7062 vs. 12AT7

[skyvia]

I have some pinched waist 7062's that I'd like to run in my Crack (currently without Speedball, but want to modify the Speedball for this tube as well). I found a post talking about different resistor values needed to run 12AT7's, and I'm wondering if they should be adjusted a bit to run 7062's.

[Paul Joppa]

I assume you meant 6072, not 7062? (The 6072 is a variant of the 12AY7.)

**Edit** My mistake, 7062 is also known as E180CC - see my later post ...

The 6072 has a mu of 44, which is too great to operate comfortably with only 70 volts on the plate while maintaining enough plate current. This is the same problem that the 12AT7 has. A plate voltage of about 70v is necessary to set the operating point of the 6AS7G/6080 output tube.

The usual adjustment is to change the plate current to obtain the desired plate voltage at 1.55v bias (which is maintained by the HLMP-6000 LED). For the 12AY7, that would have to be about 0.6mA, compared to about 3.8mA for the 12AU7. That's a pretty weak driver. You can try it using around 1500 ohms for R1 on the Speedball small board, but it's too far from the design intent  to be recommended, at least in my opinion. You may need to adjust the value of R1 after checking the voltages.

[Paul Birkeland]

I would also not recommend this.  If you want to use a different tube type, rewire the socket for the 6CG7 and try that.

[skyvia]

No, I meant 7062. Similar to a 12AT7, same pinout. Bias point of these seem to be a little bit different than a 12AT7. I have a huge stash of 12AT7's, and the pinched waist 7062's are my favorite tube, so I'd like to modify my Crack to run them. I saw another post about using different resistor values to change the bias point to be able to use 12AT7's, and I'm wondering if that will be good enough for the 7062's, or if I should use different values for them.

So I guess my question is, should I just use those values to bias for a 12AT7, and then adjust based on voltage readings, or do you have a better suggestion?

[skyvia]

This is the post I'm referring to:

https://forum.bottlehead.com/index.php?topic=13277.msg121215#msg121215

[Paul Joppa]

No, I meant 7062. Similar to a 12AT7 ...

Ah! Same as E180CC, right? Apologies for my mistake. It's quite nonlinear, being made for binary digital operation, and it's "... not intended to be used in circuits critical as to hum, microphony and noise" according to the data sheet. But some people like it, possibly for the interface-resistant cathode.

The published curves seem to indicate a current of 1.5mA, so a Speedball R1 of 560-620 ohms should put it close to 70v at the plate. Still a wimpy driver, but it might work acceptably because of the 6080's low Miller capacitance.

[skyvia]

Ah! Same as E180CC, right? Apologies for my mistake. It's quite nonlinear, being made for binary digital operation, and it's "... not intended to be used in circuits critical as to hum, microphony and noise" according to the data sheet. But some people like it, possibly for the interface-resistant cathode.

The published curves seem to indicate a current of 1.5mA, so a Speedball R1 of 560-620 ohms should put it close to 70v at the plate. Still a wimpy driver, but it might work acceptably because of the 6080's low Miller capacitance.

My apologies, I should have written it as 7062/E180CC to avoid any possible confusion.

For what it's worth, I have an E181CC, which is a lower gain version (~27mu), and I love it with 6080's. I suspect the higher gain is still restricting the 6080, and it isn't biased quite the same as a 12AU7, but it's close enough to sound really good. It's the second order harmonics that it does better than any other driver tube I've heard.

Do you have approximate values for the plate loading resistors in a stock circuit? I haven't put my Speedball in yet, and would like to modify both so I can get a better comparison.

[Paul Birkeland]

You'd want something around 35K for a plate load in the stock circuit to get close to 75V on the plate of the 7062.

[skyvia]

Thank you so much!

[Paul Birkeland]

But just to be extra, extra clear, this is not a good idea for a wide variety of reasons.  If you really like the 7062 (I have quite a few of them stashed away), the Stereomour is the place to put those to use.

[skyvia]

Could you elaborate on why it's not a good idea? Any risks to the amp or headphones by doing this?

[Doc B.]

It's not a good idea because the circuit will not perform the way it is intended with that tube in it. We have a standard for performance of our designs and a 7062 will not perform to that standard in this design. Too much gain, too much distortion for our standards. The people who designed the 7062 would likely agree that this circuit is not offering an operating point that is within the intended operating envelope of the 7062. I think the guys have explained that in pretty good detail.

Smoke it and see. There is really nothing to stop you. We can only offer our rationale behind the design and offer our opinion on whether such a change is reasonable or not. One must bear in mind that this is DIY and therefore you have to take some responsibility for your own changes to the circuit. If something doesn't work right because of that change you have to own it. That's the learning process.

IMO the Crack circuit is designed around a medium mu driver tube like the 12AU7. A reasonable sub would be the 6CG7 as PB suggests. With an adapter a 6SN7 can work. High mu triodes are not really going to work the way we intended in this amp. But if you want it to distort like a guitar amp driver it might be acceptable to your ears.

[ipetruk]

Do all suggested modifications remove the possibility to use the usual 12AU7 or is it going to sound suboptimal now? I wonder if after this mod there is a way to compare 12AU7 vs 7062?

Thanks.

[Tom-s]

Without the Speedball I can't help out.

With the Speedball you could opt for installing a R1 switch for an E80CC as has been done many times; with good samples to be found on this forum.
That modification has a +-470ohm current setting resistor in the R1 position on Speedball small board. You're now on only 1.8mA and should land on 75volts or thereabouts anode voltage for E180CC. This also allows you to run E80CC. And add the parallel 470ohm resistor with a switch for the ECC82.


Take note that I personally find some differences between datasheets on the curves with the E180CC or 6072 7062. If you have E180CC the 1.8mA looks like a better match compared to 6072 7062 sets and the suggested 1.4mA.
The only way to be sure is just to try it out, measure it and post your anode voltage here for feedback.

Edit 7062; not 6072. I'm the second person with this issue 

[Paul Birkeland]

Do all suggested modifications remove the possibility to use the usual 12AU7 or is it going to sound suboptimal now? I wonder if after this mod there is a way to compare 12AU7 vs 7062?

Yes, you can't go back to the 12AU7 without undoing the changes to use the 7062.

This is not a good circuit to compare the two tubes because there isn't sufficient voltage and the direct coupling doesn't give you any flexibility to set a decent operating point.