5 Knob Aliaser Project - Verified.

[nelson]

Hi folks,

I updated my site with an Aliaser project. It's a modified Crazy Larry which is TTG's design. I changed the envelope detector and added a full wet/dry blend control.


It's verified and working.


http://electroconducive.googlepages.com/wolfbagger-aliaserproject



I also have a few left over Pre Tinned PCB's for this project if anyone wants one just £4.50 + shipping.

[orangetones]

Nice looking work there.  Do you have sound samples?

[nelson]

If you do a search for crazy larry, I'm sure Andrew uploaded some.

[The Tone God]

Great stuff.

That looks like variegated gold leaf which is a favourite of mine. I'm kind of honoured to have an effect with that type of finish. Also the blend idea is kind of spiffy. I was planning on building another aliaser so I may have to give that try.

Andrew

[Gila_Crisis]

really cool progect
thank you!

[nelson]

Great stuff.

That looks like variegated gold leaf which is a favourite of mine. I'm kind of honoured to have an effect with that type of finish. Also the blend idea is kind of spiffy. I was planning on building another aliaser so I may have to give that try.

Andrew

You're quite correct, it's gilded in variegated gold leaf.

The blend function is nice it allows the aliasing to add far more subtle atonal ghost tones to the guitar sound.

really cool progect
thank you!

You're welcome.

I'd sure love to see any build reports based on this project.

Thanks for the kind words guys.

[The Tone God]

You're quite correct, it's gilded in variegated gold leaf.

The blend function is nice it allows the aliasing to add far more subtle atonal ghost tones to the guitar sound.

Wow real gold leaf. Now I'm even more impressed. I love the look of it.

I have a couple of boxes lying around so I thought I would throw an aliaser into one. I thought about adding a mix because on some occasions you just want that slight change.

Andrew

[nelson]

So, I got 5 boards made up for this project, as I hate etching and drilling my own boards. I've now built 3 of them.

Something about the PWM and triangle LFO feeding it didn't sit right with me. You have to "tune" the PWM control to get aliasing over the entire frequency range. Initially I thought this was just because the signal spent more time off than on.

So, then I decided to look at why and improve the oscillator.

No offence TTG, I just think there is a more intuitive way for the width control to be utilised over its entire range. As it stands it has such a short useable range in the center of its rotation.

The culprit is a result of where you've decided to place the variable resistor to change the frequency of the oscillator - from 50hz to 2.5Khz.

As the frequency of the triangle wave changes in the schmidt trigger + integrator oscillator feeding the comparator the peak to peak voltage of the oscillator changes.   

This has an effect on the comparator, resulting in a constant high or low output at many settings of the width pot. This means you constantly have to tweak the width control to get an effect, it makes the width and frequency pots too interactive, and results in a counter intuitive user interface.

I have a LFO design tool I made in open office, plug the values in and you'll see what I mean.

http://electroconducive.googlepages.com/SimpleLFODesign.ods

I'm going to use the designators in the file.

The amplitude of the triangle wave is relative to the ratio of R2/R3.

So, a simple fix is to move the 50K variable resistor to position R1, in series with a 1K resistor and replace R3 with an 82K resistor and R2 with a 100K resistor, leaving C1 as a 100nF cap. The frequency range is the same, but the amplitude of the triangle feeding the comparator is constant.  This (in theory) should result in a wider useable range on the width pot and should make the pots far less interactive.

So, to those that have built this, I suggest you try these mods.

Report back and let me know how it turns out.

I'm abandoning the design in favour of a voltage controlled clock. I think that aliasing is far more interesting when swept.

If anyone wants the remaining two boards I have they're yours for the price of postage.

[gigimarga]

Hello,

Can you explain your mods using the Wolf Bagger's schematic as reference?

Thx a lot!

[nelson]

Hello,

Can you explain your mods using the Wolf Bagger's schematic as reference?

Thx a lot!

Remove the 50K frequency pot, replace with an 82K resistor, replace R20 with the 50K frequency pot wired in series with a 1K resistor. This gives a sampling frequency of 59.06hz to 3012.05hz with a constant peak to peak voltage of 7.38v.

[gigimarga]

Thx a lot! I will try tonight and, after, i will post the results.

[gigimarga]

I made the mod, but it didn't work right: now i have a low-level mosquito like fuzz instead of the aliaser...maybe i made a fault...i will put the things back and i will see what will happen.

[nelson]

I did a spice sim on the clock for the aliaser.


So, here are the values that work in the sim and produce a clocking signal to the S+H circuit at all settings if the freq and width pots, at roughly the same range of frequencies as the original clocking circuit.

This should result in aliasing at all settings of the pots.

R20 = 100K pot + 1K resistor in series.

Freq pot = 82K resistor.

This gives a clocking signal range of 31hz to over 2.2Khz.

The original circuit had a range of 51hz to 2Khz.

All other values are the same.

To increase the frequency, increase the value of C12.

Make sure you use a Rail to Rail quad opamp!

[Andi]

D'you still have one of those boards left? I have a 5-pot enclosure waiting for something to go into it...

[nelson]

D'you still have one of those boards left? I have a 5-pot enclosure waiting for something to go into it...

Sorry Andi, they're all gone.

I could give you the eagle files and you could edit, to get it to work in your router?

[Andi]

That'd be fantastic if you didn't mind - it'd just be for me, of course, not to sell.

[nelson]

That'd be fantastic if you didn't mind - it'd just be for me, of course, not to sell.

Yeah, I'm selling two of these with the mods.

Then I have a voltage controlled clock version I'm designing.

Fun fun fun.

Email on its way.

[Andi]

You're a gent. Cheers.

[rocklander]

do you have a n00b's guide to setting the threshold pot? I've started building this, but not sure if I should be setting the threshold like I was biasing a FET, or there's something else I should be doing? tia

[tlaubs]

Hello,

Can you explain your mods using the Wolf Bagger's schematic as reference?

Thx a lot!

Remove the 50K frequency pot, replace with an 82K resistor, replace R20 with the 50K frequency pot wired in series with a 1K resistor. This gives a sampling frequency of 59.06hz to 3012.05hz with a constant peak to peak voltage of 7.38v.

Sorry to revive a zombie thread but was looking for a little help with this mod. To do this, do I just put an 82k resistor where the freq pot lug 1 and 3 was. Then replace the r20 with the 50k pot with lug 1 going to one of the resistor holes, lug 2 of pot going to the 1k resistor which then goes to the other hole of r20?

Thanks