Noise and nasty distortion from a preamp depending on op amp used

[tuckster]

Hi,

I built a bass preamp (GK 800 RB) with a few LF353 powered by a +/-15V dc-dc converter (if that matters to the problem).
There is a nasty distortion/noise almost like the input clipping of a recording interface but I also have this sound when I run the preamp for example in front of any of my combos so it's not my interface where the clipping indicator is ok.
If the boost pot is cranked the noise starts when volume is at around 3 and it disappears when volume is at 8 1/2 then the signal gets much louder without the noise to the max of the volume pot.
Out of curiosity I soldered decoupling caps to the LF353 because I noticed that there are none and the problem almost went away.
The same issue with TL062 (also used caps) but TL72 are perfectly fine without decoupling caps.
What can couse this issue?
Do you need more information?

Thanks!

[antonis]

Do you need more information?

A schematic should be helpfull but let it be..

[tuckster]

Exactly that... Sorry! The only difference is the boost is always on, I did not build in the switch. I also put an aditional boost in front of the preamp as well as a compressor but they shouldn't be part of the issue. They are connected to the 9v jack not the converter (which i did before...) and are switched via 3pdt switches.

[antonis]

Now, to focus on your querry..

Are we dealing with various op-amps decoupling requirements..??

[ElectricDruid]

Maybe run the circuit off a pair of 9V batteries wired to provide +/-9V and see what happens?

If that fixes it, the problem is almost certainly the +/-15V DC-DC convertor injecting noise. As Antonis said, better decoupling might be able to solve that.

[Rob Strand]

What can couse this issue?
Do you need more information?

Maybe you need the supply caps.  Sometimes the answer is that simple.

To me the most suspicious behaviour is the fact the noise is worse near the cente of the Volume pot.

I'd be looking at:
- The wiring of the volume pot, the high-boost switch and the mid-contour switch.
   If any of those are exposed it could promote the behaviour you are seeing.
   If they pass near the converter that could also create problems.
- Does the behaviour change with different high-boost or mid-contour switch setting?
- (As ElectricDruids posted) Is the converter causing it?
  (The two possibilities for the converter noise: junk on the supply or junk injected into the audio path.)

[Vivek]

I built some Rockman X100 Clones using DC to DC convertors/ invertors

It made strange noises, like helicopter sounds

I assumed it was heterodyning since Rockman X100 has chorus and echo with their clocks

In any case, I added lots of decoupling caps close to the opamps and clocks. The problem went away

For some builds and some versions, the problem did not go away with the decoupling caps, so I further added 7806 and 7906 regulators after the DC convertors/invertors and that solved the issue.

[Vivek]

I do feel these Baxandall inspired Hi Mid/Low mid/Bass controls will have the normal issues of

a) Low Q

b) some level of interaction if Frequencies of bands are close

However, I would love to see the freq response curves needed for a bass guitar, so I will enter this into SPICE

[DIY Bass]

Exactly that... Sorry! The only difference is the boost is always on, I did not build in the switch. I also put an aditional boost in front of the preamp as well as a compressor but they shouldn't be part of the issue. They are connected to the 9v jack not the converter (which i did before...) and are switched via 3pdt switches.

I have this on my board, which sounds like it is close to what you have done

https://schalltechnik04.de/wp-content/uploads/2016/01/GK_04_Schaltplan.png

The boost will certainly throw it into distortion.  Wiring it always on I would expect to pretty much always have a distorted signal, especially with more boost in front. 

[tuckster]

Maybe run the circuit off a pair of 9V batteries wired to provide +/-9V and see what happens?
If that fixes it, the problem is almost certainly the +/-15V DC-DC convertor injecting noise. As Antonis said, better decoupling might be able to solve that.

Will do!

Maybe you need the supply caps.  Sometimes the answer is that simple.

Caps to the IC in addition to the decoupling caps?

- The wiring of the volume pot, the high-boost switch and the mid-contour switch.
   If any of those are exposed it could promote the behaviour you are seeing.
   If they pass near the converter that could also create problems.

Everything is directly soldered to the pcb.

- Does the behaviour change with different high-boost or mid-contour switch setting?

I will try to check this.

The boost will certainly throw it into distortion.  Wiring it always on I would expect to pretty much always have a distorted signal, especially with more boost in front. 

Its almost the same preamp in the GK400RB and the GK800RB with a few values changed.
Regarding the boost function I was told in another forum:
A deactivated boost means that there is a voltage devider consisting of the boost pot which is then a fixed resistor of 50k ohms and the 5.6k Ohm resistor which gives a ratio of ~1:9. So the regular signal gets attenuated.
When the boost is activated the ratio can be changed to roughly 1:1. In another forum they called it an anti boost. The signal does not get boosted but it is brought to the regular level by turning the boost pot to max.

This is the second build of this preamp. When I had the first one I noticed that the output was so low in volume that I always had to activate the boost pot. It can be dialed in to be still clean in lower settings. But then I had no additional boost to activate to get a gritty sound. That's why I removed the footswitch and added another boost in front of the preamp.
I saw that the GK700RB has no boost switch.

I don't own a Growling Krizzly to compare.
Maybe I should also use 220uF caps on the converter.

That's my PSU:

Mistake in the schematic: I use a Tec-2 1223.

[antonis]

Power supply decouplig caps should be placed as close as physically possible to IC respective pins..

Also, for bipolar supply, it should be better to place those caps between Vcc and Vee instead of going to GND..
(which GND should be kept as "clean" as posssible..)

[tuckster]

I did put them to gnd on the lf353 and tl62 but no caps on noise free tl072

[tuckster]

Quick check:
- same result with battery as with a regular power supply.
- no change when activating the high bost but there is no change when using the mid contour as this is just distorting the signal like crazy.

[merlinb]

Some opamps have better PSRR than others. That's life.

[merlinb]

Also, for bipolar supply, it should be better to place those caps between Vcc and Vee instead of going to GND..

This is very rarely the case. If you want to keep ground clean you can run a separate ground feed to the decoupling caps (standard in power amplifiers, rarely necessary in small signal work). Rail-to-rail caps usually do nothing and sometimes make things worse.