I don't understand what the 12AT7 is designed for. The higher mu 12AX7 and the lower mu 12AU7 seem to surpass the subjective quality of a 12AT7.
So my question is, do you know how to get benefit from 12AT7 "characteristics"? have you ever heard a system that sounds very good with 12AT7? Is there any situation when you decide that you are going to use 12AT7 instead of the other tubes of the same price level?
With the Aikido, the 12AT7 is suitable for the front end (low enough gain) and also for the back end (low enough Zout, especially when I use a solid state amplifier). But it is still not good enough (BTW I own several Mullards).
Is there any circuit where 12AT7 is not inferior? Simple buffer may be?
So my question is, do you know how to get benefit from 12AT7 "characteristics"? have you ever heard a system that sounds very good with 12AT7? Is there any situation when you decide that you are going to use 12AT7 instead of the other tubes of the same price level?
With the Aikido, the 12AT7 is suitable for the front end (low enough gain) and also for the back end (low enough Zout, especially when I use a solid state amplifier). But it is still not good enough (BTW I own several Mullards).
Is there any circuit where 12AT7 is not inferior? Simple buffer may be?
By "yes," I mean "yes" to each of those questions. Yes, I know how to use 12AT7 to make good circuits. Yes, I have heard many systems that used 12AT7 that sound good.Yes, there are situations where the 12AT7 characteristics fit the application very well, but I don't know what you mean by "price level"- tubes of a given type can usually be inexpensive or expensive, according to collector value. The quality of the tube has little or no relationship to the price. Yes, it can make a good simple buffer, but there are better tubes for that.
I don't know John's crossover.
I don't know John's crossover.
What confused me was the last sentence: there are better tubes 🙁
If the better tubes are "rare" (I'm not saying expensive okay), that's fine, but if it were 12AU7... 🙁
BTW, could you share those schematics you were talking about?
The JB's crossover is using 6922 or 6DJ8 (of course not 12AT7
)
A buffer ideally uses a tube with high transconductance and high gain. The gain of a 12AT7 is very good, but the transconductance is moderate. A 6DJ8 (for example) has a better transconductance, but lower gain. Fortunately, it's linear enough to not suffer from the reduced feedback, so it makes an excellent buffer. A D3a is even better (high transconductance and high gain, couple with excellent pre-feedback linearity), but we're already well past the point of overkill for audio buffer applications.
I'm not sure why you're correlating "expensive power supply" with tube choice. Or the implication that "expensive" means "high quality" or "suited for the purpose."
I did a pretty extensive analysis of simple tube buffers including calculation of power supply requirements- you can find that on my website in the Preamp section. You might find it useful.
I did a pretty extensive analysis of simple tube buffers including calculation of power supply requirements- you can find that on my website in the Preamp section. You might find it useful.
Best use of a 12AT7/ECC81 is in a simple VHF receiver front-end - that is what it was designed for. Even there it can be bettered by the ECC85/6AQ8, as that has an internal screen.
Audio uses include MM preamp input, phase splitter. I have used it to replace the EF86 as the first stage in a Mullard 5-20. The 12AT7 is not particularly linear so you have to be aware of this. Low level signals are fine, and balanced operation gives some distortion cancellation.
I suspect that the original questions indicate some confusion about how valves work and how best to use them. 12AU7, 12AT7 and 12AX7 share a pinout and heater requirements, apart from that they are quite different and meant for different purposes. Only in the world of guitar amps would anyone think of them as being related.
Audio uses include MM preamp input, phase splitter. I have used it to replace the EF86 as the first stage in a Mullard 5-20. The 12AT7 is not particularly linear so you have to be aware of this. Low level signals are fine, and balanced operation gives some distortion cancellation.
I suspect that the original questions indicate some confusion about how valves work and how best to use them. 12AU7, 12AT7 and 12AX7 share a pinout and heater requirements, apart from that they are quite different and meant for different purposes. Only in the world of guitar amps would anyone think of them as being related.
I've got a pile of measurements which indicate the opposite.
edit: From the RLD article:
"In my measurements of a range of 12AT7s in common cathode topology with a current source as a plate load, I found that at 30V peak-to-peak (10.6VRMS), the distortion was consistently below 0.1% with a favorable distribution of harmonics. For the best specimens, THD was less than 0.03%, even with a source impedance of 1 megohm."
I have found that 12AT7 based circuits do not sound good. That's all. I don't have a space for 12AT7 because I can use many other tubes that sound better for the same application. Octal tubes sound better in general. Among 9-pin tubes, I think 12AT7 is one of the worst. May be it is about proper/improper implementation. And that is the point of the question.
I don't use MM preamp. But I will find out about Mullard 5-20 (Don't know yet what it is).
How does it compare with a properly-fed 12AX7? The datasheet (last time I looked) suggested the AT was inferior - it was meant to be a VHF mixer, after all!SY said:
Depends on the load. If neither is loaded, the 12AX7 is better, although (as you can see from my numbers) the 12AT7 is still excellent, comparable to 6SN7. Once you introduce any significant load, the high plate resistance of the 12AX7 bites you, and the 12AT7 becomes more linear.
For signal levels at or below a few volts output (i.e. line level) almost any triode, properly biassed and loaded, will be good enough as far as distortion is concerned. Problems arise from the underlined part of the previous sentence, or for people who like a bit of distortion and miss it when it is absent.
I am perfectly happy to use a 12AT7. The main issue where I used it (input stage of power amp) is common-mode distortion due to non-linear Ra but that is small enough to be ignored. The 12AT7 gave me the right combination of gain and anode impedance. A 6DJ8 could have worked, but with less gain. 12AX7 would have too much gain and far too high anode impedance - the next stage was a 12AX7 LTP PS on the edge of grid current so I needed lowish impedance. 12AU7 would have too little gain.
I am perfectly happy to use a 12AT7. The main issue where I used it (input stage of power amp) is common-mode distortion due to non-linear Ra but that is small enough to be ignored. The 12AT7 gave me the right combination of gain and anode impedance. A 6DJ8 could have worked, but with less gain. 12AX7 would have too much gain and far too high anode impedance - the next stage was a 12AX7 LTP PS on the edge of grid current so I needed lowish impedance. 12AU7 would have too little gain.
Yes, that is what it was designed for, but there are zillions of sand based life forms today that make far better VHF receiver front ends. My current favorite is a little GaAs pHEMT chip from RFMD with 0.6 db noise figure and 20+ db of gain from 50 MHz to 3 GHz.
I can find plenty of worse tubes. I currently sell 3 different audio amp PC boards. 2 of them use a 12AT7 for the input tube. Why? Because it was one of the best, low cost, widely available dual triodes that had enough gain. Loaded with a CCS it is a very linear tube. Loaded with a CCS in parallel with the grid resistor and the miller capacitance of the output tube, it still wins over about a dozen different tubes.
A CCS loaded 12AX7 is more linear than a 12AT7 with higher gain. Load it with a large output tube like a triode strapped KT88 and the 12AT7 works as good or better than the 12AX7. Compare a box full of random versions of both tubes and the 12AT7 shows less variation than the 12AX7 into this load.
- Status
- Not open for further replies.