Need help with static and hiss on a GGG TS-808 clone

[Electron Tornado]

I have a GGG TS 808 clone that I built. I added a mod for 3 different clipping settings and the "fat" mod. Unfortunately, it has a great deal of static and hiss when the effect is engaged. The drive control has the most effect on the hiss, and the volume control has a certain amount as well.

Here is the schematic:   http://www.generalguitargadgets.com/pdf/ggg_its8_sc.pdf

The voltages on the pins are all OK, and pin 4 is grounded well. Using an audio probe I found no hiss on pins 2 or 3, but a great deal of hiss on pin 1. Pins 5, 6, and 7 had some hiss, but not to the extent of pin 1.

Here are the things I've tried, none of which have resulted in any change in the level of static and hiss:

1. Used a battery instead of external power.

2. Changed the chip for another JRC4558. (I use a socket for the IC)

3. I checked and re-heated all the solder joints on the chip and all the joints in the loop between pins 1 and 2.

4. I have the clipping diodes on a daughter board, but I never hear any radio stations from them in the hiss and static. Switching between different sets has no effect on the hiss.

5. I removed the clipping diodes from the circuit.


Does anyone have any suggestions for anything else I can check or try?

[tux320]

Try shielded wire on the input and output.

[Electron Tornado]

Thanks for the tip. I can try that, but I don't have any shielded wire on hand.

I'm not sure if I understand how that will help, however. There is no noise problem from the input of the pedal to the input (pin 3) of the op amp.

[oskar]

-C4, try poking on the joints with your finger and/or stick a second cap in parallel... like in between the leads of the diods.
-Try yet another IC, any dual. Even though the chance is you would get two fawlty/ extremely noisy ones isn't big, popping in another one is so easy I really think you should have a go at it.

Now it's time to post some voltages.

[Electron Tornado]

Thanks for the tip about C4, oskar.


Here are the IC's pin voltages:

Power supply voltage = 9.2v
Pin      Voltage
1          4.6
2          4.8
3          4.4
4          0
5          4.6
6          4.6
7          4.6
8          9.2

- Pin 4 shows zero ohms between it and ground as well

- I had tried another JRC4558D earlier with no luck. I tried a TA75558P out of a TS9 as well with no luck.

- Wiggled all wires and components. Found and corrected two possible bad solder joints. No reduction in hiss.

- Put capacitor in parallel with C4. Used values 100pf, 330pf, 470pf, and 0.001 uf. This got rid of more hiss as I used larger values.


I can change C4 from a 51pf cap to something around 500pf. I think that will cut some of the highs though I'm not sure how significant it will be. Off hand, I can't recall a cap value in that position being more than 50-100pf, so I need to research that a bit to find the degree to which it might affect the highs.

[ACS]

Have you checked all the component values are what they are supposed to be?  An accidental substitution could be your culprit...

[Electron Tornado]

I can change C4 from a 51pf cap to something around 500pf. I think that will cut some of the highs though I'm not sure how significant it will be. Off hand, I can't recall a cap value in that position being more than 50-100pf, so I need to research that a bit to find the degree to which it might affect the highs.

Come to think of it, I wonder if I can mitigate any highs lost from C4 by changing C5 or C9. Or... ...I can try a small value cap to ground at the output. I've used that successfully in another overdrive pedal.


ACS, thanks for the reply. I considered that, though I recall doing a thorough parts check when I built the kit to be sure I had everything. I also checked the board against the layout.

[oskar]

Substituting caps doesn't sound like the optimal fix. I only suggested fiddling with C4 as a means to find the error...
The noise seems to be introduced around IC1a ( Try sticking a cap between pin 3 and ground, that would rule out noise introduced earlier in the chain ) and a too high amplification could lead to this.
As ACS said above. Check the values again...
If for instance R6 is 470R instead of 4k7 you would get 10 times the amplification and a minimum x100 gain...
This would definitely increase the noiselevel.

[tux320]

Thanks for the tip. I can try that, but I don't have any shielded wire on hand.

I'm not sure if I understand how that will help, however. There is no noise problem from the input of the pedal to the input (pin 3) of the op amp.

Bugger, should have read your post better

[Electron Tornado]

Substituting caps doesn't sound like the optimal fix. I only suggested fiddling with C4 as a means to find the error...
The noise seems to be introduced around IC1a ( Try sticking a cap between pin 3 and ground, that would rule out noise introduced earlier in the chain ) and a too high amplification could lead to this.

I checked the component values and everything is correct.

I changed my test set up so I can turn up volume levels without getting feedback. Then poked around with an audio probe. There is some noise between the emitter of Q1 and pin 3 that was low enough that I couldn't hear it earlier. (Lesson learned.) The noise is there whether I use external power or a battery. Everything from the input to the base of Q1 is still quiet. I re-heated the solder joints from the input to pin 3 but the noise remained.

I poked around with a cap to ground between Q1 and pin 3 with some good results. Using at least 1uf I get a decent reduction in noise. Any lower cap values don't attenuate the noise much. I have a TS9 on the bench with a JRC 4558, and when using a 1uf cap in the TS808 they sound very comparable.

Could Q1 be "noisy" and changing it be a fix, or should I simply use the 1uf cap to ground?

[oskar]

I poked around with a cap to ground between Q1 and pin 3 with some good results. Using at least 1uf I get a decent reduction in noise. Any lower cap values don't attenuate the noise much. I have a TS9 on the bench with a JRC 4558, and when using a 1uf cap in the TS808 they sound very comparable.

Could Q1 be "noisy" and changing it be a fix, or should I simply use the 1uf cap to ground?

Well, that was my idea behind the cap-test, but I really don't know. If you had result by introducing filtering on pin 3
I think it points strongly toward Q1, besides what else could it be? ...You don't happen to have a lot of wireless equipment around?

Anyway you shouldn't use extra filtering as a fix. Again it was for diagnostic purposes, to see where the noise is introduced.
You should settle with nothing but perfect. The worst type of errors that happens in professional electronic production isn't when a product won't start or breaks down when the customer gets home. The worst errors are the ones that still makes the machine operable but just degrades performance leaving the user believing it's just a bad product...

[Electron Tornado]

Well, that was my idea behind the cap-test, but I really don't know. If you had result by introducing filtering on pin 3
I think it points strongly toward Q1, besides what else could it be? ...You don't happen to have a lot of wireless equipment around?

Anyway you shouldn't use extra filtering as a fix. Again it was for diagnostic purposes, to see where the noise is introduced.
You should settle with nothing but perfect. The worst type of errors that happens in professional electronic production isn't when a product won't start or breaks down when the customer gets home. The worst errors are the ones that still makes the machine operable but just degrades performance leaving the user believing it's just a bad product...

Yeah, I really don't like having to use a band aid on a circuit.

OK, fingers crossed, it's time to change  Q1.   

[Zben3129]

1. Is it almost a "white noise" or "TV snow" kind of noise?

2. Did you directly solder one/all of your transistors?


Burning a transistor by soldering it instead of using sockets can create this exact kind of noise. Until I became very confident with my soldering skills I always socketed every transistor, just in case.


If you directly soldered them I recommend putting a socket where Q1 is and replacing it with a fresh transistor.

If Q1 is socketed, then try a fresh transistor, just because it is a very easy thing to try.



Zach

[Electron Tornado]

I replaced Q1 with a 2n3904 and got rid of almost all of the noise. The pedal is now only slightly noisier than the TS9. The transistor was soldered directly, since I'm fairly confident with my soldering. I do socket all ICs.

Just for the heck of it I tried a cap to ground on pin 3 again. That quieted it down further, but I didn't install the extra cap.

Lessons learned:

Don't assume that noise is generated in the stage where it's first heard at lower volume levels. When tracing a circuit for a noise like this, "look closely", meaning, set up the test so that gains and volumes can be turned up in order to find where noise may be soft, but still exist. A stage where amplification occurs may be only amplifying noise generated in a previous stage.


Thanks to all for their help and suggestions. 

[Zben3129]

You must have either had a bad transistor or made it bad by soldering 

I don't think theres anything wrong with direct soldering transistors, though I still socket MOSFETs and sometimes JFETs as they are more sensitive. Normal BJT's are always soldered. Sometimes one gets burned by mistake, happen to everyone.



Glad the problem is solved


Zach

[oskar]

Glad it worked.