Introducing the "One Chip Chorus"!

[Taylor]

I think there is an error in the schematic. There should be another 10uf from the junction of the 22k and 10uf to pin 14, right? I think it's there in your vero layout, but those make my head spin. In the current schem there is no feedback path for the PSO.

Here's a sim of the LFO.

It's pretty solidly a sine, so I guess the strange method of modulating contributes to its non-linearity. I think the sound is fine the way it is, but out of curiosity did you try modulating the reference voltage with it after you discovered it wasn't necessary?

[anchovie]

I think there is an error in the schematic. There should be another 10uf from the junction of the 22k and 10uf to pin 14, right? I think it's there in your vero layout, but those make my head spin. In the current schem there is no feedback path for the PSO.

Thanks for spotting that!

Updated schem, courtesy of the wonders of MS Paint:


I've added the output cap, too. Scruffie: hope it isn't a big headache to update your layout.

[iq01221]

 :lol: Maybe if you draw and picture it, it'll be easier. Oh! the good times, when I was young there were no computers for everyone...

[anchovie]

I think the sound is fine the way it is, but out of curiosity did you try modulating the reference voltage with it after you discovered it wasn't necessary?

I didn't, as the Vref pin is the ground for the LFO. I originally had it set up to inject the LFO directly into pin 6. The happy accident occurred through removing that connection with the power still on and the guitar on my lap - "Hey, those pings and clunks still have a chorus effect on them!".

[frequencycentral]

While searching for something entirely unrelated just now I came across this post and thought it was quite funny  

I dare anyone to design a lower parts count chorus.    

You two better fight it out!  

Yeah, who would have thought it? Pure genius to use the built in opamps as a PSO.

Now, I dare anyone to design a lower parts count chorus than James's. 

[Perrow]

.

While searching for something entirely unrelated just now I came across this post and thought it was quite funny  

I dare anyone to design a lower parts count chorus.    

You two better fight it out!  

Yeah, who would have thought it? Pure genius to use the built in opamps as a PSO.

Now, I dare anyone to design a lower parts count chorus than James's. 

Ok, I'll swap those two 10k resistors with a 20k trim pot set to its middle position, and the 22k+33k combo with a 50k trim pot

[Taylor]

.

Now, I dare anyone to design a lower parts count chorus than James's. 

Ok, I'll swap those two 10k resistors with a 20k trim pot set to its middle position, and the 22k+33k combo with a 50k trim pot

Eh, don't bother. I think you'll find that Bach already designed a 5-part chorus quite some time ago.

[Perrow]

Eh, don't bother. I think you'll find that Bach already designed a 5-part chorus quite some time ago.

I've heard of a one part chorus, a chorus line to be exact

[anchovie]

Ok, I'll swap those two 10k resistors with a 20k trim pot set to its middle position, and the 22k+33k combo with a 50k trim pot

...and I'll assume that everyone's using a well-regulated 5V supply and that the polarity cannot be changed, losing the 1N914, regulator and 100uF! 

The 20k pot idea is one that I was going to suggest as a possible mod to give a wet/dry balance, or as Taylor suggested you can replace the dry mix resistor with a switch to get vibrato. The 22K resistor spot does need to contain some value as if you just have a pot and set it to zero ohms, the LFO freezes and can only be restarted by backing off the pot and cycling power. Simulation suggested at least 100 ohms but I don't know what the true minimum is in practice - I settled on 22K as that gave me the most depth without getting distortion on each LFO pulse (note: noise musicians may want to exploit this "rhythmic screech" feature!).

Some other notes:

• The 10uF caps from pins 7 & 8 are obviously way higher than the datasheet values; I found in this circuit that they cut out a lot of background hiss.
• The 10uF from pin 2 can be increased to lower the LFO rate range (maybe set this to 100uF and have a switch to add 10uF in series?)
• If you don't mind a bit of background noise, the 100nF coming from the 4.7K resistor can be lowered for a brighter wet sound.

[newfish]

Just when you think you're 'done' building for a while...

I can see this going on my breadboard this evening.
...then probably getting built 

Thanks for sharing your ideas!

[~arph]

Wow nice work. Cool building block. PSO's are tricky things. Make sure that the LFO doesn't fade out after running for an hour.

[anchovie]

Make sure that the LFO doesn't fade out after running for an hour.

I accidently left the power on for 20 hours and it was still chorussing when I came back to it!

[~arph]

Very good!

[Renegadrian]

Nice project!!! the OCC!!! 

[~arph]

Yeah I'm down wit it. 

[Scruffie]

I've added the output cap, too. Scruffie: hope it isn't a big headache to update your layout.

A little, but I got there in the end, i'm not so keen on where I had to put it but ah well, aslong as it works!

Here's the new PCB & SS then





49.53 x 49.53mm

[Kearns892]

I guess Scruffie beat me to it, but I was almost done with mine, so I thought I would go ahead, finish it up, and post it. It "should" fit in a 1590A based on the dimensions on the hammond site (I'm not crazy enough to fool with the A size enclosures, so I can't say for sure). All the electros should lie flat.



Transfer is 1.3 in x 2.1 in (33.0mm x 53.3mm) 
Unverified as of yet

[Earthscum]

"Lush" knob could lose a resistor, I think...

(at 10u-10u, remove the 22k, 33k, and pot) From the 10u-10u, take a 50k pot, CCW lug, and a 39k resistor. Tie the resistor's other end to the wiper of the pot, and tie the CW lug to the VR rail. 50k and 39k= about 22k (just under). So, all the way down, you have 50k, which is just a little more than 22k+(33k and 100k)=24k, and all the way up you are sitting at just under 22k.

[Thomeeque]

22k+(33k and 100k)=24k

You have just forgot to add those 22k here, it should be 46k (just for the record, it's obviously "typo").. T.

[Thomeeque]

 Congrats James to this design, brilliant idea and it sounds really sweat!

Would you mind to make short sample with heavy distorted guitar, please?

Btw. thanks to additional 4k7 (output LPF) on the wet path and R11=R12=10k you don't get exact 1:1 dry/wet mix (wet is cca 3.3dB weaker), is it on purpose?

Cheers, T.