Bender roll call

[Kipper4]

So I'm looking to do some pitch bending with an lfo via vactrol.
So far I'm not having much luck with pt2399.
Come one you benders throw me a line.
Cheers

[Digital Larry]

Well you need to have some sort of clockable delay line, which as far as I know the PT2399 is.  Then you need to be able to tweak the clock frequency such that samples get read out at a different frequency they were read in at.  However, this arrangement has to have the clock frequency going up and down, because the difference between read-in and read-out sampling rate is what makes the pitch bend.  Also it is not practical to design an oscillator whose frequency keeps going up at the same rate forever.  So you get a vibrato, aka a chorus with no dry signal mixed in.

A DSP such as the FV-1 does pitch shifting by reading the audio bytes into a circular buffer at one rate, then reading them out faster or slower (which really means reading one sample per sample period, but having the interval be fractional - so the chip interpolates between adjacent samples to make this happen).

Are you looking for a constant pitch shift or a vibrato? 

[ElectricDruid]

No luck with the PT2399, Kipper? Let's have a look what you've got so far. There can't be much wrong with it.

Does it do anything at all? PT is fussy about its start-up conditions, as we know. It doesn't like to have a small resistance from pin 6 to ground initially. Problem is described here, about half way down:

http://sound.whsites.net/project26a.htm

There's plenty about it on other threads here too. (How did Freq Central's Little Angel deal with that? I don't remember.)

If you're going for pitch shift, you want the shortest delay you can (probably 20-30msecs with PT2399), which puts you right in the area where it'll start making a fuss about start-up.

All this makes me think I should design a voltage controlled BBD clock driver. I mean, the MN3102 can have some voltage control bolted on, more or less, but it's a bit clunky. Better would be a nice clean chip with 25-500KHz clock outputs and a 0-5V CV input. The FLANGE chip with the Depth CV at zero is almost there, but I could probably do better if the chip was doing that one job alone (would be able to sample the CV at a higher rate and so on).

Tom

[Kipper4]

I have this on the BB.

Needs some work granted, It's the early hours so please excuse the goofy scheme. There may be some mistakes.

I think it may need some smaller caps for the 10nf's and 1nf's. Because I think I'm hearing too much of the lows in the pt2399 output.
Or maybe some more filtering elsewhere.




I'll have a more thorough read of the replys when I awake and I can take it in better
I appreciate it and look forward to the read.

Cheers
Rich

[Kipper4]

Wow Larry thanks.
That's kinda a bit over my head a little. Thankfully you did a good job of explaining it.
I never even considered using the fv1 in combination with the Stompflo lfo.
I'd better not even get started with that yet. My minds already going crazy enough.
Although I do have an fv1 from another project in the parts bin.
Cheers
Rich

[Kipper4]

Tom nice helpful article. I don't know why I love Rods work.
His works make good reference point.

Reading through it. I'm sure it will help me tune this up to something useful.

I did swap the pin6 R to 1k just before bed and it's looking promising. I plumped for the 100R after looking at the little angel. But I can see that will need some tweaking.

What am I going for.
Ideally a nice lush chorus that's simple to build.
Given the number of different waves available with the stompflo. It's not going to be that simple unless I restrict the number and type of wave forms in use.
And why would I want to restrict the waves. They're there so why not use them.
Different is not always bad right?

I'm babbling now. I'll leave at that for now.
I can't wait to finish work and carry on with this.
Cheers

[Kipper4]

I couldnt resist a little tweek or two before work.
Now I'm late.
More coffee.................

[ElectricDruid]

Very interesting. I see you've gone for the Little Angel trick of modulating the Ref voltage, rather than hanging a vactrol off pin 6.

Not all the waveforms are necessarily appropriate for a chorus, but that Random Slopes waveform should be good, plus triangle and/or sine for something more normal, and maybe the Sweep for something a bit weirder. I can't imagine the ramps or square work that well. It's easy enough to connect up a chain of resistors for the Waveform control, and then use an n-way switch (where n is however many waveforms you want) to select the required points.
Example in the TAPLFO datasheet, page 9:

http://www.electricdruid.net/datasheets/TAPLFO3Datasheet.pdf

HTH,
Tom

[Kipper4]

I started off with a vactrol approach.
I've yet to experiment more with that.
My mother said I was a deviant and she wasn't far off.

It is interesting. I've tweaked some values. More later.
I go be deviant.

Thanks for the link.

[Kipper4]

Since it's really low parts. I'm thinking of adding an anti latch up.
Anyone got any experiences with these. Worth it.
Please shout up.
What works and what's not so good.

Rod Elliot has one such on the Link Electric Druid posted.

http://sound.whsites.net/project26a.htm



maybe this

http://www.valvewizard.co.uk/jennygreenteeth.html


Thanks
Rich

[ElectricDruid]

I don't know why I love Rods work.
His works make good reference point.

Because they're always excellent stuff is why! They do indeed make a good reference point.

Rod Elliot's articles are well-informed, clearly explained, and full of useful details. We should all aspire to write such good stuff! I'm a massive fan, as you can probably tell.

Tom