Shaka Tube Waveform Study

[Paul Marossy]

Yep, it's time for yet another waveform study. This time it's the Shaka Tube, since was off of my pedalboard for a day, it was begging me to check out its curves... :wink:
It's really interesting to see these waveforms that you think sound good. A lot of the time, they don't look anything like you would expect them to.

Scope was my old Tektronix Model 453 with settings as follows:
Volts/Div 0.1V, Time/Div 0.5uS, Slope set to "+". Scope was in chop mode with no input on B to give a flatline to reference to. 600Hz sine wave at 20dB attenuation was fed into the input. These are what the waveforms look like at the output of the pedal. Volume control was at 5:00 for all shots below.




Here is what the waveform looks like with gain at 7:00 and tone at 7:00. Pretty much a clean sine wave, an amplified replica of what comes out of the signal generator.



Here it is with gain at 11:00 and tone at 7:00. Bottoms are getting a little squashed.




Here it is with gain at 2:00 and tone at 7:00. Bottoms more squashed now.




Here it is with gain at 5:00 and tone at 7:00. Not much change to be seen.




Here it is with gain at 5:00 and tone at 11:00. Now the tops are getting clipped and it looks like the pulse width changes a little bit, too. This is starting to sound harsh to my ears.




Here it is with gain at 5:00 and tone at 2:00. Volts/Div switch was set to 0.2V for this shot. This sounds harsher.



Here it is with gain at 5:00 and tone at 5:00. Volts/Div switch was set to 0.2V for this shot also. This to me sounds like the old "ice pick in the forehead" sound. It also looks kind of jagged.


A couple of things stand out to me here. One is that this circuit clips the bottoms more, just like the Real McTube II does. Second, the waveforms look a lot harsher than the McTube does. Compare: http://www.diystompboxes.com/forums/stompboxforum/
I don't know if it has anything to do with the fact that this is a starved plate circuit or not, but it's interesting nonetheless. BTW, the opamp I used is a JRC4558D. Tube was a 12AX7. These waveforms are also with a 0.001uF cap to ground on the output. Switching the cap out of the circuit has a minimal effect on the waveforms. Probably a 0.1uF cap would have more effect...

Hope this was interesting.  

[george]

Paul

the link just points to the forum?

BTW what if you use TL072 for the opamp?  as this is a JFET opamp maybe this would clip a bit more smoothly than a BJT based opamp?

I say this because at one stage, when I was debugging my ST, I just had the output of the TL072 coming through.  It was distorted at higher gains, but it actually didn't sound too harsh.

Maybe what is happening with your ST is the opamp is starting to clip at higher gains.

Regards
George

[Paul Marossy]

George-

Oh, sorry about that, I think this link should work:
http://www.diystompboxes.com/sboxforum/viewtopic.php?t=23922&highlight=mctube

Yes, that's a good point. A JFET opamp may look a little more tubey on the scope. I think I'll have to try that. I like it quite a lot with the JRC4558 in it as long as the tone control is at 9:00...

[Paul Marossy]

OK, so I have a question: For all you Shaka Tube builders out there, what opamp did you end up using in yours?

[Jay Doyle]

Paul,

Thanks for the scope pics! I don't know but would it be a more representative example if the probe was placed BEFORE the tone control? That way we could see how the distortion clips without the influence of the wave shaping of the tone control?

Just an idea.

Jay

[aron]

Obviously I used a TL072.  

[Paul Marossy]

Jay -

Another good point. I think I'll have to look at what the waveforms look like before the tone control. Then I'll report back. I'll try to do that tonight.

Aron -

Gee, did you really?!  :lol:
Hey, did you try out another types of opamps in it?

[aron]

I think I did try other op amps, but to be honest, I only use a few op amps and design around them.

I really don't use the JRC4558 unless in a Tube Screamer and sometimes not even then. I don't like the tone of them but I know a lot of people do.

Ever see my Deluxe Memory Man mods?  

[smoguzbenjamin]

I use the 4558 or 4559 or 072 or 2262. Whatever shows up first.

[Paul Marossy]

OK, I took Jay's suggestion and looked at the waveforms before the tone control. I looked at them at the gain pot. Here is what I saw:


Here is the gain control at 10:00. Bottoms are getting clipped more than the tops. Pretty hard clipping going here.


Here is the gain control at 12:00. Looks almost like a square wave. Still got the hard clipping.


This is pretty much how it looks between 1:00 and 5:00. Basically, really tall square waves. Really hard clipping going on...

I'm not sure what all this means other than there must be a lot of odd order harmonics being generated at this stage. I wonder if that accounts for the odd looking waveforms at the output? Next test is to try a JFET opamp instead...

[Thomas P.]

I tried a few opamps. First the TL072 which is my first choice noramlly but it didn't sound good. I also tried some piggiback combinations but I stuck with a NJM4558 which sounded best.

[Paul Marossy]

I remember trying a few different ones, too. I liked the JRC4558D the best. I'm just trying to see if there is a consensus as to what sounds best as far as opamps go. Then there is tubes...  :shock:

[WGTP]

Your probably not taking requests, but I would like to see some of Joe Davidsons stuff like the Black Fire and the Vulcan (diode compression) stuff.  

[Paul Marossy]

Blackfire could be a possibility. Mine has a Marshall style tone stack tacked on the end of it, but I could look at the signal before the tone stack...

I was going to do my original BSIAB next. It sounds so good, it's gotta look something like the Real McTube did, I would think. I don't try to guess what they'll look like anymore...  :oops:

[Paul Marossy]

Since I just built another Shaka Tube, with a couple of slightly different parts, I thought I would check out how it compares with the first group of waveforms in this thread. The slightly different parts consist of the following: On my earlier Shaka Tube build, I used a 390pF cap in parallel with the gain control instead of the specified 250pF and a 4.7uF cap after the 220 ohm resistor connected to Pin 2 of the opamp instead of the 2.2uF as specified on the schematic. Also, I used a 50K pot for the tone control instead of the 100K as specified. On the new build, it is built exactly per the schematic.

Scope: Tek 453, 1M probe
Time/Div: 0.5mS
Volt/Div: 50mV
Input Signal: 600Hz, 20dB attenuation
Slope: +
Chop Mode on with no input on B to give ground reference line.


This is with volume at 2:00, Gain at 9:00 and Tone at 12:00


This is with volume at 2:00, Gain at 10:00 and Tone at 12:00


This is with volume at 2:00, Gain at 2:00 and Tone at 12:00


This is with volume at 2:00, Gain at 5:00 and Tone at 12:00

They look a little different, with a much more pronounced hump in the bottom of the waveform. There wasn't much change with adjustment of the tone control or with frequency. The bright switch (0.01 cap to ground on the output) didn't have much effect, either.

[Paul Marossy]

This morning I looked at the waveforms right at the output of the opamp. In my first build using a JRC4558D, with the gain at  max., it's a pretty uniform, symmetrical square wave. In my new build, there is a TL072 in it. With the gain at max, it is also a square wave, but it is asymmetrical as far as pulse width goes. The two really clip about the same, except that TL072 has some curious little spikes on both edges of the square wave, top and bottom.