version is a 6SN7 as a differential amp with a 6AS7 in push-pull and a DC servo on the 6AS7 grid bias. This is actually operating in Class A at the moment. With feedback I still think it would have enough power for average listening without adding an additional stage.

The second simulation I have a driver stage between the two - typically this can really only be used with feedback or causes overdrive distortion by soft clipping without the gain change using global negative feedback. I could make it 1:1 if feedback not used but then it would simply add noise.

This was just a second of simulation with NFB so the FFT hasn't had chance to really drop the noise floor - typically I'd run 5 minutes of simulation time without the first 30 seconds but it's very slow:
Screenshot 2020-12-02 at 11.21.53.png
 
Last edited:
You need a better window function there... got Kaiser-Bessel? In LTspice I generally let transient sim run for 20 cycles or so.

Note that K240S' are notorious for being sensitive to output impedance... you really want <10 ohms. If you must go OTL by all means (silly and wasteful as it is... tubes just don't work well at the voltage/impedance levels you need for a headphone amp), I'd recommend near Bruno Putzeys levels of feedback.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.