Guitar headphone amp with cab sim

Status
Not open for further replies.
So far I have built Runoffgroove's Ruby (with Bassman mod) and Professor Tweed pre-amp.



I will be building more of their pre-amps (probably the Umble and the Eighteen) but it would be nice to use them with headphones as well.

I am thinking about building the Condor cab sim and put a headphone stage in the same box. It will be powered by a 9V battery.

An LM386 based amp like the Ruby can easily power a pair of headphones and it is generally recommended to put a 10 Ohms resistor between the output and the headphone jack.

I think I want to base the headphone amp on the Ruby at minimum gain (nothing between pins 1 and 8). I can always put a in a resistor of 2K if it does not go loud enough but I doubt it will be necessary for 32Ohms headphones.

What I do not really want is too many volume knobs so I guess I could either skip the master volume pot of the Condor or the volume pot of the Ruby.

As I am a bit of a minimalist, I am wondering if there is any need for the input cap and fet buffer of the Ruby as the condor output comes from an opamp and has an output cap.

Am I cutting the right corners?
 
If you remove the volume pot and cap, you basically end up with just the LM386. And it can be implemented as you see fit. Removing the Ruby's cap WILL alter the low-frequency response. The -3dB point in the original schematic is 339Hz, which is fairly high, even for guitar. Since the Condor comes with a 22uF output cap, the -3dB point into 10kohm is 0.7ohm(!). There's absolutely no need for such a low frequency response. So what I'm trying to say is, it's fine to remove the FET stage, but keep the cap to maintain the same sound. And maybe experiment with different cap values to tweak it to your liking.
 
Thanks!

Removing the input stage would make the front end similar to the Little Gem (of which the Ruby is an evolution).

I do get a headache of thinking about the interaction of the output cap of the Condor and the input cap of the headphone amp section which are only separated by a pot and would basically be in series (right?). Wouldn't the significantly lower cap value of the input cap overrule the 22uF output cap?
 
Ah yes, caps in series 🙂. Ctotal = (1/C1) + (1/C2) + etc.

I'm asuming you want to keep one of the volume pots? Either from the output of the Condor or the input of the Ruby? Whatever capacitance will be placed before the pot will create a high-pass filter (f = 1/2*pi*C*R). R being the pot's total resistance, with the input impedance of the LM386 (50kohm) in parallel. Depending on the volume setting, the total input impedance will roughly be somewhere between 8.33k and 10kohm.

There's really no point in using caps in series here. Just one cap to set the low frequency point and block any DC offet is all you need. Place it before the volume pot though, since U2B (condor) will put out DC and pots and DC don't play well together.
 
This is a tricky one. The 47n cap/10k pot is what largely determines the 'sound' of the Ruby, since it makes a highpass filter with a -3dB point at 339Hz, which is high. It's probably intended to drive small speakers with only a high frequency response and to keep it well away of the speaker's Fs. If you're going to use it as a headphone amp, you might consider using a larger cap to lower the -3dB point and get a fuller sound out of it. But that's up to your personal taste. 22uF (and everything above 1uF) is unnecessary though.
 
The Ruby is actually designed for full size guitar speakers. The one I built is used with a 10" and goes low enough. I did the Bassman mod and this includes a 0,1uF input cap. The headphone amp does not need to have as much character as I will be using it with pre amps that already emulate a complete amp. An ROG pre amp + condor should be sufficient to record directly from so I do not mind if the headphone stage is relatively transparent.
 
I think I will go with a 1uF cap at the output of the condor section. I think I should also change the output cap of the amp section as this is probably optimised for an 8 Ohm speaker. My headphones are 32 Ohm which would be 16 in mono. Changing the 220 uF cap to 100 should result in a similar -3 dB point.
 
A 1uF cap should be good and allows for a polypropylene instead of an elco. I'd leave the 220uF as is. Cap distortion becomes audible at the point where the reactance starts to increase I.O.W. where they start to act as a filter instead of a DC blocker.
 
Every CR (highpass) filter has a point where the cap's reactance starts to increase and in essence forms a frequency dependent voltage divider with the resistor. This is the point where you'll see a frequency graph take a dive on a bode plot. This is also the point where there's a voltage is starting to develop across the cap.

Here's the thing. There's a LOT of ongoing (heated) discussion about the audibility of caps in the audio path and I'm not going to stir it up here, but if any distortion is induced, it is in the region where a voltage is present across the cap. By using a bigger cap, keeping the highpass frequency point well below your hearing threshold, you minimize any possible audible effects.
 
Status
Not open for further replies.