GGG EQ build hiss problem

[oskar]

What are the two inputs getting?   Well, one is getting the original signal.  The other is getting the original signal with selected frequencies attenuated at varying degrees, depending on how the pots are set.  So what gets amplified?  Well, the differences do...  including, and especially any and all noise introduced by the attenuator stages. In effect, this thing is summing up and then amplifying 5 op-amps + resistors + caps + RFI + EMI worth of noise...      Instead of this noise being canceled/controlled via feedback, it is what is getting amplified -- along with the portions of the original that are different.  The actual EQ circuit attenuates, but because of the differencing of the last stage, it effectively gives the attenuated stuff a boost... (it is even marked "boost" on the circuit).  Unfortunately, any and all noise that wasn't in the original signal also gets a boost...

If I've got the analysis correct (and please forgive me if I've got it wrong -- it has been a l o n g day), it may be quite difficult to de-noise this circuit without some design changes.

When you sum up the outputs from the 5 filter-sections, the thermic/white noise should actually be slightly suppressed... (I think, kind of... )
This is because the noise part from the OP-outputs is random and won't add up levels straight away... ( bla, bla, bla, never made it this
far in school, but I think this is where someone will stick 1/2^-1 up your nose).       

[PerroGrande]

I've truthfully never had 1/2^-1 stuck up my nose... I think this is a good thing.   

[Dave Simpson]

First of all I would like to thank all you guys for your opinions and ideas. I appreciate you taking the time to try and help me with this project. You guys are much smarter than me!

Oskar, I don't think it's a problem with the IC's. I built two of these pedals and the both have the same hiss problem. I tried putting the 220k resistor to ground and the hiss is still there. I'll take the 10uF capacitor out of the gain section again and see if it helps. I still haven't got around to trying the decoupling thing you mentioned yet, hopefully soon.

[oskar]

Oskar, I don't think it's a problem with the IC's.

Me neither, but I thought swaping IC's was so simple, clean and fast, that it was worth giving a try... but not when two circuits are having exactly the
same problem.

I built two of these pedals and the both have the same hiss problem. I tried putting the 220k resistor to ground and the hiss is still there. I'll take the 10uF capacitor out of the gain section again and see if it helps. I still haven't got around to trying the decoupling thing you mentioned yet, hopefully soon.

Right, you built two of them, I forgot that. Also Kevin B. had similar problems... The only problem I see from the voltages you provided is the not to symmetric bipolar voltage. This however is beyond my abilities...          perhaps it is perfectly well, I don't know...
I do know however that the remedy for it is lower resistance in the divider and bigger caps. Personally I would probably use the spare OP and do a little
cutting to get a more ( in swedish: matningsspänning...      )   V+/V- symmetric around ground, but... Kevin had the hiss thing going on and he used a bipolar supply...
A resistor/wire to ground from the nonused OP is clearly the way to go but I don't think it wouldn't cure much hiss. What could happen if you don't stabilize an unused OP is that it could start oscillate and draw current ( current off course generates hiss ).

Two things strike me:
1. Shielding. Did you actually build these into a box? I should've asked this way up in the thread... I've seen somewhere in another thread hiss
disappearing after building the thing into a box.
2. Current, It seems like it actually is drawing quite a bit of current. Could you meassure this?
Actually I think current could be the problem here. TL064 should be more noisy on the paper but it draws less current. You could try and swap for a
couple of these.

some earlier threads on noise...
http://www.diystompboxes.com/smfforum/index.php?topic=57427.0
http://www.diystompboxes.com/smfforum/index.php?topic=59016.0
http://www.diystompboxes.com/smfforum/index.php?topic=51583.0

[Dean Hazelwanter]

Sorry for not replying sooner. I've been taking time off and we've got a new high school coming online in less than a month, so life is hectic.

This is based on a 10 band EQ project from Oct 77 ETI magazine, that used 4136 opamps. Changes made:
- 4136 quad opamps to TL074
- 10 bands to 6 (JD rounded off the marginally useful 1st band to make it 5)
- Because there was way too much gain in the prototype, removed the 1M fixed resistor that was in their schematic from IC1 pins 6-7, changed the first resistor (between the input and IC1 pin6) from 47k to 100K, and the feedback loop resistor around IC1b from 15k to 10k.
- Changed the resistor in parallel with the output from 1M to 10K.

One other silly question: For the series resistor between IC2c pin 8 and the output, you are using a 100R (ohm) not 100K (kilo ohm), right? That one has bitten others...

Hopefully that helps. I sure don't recall major hiss problems with this (or I wouldn't have released it), but I sent the project off to JD about a year and a half ago.

[Dave Simpson]

Dean, it's not a major problem. I'm just trying to get the hiss level down to about the same level as my Boss EQ's. I know with any EQ there's always going to be a little noise no matter what. The only thing I'm having a hard time getting my head around is that even with everything turned down to zero there's still hiss coming through when the pedals on. Other than the hiss I love the sound of the EQ, blows away the Boss GE-7.

I did the 100 ohm resistor mistake with the first EQ I built. So for sure it was right after that on both pedals.

Oskar, yes they are both in enclosures.

[kvb]

I put in some tlo84 ops - there is no difference in the noise level.
the 84s are now on a breadboard.
three weeks ago, I breadboarded a partial circuit of this EQ but it did not work - so I did something else.

I am going to try again

I will experiment with larger pot values