How to calculate gain and draw loadline for tubes with cathode resistors?

[Alan0354]

Here's an interesting picture for you to look at, on this topic. Its source is probably familiar to some of the people around here, perhaps obscure to others.

View attachment 1059299

source: http://www.clarisonus.com/Archives/TubeTheory/Schade 1938 Beam Power Tubes.pdf

This is 6L6GC or just 6L6, it's very different from the plate curve I used. Don't scare me!!!! It uses 6.6K Raa.

The grid voltage does not look right either, just look at the bias of 50mA at 400V, the Vg1=-65V, that's way too high(too negative) for 6L6GC.

 

[PakProtector]

The grid voltage does not look right either, just look at the bias of 50mA at 400V, the Vg1=-65V, that's way too high(too negative) for 6L6GC.

Then to answer your question about reading only part of the document, I'd say that you will have to read enough so that those curves make sense to you. There is a discussion on DIYAudio on that effect/circuit/topology, and it also goes into cathode FB use. I have the link elsewhere, I'll post it and I suggest resurrecting that thread and asking questions you may develop from reading it.
cheers,
Douglas

 

[maxhifi]

6L6 and 6L6GC have the same curve, the difference is maximum ratings.

What that curve is, is a curve tracer hooked up to a 6L6 having 10% negative feedback to the control grid. The interesting bit is it quantifies the lowered plate resistance, and also how plate to control grid negative feedback looks similar to ultralinear operation (plate to screen grid). As for the 6.6k load, note the 400V plate voltage and 300V screen. That's right at the maximum ratings for the regular 6L6. Like Douglas said, the section about feedback in the article makes that picture fairly clear.

The article also shows curves for up to 100% "Schade" negative feedback, which I'm curious if anyone has tried in a practical amplifier. It would require a driver stage with high voltage capability, but I'm curious how it would measure and sound.

 

[PakProtector]

There has been at least one AK-er build one of my E-Linear amps who supplied the driver LTP directly from the plates of the power tubes( one that posted about it anyway...LOL ). The voltage tolerance is not quite so nutty as you might think even though the source can go to double B+.
cheers,
Douglas

 

[Alan0354]

6L6 and 6L6GC have the same curve, the difference is maximum ratings.

What that curve is, is a curve tracer hooked up to a 6L6 having 10% negative feedback to the control grid. The interesting bit is it quantifies the lowered plate resistance, and also how plate to control grid negative feedback looks similar to ultralinear operation (plate to screen grid). As for the 6.6k load, note the 400V plate voltage and 300V screen. That's right at the maximum ratings for the regular 6L6. Like Douglas said, the section about feedback in the article makes that picture fairly clear.

The article also shows curves for up to 100% "Schade" negative feedback, which I'm curious if anyone has tried in a practical amplifier. It would require a driver stage with high voltage capability, but I'm curious how it would measure and sound.

I compare the graph you have to JJ and GE 6L6GC, what a relieve, the plate curves are close enough, only the Vg1 is not the same as in the JJ and GE datasheet.

You scare me!!! If the plate curves are not the same, that means my Hammond 1650T I pick of 1.9K prg1=-8V(Vg2=400V) as Vg1=0 curve for Vg2=300V. It is very close. It is a good approximation. You can see the loadline/Vg1=0 intersection point give very close voltage and current as shown.

The bias point though is not right on the graph you showed. Your graph said for 50mA bias, Vg1=-65V. That's way too high. All the guitar amps I worked with using 6L6GC, even the lightest bias used no more negative than -52V. I always adjust to about -42V to get best sound. This is alignment more with GE and JJ datasheet.

 

[maxhifi]

Hi Alan,

Those curves are for the situation in circuit (b) in the below picture. For the practical implementation, circuit (c), the signal is AC coupled, so that curve is only valid for AC signals, and the normal pentode curves are still valid for the DC bias point.

In other words, it's nothing to worry about, it's just an interesting way to show graphically how negative feedback makes the tube appear more like a triode to the outside world. As for any discrepancies between the pentode curves in that old image, and modern curves for the 6L6, well, take the more up to date version, that image is from 1938!

Best Regards

Max

 

[PakProtector]

As promised, the link on Schade FB. There is some solid info there, and some capable folks doing some good work.
http://www.diyaudio.com/forums/tubes-valves/271134-schade-cfb-exactly-equivalent.html

Not just Schade but some solid CFB info. If acquiring a CFB OPT, be aware that they are not all as good as they claim to be. The A441 being a fine example of going too big. The Chicago BOH-series OTOH are very fine...
cheers,
Douglas

 

[Alan0354]

Here's an interesting picture for you to look at, on this topic. Its source is probably familiar to some of the people around here, perhaps obscure to others.

View attachment 1059299

source: http://www.clarisonus.com/Archives/TubeTheory/Schade 1938 Beam Power Tubes.pdf

I read the first 8 pages quickly, all I see are reasons NOT to use triodes and drive to +ve grid to get the extra power!!! I agree, distortion ( mostly high order) generated in the transition from 0 grid current region to +ve grid current. Unless one use a 0 ohm output impedance driver, there's going to be a problem. AND even at that, from the triode loadline graph I drew in post #69, you really don't get a whole lot out, still only about 1/2 of what you can get with penthode.

The article also say THD is higher when working in +ve grid region, I am not too sure I can agree as the spacing of the grid curve look quite even spaced. But that's besides the point, the grid drive require pretty much exclude all tube driving circuit that has higher output impedance. Even cathode follow is a whimp. Only way is to drive with SS driver circuit, that really complicates the design.

 

[Alan0354]

Hi Max

I skipped all the physics of the tubes, don't care to know. I read page 353, here is the graph of harmonic vs Raa plate load resistance.



This is graph for a single tube. I am familiar with this, this is the graph I used to pick the primary impedance of the OPT. This one shows 5th harmonics that the others don't. This make me question to choice of Hammond 1650T with 1.9K primary. The 5th harmonics is higher at 1.9K compare to 2.5K ( GREEN line). Of cause the 3rd is higher at 2.5K, but the 5th is 0. Really seem that 2.2K is the best compromise. That's too bad.

Can you explain Fig. 30a?



I have no idea both A and B means. Can you explain.

Alan


EDIT:

I scan through the rest of the article. What am I supposed to learn from this article? The last part is just composite characteristics of triode and penthode, it uses IDEAL curve rather than real life curves. The info about harmonic content vs primary impedance has been covered by other books like RDH4 etc. So far, the main thing I learn is right at the beginning to avoid Triode because of the +ve grid current that make it hard to drive the triode. Am I missing the moon?

 

[maxhifi]

I liked the article for a few reasons:

1. The description of how beam forming plates work, and the topographic analogies, were interesting.

2. It's by the design engineer of the 6L6, telling about why it was designed a certain way, and how to use it. I found his opinions interesting because they actually show influence on the development of the tube.

3. It shows what is "Schade" feedback. People call it that after the author of that article. And even better, it shows the curve tracer hooked up to a circuit, to clearly show the effect of feedback. It's interesting.

That graph with harmonics is for 250V on the screen, so not directly relevant. The other I have to go back to the article to look.

I'm leaving for a bit today, got a trip to NYC. I wonder if there's anywhere to buy tubes and stuff in that area? Will be back on the forum next week, or likely when I'm at airports.

 

[Alan0354]

I liked the article for a few reasons:

1. The description of how beam forming plates work, and the topographic analogies, were interesting.

2. It's by the design engineer of the 6L6, telling about why it was designed a certain way, and how to use it. I found his opinions interesting because they actually show influence on the development of the tube.

3. It shows what is "Schade" feedback. People call it that after the author of that article. And even better, it shows the curve tracer hooked up to a circuit, to clearly show the effect of feedback. It's interesting.

That graph with harmonics is for 250V on the screen, so not directly relevant. The other I have to go back to the article to look.

I'm leaving for a bit today, got a trip to NYC. I wonder if there's anywhere to buy tubes and stuff in that area? Will be back on the forum next week, or likely when I'm at airports.

Wow, I missed the Schade feedback, I have to go back and find it. Must be in the middle part I skipped.

Yeh, I am really not interested in all the secondary electrons, velocity, E field and all. I skipped the whole thing.

You have a nice trip, I'll talk to you when you come back. I did some major change in my design again, I don't like that I put all the regulated screen and input regulator supply onto the Output pcb as I really want the output pcb to be general purpose, so I redesign the whole grounding and put all the SS regulators back to the power supply board. I did not send the schematic to you, don't want to waste your time to look at something that is constantly changing.

Thanks

Alan

 

[Alan0354]

I found out what happened, my printer did not print from p358 to 364 where all the useful stuffs are. I can't read on the computer because of my neck problem, I have to print everything out to read. Sorry trees!!! I'll read it first.

No wonder!!!! I was wondering what are you talking about, all I saw is those physics of secondary electrons, beam and all.