I've designed a fairly simple guitar amp that has uses a 12AX7 preamp tube with the tone stack in the middle, a 12AU7 long-tail-pair phase inverter, into a pair of 6V6's.
With cathode bypass caps (25µF with 1.5K) on the two preamp stages, there's too much gain. I could put in a simple resistive voltage divider.
Without the bypass caps, the gain is just right.
I've heard that the bypass caps help reduce noise, but I really don't hear much noise without them.
Are there benefits to the cathode bypass caps, which compel me to use them?
Thanks!
With cathode bypass caps (25µF with 1.5K) on the two preamp stages, there's too much gain. I could put in a simple resistive voltage divider.
Without the bypass caps, the gain is just right.
I've heard that the bypass caps help reduce noise, but I really don't hear much noise without them.
Are there benefits to the cathode bypass caps, which compel me to use them?
Thanks!
PRNDL said:
You indicate that noise performance is acceptable so they aren't needed for this reason anyway. Tone and compression behavior under deliberate overdrive may change, but again if you like the result then it is acceptable. There is no mandatory reason they have to be there - so if performance is satisfactory without them they can be left out.
The bypass caps bypass AC signals which allow higher gain through the tube (as you as well have found out). In combination with the cathode resistor you can forum a hi pass filter to allow only higher frequencies to pass. So if you have too much bass you can adjust the cap appropriately.
The formula is:
cut-off freq = 1/ (2*pi*R*C)
The formula is:
cut-off freq = 1/ (2*pi*R*C)
d1camero said:
Perhaps not quite that simple, you need to know the thevenin equivalent source resistance of the cathode resistor plus the cathode's internal resistance. Internal cathode resistance is typically fairly close to 1/gm.. IIRC
Pretty sure none of this is relevant to PRNDL's original question about omitting the cathode bypass caps..
d1camero said:
Just calculate the parallel equivalent resistance using the value derived from gm (the tube's transconductance) and the cathode resistor you are using. IIRC for the typical common cathode 12AX7 it is about 1K, (assuming gm of ~1000 uMhos) so using a 1K cathode resistor would give you 500 ohms thevenin equivalent resistance and then use the equation you posted previously. Most people not "tuning" tend to just deliberately overkill on this value rather than calculate it I suspect.
Point taken on voicing tone, was not thinking of the small value caps sometimes seen in guitar pre stages cathode circuits to fine tune the sound.
PRNDL said:
No.
If you have too much gain with these cathode bypasses, and enough without them, it's a good deal easier to eliminate them than to be making either voltage dividers or adding NFB networks to make anode followers. Removing cathode bypasses also improves linearity and reduces distortion.
Craig, the electric guitar amp forms part of the instrument. High fidelity is bad. It is expected and desired that the amp changes the characteristics of the signal. Sometimes a player wants even harmonic distortion, other times odd. Sometimes heavy clipping, sometimes light clipping. Sometimes linearity is OK, other times it is undesirable (go figure). It is all about the elusive and undefinable "tone".
Lots of tricks are played to get different tone. Multiple gain stages whilst knocking down the signal each stage. Various coupling cap values. Various bypass cap values. Using carbon comp resistors in key areas, as they drift under high voltage. ad naseum.
This however does not apply to acoustic guitar amps. Those tend to be high fidelity, as the tone is in the wood and string of the instrument.
Lots of tricks are played to get different tone. Multiple gain stages whilst knocking down the signal each stage. Various coupling cap values. Various bypass cap values. Using carbon comp resistors in key areas, as they drift under high voltage. ad naseum.
This however does not apply to acoustic guitar amps. Those tend to be high fidelity, as the tone is in the wood and string of the instrument.
Depending on the character one wants to have, it is sometimes better to reduce gain of the stage at the output, that is, use a tap on the anode resistor or make a divider with the next stage's grid resistor, while the triode itself runs at full AC gain (down to the desired cut-off freq.) with the cathode bypass and produces the full open-loop curvature.
- Klaus
- Klaus
Re: Re: Re: preamp cathode bypass cap benefits
Not variations in Ra, but rather Gm. This is a problem with every active device: Gm increases on positive upswings (negative for PNP and P-Channel transistors). Adding additional resistance to degenerate the cathode/source/emitter (since Gm= ~1/Rk) reduces that variation and reduces distortion. You can also look at it as a means of introducing local NFB, which also reduces distortion.
7N7 said:
Not variations in Ra, but rather Gm. This is a problem with every active device: Gm increases on positive upswings (negative for PNP and P-Channel transistors). Adding additional resistance to degenerate the cathode/source/emitter (since Gm= ~1/Rk) reduces that variation and reduces distortion. You can also look at it as a means of introducing local NFB, which also reduces distortion.
^^^^
Hey Kevin,
That's a common explanation for emitter degeneration from solid state practice. Not so much seen in hollow state practice, but it's still relevant. (See Doug Self's Site).
Hey Kevin,
That's a common explanation for emitter degeneration from solid state practice. Not so much seen in hollow state practice, but it's still relevant. (See Doug Self's Site).
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.