An Alternate Use for Crack

[Dr. Toobz]

I had something of an interesting idea today....you see, I'm starting to play guitar (more as an excuse to build additional tube gear than anything; bass is my normal instrument) and I've been messing around with my Quickie to get a sort of battery-powered headphone practice amp. Not a lot of progress there - the 3S4's are just too wimpy to get enough gain, let alone distortion. But then I had an idea: to use the Crack as a sort of OTL headphone guitar practice amp, or line amp to go into my iMac, where I have some effects software running (Amplitube 2). I've only seen a few examples of OTL guitar amps in the wild, period, and those are for driving speakers, not phones. But sure enough, the Crack gives the guitar a really interesting, deep, clean tone. Not like a solid-state amp, not like a transformer-coupled amp....something different, yet definitely "tubey." The gain is pretty low, however. Couldn't one slip in a 12AX7 or 12AY7 in place of the 12AU7 to increase the gain? I'm not necessarily wanting to get distortion or stress out my power tube; more volume would be nice. My understanding is that in a "real" guitar amp, you wouldn't re-bias the preamp tubes; changing them within the 12A*7 family would just lower or raise the gain (and accordingly, distortion level). Is the same true for the preamp circuit in the Crack? The plate voltage between the 12AU7 and 12AX7 seems to be the same (300V max), but the former has a higher plate dissipation rating (2.75W vs only 1W for the 12AX7), which I'm assuming is why the latter (12AX7) seems to be listed as only being able to handle about 1/10th the plate current (GE tubes sheets suggest 1.2mA for the 12AX7 and 10.5mA for the 12AU7). I guess this means a direct swap is not advisable?

FWIW, I will likely build a proper tube guitar amp and effects pedals eventually, but the Crack is sitting right next to my guitar, so why not get double duty out of it? Plus, maybe BH could add the possibly of guitar use as a selling point (a unique one amongst practice amps, it would seem).

[Paul Birkeland]

You could indeed slip a 12AT7 or 12AX7 in for the 12AU7, but you would want to change R1 on the C4S boards. 

To go to a 12AT7, you would want to reduce the current a little bit.  I would try 500 ohm resistors in place of the 237 ohm resistors and see if the plate voltage is reasonably close to 70-80v. 

-PB

[Dr. Toobz]

You could indeed slip a 12AT7 or 12AX7 in for the 12AU7, but you would want to change R1 on the C4S boards.  

To go to a 12AT7, you would want to reduce the current a little bit.  I would try 500 ohm resistors in place of the 237 ohm resistors and see if the plate voltage is reasonably close to 70-80v.  

-PB

I'm using the stock Crack w/ no C4S boards; not a fan of those things ;-) So, I would assume no change to go to the 12AX7? I might as well go "whole hog" and go for the highest gain variant as a matter of experiment. Going from mu=19 to mu=100 on the preamp side of things should be really interesting, I'd think - might even really start to distort at higher turns of the pot.

Another interesting idea might be to try out the 5998 in place of the 6AS7. I think the gain is higher on the former (mu=5? vs. 2 for the 6080/6AS7). But I'm assuming this does not matter in a cathode follower, where the gain will always effectively be less than 1? 5998's seem to be be a bid harder to find and/or pricey when they are found.

[Paul Joppa]

The problem is getting the plate voltage right, which interacts with getting the bias voltage right.

In Crack (as in Foreplay) the target plate voltage is 75 volts. With a low gain tube like the 12AU7, the grid bias is around 2 volts which is reasonable and can be supplied by an LED etc.

But the bias is inversely proportional to the gain. So a high gain tube needs a very small bias to get a low plate voltage. That means you have to deal with grid current, which starts at around -0.6 volts but varies a lot with different tubes. It causes distortion when the current is drawn, which you might actually like in a guitar amp though it's a bad problem for high fidelity circuits.

The 12AX7 is a non-starter; it will drop 175v or so even at zero bias. A 12AY7 at zero bias (short out the bias LED) might work, and a 12AT7 at 0.7v bias (replace the LED with a silicon diode) might also work. These are at the stock plate current; you can look up the plate curves for other workable operating points.

[Dr. Toobz]

I had figured this might be a problem. I had remembered reading something someplace that quoted a rather low plate voltage in the "Foreplay" arrangement, as you cite here, and I couldn't get my mind around how to get the plate voltage on the 12AX7 below 100V. I had figured on grounding the cathodes and getting rid of the LEDs, but as you point out, that will still leave me with too high of a voltage on the plates. I'm curious as to where the resulting 175V plate voltage would cause problems, though - the whole direct-coupled thing is still hard for me to wrap my mind around. I'm assuming that the interaction between the two stages is why you want to keep the driver stage at a lower voltage, correct? When there's nothing to couple the two stages, what keeps things stable between tubes?

[Paul Birkeland]

The important consideration is the driver plate voltage. It sets the bias for the following stage.

With the Speedball, the Crack is less sensitive, so the acceptable range of bias voltages.

For what it's worth, the first Crack variant we tried had a 12AT7 in it, but this had way too much gain.