808 hiss reducers

[Govmnt_Lacky]

Built a TS-808 for a friend of mine and he told me today that there is a small but noticable HISS that increases with the Volume pot. The build was stock, with a JRC4558 chip.

Other than potential grounding issues... does anyone have other hiss-inducing suggestions?

I am thinking maybe a different op amp but, I am looking for past experiences. The build is off of the Tonepad circuit.

Thanks in advance 

[Mark Hammer]

I would suggest a lower noise chip, but I gather people ARE pretty attached to their JRC4558s, aren't they?  One could also suggest a feedback cap in the tone stage, in parallel with the 1k feedback resistor, but that might introduce some unwanted phase-related "smearing" of the signal because you'd need something on the order of 22nf or more to achieve the desired hiss reduction.

Which leaves us with replacing the trannies.  The 2SC1815 is a decent transistor, but there may be lower noise ones you can use for the input and output buffer stages, like a 2N5089, or 2N5210, or similar.

[R.G.]

Also remember that cold solder joints can hiss all by themselves. Remelt is quick and easy.

[Govmnt_Lacky]

I appreciate the feedback gents 

Most of the reading I am doing suggests that the IC may be the likely culprit. Some have also suggested shielded in/outs however, I wonder if the gain in this circuit warrants that 

I have several other ICs to try... as well as some 2N50XX trannys. I only wonder if replacing the IC (especially) will effect the overall tone of the pedal. The friend I made it for is a gigging musician and he knows what sound he should be getting from a TS.

I have also read that TS circuits are susceptable to hiss. Production units included.

I guess I will just have to throw some parts at it and see what turns up 

[Mark Hammer]

One last suggestion...

The 51pf cap in the feedbackpath of the clipping stage provides a gradually lower rolloff as gain goes up.  At max gain (pot set to 500k), it provides a rolloff at 5.6khz, which is great when you want to smooth out asignal.  But at lower gain (e.g., gain pot set to 250k for a gain of 65x), that rolloff moves up to around 10.6khz, which is going to let a more hiss come through, and provide a tougher challenge for that simple 1k/220nf lowpass filter to control.  If your buddy tends not to max the gain, it might be worth his while to swap that 51pf for a larger value like 82 or 100pf to take out a bit more of the hiss.

As always, such pedals are designed in the total absence of any insight or privileged knowledge about how any specific player tends to use/set it, so they design round worst case scenarios.  No reason why you couldn't tailor it to the circumstances your buddy plays under, though...right?

[ubersam]

I haven't noticed any hissing with my TS builds (which are modded for more gain than stock), could be my ears but could also be some of the things I do. I've been digging the TI RC4558 in my TS builds, they sound better to me than the JRC4558. I also like using JFET buffers (J201/2N5457) as opposed to bipolars. Another thing I do is add a small value ceramic cap (330pF~1nF) after the 1K input resistor to filter out RF. In addition to keeping RF out, it seems to helps keep any hissing down.


oh ya... I also up the 51pF cap to 100pF

[Govmnt_Lacky]

Finally sat down to troubleshoot this. I tried 5 different ICs with the following results:

NE5532 = Slight reduction in hiss but, it was definitely noticable
TL072 = No change
RC4558 = No change
TL082 = More hiss AND lots of unwanted feedback
Burr Brown = No change (really surprised at this one)

My next steps are to up the 51pF cap by adding a 47pF in parallel.

Also, I think I am going to put in shielded wire from Input jack and Output jack to the 3PDT.

To clarify the problem.. with my guitar plugged in and the effect bypassed it is dead silent. When I switch the 808 on, all of the controls have an effect on the background hiss. Volume and Gain make it louder. Tone makes it a "bit" louder as it is rotated towards the treble extreme (CW)

Does anyone have other ideas that have not been discussed?  

[Govmnt_Lacky]

One last tid-bit:

I used electros for the 0.22uF caps. The layout (Tonepad) says to use tants. In the past , I have built this circuit (albeit on a GGG layout) with film 0.22uF caps and they have no hiss.

Could the electros be causing a problem as well?

[Earthscum]

I have noticed a couple builds in the past that I used electros in being hissy. I kind of recall R.G. mentioning something about hiss being due to the leakage in electros?

One thing I tried (haven't committed it to PCB, though) is using MPSA13's for the input buffer. It did seem to reduce hiss a bit, especially after I completely re-did it for lower impedance (I think I just settled on a 2.7k Emitter resister, and a re-bias). I noticed that the overall sound seemed a bit "clearer", however subtly so. In that session I also tried an NE5532 (I was shooting for a Bass OD, as usual). Not sure what you thought about them, but they definitely don't sound TS'ish to me. The clarity all the way through with Darlington buffer and 5532 was crazy, though. Almost Hi-Fi, lol.

If it were my troubleshoot, I would be looking at everything before the gain/clipping stage first and foremost. Any hiss there is getting amplified. I like Mark's suggestion of tuning it to work best within the range that your buddy uses it. I've done this on a couple builds, one being a FF that, well... I know the guy likes to see the knob cranked. If it rips, and the knob is sitting at 12 o'clock, he'll crank it higher, so I built it with most of the gain in the upper half of rotation because I knew it was going to end up there anyways. 

[DougH]

Did you try reheating the solder joints yet?

[arawn]

TI makes an rc4560 ic - lower noise floor than the 4558 but otw supposed to be the same.

[R.G.]

Finally sat down to troubleshoot this. I tried 5 different ICs with the following results:
NE5532 = Slight reduction in hiss but, it was definitely noticable
TL072 = No change
RC4558 = No change
TL082 = More hiss AND lots of unwanted feedback
Burr Brown = No change (really surprised at this one)

Mother Nature is whispering in your ear "It's not the internal opamp noise that matters."

The TL082 comment makes me suspicious that you may be fighting a low level RF oscillation.

[Govmnt_Lacky]

Did you try reheating the solder joints yet?

Yes. This was done before I tried swapping ICs.

Yesterday I Replaced the In/Out wiring from the jacks to the 3PDT with shielded cable AND I added the parallel 47pF cap to the 51pF (98pf?). This made NO change in the hissing. All I am doing is plugging in my guitar, rolling the clean volume on my amp to about 4, and without playing a note I rotate the knobs on the TS. HISS GALORE 

So, today I worked on Part 2 of my plan. I replaced the 0.22uF electro caps with box film caps. I also swapped out both 2N3904 transistors with 2N4401 (GGG layout)

Hopefully, I will get to test it tonight/early tomorrow 

Fingers crossed 

[Earthscum]

The TL082 comment makes me suspicious that you may be fighting a low level RF oscillation.

That just made me remember... I was building something or other on the breadboard once and had some serious hissing. It would go away every time I touched one of the leads of the protection diode I was using at the input (1N4148). I tried another and it didn't do it, so I checked the hissy one with a diode checker and it acted like it conducted both ways (but with something like .8V Vf in reverse). I thought I fried it and tossed it out, and put in a 4001 so it didn't happen again.

I wonder about a couple other noisy projects I tried, now, or if that was just a highly random circumstance. Anyways, just throwing ideas out there.

[Govmnt_Lacky]

That just made me remember... I was building something or other on the breadboard once and had some serious hissing. It would go away every time I touched one of the leads of the protection diode I was using at the input (1N4148). I tried another and it didn't do it, so I checked the hissy one with a diode checker and it acted like it conducted both ways (but with something like .8V Vf in reverse). I thought I fried it and tossed it out, and put in a 4001 so it didn't happen again.

I wonder about a couple other noisy projects I tried, now, or if that was just a highly random circumstance. Anyways, just throwing ideas out there.

Hmmmm. Now that you say that, I did use some mil-spec 1N914 diodes that I have never used before. They may be my next choice if the last ideas do not work.

Thanks for the tip 

[Govmnt_Lacky]

Well.... no change in the HISS department 

Replaced the 2N3904s with 2N4401s (GGG) and I replaced the 0.22uF electros with 0.22uF box caps.

Next up is to replace the mil-spec 914 clippers with regular 914s. Also, I am going to swap out the 0.027uF film input cap (Tonepad) with a 0.022uF film (GGG)

If this does not do it, I am at a loss. I can't think of anything else that could be inducing the hiss except for the fact that it is just a consequence of using the Tonepad layout. I have built 2 from the GGG layout and they both are silent.

P.S. Another tid-bit. When I am testing, I do not even have the guitar plugged into the input. Just the cable to activate the battery. When I do plug in the guitar, there is no change in the hiss.

[R.G.]

The more parts you change with no change in hiss, the more I think you have a problem with layout, wiring or (lack of) power supply bypassing letting RF and/or ultrasonic get detected and heard as hiss.

What does your wiring look like, and how are your power supplies decoupled? How is it grounded? (unfortunate grounding can form the feedback path for causing RF)

[Govmnt_Lacky]

The more parts you change with no change in hiss, the more I think you have a problem with layout, wiring or (lack of) power supply bypassing letting RF and/or ultrasonic get detected and heard as hiss.

What does your wiring look like, and how are your power supplies decoupled? How is it grounded? (unfortunate grounding can form the feedback path for causing RF)

I am using the Tonepad layout for the TS-808 spec with no mods.

I was using a battery for the testing (as not to induce any wall wart hum). Its funny, when I tested it this afternoon I thought I had fixed it but, sadly it was just the battery slowly dying. When I plugged in the wall wart, the hissing returned.  

As far as grounding, this is how I have it wired:

DC jack connected to center lug of 3PDT
Center lug of 3PDT to Input jack sleeve
Input jack sleeve to circuit AND to output jack sleeve.

I do not wire my 3PDTs for input grounding when the effect is off. I used to but I have built dozens of circuits without it and had no problems.

EDIT: I just noticed that the GGG layout uses a 100uF electro in the decoupling. Would the absence of this cap be the cause of all of the hiss?

[R.G.]

I was using a battery for the testing (as not to induce any wall wart hum). Its funny, when I tested it this afternoon I thought I had fixed it but, sadly it was just the battery slowly dying. When I plugged in the wall wart, the hissing returned.  

Oscillation at some power supplies and not others also fits with self-oscillation. Again, not definitive, but it fits.

EDIT: I just noticed that the GGG layout uses a 100uF electro in the decoupling. Would the absence of this cap be the cause of all of the hiss?

I don't know that I'd say it's the cause of all the hiss. More like not having it lets any tendency to self-oscillate from power supply impedance problems express themselves.

If you have a 100uF and a 0.1uF ceramic, tie them across +9 and ground on the PCB. Actually, the 0.1uF should be soldered right across opamp pins 4 and 8 for best effect.

This may not be the problem, but putting those in will definitely not hurt, and will avoid other situations that could trip you up.

I hate to mention this, but I've recently recovered from a miswiring error myself. Are you *sure* all parts are inserted with the correct pinout, and that all wiring is correct? It is very easy to miss.

[Govmnt_Lacky]

I hate to mention this, but I've recently recovered from a miswiring error myself. Are you *sure* all parts are inserted with the correct pinout, and that all wiring is correct? It is very easy to miss.

I have learned after many builds that it is not wise to post until AFTER you have verified parts placement, parts values, and wiring/soldering at least 3 times at different intervals. 

Looking at the layout, there really is no "reasonable" place to drill and insert the 100uF electro on the board. I may just temporarily insert it directly on the DC jack for test purposes (if my current fixes do not pan out of course  )