This is a circuit I see used on some DIY sites, but not here on diyAudio. I'm not sure what it's called. I think it's a 'triode with an active load', but I've seen it called an SRPP -- even though I know it's not that, because the SRPP has its output taken from the cathode of the top tube.
It looks to me like a half-baked mu follower, with an inadequate load on the upper cathode follower. Or maybe it's a simple CCS made out of a triode, used as the plate load for the lower tube. But as often happens, I could be looking at it wrong.
I've modeled it in spice, where it seems to perform exactly as well as a standard common cathode amplifier with plate resistor, but works with lower B+. That could be an advantage when you don't have a high enough voltage available from your plate supply.
So what is this circuit called?
What are the disadvantages of this circuit?
It looks to me like a half-baked mu follower, with an inadequate load on the upper cathode follower. Or maybe it's a simple CCS made out of a triode, used as the plate load for the lower tube. But as often happens, I could be looking at it wrong.
I've modeled it in spice, where it seems to perform exactly as well as a standard common cathode amplifier with plate resistor, but works with lower B+. That could be an advantage when you don't have a high enough voltage available from your plate supply.
So what is this circuit called?
What are the disadvantages of this circuit?
Attachments
Last edited:
Member
Joined 2009
Paid Member
It's just a triode wired up as a current source load - you'll see plenty of examples on this forum using JFETs instead of Triodes.
You posted that just as I edited my OP, as that very thought had somehow materialized out of the fog of my brain. 😱
So, the reason I get similar performance from this circuit as compared to the same circuit with a plate resistor is that a triode only gives you a low impedance (weak) CCS?
--
So, the reason I get similar performance from this circuit as compared to the same circuit with a plate resistor is that a triode only gives you a low impedance (weak) CCS?
--
Last edited:
Yes. Only in Spiceland will the upper and lower triodes even-order distortion cancel out exactly.
In practice, the cancellation can be quite good IF the stage is unloaded.
Trivia: This circuit was originally developed to reduce sensitivity of DC coupling to variations in heater voltage.
yes, this is what i used n my 5894 SET....
12at7 on B+ of 400volts, feeding a 6CG7 cathode follower which in turn fed paralleled grids...
the amp has since been sold, someone dropped by the house and when he heard it,
twisted my arm for it, he is now selling his 2A3 set to his friend...
12at7 on B+ of 400volts, feeding a 6CG7 cathode follower which in turn fed paralleled grids...
the amp has since been sold, someone dropped by the house and when he heard it,
twisted my arm for it, he is now selling his 2A3 set to his friend...
Since the cathode follower is very easy to drive, why use this stacked configuration instead of a plain resistor?
There is no even order distortion cancellation in this circuit (as would be in a properly configured srpp).
As SY mentioned, the basic advantage of this circuit is that, due to the symmetry, the output dc voltage is quite stable, which is desirable in dc coupling to the next stage.
The use of an extra triode, and the low drive capability / high Zout.
the active load gives the amplifying triode more gain since a plain resistor may not have the very high impedance of the current source load...
Member
Joined 2009
Paid Member
Broskie uses this front end in his 'Aikido' amplifier - if I remember correctly he had a few words to say on the advantages.
If you use one half of a dual triode as the gain stage, and the other half as the follower, you can many times bootstrap the gain stage from the follower's output, doing away with an extra CCS.. 🙂
Member
Joined 2009
Paid Member
But the follower needs a CCS load or a large resistor (not so much a problem in your L-W amp because you're running with 500V B+)
Another advantage is that there is no cap anywhere, and the stage is fully dc coupled. There is no Lower cut off frequency or phase shift..
Also very easy to wire up.
It might depend on what spice model(s) you're using. I used a model for the 12AX7 that I got from this forum, and the distortion and gain are pretty much the same whether I use this 'active load' circuit or a plate resistor, with the operating points kept constant. (I raised the B+ to allow use of a 220k plate resistor at the same plate current and plate voltage.)
So, I think SY got it right here. The CCS is a weak one, so this circuit is not good at cancelling even order distortion. But it does allow you to use a lower-than-optimal B+.
Yes, I know a DN2540 is much better. It is, I know. Yes.
In this case, it turned out in the spice sim that gain was ever so slightly lower from the active load version than from the plate loaded version. But that was with an unbypassed cathode resistor. Is the half-mu stage meant to be done with a cathode bypass cap?
I was asking about the topology, not about the limitations of a 12AX7. Let's say I asked this question about this circuit made with a 5687...
As mentioned, sometimes you end up with an extra triode in a twin-triode envelope, which this circuit can make useful.
Perhaps this circuit would be best done with a triode-pentode like 6GH8A or whatever. (Many are still cheap!) The pentode could be the CCS on top of the triode. More complex, though, because of the screen supply and decoupling. But not much.
--
Last edited:
I thought of another use for this stage... What if you used two of these to make an LTP? The CCS in the plate is weak, so will have to yield to a strong CCS in the tail, but you'll still get the advantage of being able to use a lower B+, which is very important if you want to maximize voltage swing (as in an LTP driving PP output grids in a power amp).
I would not use 12AX7s for this. I'd use something like 12AT7, maybe. Certainly 6SN7, 5687 or 12BH7A. Or... Use a 6DJ8 to get a low-voltage LTP going for driving PP EL84s.
--
I would not use 12AX7s for this. I'd use something like 12AT7, maybe. Certainly 6SN7, 5687 or 12BH7A. Or... Use a 6DJ8 to get a low-voltage LTP going for driving PP EL84s.
--
Last edited:
- Status
- Not open for further replies.