DIY guitar preamp in stompbox makes oscillate as hell

[freebird92]

Hi. In these days I play with some solid state guitar amp.

I take the preamp section of famous solid state amp. Make the preamp into tiny pcb board that could fit in stompbox.

When I play these preamp with IR loader - these are really just like amp in the pedal board



In this time I take Fender Deluxe 112p. To make this work with 9V DC power supply, I apply 3w 15v dualrail DC-DC converter.

I make GERBER with Eagle, ordered the board, assembled the board and connect jack. And I heard some NOISE like hell.

This noise is about 10kHz.

The voltage of VCC VCC- are droped to 11V , -11V. When I put off the OPAMPs, the voltage restored to +-15V
I assemble 2 boards to check there are some cold joints or some mistake in my assembling. But both oscillate as hell. Only really small guitar sound and spring sound deliver to next stage.

To check power supply I use 3 power supplys. One of them is 60w 12V dc adaptor.

But every try I take I could hear the same sound. The oscillate from Hell.

I think I made some mistake in drawing board and schematic.

How can I make this work right?

Thank you.



1. original schematic: Fender Deluxe 112P original schematic
2. my dcdc converter: 3W DC DC dualrail 15V converter
3. my board and schematic of preamp:




View attachment del112p schematic.png

Fender Deluxe 112P original schematic

 

[Bo Deadly]

Try running it on two 9V batteries temporarily just to see if the DCDC converter might be responsible. I don't see an immediate problem with the DCDC converter but they can be sensitive. Especially the little ones like that one.

It could be a ground issue. Presumably you have a ground plane. If something is not grounded properly, that can be a source of oscillation.

I assume the DCDC converter input circuitry is completely isolated from the rest and that the ground plane does not go anywhere near it?

Check any high gain / high impedance sections carefully.

Disconnect everything that you can like the reverb to see if it influences the oscillation. Does anything change the frequency of oscillation?

Try momentarily shorting inputs to see if it influences the oscillation.

For example, the first amp looks like it's using some kind of positive feedback. Try grounding the intersection of C1 and C2.

Do you have a scope?

 

[freebird92]

Try running it on two 9V batteries temporarily just to see if the DCDC converter might be responsible. I don't see an immediate problem with the DCDC converter but they can be sensitive. Especially the little ones like that one.

It could be a ground issue. Presumably you have a ground plane. If something is not grounded properly, that can be a source of oscillation.

I assume the DCDC converter input circuitry is completely isolated from the rest and that the ground plane does not go anywhere near it?

Check any high gain / high impedance sections carefully.

Disconnect everything that you can like the reverb to see if it influences the oscillation. Does anything change the frequency of oscillation?

Try momentarily shorting inputs to see if it influences the oscillation.

For example, the first amp looks like it's using some kind of positive feedback. Try grounding the intersection of C1 and C2.

Do you have a scope?

Thank you for your kindness.

I have ground plane. I isolate the power supply negative from the board's ground plane. Because i use isolated dc-dc converter.

I connect dc-dc convert's common(0V/Ground) and 9V power supply's negative(0V) to test. But the oscillation occurs.

When I tested without spring reverb same oscillation occured. And I also tried pull off opamp one by one. The supply voltage increase a bit but oscillation still occurs. : (

When the input pin connected to ground(0V of the board) or It stops to oscillate. When I disconnect power supply from the board it also stop osscillation.

I have a multi meter but I don't have a oscilloscope. : (

I'll try with batteries and report the results! Thank you!

 

[abbey road d enfer]

In this time I take Fender Deluxe 112p. To make this work with 9V DC power supply, I apply 3w 15v dualrail DC-DC converter.

Have you checked the specs for the DC/DC converter? They usually indicate a maximum value for the capacitive load. I see you have about 180uF there. Seems rather high to me.
Generally, opamps rail bypass uses 0.1uF ceramic caps, not 10uF electrolytics.
Also, have you checked that all voltages are correct? Is the reverb tank connected?
Interestingly enough, as far as I can see, according to both schematic and layout of the Fender Deluxe 112, there are no individual rail bypass caps for the opamps, only a centralized pair of 47uF electrolytics.

 

[freebird92]

Have you checked the specs for the DC/DC converter? They usually indicate a maximum value for the capacitive load. I see you have about 180uF there. Seems rather high to me.
Generally, opamps rail bypass uses 0.1uF ceramic caps, not 10uF electrolytics.
Also, have you checked that all voltages are correct? Is the reverb tank connected?
Interestingly enough, as far as I can see, according to both schematic and layout of the Fender Deluxe 112, there are no individual rail bypass caps for the opamps, only a centralized pair of 47uF electrolytics.

Datasheet said "Capacitive Load for +-15V is 440 / 440 max".

I suppose that I could attenuate ripple when I use bypass cap. So I put some electrolyric caps with 0.1uF MLCC caps on both rails.

Listening to you, there may be too many capacitive loads. I test the board after put all electrolyric caps off and report!

Thank you for your kind advise.

 

[abbey road d enfer]

Listening to you, there may be too many capacitive loads. I test the board after put all electrolyric caps off and report!

Leave at least one pair!

 

[freebird92]

I remove all caps but It doesn't work. I try 9V battery but It doesn't work,

after remove OP7 the voltage goes to 15 / -15V

So I connect the jack into R18 R17 and play. The result is ... attached audio file

As attached audio 17sec. Reverb driver works. When I turn the reverb pot clock wise. The noise starts to wet.

I think the only thing I could do ... Go back to first step and draw the schematic once again.

This is my first board that has dual rail 15V. And I made some other dual rail 15V boards with some expectation. So this time It make me frustrated : (

I really appreciate all of your helps.

 

[abbey road d enfer]

after remove OP7 the voltage goes to 15 / -15V

What is OP7? TL072?
Check voltages on ALL pins with the opamp removed, then check with the opamp in place. Report here.

So I connect the jack into R18 R17 and play. The result is ... attached audio file

I can't open it, don't know why...

 

[freebird92]

What is OP7? TL072?
Check voltages on ALL pins with the opamp removed, then check with the opamp in place. Report here.

I can't open it, don't know why...

The OP7 in shcematic is yes, TL072.

Voltage of of OP7 pin (removed)
1 0V
2 0V
3 0V
4 -14.91 3.7VAC
5 0V
6 0V 15mVAC
7 0V 15mVAC
8 14.92

Voltage of of OP7 pin (fit TL072)
1 0V
2 0V
3 0V
4 -11.91 0.382VAC
5 0V
6 0V
7 0V
8 11.9 0.375VAC

You can download the file in here(my blog): 퇴역군인 : 네이버 블로그

 

[chilidawg]

Just a suggestion, if you revise you design:

in your current schematic OP2_A goes to OP7_A, then OP7_A goes back to OP2_B.
You could do this instead: OP2_A goes to OP2_B, then to OP7_A.

Always use all the op-amps available in one chip, before moving on to the next one. This simplifies things and make it easier to troubleshoot because you don't jump forward and back again.

 

[alphasnk]

Would a LC filter work there ? I had to use that on some dc/dc converters..

Edit: You could start a song/album with that whine/noise

Good luck!

 

[freebird92]

Just a suggestion, if you revise you design:

in your current schematic OP2_A goes to OP7_A, then OP7_A goes back to OP2_B.
You could do this instead: OP2_A goes to OP2_B, then to OP7_A.

Always use all the op-amps available in one chip, before moving on to the next one. This simplifies things and make it easier to troubleshoot because you don't jump forward and back again.

You're right.

I consider it as just copying schematic.

So I never thought about troubleshooting or debug.

If I draw a revision schematic I remember your advise. Thanks a lot.

Would a LC filter work there ? I had to use that on some dc/dc converters..

Edit: You could start a song/album with that whine/noise

Good luck!

I used dc-dc converter with LC filter in other design. And I couldn't find any issues about ripple or noise.

But in this time I don't know that works or not. I don't have oscilloscope yet.

Thank you for helping me think positively. : )

 

[Newmarket]

Would a LC filter work there ? I had to use that on some dc/dc converters..

Edit: You could start a song/album with that whine/noise

Good luck!

I think OP has stated that it doesn't work with batteries either ?

 

[chilidawg]

I don't think the oscillation has anything to do with having a LC filter or not, because when OP7 is present, the supply drop by 3V. That's a lot.

Either the 3W DC-DC converter doesn't have enough juice, or the circuit involving OP7 is not correctly drawn/designed.

 

[abbey road d enfer]

I don't think the oscillation has anything to do with having a LC filter or not, because when OP7 is present, the supply drop by 3V. That's a lot.

Either the 3W DC-DC converter doesn't have enough juice, or the circuit involving OP7 is not correctly drawn/designed.

+1.
In addition:
Voltage of OP7 removed) pin 4 -14.91 3.7VAC is very wrong.
Suggests some kind of oscillation. Maybe something is loading the output; could be an electrolytic cap reversed?
That's where an oscilloscope is very handy.

 

[freebird92]

I think OP has stated that it doesn't work with batteries either ?

Thank you for your reply!

You mean supply board's power with batteries? Or supply opamp directly through batteries?

I don't think the oscillation has anything to do with having a LC filter or not, because when OP7 is present, the supply drop by 3V. That's a lot.

Either the 3W DC-DC converter doesn't have enough juice, or the circuit involving OP7 is not correctly drawn/designed.

+1.
In addition:
Voltage of OP7 removed) pin 4 -14.91 3.7VAC is very wrong.
Suggests some kind of oscillation. Maybe something is loading the output; could be an electrolytic cap reversed?
That's where an oscilloscope is very handy.

No electrolytic cap was reversed. And the last test was without electrolytic caps.

So. I Assembled only few components and tested. Power supply section and the first stage.

input signal goes through OP2_A and pin 1 connected to c4 - output. To turn off other section of opamp, I tied opamp's pin 6, 7 and connected to GND.

In this time oscillation still occurs.

I think my design, especially dc -dc converter and power section, was wrong.

I am looking for some oscilloscope in used market but it is hard to find. (In my region there is almost no used market for electric DIY.)

Thanks a lot!




My previous design works well. pic1 is almost same design but single supply(24V). It works well.

Pic2 is dual rail 12V supply but without isolation. And It works well tool.

I wonder what is difference between them and what makes my recent design wrong.

pic 1


pic 2

 

[Bo Deadly]

Your DCDC converter is +-15 @ 100mA but your 4 amps only need a total of about 8mA. It could be that when you remove one of the amps, the load drops (to 6mA) and the SMPS is becoming extra unstable.

Tack on some parallel 2k2 (so 1k1) resistors across +15 and 0V and 0V to -15 and see if that helps stabilize the SMPS.

Or remove the inductors and insert some 9V battery clips to make +-9V.

You have a very low load for that SMPS and a lot of capacitance. That could be a bad combination.

If you are seeing 3.7VAC on the negative rail under any conditions, something is very wrong with the supply.

 

[ccaudle]

You mean supply board's power with batteries? Or supply opamp directly through batteries?

Use two in place of the switching regulator. With two batteries you only have +/- 9V, but the op-amps will still work. That will eliminate any problems with the regulator as being the cause.

It could be that when you remove one of the amps, the load drops (to 6mA) and the SMPS is becoming extra unstable.

Possible, but these models are not supposed to do that. The intro paragraph on the datasheet says "The converters are fully regulated over 0 - 100% load" and later "Minimum load: not required," so they are supposed to stay in regulation even with no load. They could drop the switching frequency at light loads and get down into the audible range.
That is the kind of info you can find in about 15 seconds with a 'scope.

Wiring in the batteries as supply should indicate whether the power supply is the place to keep investigating or not.

 

[freebird92]

Your DCDC converter is +-15 @ 100mA but your 4 amps only need a total of about 8mA. It could be that when you remove one of the amps, the load drops (to 6mA) and the SMPS is becoming extra unstable.

Tack on some parallel 2k2 (so 1k1) resistors across +15 and 0V and 0V to -15 and see if that helps stabilize the SMPS.

Or remove the inductors and insert some 9V battery clips to make +-9V.

You have a very low load for that SMPS and a lot of capacitance. That could be a bad combination.

If you are seeing 3.7VAC on the negative rail under any conditions, something is very wrong with the supply.

Use two in place of the switching regulator. With two batteries you only have +/- 9V, but the op-amps will still work. That will eliminate any problems with the regulator as being the cause.



Possible, but these models are not supposed to do that. The intro paragraph on the datasheet says "The converters are fully regulated over 0 - 100% load" and later "Minimum load: not required," so they are supposed to stay in regulation even with no load. They could drop the switching frequency at light loads and get down into the audible range.
That is the kind of info you can find in about 15 seconds with a 'scope.

Wiring in the batteries as supply should indicate whether the power supply is the place to keep investigating or not.

To sum up, the the first thing I have to do is supply opamps with 2 batteries to verify circuit works right without DC-DC converter.
(I tried 12V dualrail dc -dc converter that works right in other circuit, but it doesn't in this circuit.)

Since I don't have battery snap or clip wire for test I have to buy it. When I get battery snap I'll test and report here!

Thank you!

 

[smallstepsmastering]

Not sure if this applies to your design, but I do know of voltage inverters/charge pumps in other guitar pedal designs (MAX1044 and others) that naturally oscillate at 10KHz unless pin 1 & 8 are connected to each other. Worth checking the data sheet for anything that might allude to this.