Configure Stereomour for headphones?

[Chris]

Doc, just have to say , your FISH quote is awesome!! thanks for that...

[Paul Joppa]

Just in principal, we have done the two integrated amps because it is a bit more economical, especially if you don't use a preamp. But it's less flexible, and sacrifices the sonic advantages of monoblocks to achieve those economic benefits.

The hum from AC heaters in indirectly-heated tubes will vary a lot between individual tubes. And, since it comes partly from leakage currents, it will change as the tube ages. Of course it is also highly dependent on the sensitivity of whatever follows the amp in question. We have a lot of experience with hum from AC-heated 12AU7s, since they were in the original Foreplay - that's nearly 15 years ago, I think. Many owners have been very happy and had no hum problems at all, but a good number did have hum problems, and there were many solutions developed over the years. DC heaters seemed to be the most reliable, which is why that became part of the circuit with the major re-design of version III.

[Dr. Toobz]

Paul, what if I tied the output signal from the selector switch decks in the Steremour to the third set of RCA jacks, normally used for input, with a 1k resistor, as you've suggested long ago for adding a sub output to the Foreplay? That way, I'd still have my two inputs (one for a digital source, one for the Seduction), and would gain a "tape out" to go into the Crack, which I could find room for (somehow) on my rack. Since the speaker amp wouldn't be on when I was listening to headphones with the Crack, I would assume this would not cause loading problems, and I wouldn't have to wire a cut switch. Yet, I could still switch between sources on the Stereomour.

[Paul Joppa]

Dr. Toobz, that should work. It would place the 100K controls in the Stereomour in parallel with those in the Crack, for a net 50K, but that's an easy load for Seduction and pretty much anything digital.

[fonslet]

Hi guys

I've been looking at the Stereomour for a very long time and the only thing that's kept me from pulling the trigger on it, has been the lack of a headphone output.
I own a pair of Grado SR325is, and using the specs of these I did some calculations.

Before I go on I should warn you, that I have never built anything, nor worked with electronics before, so please just tell me if the following is BS:)

OK here we go:
The Grados are 98dB sensitive @ 1mW and are 32 ohm impedant.

The SEX amp rated at 0,3mV hum will produce 2,8nW into 32ohm. This should produce roughly 41dB hum in the grados.

The Stereomour amp rated at 1 mV hum will produce 31nW producing roughly 51 dB.

This I guess could be a problem.

I was thinking if this could be solved by dividing the voltage in by using resistors like in this pic:
http://en.wikipedia.org/wiki/File:Resistive_divider.png [nofollow]

Using values of R1=6 ohms and R2=2 ohms and letting the headphone sit in parallel with R2, a 32 ohm headphone would recieve only ~1,5% of the power put into the devider.

The impedance seen by the amp is now only ~8 ohms and the hum is therefore larger by a factor of 4 (124nW). However only 1,5% of this will go to the headphone and the overall hum will be ~1,9nW. This is even lower than with the SEX.

The dampening factor should still be ok at a value of appx 2.

The downside is that you lose a lot of power. Max output would be ~50mW.

So what do you think? Could this work or is it just crazy ramblings?
Will 50mW be enough power for my grados?
Will running the amp this hard cause the OT's to saturate and thus make me lack low end?

Thank you guys for a great forum. Though this is my first post ive been reading end enjoying it for at least a year now.

[Paul Joppa]

Your calculations look right to me.

The most flexible approach would, in my opinion, be to use a low-value potentiometer to make an adjustable voltage divider. This way headphones of different sensitivity could be used.

For an example, Radio Shack sells a 25-ohm 3-watt wirewound rheostat. A pair of those, in a small project box with wires to the amp output and a jack for the headphones, would do the job well.

[fonslet]

Great! Thanks for the reply.

I placed the order this weekend.
Good idea with the box. That way the pots could even act as "volume control" and lower the hum even more at low volumes.
I'll be sure to let you know how it all works out:)

[Jim R.]

I just found a little headphone adaptor box I forgot I had.  This one has a 4-gang wirewound pot, an impedance selector switch and a headphone jack, as well as two pairs of binding posts for the input.  If I'm remembering correctly, this box was designed for use on up to a 10 watt amp (though much less is recommended), connecting directly to the outputs of the amp, setting the load the amp sees with the impedance selector (not the impedance of the cans), and allows you to adjust the volume at your seat -- of course after setting an appropriate level on the amp.

I'vee used this with my sennheiser hd-650s (which I don't own anymore) and with a variety of small tube amps -- miniwatt s1, audio paradise mp-301, etc., but never tried it with anything else -- I'm guessing because of the noise issue from typical power amps.  Both the aforementioned amps had snr of 90 dB or greater so they were ok, but I think even with my also now gone Carina, that the noise was a bit too much -- and that was a very quiet amp on even 100 dB speakers.  Though it's been so long, I can't remember if I actually tried the Carina or not.

Maybe I'll rework this box a bit -- better, captive cables, etc. and see how well it works with the s.e.x. and stereomour.  As I said, it looks like there are 4 pots on one shaft, so it may be setup as an L-pad.

-- Jim

[fonslet]

So.. A lot of time has passed and I realised I never got back to the forum to report my results..
After all this I ended up going a different way.
My good friend Theis looked at the circuit and proposed making an OTL headphone output. It was a very easy fix. 4 resistors, a switch and a jack socket.

https://www.dropbox.com/s/0iunv5wspwfv84c/OTL%20losning.JPG [nofollow]

With the values given (and if im correct ), the headphones will "look" into 1800 ohm // 16 ohm = ~15,9 ohms. In this configuration the amp should be able to deliver approx. 3 volts to the headphones. For my Grado's that will result in 300 mW which is more than enough.
The sound is very pleasing. A very well integrated treble with no listening fatigue. No hum at all. I don't have much experience with designated headphone amps, so I can't tell you if the sound would be pleasing to a more experienced headphone listener, but to me it definetely is..
I went with a switch rated for 250 and have had no problems. However if you use less sensitive headphones and need to crank it all the way up, you should probably go with one rated for 400V.
BR Jesper

[Paul Birkeland]

FWIW, you've decreased the load impedance on the 2A3 from 4K to 1K.

This will increase the power output available at the output of the 3.3uF parallel feed cap (which you don't need) at the expense of producing more distortion (which you also don't need).

You will also be down 3dB at 50Hz because of the filter formed by the 3.3uF cap and 1K load.

As an experiment, try bumping up that 1K resistor to 4K and observe what changes.

Compared to the transformer, you are attenuating 36dB of the output with the resistor divider, if you wire the output transformer for 2 Ohm operation, you will step the voltage down by 33dB, but at the same time nearly all the power will still be available (not burned up by resistor heat).

Please do post your impressions of these options.

-PB

[fonslet]

We did a similar calculation but added the plate impedance of ~700 ohms giving a total load of 1,7K and a -3dB point at 28 Hz.
The 1K resistor was a compromise because I was afraid of losing to much power. However power does not seem to be a problem, as i listen with the amp turned down very low.
I am going to change the resistors for higher quality ones soon and will follow your advise to go with higher values.
Thanks a lot for your input!
Jesper

[Paul Birkeland]

The plate impedance is what sees the load, not actually part of the load.

I'd highly recommend the 2 Ohm taps on the output transformers.

-PB