Anybody interested in doing some circuit analysis?

[isaac42]

I've been looking over the schematic for my Peavey T.B. Raxx preamp with the view of determining its low end cutoff. Mostly it looks to me to be rolling off really low, around 5Hz or so. But there are a couple of locations where it's not clear to me exactly what's going on.

 

I tried to upload the schematic, but the forum says it's too big. It's Peavey drawing #98810270A, T. B. Raxx. There's a link to download the schematic here. It's in the sixth post down, by Vortexion. I'd be happy to send the schematic to anyone who wants it.

 

Here's what I'm seeing: all four tube stages have the cathode resistor bypassed down to 5Hz. That's good. But it looks to me as though C104, R100 and R104 form a high pass filter at either 72Hz or 144Hz, depending on which input is used.

 

On the control grid of the second stage V1B, the 1M resistor R119 is bypassed by C131. 884Hz? Or should I be including the grid resistance? Either way, I'm not sure what's going on there.

 

Then at the post gain pot, VR101, C101 looks as though it should roll off the lows at about 67Hz (140Hz with the GAIN switch cut in). However, I probably should be taking R139, VR100, VR104, VR107, R112 and R117 into account as well. That should put an equivalent 122K in parallel with VR101, further raising the cutoff frequency.

 

I'm fairly confident of all of the above. but what's going on with the post EQ section? Coming out of the post gain control, the signal goes through VR100, VR104 and VR107 to the outputs. However, the wipers of those pots feed the post gain EQ section and the fourth amplification stage V2B. The output of that stage is then fed to the outputs via C112. That seems a bit weird to me, but it seems to work. Despite the odd labeling (BOTTOM, BODY and EDGE), I've read that this is a conventional boost/cut tone control. Is this the way these usually work? I admit to being not at all good at understanding tone controls.

 

So, feel free to tell me where I've gone wrong and what I've got right.

[WRGKMC]

 

 

I'm fairly confident of all of the above. but what's going on with the post EQ section? Coming out of the post gain control, the signal goes through VR100, VR104 and VR107 to the outputs. However, the wipers of those pots feed the post gain EQ section and the fourth amplification stage V2B. The output of that stage is then fed to the outputs via C112. That seems a bit weird to me, but it seems to work. Despite the odd labeling (BOTTOM, BODY and EDGE), I've read that this is a conventional boost/cut tone control. Is this the way these usually work? I admit to being not at all good at understanding tone controls. So, feel free to tell me where I've gone wrong and what I've got right.

 

It looks like a second tone stack to me. The key there is it occurs after additional gain staging. The amplification factor is greater so those components react differently with higher voltages then they do at lower voltages. My guess is the second tone stack has a broader, lower Q filtering then they do in the first stack. The first stack receives the signal directly from the tube with only a coupling cap, but it a much lower gain signal. Without digging out a calculator I believe this would produce sharper Q filtering. The second tone stack would be broader. Much of this will do with the resistors used and how much of the filtered signal is mixed with the unfiltered.

 

Where a tone stack is placed between the amplifications in going to influence how frequencies get cut and how they sound to your ears. The capacitive charging and discharging of caps will take longer with higher voltages and that will affect the filter curve. The ears also have a different response to frequencies at low and high gains so having a second tone stack at a different gain level is going to change our perceived loudness of those frequencies.

 

Like I said I could probably figure it all out on paper, but I spend to may years doing hands on to be bothered. When I want to know what an amps doing, I just pump pink noise in and see what it does to that pink noise viewing a frequency analyzer. I can even go to the limbs of the tone stack and view what's going on with each one. If I wanted to change the tone stack I stick a decade box in there and select different valued components to find the sound I want.

 

This is a pretty simple preamp however. Pretty much straight forward road map. The only flaw I see ins the interconnects between the two boards. I know why they do that but if this is a plugin type board it can be a can of worms. The Tube board gets hot, The convection air flow draws in dust, Metal connections expand with heat and you wind up having an intermittent nightmare. Of course if they are soldered, then that problem is a non issue.

[isaac42]

Okay, my first error is using R100 and R104 to calculate the low cutoff of the input instead of R120. R120 is 1M, so R100 and R104 are pretty much negligible. F=1/(2pi*C100*R120)=1.6Hz. Including R100 and R104 in the calculation would lower the cutoff slightly, but it's already plenty low, so that's not an issue.

[isaac42]

Time to take another look at C101. That's the cap blocking the DC from the output of the third gain stage, V1A.

 

In my initial analysis, I looked at C101 and VR101 mainly in isolation. Then I added in R137, R139 and the pots in the post-gain EQ section. What I didn't do was to include R136, R115, R112, R117 and the plate resistance of V1A. What happens when I do?

 

On the one side, we have VR101 at 50K. With the GAIN switch closed, R137 is in parallel for a total of about 24K. Beyond that we have R139 at 10K, VR100, 104 and 107, each 250K for 83K in parallel, and the input resistance of whatever is downstream in parallel with R112 and/or R117. I'm going to simplify and assume that's going to be a 100K equivalent. On the other side, we have R136 in series and R115 and Rp (V1A plate resistance) in parallel with each other but in series with R136 and C101. Rp for a 12AX7 is 70K, and I'm going to ignore R125 as being too small to matter. That gives us 100K in parallel with 70K, or about 40K, in series with 100K, for 140K on that side.

 

Overall, that gives us VR101 at 50K, that whole mess on the right estimated at 100K, and 140K on the left, all in parallel. That's an equivalent resistance of about 27K. C101 is .047μF, so that gives us a cutoff of 125 Hz. That's higher than my initial calculation, and switching in R137 with the GAIN switch makes it worse. I'm thinking that replacing C101 with a .47μF cap might be a good idea. I don't see how it could hurt, anyway.

 

[isaac42]

Additional thoughts on C101. The voltages coming in from the EQ amplification stage V2B across C112 are going to play havoc with yesterday's analysis. I think I'm going to have to ignore everything to the right of VR101. I think that's fair, as the voltages across C112 would tend to minimize current flow through the EQ pots VR100, 104 and 107 and therefore through R139 as well. Any such reduction would reduce the effects of that branch on the frequency response. That changes things a bit. Now, on the right we have either 50K or 24K, and on the left still 140K. The total resistance is therefore either 37K or 20.5K for a cutoff of either 91.5Hz or 165Hz. Still not real good for a bass amp, IMO.

[vince-jelly]

Hello!

Very interesting stuff...

Please forgive me for my bad english, but...I was wondering why you are considering the left part of VR101 and the right part of it like parallel circuits, I mean....Doesn't the signal follows from the pot VR101 to the next path, the active tone circuit one...?

It enters V2B, then it goes to c112 (which is to be considered within its connections to the parallel resistors VR100, VR104 and VR107 to ground, or, better, then to VR101 and to ground..), then the signal goes to the output jacks...Isn't it?

If you use this calculator, http://www.ampbooks.com/home/amplifier-calculators/coupling-capacitor/

it shouldn't roll of as much low end...this calculator allows you to enter also output impedances, which should be very important when you calculate coupling capacitors after tubes...try it, for example, considering C101 and the parallel resistors R137 and VR101..: if you enter 100k (and more, because of the output impedance of a tube), which is the value of R136, as the output impedance in the calculator, then C101 and, let's say, 22K as the "volume control" in the calculator, it shouldn't roll off that much bass....

 

[isaac42]

It looks to me as though the signal follows two paths after VR101, R139 and the VR100, VR104 and VR107 combination. Going straight through those three potentiometers, the outputs are combined and go to the right side of C112. Coming from the wipers, the signal goes through the tone shaping circuits, then to V2B. From the output of V2B it is coupled via C112 to the output. I may be wrong about that, and I've expressed my confusion already, but that's what it looks like to me.

 

You're quite correct to question my series/parallel assumptions. It's obvious that R136 has to be in series. So I should have used R136 in series with the R115||Rp combination, but that should be in series with VR101 or VR101||R137, not in parallel. So, 140K in series with 50K or 24K rather than in parallel for a total resistance of 190K or 164K. That works out to either 18Hz or 20.64Hz. Much better. Also much more in line with the calculator you linked to.

 

Thank you for pointing out my error. Looks as though there's no need to replace C101 after all.