eliminating noise w an LM567

[baremountain]

So I spent some time this weekend playing with an LM567 IC, which is a Tone Decoder, basically used to modulate incoming signal, so (pls forgive me if this is nonsense) I'm assuming it is generally used as some kind of modulator for microprocessors and whatnot to decode signals on a frequency spectrum. It's super cool and can make some wild sounds, but there's a constant hum coming out of it at the chip's modulation frequency, and I'm wondering if there's some kind of analog way to filter that out. I've come up with a few extravagant solutions that involve using a second chip & dual gang pot to keep the frequency sweeps of both chips, then send the one w no guitar input to an inverting op amp and basically use phase cancellation to kill it, but that would complicate my switching options greatly & also I worry that'd kill the effect by cancelling out the modulation pitch.

I'm also wondering if this might be a problem with breadboarding certain ICs. I've had a lot of trouble getting certain designs to sound nice and noise-free on breadboards. Mainly when I use things like an NE555, PT2399, or the LM567 in this case. In fact, I had a 555 hooked up on the same breadboard and the oscillation pitch would carry over into the output audio just from sharing the same power and ground rails. Once I disconnected power it stopped. Is this normal for some/all breadboards or just the cumulative result of proximity of long exposed leads experiencing dynamic electromagnetic moments and all that good stuff? And is there a way to get rid of it?

[mathias]

Is it possible that your signal is bleeding to ground on your breadboard? Have you tried a different breadboard?

[crochambeau]

What sort of power supply are you using?

I don't think the LM567 (or the other ICs you mention as being problematic) provides common mode rejection like a bog-standard op-amp, so any filth on the power rails will wind up amplified.

[drolo]

I'm really not too familiar with that chip so I have no specific advice for that, but just in case it is bleeding from the chip through the power lanes, you could try to decouple the power from the chip and the rest of your circuit. adding a small resistor and a big cap for each lane (let's say 47R and 220uF). perhaps that will help.
But it could also be due to all the leads flying about unshielded. That you will only know once you box it up I'm afraid :-(

[baremountain]

Is it possible that your signal is bleeding to ground on your breadboard? Have you tried a different breadboard?

I'm not sure, I get plenty of signal out & no real loss along the way, so I'm not inclined to think that's the issue. I haven't tried a different breadboard, but that might be a good test.

What sort of power supply are you using?

I don't think the LM567 (or the other ICs you mention as being problematic) provides common mode rejection like a bog-standard op-amp, so any filth on the power rails will wind up amplified.

I'm using a One-Spot run thru a tester box I built.

I'm really not too familiar with that chip so I have no specific advice for that, but just in case it is bleeding from the chip through the power lanes, you could try to decouple the power from the chip and the rest of your circuit. adding a small resistor and a big cap for each lane (let's say 47R and 220uF). perhaps that will help.
But it could also be due to all the leads flying about unshielded. That you will only know once you box it up I'm afraid :-(

Yeah, I might try this though, just decoupling at the power pins first. I'll report back after I try that this afternoon.

[crochambeau]

One-Spot should run quiet.

I also just re-read where you referred to this "hum" following the pitch, which indicates it being carrier bleed... I blame lack of caffeine for my earlier comment, clearly not a power supply issue.

Isn't this a touch tone telephone part? Since those are actuated when a button is pressed they might not have worried about providing a way to trim out bleed on output. I don't think these were designed to support analog signal, so this might just be a feature.

Good luck with your experiments!

[baremountain]

Yeah that was my other thought. The 555 bleeding into the signal is what makes me think maybe it's coming through even stronger due to a my breadboard or something. I suppose I'll settle on a design and board it up to verify. Also for what it's worth, the ringing is much more noticeable when I use by Jaguar (single coils) vs my Les Paul (humbuckers).
I'm guessing you'd need some kind of digital frequency follower to be smart about noise reduction if that is indeed the case??

[crochambeau]

I'm guessing you'd need some kind of digital frequency follower to be smart about noise reduction if that is indeed the case??

It can be done with analog circuitry, but unless you want to throw thousands of hours at it, a digital solution of some sort would be the way to go.

[moid]

I'm not sure if this will help you, but a nice person on another forum recommended to me that I should swap NE555 for TLC555 because the NE555 has a problem with audio squealing / power that is resolved in the TLC555. They have identical pinouts so you can just swap them. The TLC (at least in the UK) are more expensive, but I'm going to try them soon as a hopeful fix to a breadboarded pedal I'm building at the moment. And if that doesn't work I may just swear a lot. In a manly fashion of course.

[culturejam]

I don't think you can totally stop-band reject the carrier signal on the LM567 in a way that follows the change of frequency as you adjust the carrier. Well, it's probably possible, but the solution would be much more complex than the rest of the circuit (I think).

[culturejam]

BTW, back in 2010, I did a circuit with the LMC567 (note the C), which was based on Tim Escobedo's Thing Modulator. It worked okay, and offered some noise/carrier rejection. Schematic is attached.

[baremountain]

Badass, I'll board this out and see how it sounds. Thanks!
Also I was searching differences between the LM567 and the LMC567 and found this, it might have some useful info about killing noise. http://www.diystompboxes.com/smfforum/i ... ic=67470.0

[false]

The datasheet makes it look like the VCO signal can be tapped off pin 6. You could check this with an oscilloscope. If so, invert it and mix it with the signal output to reduce the carrier. Similar to cancelling clock signals in BBDs.

[culturejam]

The datasheet makes it look like the VCO signal can be tapped off pin 6. You could check this with an oscilloscope. If so, invert it and mix it with the signal output to reduce the carrier. Similar to cancelling clock signals in BBDs.

That's a damn good idea!

[baremountain]

The datasheet makes it look like the VCO signal can be tapped off pin 6. You could check this with an oscilloscope. If so, invert it and mix it with the signal output to reduce the carrier. Similar to cancelling clock signals in BBDs.

That's a damn good idea!

Wouldn't that get rid of the signal modulation that I want to come through too though?