Anderton Tremolo Help please?

[MarkDonMel]

Hey all,

I finished the anderton tremolo last night using the resources on GGG.  I couldn't test it out though because it was too late, so instead I checked continuity and stuff on all the solder points and connections, you know just kind of tinkering because I was itching to play it so bad, but it was too late to play it... gotta time these builds better.

Anyways, plugged it in just now on my lunch break, and it sounds ok, but there are two problems I was hoping someone could help me troubleshoot.

1.  The intesity is really smooth and nice sounding on one end of the pot turn, but turning the pot the other way doesn't only increase the intensity (which I know is normal and desired), but makes the tremolo effect uneven, like a little more silence than signal going through.  I guess each cycle contains a longer length of silence than of guitar string vibration.

2.  I have read about this before, but there is a soft clicking on each cycle.  It isn't harsh, when I first turned the box on, it reminded me of the very beginning of immigrant song... I think that's the one, almost sounds like a soft "pat" noise.  No static or noise just that "patting" sound.

Any advice where to begin, or anyone had the same problems?

Thanks much, Matt.

[MarkDonMel]

Please?

[MarkDonMel]

Ok, well thanks just the same.  :wink:

Matt.

[Peter Snowberg]

Sometimes an answer just takes a couple days to get, especially when it's a three day weekend with Valentines Day in there.

Hold tight, I'm sure somebody will pipe up. I've never built one of those so I can't be of much help.

Take care,
-Peter

[fishdds]

I'm building one of these right now, but I've got some other projects going on also.  I'm ready to etch the board.  Had to fiddle around with the transfer image to get it the right size upon printing.  32% was the magic number.  Acrobat makes things more consistent than a gif image like this.  Oh well, I'm interested in your questions, because you know Murphy's Law, so I'll probably run into these problems as well.  I'm just not as far along as you are yet.  Do you know if I can use the Clairex CLM6000 from Small Bear instead of the CLM600 optocoupler?  I can't find that part anywhere.

[MarkDonMel]

I think they are the same part.

I bought all the parts for this project a few years ago after I got anderton's book.  It was too difficult for me back then so I never built it.  I have a few of those clm6000's and looking back I wish I had bought more, oh well.

Thanks, Matt.

[Jason D]

I built an Anderton Trem, a few weeks ago. And, I have ran into several problems. Most of which I am thinking have to do with the fact that I am powering it with a MAX1044 to get the -9v/+9v. I haven't noticed the uneveniness in the sound when turning the depth control. Put if I turn my volume level up on the circuit I get the ticking. But, then again I also up'ed the gain on the opamp to make sure there was no volume loss.

For me the most annoying part is that I get the ticking when the pedal is bypassed! It goes away if I turn down the volume knob on my guitar. How strange?!

Also I used a Vactec VTL5C3 for the optocoupler. And pcb layout was my own design to make it as small as possible using smd for the TL072, and the 4049 chips.

[gez]

Could be stray capacitance.  Have you tried using shielded wire for all the audio connections?  Sometimes this is all that's needed.

[gez]

1.  The intesity is really smooth and nice sounding on one end of the pot turn, but turning the pot the other way doesn't only increase the intensity (which I know is normal and desired), but makes the tremolo effect uneven, like a little more silence than signal going through.  I guess each cycle contains a longer length of silence than of guitar string vibration.

I can't see a intensity pot on the schematic I'm looking at.  Do you mean the frequency pot?

Assuming you've got the thing wired up ok, and you DO mean the frequency pot, then it could be a simple matter of adjusting the mark-space ratio.  It might be that you're noticing it one end of the pot more than the other because of the faster speed making the problem more pronounced/noticable?  Could you give a little more detail please?

One of the drawbacks with using CMOS for this application is that you can get a slightly uneven mark-space ratio.  This gives more on time than off (or the other way round) so to speak.  Before you try the following, pop in another chip to see if it resolves the matter.  

A simple fix would be to replace the 2k2 "R8" with a 1k2 resistor, disconnect the pot (R9) attached to it and wire up a 2k trimpot as follows:  wiper connects to 1k2‘s free end, each outer lug connects to a diode (facing different directions - doesn’t matter which) then the diodes combine and attach to the 50k pot.

Set trim control in middle and turn (slightly!) till you get a 50:50 mark space ratio.  At 9V the inclusion of the diodes shouldn’t screw with the frequency range too much.

PS I has to put R8 in inverted commas as if you do it in brackets it turns out as (R8)

[gez]

Had to pause for supper!

PS The trim will only affect the mark space when the LFO is at the faster end of its range (and it should be set for this when you adjust the trim),  but if this is where you have the problem it should iron out the difference.

When I said try replacing the chip, I meant one from a different manufacturer if you can.

One thing I’ve noticed with CMOS triangle LFOs is a slight glitch when the Schmidt changes states.  The triangle ramps up and then at the top of it’s peak, as the Schmidt changes over, there’s a slight plummet downwards before the output gradually slopes back downwards.  I’ve never had a problem with this, but the difference in amplitude between each half of the triangle might be creating the uneven mark-space ratio.  

The cause of this MIGHT be the fact that CMOS inverters are pretty low gain amplifiers (well, at 9V at any rate) so the feedback might not be strong enough to hold the input of the integrator steady when the Schmidtt switches over.  Perhaps including a huge value resistor in series (edit: I meant PARALLEL!  :oops: ) with C3 (10u)  would help to keep the amp bias stable.  10Meg is a value I’ve seen when this is done, but I can’t help feeling the value would have to be a little smaller than this.  You could try croc-clipping different resistors in parallel with C3 (in the meg range) to see if this helps (this is a long shot mind!).

[The Tone God]

I played with one of these years ago and ran into the same problem, ticking. One of the things I found was that some of the ticking was coming through the speed control wiring into the audio. Being lazy I designed a board that seperated the LFO from the rest of the circuit, put a ground plane all around it, then interfaced the speed control through an opto-coupler. That got rid of the ticking.

Andrew

[Jason D]

I played with one of these years ago and ran into the same problem, ticking. One of the things I found was that some of the ticking was coming through the speed control wiring into the audio. Being lazy I designed a board that seperated the LFO from the rest of the circuit, put a ground plane all around it, then interfaced the speed control through an opto-coupler. That got rid of the ticking.

Andrew

I thought about trying the sheilded cable idea. But, then I saw the Tremolo LEDs from pisotones I thought I might breadboard it before working on the Anderton more. Making a seperate board for the oscillation seems like a great idea. Also I thought about seeing if there was a way to get +/-9v from 18v power supply. Maybe use a dual opamp, 1 side inverting to get -9v, the other side noninverting to get +9v. Would that work? Or, would that just cause more problems?

[Chico]

I built one of these a while back.  It is a pretty cool tremolo.  

This device is sort of prone to ticking unless you are careful.  Mine does not have the ticking problem you speak of, but I did note it when working thru the circuit.  You can design it out with a little experimentation and patience.

When you do the layout, you need to be careful to keep the LFO components close together and as separated from the audio as possible.

Double check that you do not have the triangle and square signals that couple to the spdt switch close to your inputs and outputs.

Also, check to make sure you have at least a 1k resistor between the power and pin 1 of the CMOS chip.  I think that I used a 1.5k.  If you are at your wits end, play with this resistor.

Also, there is not alot of gain in the audio part of this circuit.  Being a lasy person, I opted to mod the circuit for +9v standard stomp box voltage instead of messing with a dual polarity supply.  I know that the headroom is limited in this capacity, but it sounds fine to my ears.  There are several good posts about how to go about doing this.  Basically, each op amp is inverting, so tie approx 4.5 volts to the + pins of each op amp instead of ground.  I built a garden variety voltage divider using 2 10k resistors in series and a 10uf cap.  Again, not the most stable Vref signal, but it works well in this application.

Good luckand let us know how the build comes along.

[The Tone God]

I thought about trying the sheilded cable idea. But, then I saw the Tremolo LEDs from pisotones I thought I might breadboard it before working on the Anderton more. Making a seperate board for the oscillation seems like a great idea. Also I thought about seeing if there was a way to get +/-9v from 18v power supply. Maybe use a dual opamp, 1 side inverting to get -9v, the other side noninverting to get +9v. Would that work? Or, would that just cause more problems?

Sheilded cable may help. I didn't make a seperate board for the LFO. I did this all on the same board.

Forget the split supply. In this situation there is no major reason to use it. I converted mine to single supply and there was little change in performance. I suggest you do the same.

I just recently put up a quick article about converting opamp circuits to single supply called Opamp-eration. Its a good place to start.

The Tone God's Domain

Andrew

[Jason D]

Thank you (everyone) for the help, I will work on it this week and see what happens.

[Mark Hammer]

Keep your expectations about what it will sound like at faster speeds reasonable.  The CLM6000 is blessed/cursed with a sluggish decay time.  That helps with things like compressors or envelope-controlled filters, but it gets in the way when you want to modulate things quickly with it.  I have not verified it on a scope, but my guess is that it distorts the shape of the modulation waveform at higher frequencies (e.g., >3hz or so), especially if you go for square-wave modulation.

Given the maleability of the design, if you really want to get deep modulation at higher frequencies, and very obvious square-wave on-off modulation, consider switching to another optoisolator with a quicker recovery time.  Some of the Vactrols are faster than a CLM6000.  Even if the resistance range is entirely different than the CLM6000, it is a piece of cake to tweak the various resistance values between IC1a and 1b to get the desired degree of modulation with the different photocell.

[Jason D]

I just breadboard the Anderton using the idea from the previous post. I used the Vref from a voltage divider circuit for the opamp. And I kept the LFO away from the audio part of the ckt. Works great, no ticking (on or bypassed)!!

Also, concerning the opto, I used the LDR from a broken Vactec VTL5C3, and a clear yello LED. I used a good Vactec VTL5C3, before, and I didn't notice any uneveniness. So, maybe it is the CLM6000 's response that causes the change.

[Chico]

I think that I used an NSL32.  IT sounds cool.

If you like tremolo, also build the EA.  The EA and the Anderton are two different beasts, each with their own unique place.  I like the EA for more subtle work, esp with slide.  I like the Anderton for that square wave, really strong effect.

[Jason D]

Finished the pcb layout and made a board. Replaced the first one I made using the MAX1044. The new one works great! NO ticking unless you turn the level up high. But, I used 100k for the feedback resistor on the output side of the dual op. I know thats high but if I turn my level to 12:00, I get what sounds like unity gain out.

I can make the board available to whoever may want it. I put mine in a 125B box, so it could possibly fit in a 1590B box.