Capacitor value between preamp "stages"

[surfsup]

A question for the experts please. In the schematic:
http://ax84.com/p1x/AX84_P1x_101004.pdf

[img:488:429]http://chicagocadcam.com/ChrisHahn/ax84-p1x-pre.jpg[/img]

for the preamp stage, there is a single 12ax7 tube with two triodes. The Ra=100k and the resistor between "stages" is 220k. To achieve a cutoff frequency of 100Hz into the second triode, the value for C should be:

C=1/(6.28*F*R)=1/(6.28*100*69) = ~0.022uF

I use 69 because it is a parallel configuration of resistors. Did I calculate this correctly based on this simple schematic? I wanna be sure I understand what resistance to use. I am assuming we use the "source resistance" which is the value of the parallel resistors should the anode be grounded directly.?

[daydreamer]

hi surfsup,
I don't have any answers for you, but I'm studying this design as a first build as well. I'm new around here also.

There is an extensive 'brains trust' on this forum, and friendly too !

Andy

[surfsup]

Andy, good luck. My brain is overloaded for sure.

Just to let you know what I'm thinking, I am trying to design the amp from scratch. I figured out that if I bias the tube at -1.5V and do the load line, I get the same cathode resistor value, and calculated these values for the first half of the 12ax7 (if you want to compare numbers);

275V supply with 100k resistor gives 2.75mA (left side of load line)
275 is right side of load line. Draw load line
Bias Point = -1.5V (not cold, not hot, in middle)
Operating V (Va) = 185V (from 12ax7 graph)
Operating I (Ia) = 0.98mA (from graph, schematic shows 181 = 0.93mA)
Rk (cathode resistor) = ~1.5k
Ck (cathode cap) = 1uF (1uF will cut out frequencies below 100Hz, I think, due to R=1.5k)
Gain (A) = 62 (1Volt swing Vp-p = 212 - 150 = 62)
2nd distortion percent = 6.5%
Decibel gain = 36%

So far all the numbers that I calculated that are given in the schematic match up.

What I am asking the forum now is it "seems" a capacitor is used between the preamp stages to cut out frequencies and not feed low-frequency gain to the next stage. There is a 0.022uF cap there (C15). I am looking at getting that value based on the equation I posted earlier. I hope someone can confirm my thinking. If you can that would be great too! (If I'm right, it seems both caps C15 and Ck (C16) are acting as hipass filters removing 100Hz or lower).

All in all I think I'm starting to figure this stuff out, but maybe that means I'm in the danger zone!

[Structo]

That is a good way to learn.

Work the math to see how it all comes together.

You will begin to see how it works.

I think you will start to see some common values that are used in amps.
The bypass and coupling caps are used to "tune" the amp because they limit or pass certain frequencies.
The larger the cap the more bass.

[surfsup]

Tom do you have any advice for me on my math? I would certainly appreciate it.

I guess my next question would be regarding the second triode, the voltage reading in RED is 136. I assume this is due to the current from teh first triode stage entering the screen/grid. Is that correct? How is the 136 calculated? I'm having a hard time finding anything on the internet explaining the origin of this number. Thanks in advance...

[ChrisM]

.....

[Jana]

The source impedance is not going to be 100K. It is going to be some other value based on what the ax7 and the plate resistor create. I am *guessing* it is going to be about 60 to 70K.

The load impedance is going to be 220K + 1meg. The loading of the second grid will be quite high and shouldn't affect your calcs at the frequencies you are at.

Also, C14 and R22 form a high pass filter.

Don't ask me for anything further than this--my simple mind doesn't have the answers.

[daydreamer]

No idea yet on the maths, but word is that is a very "trebly" amp, lot of high end. You have probably uncovered why...thanks! My knowledge is still on a step curve..you are way ahead of me
Andy

[surfsup]

Jana thanks so much for confirming! My math actually calculates 69k Ohms. All the numbers so far match on the schematic to what I came up with. Additionally, you said:

The load impedance is going to be 220K + 1meg.

I thought the "source resistance" to be used for the next stage was determined by grounding the anode of the tube directly. If this is the case, isn't the 1M at the input bypassed? (so I should use the 69k value?)

Also, the voltage reading of the next triode in RED is 136. I assume this is diff. from the first due to the current from the first triode stage entering the screen/grid. Is that correct? How is the 136 calculated?

Man, I'm starting to sound like an amp builder! Here's the schematic again for reference:

[img:488:429]http://chicagocadcam.com/ChrisHahn/ax84-p1x-pre.jpg[/img]

[markr14850]

Also, the voltage reading of the next triode in RED is 136. I assume this is diff. from the first due to the current from the first triode stage entering the screen/grid. Is that correct? How is the 136 calculated?

The second stage cathode resistor is smaller than the first stage's, resulting in a hotter bias. More current flow leads to a larger drop across the plate resistor.

[Ken Moon]

Have you seen the P1-eX Theory of Operation document? I think it answers most if not all of your questions:

http://ax84.com/static/p1x/p1-ex-theory.pdf

Anything not covered there is covered in Merlin's preamp book, including all the math:

http://shop.dobermanamps.com/product-p/book-dvpgb.htm

Merlin also has a lot of this info posted on his website - I can't get to it from work, but I think he even has a downloadable first chapter to his book:

http://www.freewebs.com/valvewizard/

[surfsup]

Ken I'll check that out now thanks. I ordered the intro and preamp books from Merlin but haven't gotten a shipment notification yet. Ordered two weeks ago. Not sure what's up with that...
Mark, thanks so much:

The second stage cathode resistor is smaller than the first stage's

totally missed that.

resulting in a hotter bias. More current flow leads to a larger drop across the plate resistor.

When I put in values for a hotter bias (-2V) I get from graph a 215Va and 0.7mA Ia which results in a Rk=2.8k (higher value, not the 604Ohm). So now I'm confused. I am thinking I have to add the mA from the first stage into the calculation, but I'll look at the PDF Ken posted.

[vibratoking]

You are getting the right idea. I can't do a complete check here from work, but here are some comments off the top of my head.

The impedance looking back is basically the internal resistance of the V4b in parallel with R23. The internal resistance is probably about 70k or so. You apply what is called an AC ground to determine this. Power supply nodes and ground nodes are all considered AC ground.

The bias current is set by the cathode resistance for each device. Note that V4b is 1.k and V4a is 604.

C16 and C12 provide more gain at high frequency - a high frequency boost.

C14 is basically a bright cap configuration. R22 plus the resistance between pins 1 and 2 of the pot along with the input capacitance of V4a form a low pass filter. The input capacitance is a Miller cap, so it is dependent on the gain of V4a. The problem is that the resistance between pins 1 and 2 changes and causes the cutoff frequency to change as the pot is adjusted. This has the effect of lost highs as the pot is adjusted to attenuate the signal. C14 bypasses some of the highs directly to the V4a grid so that they are not lost. Some people like this and some don't.

That's all I have time for at the moment.

See "Designing Common-Cathode Triode Amplifiers" on the Aiken sight. It answers most of your questions.

[C Moore]

Ken Moon -
Thanks for that AX84 link. I have been looking for that
Thank You

[surfsup]

hey hired hand, if you're going to go through the PDF, I found an error already (well, more an inconsistency).

The schematic I posted is the most recent at AX84.com. It shows R18 as Ra=100k

The load line in the PDF Ken posted is using R18=120k which is confirmed in Fig. 15 on page 16, where the schematic is actually DIFFERENT than the one I posted (those sneaky little ...)! Took me a minute to figure that out (but I did figure it out!). Here's what I'm talking about:

[img:496:367]http://chicagocadcam.com/ChrisHahn/stage2.png[/img]

I think it has obviously changed since this PDF was first written. The tone stack resistors and pot values are also different. i hope that helps!

BTW, my math I posted earlier for everything is right (for the first triode), so if you wanna compare notes I doublechecked it through page 14 already. (I have already built a spreadsheet to automate all the calculations based on a few inputs from me , so it only took a few seconds)