5879 OD & CF Question

[tubenit]

I don't really understand the "unity gain" concept that I read about regarding cathode followers.  I have presumed that means that the cathode follower does not add any gain?

Would the cathode follower here reflect the same gain of the 5879 whether I used a 12AY7 or a 5751?   OR would the 5751 increase the gain coming off the OD level pot compared to a 12AY7 in the CF position?   

I can not really "test" this effectively since the second half of the cathode follower tube is the return of the FX loop.  In other words, I hear more gain with a 5751 then a 12AY7 but I don't know if that is due to the triode in the FX return?

[tubenit]

I originally tried a cathode follower after a 5879 and it seemed to smooth the overdrive somewhat so I've continued to use that.

However, when I did that ............... I did not pay attention to whether there seemed to be the same amount of gain or less gain?

So, my question is :   Does the CF behind the 5879 tube somewhat "neutralize" the FAT overdrive that the 5879 tube gives and make it more
12A_7 like in gain and OD fatness ?

In other words, with the attached schematic,  would one hear a fatter OD tone and more gain without the cathode follower?  (as shown in the lower
schematic drawing)

[jazbo8]

The cathode follower has less than unity gain but it acts a buffer between the preceding stage, in this case the 5879, and the following stage, so it will certainly have an effect on the tone - but I'm not sure you can attribute the change in the tone to the "gain" alone though...

[kagliostro]

CF acts as an impedance adapter, so its benefit is used in various situations, one can be as driver of an external line (FXLoop) other can be after a pentode before a TS as the pentode didn't like TS as load, but there are a lot of situations where it improves the circuit


About gain ... the signal level outcoming  of a CF is equal or less to the input signal so you can espect the same signal or a bit less


In your situation my opinion is that it optimizes the signal transfer from the 5879 to the following stages, pentodes are more sensible than triodes about the impedance load


Franco

[sluckey]

Technically speaking, gain is equal to voltage output divided by voltage input and has nothing to do with tone (more on the tone statement later). If you apply a 1V signal to the input of a device and get a 25V signal out, the device is said to have a gain of 25. Unity gain simply means that the gain of a device is 1. IOW, if you apply a 1V signal to the input you will get a 1V signal at the output.

A characteristic of a cathode follower is that the output signal voltage will always be slightly less than the input signal voltage. The output will typically be 90% of the input. Said another way, if you apply a 1V signal to the grid you will get 0.9V signal at the cathode. Using the gain formula shows that this cathode follower has a gain of 0.9. Many people, including myself, just ignore the slight loss and call this "unity" gain just for convenience.

So what good is a cathode follower if it puts out the same as you put in? Low impedance! Even though the output voltage is the same, the output impedance is low (as compared to the high impedance of a tube's plate). A low impedance CF is capable of driving a heavy load such as a lossy tone stack without much loss in signal amplitude. Try driving the same lossy tonestack from a tubes high impedance plate and the signal amplitude will be knocked down considerably. Another advantage of a CF is that a low impedance output is less susceptible to noise or other electrical interference than a high impedance plate circuit. A good example of this is a CF driven FX loop Send, where the FX device may be located some distance away. The CF will get the "Send" signal to the FX device with minimal loss and good noise/interference rejection. Another real world application of a CF is a juke box that has the volume and tone circuit located remotely, maybe behind a bar. And I just ran across a reverb recovery circuit in that AO-63 amp that used a CF because the low level reverb recovery signal had to travel about 12 feet (6 feet each way) from the amplifier to the reverb switches on the console and back.

Gain and tone. "Gain" has come to mean something else to musicians, especially gitpickers. All musicians know that more gain is more hair, grit, growl, distortion, etc. All things that sound differently to each individual. A musician's definition of gain is complicated and very subject to interpretation. But in the electronics world, gain means exactly one thing and everyone should be in agreement, especially on test day. That said, a cathode follower has a gain of 1 which should not change the sound. But, just placing more components in a signal path will alter the sound. More importantly, the way a CF interacts with the following circuit can have a big impact on what we hear.

     *** CATHODE FOLLOWER --> UNITY GAIN --> LOW IMPEDANCE OUTPUT ***  (put this in the vault)

[HotBluePlates]

I originally tried a cathode follower after a 5879 and it seemed to smooth the overdrive somewhat so I've continued to use that.
...
So, my question is :   Does the CF behind the 5879 tube somewhat "neutralize" the FAT overdrive that the 5879 tube gives and make it more
12A_7 like in gain and OD fatness ?

In other words, with the attached schematic,  would one hear a fatter OD tone and more gain without the cathode follower?  ...

I had to read you initial questions a few times to gather what you're observing/asking. If I have it right, you noticed (or remembered) there being more or different distortion before using the cathode follower.

Others covered the gain of the follower itself, and the low output impedance. Now consider the input impedance and how that effects the stage before the follower.

In your drawings with/without the follower the 5879 has a 56kΩ plate load, which is principle in determining the gain of the 5879. The usual rule of thumb for calculating gain of a pentode is A = Gm * RL, Gm being transconductance at the operating point and RL being the plate load. However, resistance from grid to ground of the following stage is in parallel with RL (through the coupling cap), and effectively reduces the value of RL (true for triode or pentode stages).

In your sketch without the follower, after the 5879 there is ~320kΩ to ground through the resistor and OD Level pot. That reduces the effective load of the 5879 to ~47kΩ, a ~16% reduction. In contrast, the input impedance of the cathode follower is very high, enough that it does not reduce the effective plate load resistance of the previous stage. So you can expect the non-follower circuit to have 16% less gain.

Now I haven't plotted loadlines for your 5879 stage (which are a pain with pentodes because we need to have curves based on that stage's screen voltage), but I do want to point out something... Yes, with the cathode follower the gain of the 5879 is not reduced due to loading effects. That said, the angle at which the loadline cuts across the gridlines is different for the two cases. There is some optimum loading of the 5879 which will produce the largest output signal and least distortion. Going in either direction above/below that will reduce output capability and increase distortion, but one may tend to emphasize even harmonics while the other emphasizes odd harmonics. Or perhaps both are present but one load tends to have the operating point closer to where the gridlines crowd together below the knee (heavy distortion) while the other emphasizes crowding of gridlines near current cutoff.

Point of the above is that the changed effective load could be changing the way the 5879 distorts, and not just how much voltage gain it exhibits, which then accounts for the differing perception of "fatness" for the arrangement you otherwise expect to have less gain (and "gain" itself isn't equal to distortion; we can cause a stage to distort while offering zero gain).

The overall idea is that a gain stage is impacted by the load which follows it; sometimes a heavier load after a stage reduces overall gain, but also causes the stage to distort in a way it otherwise wouldn't.

[tubenit]

Gentleman,   

THANKS for all of you for your help and explanations and especially to Sluckey and HBP for the detailed information.  I think (?) I actually understand the majority of what you guys have said and it is helpful to me in shaping how I have been thinking about this circuit! 

I am going to leave the topology as is with the cathode follower.  Somewhere down the line I may revisit plate and cathode resistor values around the 5879 and the 12A_7 CF & see what happens?  IF there is any positive changes in tone, I will report back. 

As typically goes for me,  I experiment ALOT because I lack the knowledge in this field.  Results are I probably keep 1 out of 4 experimental mods in place and reverse the other 3 back to "original". 

I am truly grateful for this forum and the privilege to continue learning from such knowledgeable individuals! Thank you for taking the time to post on this thread.

My appreciation and respect, Tubenit

[shooter]

+1
I could just feel the cobwebs dissolving off 'ol memories!

(put this in the vault)

Please don't lock up your knowledge