9V Sample and Hold

[soggybag]

I built Tim's idiot wah on the breadbaord. And after a false start I got it working tonight. I sampled LDRs for the resistance to ground and used an LED connected to the sample and hold. I'm getting mild Sample and Hold from the 9V!

The effect at this point is pretty mild still. The range needs to be boosted. There is a lot of variation, LDR, LED drive out of the Sample and hold, noise going into the sample and hold. And then there's clock noise. I seem to be getting a pretty noticeable ticking. Definitely more work to done.

[nordine]

i suggest trying another wah, as the colorsound inductorless wah

it'll sound better and fuller (do the resonance mod), and that will encourage you to go more on

for the ticking, i'd stick a 100ohm+470uF on the power rail of the oscillator

try some high hfe transistors also

[soggybag]

Thanks for the reply. I had been thinking about the Colorsound Wah. I have it printed along with the Idiot Wah.

I did a few measurements and tests tonight. The sample and hold is working pretty well. The second op-amp buffering the output of the Sample and Hold puts out from about 1V to 5V. Which seems like a pretty good range to work with.

I also experimented with several LDRs. I tried a few random grab type LDRs and VTL5C3 which seems to be the best choice. It was also easier to test with. No need to fuss with bending leads to point it at the LED. I also don't need to turn off the lights to test it.

So far I can hear the effect but the range is very narrow making the effect sound very subtle.

The clock is still pretty loud. I get a tall narrow spike coming through the sample and hold on the scope. I suspect this is not the problem as the LED/LDR might smooth this out. I suspect the clock noise due more to the fact that everything is on the breadboard and clock ic is not isolated from the rest circuit. I'll have to try your power filtering ideas. A low power op-amp might be a good thing to test with, right now I'm using a 1458.

Lot's more work to do. But I'm pretty optimistic. Everything seems to be working pretty well so far.

[soggybag]

Just tried out the Colorsound wah with the Sample and Hold. Great results! Sounds pretty good, best this project has sounded yet. Almost as good as the FSH-1 I built.

I think the wah sweep could still use some tuning up. Seems to be on the high end of the sweep more than the low end. I'd like to get a little more of the low end into it. I think this would put the effect in a more useable range for guitar. Seems like the kind of thing you'd use strumming open chords.

Less clock noise also. I'm guessing the MPSA13 of the Idiot wah has more gain which picked up more of the clock. I'm using a 2N5088 now. There is a very faint ticking now. Everything is still on the bread board, the breadboard may be contributing to the noise somewhat. So I'm pretty optimistic that the clock can be negated with a careful PCB layout.

[Brian Marshall]

a low power clock and some power decoupling will usually do the trick.... and keep that clock completely sepparate.

if you use transistors in a common emitter type topology, they will likely bleed through any noise in the power supply... more so than a common collector amp anyways.

[niggez]

Is there any chance you could post some samples of what your sample and hold sounds like?

[nordine]

I think the wah sweep could still use some tuning up. Seems to be on the high end of the sweep more than the low end. I'd like to get a little more of the low end into it. I think this would put the effect in a more useable range for guitar. Seems like the kind of thing you'd use strumming open chords.

the "more on the high end" issue is an offset balance matter

try to get the voltage more to the negative side using an extra opamp, or altering an existing opamp on your circuit

try also geting some soundsamples  ..we are all droolling on expectation

[soggybag]

I will try and get some sound samples up. Maybe Friday or Saturday.

I'm guessing the low end of the filter sweep is at the high end of the resistance for the wah LDR. I was thinking I might be able to adjust it with a trimmer on the LED driving the LDR. As it is a I have the last op-amp from the S&H feeding the LED with a 1K to ground. I was thinking of adding a 10K trimmer in series with the LED.

I'm using a VTL5C3 right now. Another factor might be the speed of the LDR. I ordered some VTL5C9s from SmallBear. I think these have a faster response time. I'll test these out when they get here.

[Brian Marshall]

I think the wah sweep could still use some tuning up. Seems to be on the high end of the sweep more than the low end. I'd like to get a little more of the low end into it. I think this would put the effect in a more useable range for guitar. Seems like the kind of thing you'd use strumming open chords.

the "more on the high end" issue is an offset balance matter

try to get the voltage more to the negative side using an extra opamp, or altering an existing opamp on your circuit

try also geting some soundsamples  ..we are all droolling on expectation

yep, after re-reading his post, he said the output of the S/H was 1V to 5V.  It should be centered around half supply, so unless the battery is almost dead, it may be a good idea to try an opamp with tighter offset specs.

I've been using a lot of cmos opamps lately, especially TLC2262's.  I don't remember the exact spec, but it's much more accurate than the tl072.  Of course they are a bit more expensive.

[soggybag]

I'm using a power supply at home so the supply voltage should be about 9v.

I was thinking I could get the right range by adjusting the output from the S&H and the input to the S&H from the noise source using trimmers. This is how the FSH-1 works. I guess this might be a problem as the battery dies.

[nordine]

why don't you invert the CV signal feeding the LED?

[soggybag]

I don't think this will work. The last op-amp stage needs to have a high impedance input. In an inverting configuration the "sample" would leak out through the feedback resistor.

I think I can get it the filter happening in the right zone with some adjustment of the LED current. I think the range of the width of the sample range I think can probably be handled by adjusting the level of noise fed into the "sampler".

[Patrick from Davis]

Oddly enough I was trying to build something like this myself.  I started with this which is a dual 9 V battery operated version of his 12 V sample and hold design for his home made synth stuff here (second project down.)  I was going to use the Bipolar power schematic from tone pad to only use one battery, and then I was going to ask for help on stuffing it into a wah enclosure and using an LDR to take care of the 1 Meg pot that adjusts the sample rate. 

The Sample and Hold chip is here from analog devices.  They are very nice.  If you call them, they'll email you a form, and you can get some chips for prototyping.  My name and address got a little fudged from the communication between California to Thailand to Singapore, but it did show up.  As far as the Schmitt hex trigger, I found them at Mouser.

If anyone has some advice on getting an LDR to go from 10k to 1 M I'd appreciate some help there.  I don't know much about LDR's and need help figuring out the next step there.  But it is another sample and hold concept.  The samples that are on the Bug Brand site sound like a bit crusher, but have a very cool circuit bent sound to them.
Patrick from Davis

[soggybag]

Here's a link to a sound sample. I'm not very good with this recording stuff so I think the levels are a bit low. This is my guitar into the S&H on the breadboard into the Repeater (it's a loop sampler thing) then converted to mp3 and uploaded.

Edit: It seems this upload did not work for some reason. I'll try it again tomorrow.

[soggybag]

Here's a sound sample. I'm working on an updated schematic I'll post it soon. It is pretty much what is posted here with Color Sound wah tacked on. The wah is controlled by an LDR/LED which is connected to the output of the S&H.

http://www.super-freq.com/wp-content/uploads/2008/03/track2.mp3

[Ry]

Great work!  That sounds fantastic.  I will have to build one and add it to my two versions of the FSH-1.

Ry

[soggybag]

Thanks, I think this could sound better. I think the control voltage is not falling in the optimal zone to get the most out the effect. I want to add a couple pots so I can play with some of the parameters.

[Ry]

I agree that it could have a little more frequency range, but it sounds great.  What range of speeds can you get from it?

[Brian Marshall]

for more range try a vtl5c4.  i believe the vtl5c3's like to see a lot more current.

[soggybag]

I agree that it could have a little more frequency range, but it sounds great.  What range of speeds can you get from it?

The speed range is pretty good from about half a second to about 6 to 10 per second would be my guess. I'll post another sound sample adjusting the speed.