headphone/guitar amp

I'm trying to figure out the best (easiest) way to mod this headphone amp to also accept a guitar input (so I can "jam along" with my favrite CD's!) .


This is the Cavalli/Jones headphone amp I recently built. This is only half the circuit as it is in stereo. I would like to just do this:


But I don't know if it would load down the line input too much, or what value the guitar volume pot should be. Any thoughts?

:guinness:
Joel

Cavalli hangs on Headwize.com DIY forum, and would be the best guy to re-think this plan for guitar input.

But I think the problem is that a guitar needs more gain than a CD player, and this very-simple amp has barely enough gain as it is.

Do you want to add a 12AX7 (or 6AV6), or a TL071 chip?

If you have an FX pedal that has WAY too much gain, that might work too.

Yeah, I didn't want to hit headwize too hard with their bandwidth issues. An extra tube was my bosses thought as well. I'm mostly a jazz guy, so I wasn't looking for any "phat" tude distortion or anything-just something for while I'm in the practice room.

Joel

[quote author="Mbira"]But I don't know if it would load down the line input too much, or what value the guitar volume pot should be. Any thoughts?[/quote]

It shouldn't load down the line input; the grid of the input tube is a virtual ground, which will isolate the inputs from one another -- imperfectly, but enough.

What it will do, though, is load down the guitar pickups, probably way too much. When set for maximum gain, that 68k resistor is the load on the guitar, since it's running to the virtual ground at the input grid. And turning the pot down from maximum will give you a better load on the guitar (if the pot is nice and big), but less gain for the guitar from the amplifier. It's a lose-lose situation. An amplification stage for the guitar would help a lot. But...

How well does this amp do driving headphones by itself? It strikes me that a 6DJ8 is a little weak to be driving loads which are that low-impedance.

Peace,
Paul

> How well does this amp do driving headphones by itself? It strikes me that a 6DJ8 is a little weak to be driving loads which are that low-impedance.

We really should step over to http://headwize.com DIY forum. The forum traffic is not the bulk of Cmoy's bandwidth troubles. (I suspect over-active search engines.)

The DJ seems to be a fine tube for hi-Z phones (Mbira didn't say his Z). What is astonishing is that it also works for lo-Z, for many people. If you get the White Cathode Follower really balanced, it can put over 40mA peak in lo-Z phones. In 32Ω that is 25 milliWatts, ample for many phones in home settings (this isn't a pocket-amp). Personally I like to have more on tap, but many people like it this way.

At small values of feedback resistor, the output impedance is IIRC under 32Ω, which is enough damping for any 32Ω phone I have seen.

Of course at that feedback it is about unity gain. Input sensitivity is about 1V RMS. This is roughly true too for hi-Z loads and large values of the feedback cap: about 1V input is needed.

And with the suggested feedback impedances, you need about 100uA of drive.

Guitar might need 10mV sensitivity, 100K load, not more than 0.1uA of drive available. There simply is not enough power from a pickup, even a hot hard-played pickup, to drive this amp. Not to mention probable loss in a mixing network.

The no-fooling fix would be a full Fender-like input strip: two half of 12AX7 with a tone-stack between. Mixing should probably be done in the second grid, but might be done between second Plate and the input of the C/J amp.

Since we need a gain of 30-200 with high input impedance and low output impedance, it isn't going to happen in a single triode, pentode, BJT, or FET. However the levels are tame enough to use 741 or TL071. Gain structure and noise is a problem: guitar preamps really do like two stages with a volume control between, or some clever variable gain stage. Since some band-shaping may also be good, two stages may be the preference. Type of amplifier is also guided by available power: no +/-15V rails here! Guitar-amp design, even a tame unfussy application like this, is a lot of personal taste blended with expediency.

Horsing the dead beat:

> How well does this amp do driving headphones by itself?

Estimated power versus impedance of several common headphone amp topologies:


Curve "g" is the revised Morgan-Jones, as Mbira made.

"b" is the Cmoy, 9V battery and an op-amp, which is very popular.

"f" is taken from a hi-fi cassette deck I had. No headphone volume control and it didn't need one(!), but was adequate for many uses.

Somehow I omitted the ubiquitous two-AA WalkMan and its many tape/CD/MP3 kin. It makes about 30mW in 32Ω, 3mW in 300Ω. The M-J is actually more powerful than a Walkman above 40Ω!

"d" and "h" may be typical of the heftier commercial headphone amps.

The "*"s are maximum power ratings of several typical headphones. The M-J is not over-powered for cans under 100Ω, but is kick-butt for 300Ω and higher, finally leveling off and falling for several-KΩ and nearly-a-Watt. (It isn't the worst 600Ω line-driver you ever met.)

Nice PRR-sorry for the delay in replying. I am using Sony MDR-v600's. They are rated at 45 ohms.

Joel.

I've got midterms right now, so I'll be a little scarce for a few days...