Oh EC = Error Correction, a correction scheme usually for output stage of unity gain. I can give you the link if you ask me.
I don't see where I got mixed up. Please explain it.
Did you understand my explanation of why good curves matter when doing open loop circuits?
Did you understand my explanation of why good curves matter when doing open loop circuits?
Yes I do, and you pointed out the only way to correct (distortion etc) is GNFB, that is not all correct, because EC can do better and EC is not NFB, most people got mixed up. Sorry you "got mixed up", not offense intended, actually I don't know if you heard anything about it, just assumed you know so my mistaken, I should write in full not abbreviation.
Okay, so how do you use this error correction scheme to linearise the 12AU7, and if it's effective, why not use it on the 6SN7 to remove even more distortion (since it produces it less in the first place)?
If it is a feedback scheme that trades gain for linearity, wouldn't it be wiser to put a higher gain device inside it? Or go the opposite way, and start with a more linear (better curves) tube that doesn't need measures like that?
If it is a feedback scheme that trades gain for linearity, wouldn't it be wiser to put a higher gain device inside it? Or go the opposite way, and start with a more linear (better curves) tube that doesn't need measures like that?
I don't dispute it. It could be any tube that people like, hopefully we can sell it without insist on our own choice. It's user driven market, please read what he says on this: Error-correction power amp | Linear Audio NL
It's in the process of sim I use 12au7, I would plug many other tubes and try to come a better views.
"If it is a feedback scheme that trades gain for linearity, wouldn't it be wiser to put a higher gain device inside it? Or go the opposite way, and start with a more linear (better curves) tube that doesn't need measures like that?"
Output device needs not have high gain the sch works best if gain <1, it happened this output stage gain is <1, I agree it would be unstable if too much gain, so EC has a limitation, right on. Again I am not dispute your sch yet because I have not sim, and presumed you have sim mine. I not doing any commercial product, it's here for free what about yours?
Edit: EC doesn't trade gain for anything I hope..
Last edited:
It would be a very boring world if everyone insisted on the 6SN7 for everything. The 12AU7 isn't that bad, really. It typically poduces 0.3% THD at 10Vrms with a CCS load, whereas the 6SN7 produces perhaps 0.1%. I wouldn't crow too much about a difference like that.
I couldn't agree more. However I don't think it would be smart to flood the conversation with mentions of linear tubes. One could use 6N3P, 6S45P, 6E5P triode strapped, 2C22, 2C51, 6DJ8, ECC88, 76, 71A, 45, 27, 26 or any other tube with good curves.
The point I was making was about design philosophy; 6SN7 was a well known (known to the op earlier in this thread) example.
That is well and good for you. I would strongly disagree. My main thing is headphones, and in my experience - even with my humble AKG 601 phones - the difference in distortion you mention would be very noticeable. Not at all with speakers, but headphones are very unforgiving. One of my customers has Sennheisers with multiple times the price compared to my AKGs and the differences come up even more.
The basic design philosophic point I raised is that if you have a feedback loop where you trade gain for linearity, is it not wisest to choose a higher gain device rather than a lower gain device, and if you don't have such a loop, is it not wisest to choose (within reasonable parameters, price, size, power etc) the most linear device (best curves) you can find? Presupposing you care about linearity of course.
Choosing a low gain device with not so good curves will not bring best results in either scheme, and with the same price (even lower in some cases) you can get a linear device. Or a high gain device.
- Status
- Not open for further replies.