The mu you use is the mu for the valve at that specific bias point you have chosen. This is not the voltage gain you would get at that bias point in a grounded cathode circuit - that would be to confuse (valve) mu with (circuit) voltage gain. As it happens the 12AX7 has a fairly flat mu curve so you can assume mu=100 without too much error. If your bias point is significantly different from the 'standard' values for a 12AX7 then you will need to get the actual mu from the datasheet curves or measure a few real valves and take the average. (Note: you will probably measure voltage gain, then back-calculate mu from that).
Member
Joined 2009
Paid Member
Thanks!
I may try this out, if I do I will use the russian equivalent, the 6H23Pi (6n23p) and I calculate the same as you.
I'm not sure that the Zp here is an issue, yes it's not something to make a line stage driver out of - you'd need to buffer it with a CF and then you have put yourself on course to build something like an Aikido. But most likely I will use it in a hybrid amplifier, where the SS output stage represents a load of 5 X the Zp.
And with this "lousy CCS" (SRPP with the output taken from the plate of the lower triode) the effective load resistance (Ra) will be roughly half that of the "typical, nice working point." Which was my whole point, actually.
Is that valid?
Yeah, I'm easily confused. I did mean Zout.
Zp then is what exactly?
If Zout = Ra'/2, then what is Ra'? The apparent load on the lower triode? If so, that would be about 6500R for this circuit with a 6DJ8 (or 6N23P). That means even worse than with a resistor in the plate.
What are we gaining from using this circuit?
--
Last edited:
Are we saying that the "lousy CCS" circuit is more linear than a standard common-cathode amp using the same tube?
What if you take a 6DJ8 (for instance) and run it from a high voltage supply, like +400V? Then to get your typical, nice working point you'd need a 56k plate resistor. Now you're getting that load line really flat. Low distortion that way too, correct? And the Zout would be about 3300R.
--
@DF96 -- Yes, now that you mention it, 12AX7 has a really flat line for mu in the mu/gm/rp graph. I guess that's a good thing in a high-mu tube like that.
What if you take a 6DJ8 (for instance) and run it from a high voltage supply, like +400V? Then to get your typical, nice working point you'd need a 56k plate resistor. Now you're getting that load line really flat. Low distortion that way too, correct? And the Zout would be about 3300R.
--
@DF96 -- Yes, now that you mention it, 12AX7 has a really flat line for mu in the mu/gm/rp graph. I guess that's a good thing in a high-mu tube like that.
Last edited:
Possibly, under certain conditions. There are two ways to get lowish distortion: very high anode load so you get the bare mu distortion of the valve, or an active load which is not necessarily very high in impedance but compensates for valve non-linearity (at least for even-order distortion). A valve which has low intrinsic distortion will probably be better using the first method. A valve which has significant second-order distortion may be better using the second method although it won't reduce odd-order or higher orders.rongon said:
Ra (anode resistance) = Rp (plate resistance) = mu/gm (a property of the valve at a particular bias point)
R'a = R'p = Ra + (mu+1)Rk (effective anode/plate resistance with cathode degeneration - a property of the valve in the circuit)
R'a is not the load applied to the valve but the output impedance presented by the valve, except for the balanced SRPP where each valve is the load for the other valve.
revintage,
Be careful not generate more confusion.
For the text we are referencing output impedance is defined as Zp.
Also note that for a circuit where the Rk in not bypassed the cathode resistor contributes to the output impedance. The formula used by rongon to calculate Zp is correct. See the text, formula (92).
rongon,
To calculate the output impedance of the circuit with a plate resistor in place of the top triode see Fig. 11.12 on page 426 and formula (11) on page 429.
DT
Member
Joined 2009
Paid Member
The 6H23Pi looks very similar (Ua 90V - 120V) but the curves suggest Ia more like 10mA.
Last edited:
QUOTE]revintage,
Be careful not generate more confusion.
For the text we are referencing output impedance is defined as Zp.
The formula used by rongon to calculate Zp is correct. See the text, formula (92).
Be careful not generate more confusion! Also be careful not to use wrong expression from old textbooks! Outputimpedance that Rongon calls Zp should be Zout.
About
Be careful not generate more confusion.
For the text we are referencing output impedance is defined as Zp.
The formula used by rongon to calculate Zp is correct. See the text, formula (92).
Be careful not generate more confusion! Also be careful not to use wrong expression from old textbooks! Outputimpedance that Rongon calls Zp should be Zout.
Who could be confused by that. Noone has talked about anything else.
About
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.