Just wondering if anyone has used this type and what their experience was with it.
It looks like a 6SN7 type of octal twin-triode but with a 12.6V 150mA heater.
Maybe that makes it more like a 12AU7 in an octal bottle. ??
However, the Cag is high, like 3pF, as compared to 12AU7 Cag of only 1.5pF, and 6SN7 Cag of 4pF.
12AH7 datasheet: http://www.nj7p.org/Tubes/PDFs/Frank/hb3/12AH7GT.PDF
Thanks.
It looks like a 6SN7 type of octal twin-triode but with a 12.6V 150mA heater.
Maybe that makes it more like a 12AU7 in an octal bottle. ??
However, the Cag is high, like 3pF, as compared to 12AU7 Cag of only 1.5pF, and 6SN7 Cag of 4pF.
12AH7 datasheet: http://www.nj7p.org/Tubes/PDFs/Frank/hb3/12AH7GT.PDF
Thanks.
Last edited:
12AU7 does the same thing but better, The 12AH7 has an odd pin out,
not compatible with the other octal twin triodes. So not easily replaced if it quits.
I looked at this briefly for a project a few years ago, but quickly rejected
it for these reasons.
not compatible with the other octal twin triodes. So not easily replaced if it quits.
I looked at this briefly for a project a few years ago, but quickly rejected
it for these reasons.
Yes, I've tried it. It's not very pleasant-sounding, kind of edgy, which leads me to believe it's not very linear.
Well, not as linear as a 6SN7 to my eye. Anyway, to me it didn't sound as good as a 6SN7 or a 6J5, which was what I was looking to sub ir for--an octal dual triode with a bit less gain.
What did you try it in? The pinout is not compatible with other octal twin triodes.
Did you construct an adapter or simply wire it in? 🙂
Yep, looks good. Could be fun in a circuit needing less gain than a 6SN7, allowing of course for higher internal resistance (lower gm) and lower maximum plate dissipation. Less $$ than a NOS 6SN7 too 😉
I tried it in a no-feedback push-pull amp I was breadboarding, as a long-tailed driver. I was looking for something with lower gain than a 6SN7. Maybe with feedback it would sound better. The OP asked if anyone had tried it, and I answered. ;-)
12SN7’s cost less than 6SN7’s too. If youre willing to accept non 6 volt tubes….
The curves “look“ linear, and mu is reasonably flat across plate voltage. They would be worth adding to my next vacuumtubes.net order. SN7’s are hard to beat though, everything else small signal or low power with 20 mu or less that I’ve run into pretty darn nonlinear. Probably don’t want a 300B in your preamp.
I came across this the other day, too, and just ordered a few. It's similarity to the 12SN7 is striking (and the 6AH7 to the 6SN7). The Ken-Rad data sheet I saw was dated 1941, the RCA sheet dated 1955. I'm going to make some broad assumptions about the tube. From what I gather, Ken-Rad was a lower tier player in the tube world. I wonder if they 'developed' the 6AH7/12AH7 as an alternative to the 6SN7/12SN7 to create a market niche that their marketing people would try to exploit. However, once Ken-Rad was sold to GE (early 50s) I'm guessing that GE concluded they didn't need a competitor to the 6SN7 and never pushed it. Still, an interesting alternative to the 6SN7/12SN7.
I believe 12AH7 were popular in Connsonata organs back in the day. I picked up a stash from the estate of an old technician. Never used them though.
Here is an article (page 49): https://worldradiohistory.com/Archive-Radio-Electronics/50s/1952/Radio-Electronics-1952-02.pdf
Here is an article (page 49): https://worldradiohistory.com/Archive-Radio-Electronics/50s/1952/Radio-Electronics-1952-02.pdf
Last edited:
Thanks everybody. I'm not sure there's any kind of consensus on this tube type, but I appreciate the info!
I picked up a sleeve of them for really cheap, and I was thinking of making a line preamp with NFB to bring gain down to about 2X. I like the fact that it uses a 12V heater with low current demand (12.6V 150mA). I might try to do this line preamp with AC heaters using an old Rat Shack 12.6VCT 450mA transformer I scrounged up. Cheeeeeeaaaaaap.
I picked up a sleeve of them for really cheap, and I was thinking of making a line preamp with NFB to bring gain down to about 2X. I like the fact that it uses a 12V heater with low current demand (12.6V 150mA). I might try to do this line preamp with AC heaters using an old Rat Shack 12.6VCT 450mA transformer I scrounged up. Cheeeeeeaaaaaap.
12AH7 versus 6SN7
12AH7
1/2 of the 6SN7 heater watts
1/2 of the 6SN7 maximum plate dissipation
Similar: Gm, plate resistance rp, and u.
If you use it for audio, don't worry about the 3pF versus 6SN7 1.5pf
12AH7: u = 16, and u +1 = 17, (miller capacitance + 1) gives 3 x 17 = 51 pF, Xc = 156k Ohms.
6SN7: u = 17, (miller capacitance + 1) gives 3 x 18 = 27pF, Xc = 295k Ohms.
Most audio signal sources can drive either 295k Ohms capacitive reactance, or 1565k Ohms capacitive reactance. If not, fix the audio signal source first.
How does it sound . . . how correctly do you design the circuit parameters around the tube?
If you want to reduce the circuit gain . . . you can split the self bias cathode resistor into 2 halves.
Example: If the self bias resistor needs to be 1k, then use two 500 Ohm resistors in series. Then put a bypass capacitor across only one of the 500 Ohm self bias resistors.
That will have less gain than a single 1k Ohm self bias resistor that is bypassed by a bypass cap.
Just my opinions
12AH7
1/2 of the 6SN7 heater watts
1/2 of the 6SN7 maximum plate dissipation
Similar: Gm, plate resistance rp, and u.
If you use it for audio, don't worry about the 3pF versus 6SN7 1.5pf
12AH7: u = 16, and u +1 = 17, (miller capacitance + 1) gives 3 x 17 = 51 pF, Xc = 156k Ohms.
6SN7: u = 17, (miller capacitance + 1) gives 3 x 18 = 27pF, Xc = 295k Ohms.
Most audio signal sources can drive either 295k Ohms capacitive reactance, or 1565k Ohms capacitive reactance. If not, fix the audio signal source first.
How does it sound . . . how correctly do you design the circuit parameters around the tube?
If you want to reduce the circuit gain . . . you can split the self bias cathode resistor into 2 halves.
Example: If the self bias resistor needs to be 1k, then use two 500 Ohm resistors in series. Then put a bypass capacitor across only one of the 500 Ohm self bias resistors.
That will have less gain than a single 1k Ohm self bias resistor that is bypassed by a bypass cap.
Just my opinions
Last edited:
I was thinking of making an 'anode follower' using one section of the 12AH7 DC-coupled to a MOSFET source follower. Then wrap NFB from the output (MOSFET source) to a virtual ground at the grid of the 12AH7, through a resistive voltage divider. Use enough NFB to bring the gain down to 2X or 3X.
That won't be enough NFB to require much if any compensation. I figure about 16dB of NFB will be required.
There would be zero worries about input capacitance, as Miller effect will be brought down by the NFB gain reduction.
The only challenge will be keeping the input impedance high enough for wheezy sources. But my sources are all low impedance, so that is also not a worry.
Or... Is there an advantage to using cathode degenerative feedback instead?
At any rate, I think my question has been answered. 12AH7 is a reasonable tube. The low current for the heater is a plus in my book.
Oh... Typical Cag capacitance of a 6SN7 is 4pF, not 3, so
6SN7: u = 17, (miller capacitance + 1) gives 4 x 18 = 72pF
That won't be enough NFB to require much if any compensation. I figure about 16dB of NFB will be required.
There would be zero worries about input capacitance, as Miller effect will be brought down by the NFB gain reduction.
The only challenge will be keeping the input impedance high enough for wheezy sources. But my sources are all low impedance, so that is also not a worry.
Or... Is there an advantage to using cathode degenerative feedback instead?
At any rate, I think my question has been answered. 12AH7 is a reasonable tube. The low current for the heater is a plus in my book.
Oh... Typical Cag capacitance of a 6SN7 is 4pF, not 3, so
6SN7: u = 17, (miller capacitance + 1) gives 4 x 18 = 72pF
Nah.
It's not nearly as linear as a 6SN7 or 12SN7. It's more like that awful 12AU7 than a 12SN7.
Besides, I'm talking about using it with negative feedback...
It's not nearly as linear as a 6SN7 or 12SN7. It's more like that awful 12AU7 than a 12SN7.
Besides, I'm talking about using it with negative feedback...
Didn’t look anywhere near as bad as the AU7 to me. There are better TV tuner dual triodes than the 12AU7 (worse ones too). But if youre going to be using NFB, what’s wrong with using an SN7 Other than price? And if you are using NFB you may be better off starting with a high mu triode and more NFB than a low mu with less. Hell, I’ve been known to take pentodes and stick big ol’ cathode resistors in them and dial the gain down to what you can get out of a low mu triode. Nice ‘n linear, courtesy of the feedback.
It's pretty simple...
The green trace is the output at 'OUT'.
The tan trace is the signal at the plate of U1.
Frequency response should be -0.05dB at 20Hz, 20kHz.
Rload and Ccable are tacked on to the output as a torture test. The above predicted readings are taken with those in place.
The green trace is the output at 'OUT'.
The tan trace is the signal at the plate of U1.
Frequency response should be -0.05dB at 20Hz, 20kHz.
Rload and Ccable are tacked on to the output as a torture test. The above predicted readings are taken with those in place.
Last edited: