ampage Tube Amps / Music Electronics	For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Re: Request for dish

11/20/1998 9:56 AM
Dean Hazelwanter	Re: Request for dish I was also wondering about what exactly this device does. Is it the audio equivalent of 'pong', with the signal being bounced between 2 outputs, or is there any phasing or delay involved? Under the assumption that it's just a 'pong' e effect, similar to continually turning a stereo balance control between it's extremes, I attempted to build one. I started off with a schematic of Craig Anderton's tremolo circuit using CLM6000s to control the amount of signal passed between 2 halves of a dual opamp. Then I reasoned, that since the LFO was made from a 4049 hex inverter, and that 3 gates were left over, use them to invert the LFO output to drive a second tremolo circuit. I did a board layout using EasyTrax, and built up a unit. It works pretty good, although sometimes I get some LFO clock noise bleeding into the audio path. I scoped out the circuit, and found the CMOS V+ (which is isolated from the opamp V+), actually oscillates quite a bit. The opamp V+ stays nice and clean. I think it's caused by the proximity of the depth pots (in the audio circuit) cableing to the clock circuitry. I know I was successful when I let my 14 year old son try out the effect connected to the 2 identical hybrid amps I made, with about 6 feet separation. He started playing it, and got this big smile on his face and remarked 'Frick! That's pretty messed up!'. Anyway like I said, when the clock noise isn't there, which is most of the time, it sounds really good. I'll probably try shielding the depth control cables this weekend.

11/20/1998 9:56 AM

Dean Hazelwanter

Re: Request for dish

I was also wondering about what exactly this device does. Is it the audio equivalent of 'pong', with the signal being bounced between 2 outputs, or is there any phasing or delay involved?

Under the assumption that it's just a 'pong' e effect, similar to continually turning a stereo balance control between it's extremes, I attempted to build one.

I started off with a schematic of Craig Anderton's tremolo circuit using CLM6000s to control the amount of signal passed between 2 halves of a dual opamp. Then I reasoned, that since the LFO was made from a 4049 hex inverter, and that 3 gates were left over, use them to invert the LFO output to drive a second tremolo circuit. I did a board layout using EasyTrax, and built up a unit. It works pretty good, although sometimes I get some LFO clock noise bleeding into the audio path. I scoped out the circuit, and found the CMOS V+ (which is isolated from the opamp V+), actually oscillates quite a bit. The opamp V+ stays nice and clean. I think it's caused by the proximity of the depth pots (in the audio circuit) cableing to the clock circuitry. I know I was successful when I let my 14 year old son try out the effect connected to the 2 identical hybrid amps I made, with about 6 feet separation. He started playing it, and got this big smile on his face and remarked 'Frick! That's pretty messed up!'.

Anyway like I said, when the clock noise isn't there, which is most of the time, it sounds really good. I'll probably try shielding the depth control cables this weekend.

11/20/1998 10:03 AM

R.G.

quote:
"It works pretty good, although sometimes I get some LFO clock noise bleeding into the audio path. I scoped out the circuit, and found the CMOS V+ (which is isolated from the opamp V+), actually oscillates quite a bit. "

You might try using a 47 to 100 ohm resistor to damp the CMOS V+, with a 10uF on the bulk side and a 0.01uf or 0.001uF on the CMOS side. Should keep the CMOS V+ clean. If that doesn't work you could linear regulate the CMOS supply to +5V with a 787L05 if you can scale the outputs to the LEDs OK at 5V. Should cut power dissipation as well. Do you have all the inputs to the CMOS chip terminated?

11/20/1998 1:03 PM
Dean Hazelwanter	What do you mean by 'damping' the CMOS V+ with a resistor? The schematic I have has the opamp V+ connected to the incoming V+, with a 1k resistor that then goes to the CMOS V+. This CMOS V+ then has a 100uf filter cap, which I tried increasing all the way up to 1000uf. There was some improvement in the clock-bleed, but not much, so I thought maybe I was drawing too much current for the 4049, since it's driving 3 LEDs. I piggy-backed 2 4049s, and that didn't help at all. I then replaced the 4049AE with a 4049UB (unbuffered). This seemed to take care of the problem 100%, at 4AM when I was testing it - crystal clear under all conditions and settings. When I tried it again the next day, the problem was back, but not nearly as bad. That's why I'm not sure whether the audio is picking up clock noise, which is causing the CMOS V+ to oscillate, or if the CMOS V+ oscillation is the only problem, which then modulates the LED voltage in the CLM6000s, which in turn modulates the audio. The strange part is that the main (opamp) V+ is clean as a whistle, but on the other side of the 1k isolating resistor (CMOS V+), bad oscillation. And yes, all unused gates are tied to CMOS V+. The regulation of the CMOS V+ is a good idea, I may try it this weekend. It's either that or finish the remaining 5% of the PCB layout of the Craig Anderton Phase Shifter, using standard opamps (TL072 etc and TL074 etc). I just found this forum, so don't be surprised if any more requests for help are seen; I still need to figure why my DIY all-transistor Big Muff has a sort of noise gate effect. Thanks in advance for all input.

11/20/1998 1:03 PM

Dean Hazelwanter

What do you mean by 'damping' the CMOS V+ with a resistor? The schematic I have has the opamp V+ connected to the incoming V+, with a 1k resistor that then goes to the CMOS V+. This CMOS V+ then has a 100uf filter cap, which I tried increasing all the way up to 1000uf. There was some improvement in the clock-bleed, but not much, so I thought maybe I was drawing too much current for the 4049, since it's driving 3 LEDs. I piggy-backed 2 4049s, and that didn't help at all. I then replaced the 4049AE with a 4049UB (unbuffered). This seemed to take care of the problem 100%, at 4AM when I was testing it - crystal clear under all conditions and settings. When I tried it again the next day, the problem was back, but not nearly as bad. That's why I'm not sure whether the audio is picking up clock noise, which is causing the CMOS V+ to oscillate, or if the CMOS V+ oscillation is the only problem, which then modulates the LED voltage in the CLM6000s, which in turn modulates the audio. The strange part is that the main (opamp) V+ is clean as a whistle, but on the other side of the 1k isolating resistor (CMOS V+), bad oscillation.

And yes, all unused gates are tied to CMOS V+.

The regulation of the CMOS V+ is a good idea, I may try it this weekend. It's either that or finish the remaining 5% of the PCB layout of the Craig Anderton Phase Shifter, using standard opamps (TL072 etc and TL074 etc).

I just found this forum, so don't be surprised if any more requests for help are seen; I still need to figure why my DIY all-transistor Big Muff has a sort of noise gate effect.

Thanks in advance for all input.

11/20/1998 2:48 PM
R.G.	OK, from what you describe, the "damping" won't help. I was thinking that the power supply itself could be ringing, ran into that one myself a ways back. Theory 2: The CMOS is oscillating when it's transversing the linear region of the gate; you may need to compensate the CMOS inverters -OR- the CMOS is oscillating because of capacitive effects or surface contamination of the PCB. This would fit with the 4049UB making it better, as the inverters have lower gain. Clean the @#@$# out of the PCB to get all traces of flux off the board, then solder a 0.01 from V+ to V- pins on the chip (just for luck!), then mess with 50pf caps on the feedback resistors or 50pf to ground at odd places (this is my standard first approach to compensation - after this fails, I think

11/20/1998 2:48 PM

R.G.

OK, from what you describe, the "damping" won't help. I was thinking that the power supply itself could be ringing, ran into that one myself a ways back.

Theory 2: The CMOS is oscillating when it's transversing the linear region of the gate; you may need to compensate the CMOS inverters -OR- the CMOS is oscillating because of capacitive effects or surface contamination of the PCB. This would fit with the 4049UB making it better, as the inverters have lower gain. Clean the @#@$# out of the PCB to get all traces of flux off the board, then solder a 0.01 from V+ to V- pins on the chip (just for luck!), then mess with 50pf caps on the feedback resistors or 50pf to ground at odd places (this is my standard first approach to compensation - after this fails, I think ;-)

11/22/1998 1:04 PM
Dean Hazelwanter	I cleaned the @#@$# out of the PCB, no difference. I added a .01uf cap across CMOS V+ and V-, no difference. Remember I mentioned that after changing to a 4049UBE, the problem seemed to go away, and this was at about 4AM? Well, at that time it made sense to back out some of the other changes, including changing back from the 1000uf cap across CMOS V+, to the original 100uf cap. The circuit still seemed ok with 100uf, so I left it that way. When I tried the circuit again the next day, the oscillations were back, but I forgot about this cap change until Friday night when I looked at the problem again. I changed the cap back to 1000uf, and it works great at all settings - no more clock modulation, and the CMOS V+ is nice and clean! Sorry to lead you down the garden path, and thanks for all your suggestions!

11/22/1998 1:04 PM

Dean Hazelwanter

I cleaned the @#@$# out of the PCB, no difference. I added a .01uf cap across CMOS V+ and V-, no difference.

Remember I mentioned that after changing to a 4049UBE, the problem seemed to go away, and this was at about 4AM? Well, at that time it made sense to back out some of the other changes, including changing back from the 1000uf cap across CMOS V+, to the original 100uf cap. The circuit still seemed ok with 100uf, so I left it that way. When I tried the circuit again the next day, the oscillations were back, but I forgot about this cap change until Friday night when I looked at the problem again.

I changed the cap back to 1000uf, and it works great at all settings - no more clock modulation, and the CMOS V+ is nice and clean! Sorry to lead you down the garden path, and thanks for all your suggestions!

11/22/1998 4:16 PM
R.G.	But I like garden paths... I wonder if the problem is really just that you have a bad 100uF capacitor. I suppose that the difference between 100uF and 1000uF might stabilize things, but for CMOS, even in linear mode, it's hard to think that getting much bigger than 100uF would stabilize it. Just a thought. Congratulations - you nailed it.

11/25/1998 4:56 AM
Tobias Karlsson	This unit sounds VERY interesting, both the Flying Pan and the one Mr Dean Hazelwanter is experimenting with! I have for a while been thinking of a circuit that could make my guitar sound (with a two amps setup) like the first "song" (EXP) at Hendrix Axis: bold as love album. The guitar sounds like a moon rocket, coming in from the left side, rise in volyme. move over to the right, fade away as it moves over and starts to rise at the left side again. This would be very cool to do live with a two amps setup and controll the moving speed with my foot on a rocker! Is this what the Flying Pan or Mr Hazelwatnters tremolo experiment do? I'v been thinking to write about it here and hope that someone of you good elekticians run out of ideas and could take a shot on this. But as you all know, the flow of ideas seems to never run dry! I would construct a circuit like this immediately if I only knew how! Well, I can always hope.... B.T.W, a good name for this unit could be Rocket Rocker, eeeehh, maybe a little bit silly...

11/25/1998 4:56 AM

Tobias Karlsson

This unit sounds VERY interesting, both the Flying Pan and the one Mr Dean Hazelwanter is experimenting with!

I have for a while been thinking of a circuit that could make my guitar sound (with a two amps setup) like the first "song" (EXP) at Hendrix Axis: bold as love album.
The guitar sounds like a moon rocket, coming in from the left side, rise in volyme. move over to the right, fade away as it moves over and starts to rise at the left side again.

This would be very cool to do live with a two amps setup and controll the moving speed with my foot on a rocker!

Is this what the Flying Pan or Mr Hazelwatnters tremolo experiment do?

I'v been thinking to write about it here and hope that someone of you good elekticians run out of ideas and could take a shot on this.
But as you all know, the flow of ideas seems to never run dry!
I would construct a circuit like this immediately if I only knew how!

Well, I can always hope....
B.T.W, a good name for this unit could be Rocket Rocker, eeeehh, maybe a little bit silly...

Page 2 of 4