EASY VIBE Help please... (LFO problem using R.G. Keen's ramp up/down mod)

[orangetones]

I am having troubles with the easyvibe ramp up/down modification that I have drawn up a layout for.  I have populated the board, but am not getting the lfo to work, nor am I getting any signal through the pedal.  Although, mysteriously I get signal when I first hook up the battery, I can hear a ticking of the LFO, but then the signal gets distorted and fades.

Is there anyone who could look over my layout?  I am pretty sure that I correctly added (or rather replaced) the original LFO with the modified one from R.G.'s website.

Here is a link to what I assume is a correct replacement of the R.G. modified LFO into the Easyvibe circuit:
http://www.aronnelson.com/gallery/main.php/v/Schematics-etc/orangetones/easyvibe-modschem.jpg.html?g2_imageViewsIndex=1

Here is the R.G. mod:
http://www.geofex.com/FX_images/ramp-lfo.pdf

Here is the original circuit:
http://www.hollis.co.uk/john/easyvibe.jpg

Here is a link to my layout:
http://www.aronnelson.com/gallery/main.php/v/Schematics-etc/orangetones/

Voltages are as follows:
V+: 8.71 from a 9v battery
VBias: 8.2

IC1
1: 7.43
2: 7.43
3: 8.24
4: 8.7
5: 8.24
6: 7.43
7: 7.43
8: 7.33
9: 7.41
10: 7.90
11: 0
12: 7.58
13: 7.22
14: 7.22

IC2
1: 7.43
2: 7.43
3: 8.24
4: 8.7
5: 8.24
6: 7.43
7: 7.43
8: 7.43
9: 7.66
10: 4.12
11: 0
12: 4.019 (this goes to 0 when I adjust the speed pots fully the opposite way)
13: 2.698 (goes to 0 when I adjust the speed pots fully the opposite way)
14: 7.41 (starts to drop steadily when I adjust the speed pots fully the opposite way)

Q1 on my layout (not on R.G.'s)
C: 8.6
B: 8.6
E: 8.0

Q2 on my layout (not on R.G.'s) - these measurements vary as I adjust the speed pots
C: 0
B: 7.0
E: 0

[R.G.]

Divide and conquer.

We're pretty sure the signal path can be made to work, right? So concentrate on the LFO.

I haven't run through your layout or voltages yet, but the thing to do is:
1. Get the power supplies right. Do you have a fresh new battery or regulated power adapter? Does it put out 9+V as measured by your meter? Yes, measure it.
Now hook it up to the board. Does 9V appear everywhere it should? All IC power points? If not, why not? Turn off power. Now use your meter set to ohms and see if "ground" really connects to every place you think it does. Only after you KNOW the power is correct can you find out anything about the circuit. It's wasted effort do do anything else first.

2. Now look at the circuit. Does your bias supply work? Do all the places where Vbias goes actually measure the same?

3. Once you have the basics set up and have some chance of debugging, start looking at IC pins and transistor pins. What does an NPN base do versus an emitter? Collector? Do yours actually act that way? Are your IC pins sitting at the voltages you'd expect if they really operated?

Give this a run (yep, I'm trying to stretch your mind) and let us know back what happens.

[orangetones]

R.G. Thanks for the quick response!  I actually figured it out.  I had one of the LED's (of the 2 that are with the Diode maintaining constant voltage even when the battery drops) backwards.  All is well now, pretty much.  The ramping up and down works.  however, when yo uadjust the speed pot you are currently using it of course reacts just as it would when you switch to the other pot, so if you have the ramping speed set slow, as you turn the speed pot the speed changes slowly.  Kind of odd, but I guess workable.

I find the log taper of the 2 speed pots concentrates the speed change over a small portion of the pot. Might be a good idea to try another taper.  Or is it possible that I just need to reverse the ground and 9V+ lugs?  Hmmm...

My next quandry is the nasty ticking from the LFO.  I think there are some threads about that here though.  Have to go searching.

I was also going to toy around with a larger drive trimmer to get the LED's coupled with the LDRs to get closer to shutting off a minimum brightness.  I thought I had heard that in the Univibe setting the bulb to be very dim, almost off at its lowest point produced a better swing.

Finally, I don't seem to get a good sound out of the higher strings on the guitar (i.e. it seems to lack a lot of treble up there).  Any thoughts?

Thanks for all the help!

[R.G.]

R.G. Thanks for the quick response!  I actually figured it out.  I had one of the LED's (of the 2 that are with the Diode maintaining constant voltage even when the battery drops) backwards.  All is well now, pretty much.  The ramping up and down works.  however, when yo uadjust the speed pot you are currently using it of course reacts just as it would when you switch to the other pot, so if you have the ramping speed set slow, as you turn the speed pot the speed changes slowly.  Kind of odd, but I guess workable.

Yep, that would have kept it from working, all right. Good detective work. I've thought about the speed pot thing and I don't see any simple way to make speed pot changes be instant but flipping between them be slow. Doesn't mean there isn't a way; I just don't see one yet.

I find the log taper of the 2 speed pots concentrates the speed change over a small portion of the pot. Might be a good idea to try another taper.  Or is it possible that I just need to reverse the ground and 9V+ lugs?  Hmmm...

Experiment, of course; I think linear might be better. You may even need reverse log. If you go with linear, you can taper them with tapering resistors to taste.

My next quandry is the nasty ticking from the LFO.  I think there are some threads about that here though.  Have to go searching.

I have looked at your layout now, and have some suggestions. First, your layout has some issues with power supply distribution. Unfortunately, you have used part of each LM324 for audio, part for the LFO. That's good for dividing the audio from LFO on the board, as it neatly isolates the audio on one side, but bad in that the power and ground for LFO is inherently coupled to the power and ground for audio because it's into the same chip. It keeps you from decoupling or isolating the power and ground traces. This would have been easier if one LM324 was used entirely for LFO, then other opamps for audio. That would also let you use an opamp with better audio response for the four audio opamps. I know why you did it - the number of opamps comes out even this way and produces a neater, tighter layout. Not criticism, just suggestions for your next board layout. It takes practice before one's brain can do all this at the same time, which is what layout involves.

But it may not be a disaster. Maybe we can quiet things down a bit. Here are some things to try. The ticking has to be originating in the sudden voltage change on the output of the Schmitt trigger opamp, perhaps to a lesser extent in the sudden change in direction on the peak/valley of the triangle. It can be coupled into the audio circuit by power and/or ground lines, or by the bias voltage. The first thing I'd do is to try to slow the transition of the Schmitt trigger down by putting a smallish capacitor from the output to (-) input. Try 100pF, and increase it in steps  until you either kill the ticking or it quits working alltogether.

Next, I'd make a second identical Vbias for the LFO. The audio and LFO don't need exactly the same Vbias, it's just convenient. In your layout, the Vbias for the LFO comes across between the two LM324s to the LFO. Remove that jumper, isolating the LFO Vbias, then construct a new Vbias for the LFO in the empty space by C2.

There are other things to do, but I think these may help a lot. Let me know what happens.

I was also going to toy around with a larger drive trimmer to get the LED's coupled with the LDRs to get closer to shutting off a minimum brightness.  I thought I had heard that in the Univibe setting the bulb to be very dim, almost off at its lowest point produced a better swing.

Maybe, maybe not. The Univibe phase stages are very different, as is the light source. It's worth experimenting, but remember the differences.

Finally, I don't seem to get a good sound out of the higher strings on the guitar (i.e. it seems to lack a lot of treble up there).  Any thoughts?

Solve that one last. 

[orangetones]

Thanks R.G. I will give those suggestions a try.  I was wondering though, after reading up about schmidt triggers, I am still a little confused as to which part of the circuit that is.  I am not an expert in audio by any means, but very willing to learn.  I'll let you know later on how the modifications go.

Thanks again,
Steve

[R.G.]

Thanks R.G. I will give those suggestions a try.  I was wondering though, after reading up about schmidt triggers, I am still a little confused as to which part of the circuit that is.  I am not an expert in audio by any means, but very willing to learn.  I'll let you know later on how the modifications go.

No problem; none of us are experts on everything. In your layout, the Schmitt is the opamp section on pins 8, 9, and 10 of IC1.

A Schmitt trigger is a comparator which is set up to modify its own reference voltage to make it harder to change. In my schematic it's the right-most opamp in the LFO. As you can see, the signal is applied to the (-) input, while the (+) input is tied to the reference voltage through a 47K, and to the output by a 100K. The output swings from nearly ground to nearly V+, so the 100K and 47K resistors add or subtract about 1/3 of the power supply voltage to/from the reference voltage depending on which way the output is. If the input on the (-) input is near 0V, then the Schmitt output is high. The 100K/47K cause the (+) pin to see a voltage of Vbias+3V. So the (-) input can wander around all it wants below Vbias+3 and never change the output. When it gets above Vbias+3, it causes the output to go low. That has the side effect of changing the voltage on the (+) input to Vbias MINUS 3V, which (a) causes the output to really bang to ground hard by the trigger action and (b) leaves the (+) input sitting at Vbias - 3V so the (-) input can now wander around all it wants and nothing will happen until it gets below Vbias-3V.

In effect, the circuit causes a snap-over effect. the input has to go a certain amount to the other side of Vbias than it's on to make the thing switch. Once it switches, the trigger point flips to the opposite side, so that there is a hysteresis effect. In this case, the input voltage is a ramp and the output of the Schmitt causes the ramp to change directions. So the ramp goes up until it trips the Schmitt at Vbias +3, gets its direction changed, and ramps down until it trips the Schmitt again at Vbias-3. The Schmitt trip points determine the height of the triangle from peak to peak, the Vbias determines the middle position of the triangle.

All that's left of the LFO to describe is that the integrator (pins 12,13, 14 of IC2) is set up to integrate either up or down (determined by whether Q2 is switched on or off) at a rate determined by the currents through R20, 25, 26, 27. These currents are determined by the resistor ratios and the voltage fed to R20 and 27. This is a voltage controlled oscillator; it has some other neat tricks that can be played, but it's easy to work with and has a much wider range than most other triangle/ramp LFOs.

[RickL]

Well I was going to report that I got mine working but I guess I'm a little late. I built it this week when I was out of town on a course and only got to test it a couple of minutes ago. FWIW I tested it using VTL5C3s for the LED/LDR combos but I intend to use home brew combos for the final version if I can get acceptable sound out of them. The VTL5C3s do sound good.

I should note that I built mine using the first version of the layout that was posted. I justed subbed a 500k pot for the trimmer.

I'll echo the comment about the rate change pot. I only found about the first quarter of the range usable (from 0 ohms to maybe 100k or so) and will probably end up using a 100k pot for the control. Within that range it works quite well, from almost instant change to a long, slow ramp. The last bit of change seems to take a fair amount of time, but I suppose that makes sense since it looks like the change is based on the charge/discharge time for the capacitor.

Off topic, but I also added switches on my build to C1, C3 and C4 so I could have all four phasing caps at 0.22uF. I prefer the sound of the staggered values but it does give some variation for the cost of 3 caps and 3 switches.

[orangetones]

Hey there Rick.  Glad to hear you got one working.  The second layout is really just a movement of some parts around so that you could better fit the large electrolytic caps on and so that the wires for the pots/switches were all lined up nicely.  No change other than that.  Are you experiencing problems with a ticking LFO?  Are you using LM324s?  Anything you are doing that is a substitution?

R.G.

I tried the cap thing across the Schmidt Trigger.  No luck.  even up to 0.4 I get no real improvement.  I notice a little, but no real improvement.  If I use a 1uF electro it kills the LFO altogether.

I tried to set up another bias votage.  Just removed that jumper and then did a little bias network on my breadboard and connected to the main board.  Still have the ticking.

Hmmm... what else can we do.  I could always redesign the board!  But finding a solution that works with this already populated board might be best.

Thoughts?  and thanks again for the help.

[R.G.]

This is not a final fix, just an experiment to gather data. Can you temporarily solder 47K resistors in parallel with R21 and R22 (both 10M)? Also a 0.1uF in parallel with C11.

I think you may be picking up radiated LFO edges in the very high impedance at Pin 10. This test will tell us whether that is true or not. It will also cause treble loss from the lower impedance and bass loss from C11, but we can fix those if it gets rid of the ticking.

[RickL]

I didn't notice any ticking from the LFO. I used LM324s. Except for the phasing cap mods I built it stock.

[orangetones]

This is not a final fix, just an experiment to gather data. Can you temporarily solder 47K resistors in parallel with R21 and R22 (both 10M)? Also a 0.1uF in parallel with C11.

Done.  Now there is not the loud clicking like tick, but there is a lower pulsing throb (although at a much lower volume).


In relation to his build, are there any reasons that Rick might not be experiencing the ticking (Rick could you confirm this?)?
The newer layout I am using has the the ground filled in in the empty spaces and a couple of components moved.

Here is the layout that Rick used and the PCB transfer:
http://www.aronnelson.com/gallery/main.php/v/Schematics-etc/orangetones/easyvibe-ramp.jpg.html
http://www.aronnelson.com/gallery/main.php/v/Schematics-etc/orangetones/easyvibePnP.jpg.html

Here is the new layout and the PCB transfer I used:
http://www.aronnelson.com/gallery/main.php/v/Schematics-etc/orangetones/easyviberamp-nov21.jpg.html
http://www.aronnelson.com/gallery/main.php/v/Schematics-etc/orangetones/easyviberamp-nov21pnp.gif.html

Thoughts on that?

Here is a little SWF file taht will allow you to toggle between the two different layouts.  It might help in determining any issues that could have arisen from the newer layout.
http://www.aronnelson.com/gallery/main.php/v/Schematics-etc/orangetones/boards.swf.html

I realize that I names the parts differently in this new layout.  I wanted the parts grouped by value.

[R.G.]

Hmmm. Is your input wire shielded?

The loss of the "tick" makes me think that one of my suppositions was right - the 5M input impedance was letting it pick up LFO edges capacitively.

The throb is something else. Is the sound of the throb constant with LFO speed, or does it change? With no signal into the unit, measure the DC voltage on pins 7 and 8 of IC2 and see if they blip up and down during a throb or not.

I'm not sure why Rick would not have the issue. It's always possible that there is another unsuspected error in yours.

In light of that, can you measure the no-signal DC voltage on all pins of the two ICs?

[orangetones]

R.G. here are the pin voltages.  No jumping around on pins 7 or 8 of IC2.

My input wiring is not shielded.  I usually don't use shielded wiring, though maybe I should.

V+: 8.63 from a 9v battery (I'm sorry, but I don't have any better 9 volt batteries right now)

IC1
1: 3.875
2: 3.875
3: 3.862-3.868
4: 8.63
5: 3.862-3.868
6: 3.875
7: 3.875
8: 2.07-2.30 (on a fast speed setting, jumps around too much to measure on slow speed)
9: 5.15-5.27 (on a fast speed setting, jumps around too much to measure on slow speed)
10: 3.52-3.55 (on a fast speed setting, jumps around too much to measure on slow speed)
11: 0
12: 4.61-4.66
13: 4.61-4.66
14: 4.61-4.66

IC2
1: 3.875
2: 3.875
3: 3.862-3.868
4: 8.63
5: 3.865-3.868
6: 3.875
7: 3.875
8: 3.872
9: 3.872
10: 3.855
11: 0
12: 0.510 (slow speed) 3.55 (fast speed)
13: 0.510 (slow speed) 3.55 (fast speed)
14: won't read (jumps around too much for the meter on slow speed) 5.15-5.29 on fast speed.

[orangetones]

Oh, yes, the throb is in time with the LFO spped.  Forgot to mention that.

[R.G.]

Oh, yes, the throb is in time with the LFO spped.  Forgot to mention that.

Did you mean to say "the throb was in time with the LFO and it stopped."?

Or is it still there?

[orangetones]

The throb mentioned above is still there.  The ticking is gone, but yes, there is a throb there and it is in time with speed of the LFO is speeds up when the LFO speeds up, and slows down when I slow it down.

[orangetones]

BUMP

[tandtgr]

I built the easyvibe, with this scheme and layout:
http://www.geofex.com/PCB_layouts/Layouts/easyvibe.pdf

but I have the LFO ticking problem. I added a wallwart with a 10R in serie to 9v+ and a 470uf cap to ground....but nothing,I still have this problem, including feeding the circuit through a battery....a solution?

[Jered]

  Not trying to jack this thread but I have a question. Is there a prefered resistance for the LDR's? I'm going to build this and just looking for advice from previous builders. Thank you.
  Jered

[nico13]

I built the easyvibe, with this scheme and layout:
http://www.geofex.com/PCB_layouts/Layouts/easyvibe.pdf

but I have the LFO ticking problem. I added a wallwart with a 10R in serie to 9v+ and a 470uf cap to ground....but nothing,I still have this problem, including feeding the circuit through a battery....a solution?

I used this schematic also (I just used another layout because I perfboarded it) but I have no ticking problem.

You may try to use a regulated power supply (mine is a Zoom walwart that I regulated with a 7809 and some caps to ground).

Nico.