• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Best/favorite CF NOS noval twin triode?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
No. Cathode followers can be designed well (with all due modesty, the CF line preamp design I published is a good example). It's just that your proposed CF isn't. To get that 0.08dB extra gain (inaudible) compared to a properly designed CF with an appropriate tube, you're picking a very inappropriate tube, then using it in a non-optimal way. It compromises both the load-driving ability and distortion for a demonstrably inaudible benefit.
 
OT This is an outrageous!

SY said:
No. Cathode followers can be designed well (with all due modesty, the CF line preamp design I published is a good example). It's just that your proposed CF isn't. To get that 0.08dB extra gain (inaudible) compared to a properly designed CF with an appropriate tube, you're picking a very inappropriate tube, then using it in a non-optimal way. It compromises both the load-driving ability and distortion for a demonstrably inaudible benefit.

:no: :no: :no: :rolleyes:
 
Administrator
Joined 2004
Paid Member
Re: D3a, E280F

panos29 said:
Not exactly cheap and not even double triode, but D3a and E280F are definitely excellent (maybe the best?) for cathode followers when triode strapped and even better in pentode mode CF.

Ssssh! :D

Another contender as a dual triode is the 5687, not cheap but not terribly costly either, high GM, good plate dissipation rating, and likes to be run at high-ish currents. Certainly seems like a good choice for a CF to me.
 
Administrator
Joined 2004
Paid Member
Re: #23

oldeurope said:
Hello Kevin,
this one has a low µ of 17. This gives a loss in gain. Maximum gain is 0,94.
:att'n: In some applications it is very importand to get a gain close to 1, e.g. bootstrap.
The ECC83 provides you up to 0.99. ;)

Kind regards,
Darius

I understand that, OTOH if you need to drive 10K with good linearity or typical headphone impedances the 12AX7A won't do that at all.

My experience with 12AX7A and CF applications exactly mirrors SY's and I have all the same objections. I stopped using them nearly 10 years ago for everything but error amplifiers in power supplies because frankly there are many tubes I feel are better for specific applications.

The 12AX7A was a low cost tube intended for cost sensitive applications, it is a jack of all trades and master of none. There are better choices in almost all cases where cost and ready availability are not the determining factors.

In a situation where gains approaching 1 are required you really will get better performance with a high gm pentode (pentode connected) - most other applications don't require it.

Are you using mu followers by any chance??
 
ECC83 and equivalents...

Hello Kevin,

Originally #23 posted by kevinkr

My experience with 12AX7A and CF applications exactly mirrors SY's and I have all the same objections. I stopped using them nearly 10 years ago for everything but error amplifiers in power supplies because frankly there are many tubes I feel are better for specific applications.

Oh, this makes me sad. :( SY wasn't able to use the ECC83 because he had some grid
current.
Originally # 13 posted by SY


At 2mA, you'll have some grid current issues with actual 12AX7s.

The gain for an ECC88 in this position is 0.97. The difference between that and a tube with a mu of 100 is less than 0.1dB.

I explained him how to use the ECC83 without grid current. See #14 #16
I am sure this will help you making low loss triode cathode followers, too.

Kind regards,
Darius
 
Administrator
Joined 2004
Paid Member
I guess you totally missed the point SY and I were trying to make. I rest my case. :rolleyes:

You may now continue to use the inferior performing 12AX7A to your heart's content, and I will stick to the results 28 yrs of engineering experience and what my FFT analyzer, not to mention my ears tell me.

FWIW if you ever change your mind take a look at some of the tubes the German Post Office came up with late in the tube era, and then still tell me the 12AX7A is better.

:D :D :D :D
 
dsavitsk said:
How about a 6DT8 or 12DT8. mu = 60, Gm=5500, and they are like $3 each.

I'm with this suggestion. My first instinct was to suggest the 12AT7/6201... sort of the "industry standard" for this, since they were MADE for "cathode-drive" (ie, cathode follower) operation... it's specifically stated in the datasheets. However, a 6DT8/12DT8 is functionally identical to the 12AT7, save for having a different pin out and greater heater/cathode voltage-difference handling... DEFINITELY useful in a cathode follower under high-voltage-swing conditions. And, with the mu and gm of the 6DT8, you can get right at .95 gain... at MUCH higher current demands and much LOWER output impedance than a 12AX7 or its ilk. Not to mention the plate curves are better for the 6DT8/12AT7/6201.

OTOH- if you need mega-swing or mega-current capacity, the 6BL7 is an absolute monster as a cathode follower. +-200v? It'll do it. It'll take in excess of 15ma per section to do it right, but output impedance is like, just over 100 ohms!! Gain is only about .9... but as long as that's usable, you can run ANYTHING off of them...

One other good choice, though more expensive, is the 6n30P.... it'll definitely work well. Very high transconductance. It will lose a bit more gain than a 12AT7 or 6DT8 (about .9 vs. .95), but if you want linearity over a WIDE range of voltage swing... THIS tube is just about as good as it gets, IMHO, short of the rare 5842...

Regards,
Gordon.
 
Originally #25 posted by kevinkr

...The 12AX7A was a low cost tube intended for cost sensitive applications, it is a jack of all trades and master of none. There are better choices in almost all cases where cost and ready availability are not the determining factors.

In a situation where gains approaching 1 are required you really will get better performance with a high gm pentode (pentode connected) - most other applications don't require it.

Are you using mu followers by any chance??
Originally #28 posted by kevinkr
I guess you totally missed the point SY and I were trying to make. I rest my case. :rolleyes:

You may now continue to use the inferior performing 12AX7A to your heart's content, and I will stick to the results 28 yrs of engineering experience and what my FFT analyzer, not to mention my ears tell me.

FWIW if you ever change your mind take a look at some of the tubes the German Post Office came up with late in the tube era, and then still tell me the 12AX7A is better.

:D :D :D :D

Hi Kevin,
I like more technical discussions. I am not a tube dealer. :rolleyes:

BTW: Did you read post #1 ?

D.
 
Administrator
Joined 2004
Paid Member
oldeurope said:



BTW: Did you read post #1 ?

D.


Yep, right from the very beginning. Just because he commented that he knew about the 5687 wasn't necessarily going to prevent me from promoting that choice. :D

BTW I am not a tube dealer either, at this point no more than a hobbyist really, although I do hawk my old books still which incidentally do feature a lot of 12AX7A based circuits.

I used to supply small quantities of tubes to my old clients, and some casual buyers, but the current pricing regimen at my wholesale supplier makes it impossible to compete on price so I no longer sell any tubes at all.
 
Originally #1 posted by leadbelly
What I am looking for is a recommended NOS twin triode in the "standard" noval pinout (12AU7 etc.) for a cathode follower.

I know about the standard forum recommended 6N6P (not NOS) and 5687 (different pinout). I know 12AU7 is often called an "OK" CF tube but I was wondering if there was something better. Larger heater draw or different voltage is no problem.

Hello leadbelly,
please tell me the signal voltage level and the load resistance.
Line level? Is it used to drive a 50KΩ input?
In this case the ECC83 and similar tubes are the best. ...


Kind regards,
Darius :)
 
How to find the best ...

Cathode follower means that the signal voltage at the
cathode follows the signal voltage at the grid.
A perfect cathode follower does this job exactly,
means gain = 1.
A triode cathode follower with unity gain makes no distortion.
Thus try to find one that is as close as possible to
unity gain.
The gain of a cathode follower is simplified calculated in the following way:

vu = (1–1/µ) x (r(r+1/s))

vu: [this is the voltage gain of the cathode follower]
(1–1/µ): [open output gain]
r(r+1/s): [gain reduce caused by load]
s: [transconductance]
r: [small signal load resistance at cathode]

link cathode follower gain
link asymmetrical cathode coupled stage gain calc. example
hope it helps. :)

Kind regards,
Darius

edit: links added
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.