Digitally Controlled Analog Tremolo?

[bigjonny]

Anybody tried it?  I am thinking something like the EA tremolo, but with a digital oscillation section that controls the FET.  The main advantage would be something akin to the digital delay pedals with the "Learn Tempo" buttons on them — tap the button twice, and you've set your new delay time.  Well on the digitally controlled tremolo, you'd tap the button twice and set the wavelength of the LFO, so if your band changes tempo, it's easy to make the trem catch up.  Also, seems like it'd be easier to toggle between Square, Triangle and Sine oscillation?

I saw a post that discussed digital oscillators a bit, here, but no infomation on how to make a digital circuit.

Anybody have any online clues to digital circuit design?  I understand programming and digital processing, but have never programmed a CMOS, for instance.

[travissk]

Depending on how many features you want to add, a microcontroller might be a good option.

[idlefaction]

yeah, you can get some really cheap micro's these days.  the Atmel AVR's have In System Programming which you can do via your parallel port with a cheap DIY cable, suck very little power and are cheap as chips, they'd be a good bet;  they have PWM output also so with a bit of effort into learning how to code on one you could do it pretty easy really.

AVR, clock crystal and caps, LPF for the PWM output (LFO signal), momentary switch attached to one input, there ya go!  maybe even an analogue in so you can select between output waveforms like you were saying.  they come in 8-pin DIP as well.  good luck!  :-)

mmm look at all them TLA's!!  :-D

[ExpAnonColin]

It also might be good for making the LFO more amplitude stable, since lots of analog LFOs amplitudes vary.

Where's that DSP LFO, peter?!

-Colin

[austin]

I've been designing one in my spare time over the past two years.  I originally went with a smaller AVR, then upgraded to the Mega8 AVR.  If I were starting over, I'd probably go with one of the texas instruments MSP430 series or the new DSPic.  The chip is pretty much maxed out as is in terms of free clock cycles.  If at all possible, you probably want something that can do math.  There have been so many cases where I wish I could do division and had to think up some way to get around it.  If you have the time, sign up for UCSD's embedded systems courses.  I took real-time programming and it was extremely helpful.

[R.G.]

Easy.

Microchips PIC or equivalent.
Read the speed pot
Read the waveform switch
Read the tap tempo switch
On regular fractions of the LFO waveform, look up the nominal output in a table, spit it out to the PWM output.

If digital step jaggies are a problem, use a pin to modulate a smoothing filter frequency.

[bwanasonic]

Which reminds me, I keep meaning to go to GC and check out some those new Line 6 stompboxes. I read the reviews at Jack Orman's site, and it made me curious. The trem pedal looks interesting. I like my Supa-Trem a lot, but tap tempo would be a useful feature.

Kerry M

[gez]

If you're going to go to all that trouble to do a digi LFO, at least have the courtesy to use an OTA plus op-amp buffer(s) for the audio, instead of the bleedin' EA trem set up (you'll get a more linear representation of your LFO's waveforms)!  

[StephenGiles]

gez said:

instead of the bleedin' EA trem set up

..reminds me of the rescue worker talking to a victim of a road accident somewhere in Essex....."where are you bleeding from" he said to the girl.....she replied...."Hornchurch if you must know, wossat go''a do wiv you?"

Stephen

[Chico]

I do not have a digital Trem per se, but I did build a digital LFO from an ATMEL AVR chip.  I use it all the time for prototyping.  

It has two knobs, one for voltage - current range 0-5 volts, and 1 for frequency, current range is 0.1hz - 32 hz.

There are four outputs that are simultaneously available, each may be selected from a list of wave shapes including sin, tri, saw, ramp, rev ramp, hypertriangular (parabolic) square, plus a hand full of custom wave forms.

The outputs are also available in quadrature.

http://www.circuitmuse.com/circuitmuse_006.htm

Finally, it includes a 16x2 LCD display so that I know what the thing is doing.

The great thing about the AVR stuff, is that it is relatively cheap for one off DIY stuff, reprogrammable, there are resources galore on the web, and few extra components are required to get the things up and running.

Since this is a desktop unit and not in a stomper (I have one of those in the works), I decided to use a quad 8-bit DAC in lieu of PWM.
Best of luck on your project.

Tom

[gez]

..reminds me of the rescue worker talking to a victim of a road accident somewhere in Essex....."where are you bleeding from" he said to the girl.....she replied...."Hornchurch if you must know, wossat go''a do wiv you?"

Must have been all the re-runs of 'Budgie' that Ch4 screened in the early 80s, I seem to have used it all the bleedin' time (minus the ..'ell Hazel) since!

[bigjonny]

Hey wow!  Well, lots of great responses â€" definitely have some learning to do!  I'll update what I decide, but at first glance a variant of Chico's project looks pretty appealing.  R.G.'s suggestion also seems interesting (and simple) â€" I'll have to decide which I care about more: flexibility in use, or simplicity of build.  Getting the perfect osciallation waveform doesn't interest me so much.  There's a lot to to be said for controlled imperfection in audio!  I just want that durn Tap Tempo function and waveforms that are close-enough to Square, Sine and Triangle.

That said...

...have the courtesy to use an OTA plus op-amp buffer(s) for the audio...

gez, I will definitely consider this â€" I just picked the EA, cuz I actually understand the whole circuit.

If you have the time, sign up for UCSD's embedded systems courses.

I'll have to look into that since I do work at UCSD!  But, â€"sighâ€" no time, for now...

Finally, to avoid any questions about the jargon delved out thus far:
PWM: Pulse Width Modulation
DAC: Digital to Anolog Converter
LFO: Low Frequency Oscillator
LCD: Liquid Crystal Display
DSP: Digital Signal Processing
FET: Field Effect Transistor
DIY: Do-It-Yourself  

[Ge_Whiz]

(you'll get a more linear representation of your LFO's waveforms)!  

Linear is good? Isn't that the opposite of "mojo"?  :wink:

[bigjonny]

Thought I'd update this thread.

I was reading in another thread about how Joe Davisson's site is down, which was a pity as I was curious about his Vulcan distortion pedal.  Joe responded in that thread, and mentioned it was fine to email his schematics around, but to please not post them on any websites.  Well, I took Gringo up on his kind offer to send anyone his copies of the Davisson schems and Gringo kindly sent all he had (thanks again, Gringo).  I was browsing through them last night...

Turns out Joe had the "digitally generated LFO" idea for tremolo quite some time ago, and designed a pedal around it: the MOS-Vibe.  It even has three selectable waveforms that are similar to: Square, Sawtooth, and Sine.  I searched this site, and there are very few entries about it, and no build reports.  Anyone build it?

PM me, and I can send the schem to any who are interested.

BTW, Joe, if you read this, my compliments.  You have some pretty swell looking designs.  I only recently became aware of your work for some reason.  Can't wait to try some of 'em out!

[Paul Perry (Frostwave)]

If you're going to go to all that trouble to do a digi LFO, at least have the courtesy to use an OTA plus op-amp buffer(s) for the audio, instead of the bleedin' EA trem set up (you'll get a more linear representation of your LFO's waveforms)!  

Come on Gez, if you have done all that programming dont waste it by putting OTA noise in! use a PWM output to drive athe diode in a vactrol to control the audio. Nice & quiet.. :wink:

[Paul Perry (Frostwave)]

PWM: Pulse Wave Modulation
[

err......... Pulse WIDTH modulation? surely.

[RickL]

I've built the MOS-Vibe. It was about 50 stompboxes ago so I don't remember much about it and I won't be home for a couple of days so I can't even go dig it out of whatever storage box it ended up in.

What I can confirm is that it works, the three waveforms give distinctly different trem sounds and I don't remember specific problems (i.e. no horrible ticking). If you're still interested I'll see if I can find it when I get home and give you a more detailed description.

Rick

[bigjonny]

PWM: Pulse Wave Modulation
[

err......... Pulse WIDTH modulation? surely.

Oopsie.  Thanks Paul Perry!  I edited the post where I wrote that (to avoid confusion).

[bigjonny]

If you're still interested I'll see if I can find it when I get home and give you a more detailed description.

Regardless of what I want, I think that'd be a good contribution to this site, since there seems to be no build reports on the MOS-Vibe, here yet.

Based on what you said so far, I'll probably build one anyhow, but I actually would appreciate your description/opinion, RickL!  Thanks for the offer.