Ring Frobnicator

[dustfilledhobo]

Two days ago I built the Ring Frobnicator from the geofex.com schematic.  Initially only the LED from the millennium bypass was lighting on, then I checked all of my soldering, fixed a trace to get the two internal LEDs to light, then I discovered that my panel LED (LED3) was in backwards, fixed this, now it lights up.  I also put my transistor in backwards, this is now fixed.  Anyway, I had a few questions about getting the LFO to work. 

Here's what is happening now:  I followed the debugging guidelines on the geofex website and have determined that the transistor is working.  I have read all of the posts about this circuit on this forum and have seemingly conflicting information.  In one post I read that the voltage across the LED's need to be either 3.2-3.4 V, in another it said that the voltage needs to be 5.1 V.  I am interpreting this to mean Vb needs to be 3.2-3.4 V (or 5.1 V), is this true?  If not what should Vb equal.  I am using a 9V battery which currently has 8.1 V in it.  My Vb is currently set to 3.3 V and none of the LEDs oscillate, should they?  Also, is there a chance that I burned out one of my ICs by placing the two components in backwards?  If so, how can I verify that my chips are working, are there some voltages that I can check?

Thanks in advance for any help that you can offer.

Dusty   

[soggybag]

I built this and it does work. I used the GEO layout. I seem to remember having trouble if the LEDs were not the same type. Something about using different LEDs upsets the bias. I never quite understood bias in this. The original scheme shows bias as ~3.2V so I would go with that. There are two LEDs setting the bias, and the third is in parallel with the 914 diode going into the 3080s Iabc pin.

[dustfilledhobo]

I built this and it does work. I used the GEO layout. I seem to remember having trouble if the LEDs were not the same type. Something about using different LEDs upsets the bias. I never quite understood bias in this. The original scheme shows bias as ~3.2V so I would go with that. There are two LEDs setting the bias, and the third is in parallel with the 914 diode going into the 3080s Iabc pin.

Am I missing something? LED3 is in PARALLEL with the 914 diode? Did you mean series?  I'm really confused, was there a modification to the geofex schematic? (http://www.geofex.com/PCB_layouts/Layouts/frobn.pdf)

Thanks for your post, and your verification of the circuit, I measured some voltages, maybe someone can verify that these are correct:

All measurements were taken wen my "9V" battery measured about 8.1V

Q1 (NPN 3904)
E = 2.98 V
B = 3.50 V
C = 8.07 V

U1 CA3080E (http://www.futurlec.com/Linear/CA3080E.shtml)
1: no connection
2: 3.25 V
3: 3.22 V
4: 0.00 V
5: 0.71 V
6: 3.51 V
7: 8.09 V
8: No Connection

I don't fully understand op amps, but from what I have gathered the inverting and non-inverting are supposed to "want" to be the same, is this correct? If so, pins 5 and 6 in the TL062 disagree.

U2 TL062 (http://www.futurlec.com/Linear/TL062CP.shtml)
1: 7.26 V
2: 3.33 V
3: 3.27 V
4: 3.81 V
5: 6.51 V
6: 3.26 V
7: 7.41 V
8: 8.08 V

Thanks again.

Dusty

[soggybag]

You are correct, I made a mistake the LED is in series with the 914.

It's my suspicion that the bias voltage needs to be lower to get the low end of the LFO low enough to get the signal out of the 3080 to 0. Feel free to correct me if I'm way off. Here's my theory. The LFO goes up and down with a comfortable margin between the rails. If the low end of the swing is not low enough at the Iabc pin volume out of the 3080 won't reach 0.

From R.G.s article on the 13600, which I'm guessing is similar to the 3080, the Iabc input looks like two diode drops. So the low signal needs to get down to about 1.4V or less to get the signal out of the 3080 to completely shut off. Lowering the bias to ~3.2v gets the low part end of the LFO to ~1.4v.

I'm also guess there is something to this where the diodes provide a constant voltage that will not change as the battery depletes. With a resistor in the bias network the bias voltage would vary.

My experience with Frobnicator was that it worked pretty well. It made a good tremolo. The "ring mod" mod was not the most exciting ring mod sound. Like I said earlier I remember having some trouble getting it work when the diodes weren't matched. When I used diodes of all the same type it started working.

[dustfilledhobo]

Thanks for taking the time to help me with my problem.

It's my suspicion that the bias voltage needs to be lower to get the low end of the LFO low enough to get the signal out of the 3080 to 0. Feel free to correct me if I'm way off. Here's my theory. The LFO goes up and down with a comfortable margin between the rails. If the low end of the swing is not low enough at the Iabc pin volume out of the 3080 won't reach 0.

When you say "bias voltage" you are referring to Vb not the "Amplifier Bias Input" (pin 5 on the 3080), correct?

From R.G.s article on the 13600, which I'm guessing is similar to the 3080, the Iabc input looks like two diode drops. So the low signal needs to get down to about 1.4V or less to get the signal out of the 3080 to completely shut off. Lowering the bias to ~3.2v gets the low part end of the LFO to ~1.4v.

Assuming that the bias voltage is Vb, I believe that my Vb=3.26 V(pin 6 on TL062).  Is this still to high?

I'm also guess there is something to this where the diodes provide a constant voltage that will not change as the battery depletes. With a resistor in the bias network the bias voltage would vary.

My experience with Frobnicator was that it worked pretty well. It made a good tremolo. The "ring mod" mod was not the most exciting ring mod sound. Like I said earlier I remember having some trouble getting it work when the diodes weren't matched. When I used diodes of all the same type it started working.

All of my LEDs are the same.

Thanks,

Dusty

[dustfilledhobo]

I fixed a grounding problem, but my LFO will still not oscillate, so I re-measured all the voltages on the chips and and transistor to see if anyone could see anything odd that might help point me in the correct direction.

The transistor and CA3080 voltages did not change (but they are provided below for quick reference), but the TL062 changed a lot.  The voltage that concerns me is the .011V on pin 2 of the TL062, it seems like it shouldn't be that low.  Thanks in advance.

Schematic (Click Here)

U1 CA3080E (Datasheet)
1: No Connection
2: 3.25 V
3: 3.22 V
4: 0.00 V
5: 0.71 V
6: 3.51 V
7: 8.08 V
8: No Connection

U2 TL062 (Datasheet)
1: 3.25 V
2: 0.11 V
3: 3.27 V
4: 0.00 V
5: 7.11 V
6: 3.26 V
7: 7.43 V
8: 8.09 V

Q1 NPN (3904)
C: 8.07 V
B: 3.50 V
E: 2.97 V

Thanks

Dusty

[DougH]

I played with this one for a while and never could get it to work. My problem was with the OTA- way too much carrier bleed to be useful. I did get the LFO working but I seem to remember having to tweak the bias voltage (Vb) to get it to oscillate. Try playing with the LEDs and R11. Unfortunately this one really needs to be breadboarded before you commit to a build, IMO. I believe the problem is that the LEDs are poorly specified. The schem indicates a target Vb of 3.2v but I don't remember if that was the voltage that worked for me or not.

Incidentally, if you get the LFO working- it works well with the audio section of the "Nyquist Aliaser". This is a fairly simple ring mod circuit that works pretty well:

http://www.lynx.net/~jc/NyquistAliaserSchem.gif

The top part of the schem with the FET switch is the audio portion. Remove the R10 and C4 from the output of the LFO (don't remember if the diode and LED are required or not) and use that to drive the gate of the FET switch on the Aliaser. This makes a nice and simple sawtooth wave Ring Mod that works well. I tried it leaving in the filter for the pseudo-sine wave LFO but there wasn't a high enough voltage level to drive the FET. There may be ways of adjusting that but I was just messing around on the breadboard and came across this.

[soggybag]

I had similar troubles with getting the oscillator working. The bias needs to be right. I remember mine worked right off. But I later swapped on LED for a different type of LED and that made it stop working.

Yes, I'm using bias to refer the to the Vb or ~3.2v on the schematic. This is also list as Ground on the schematic.

I remember someone else, in a post on this forum, saying they added a trim pot in series with the two LEDs in the biasing network to adjust the voltage here.

I remember the Trem sound was pretty good. With no bleed. The ring mod sound was not super exciting and it had some bleed that I could never trim away. The trimmer could get it pretty quiet but it would never quite go away.

At the moment I'm working on a 13600 version this project. So far the Frobincator sounds better than what I have.

[dustfilledhobo]

I had similar troubles with getting the oscillator working. The bias needs to be right. I remember mine worked right off. But I later swapped on LED for a different type of LED and that made it stop working.

Which LEDs did you use, I used these: http://www.allelectronics.com/make-a-store/item/LED-1/T-1-3/4-RED-LED/-/1.html

Yes, I'm using bias to refer the to the Vb or ~3.2v on the schematic. This is also list as Ground on the schematic.

I don't understand your second sentence, are you saying that Vb = ground?  I connected all of the components together that connected to the down pointing arrow (which I interpreted as ground), and I connected all of the components connected to the Vb's on the diagram together.  Are you suggesting that I need to connect my ground to my Vb?

I remember someone else, in a post on this forum, saying they added a trim pot in series with the two LEDs in the biasing network to adjust the voltage here.

My Vb is already close to 3.2 (I think it is reading at 3.25 V right now).  I read the same post that you did that mentioned a range from 3.2-3.4V for oscillation, so I use a pot and tried every value in this range and still no oscillation. 

At the moment I'm working on a 13600 version this project. So far the Frobincator sounds better than what I have.

I don't know what 13600 is, but I would be interested in hearing more about this, do you have a post somewhere on this forum explaining it?

Thanks for taking the time to help me.

I fixed a previous connection problem and re-measured the following values.  I have also include a few measurements on my diodes.

Battery Voltage: 8.1 V

Schematic

U1 CA3080E (datasheet)
1: no connection
2: 3.24 V
3: 3.22 V
4: 0.00 V
5: 0.66 V
6: 3.24 V
7: 8.05 V
8: no connection

U2 TL062 (datasheet)
1: 7.31 V
2: 0.11 V
3: 3.25 V
4: 0.00 V
5: 7.11 V
6: 3.25 V
7: 7.40 V
8: 8.06 V

Q1 3904
E: 2.82 V
B: 3.24 V
C: 8.05 V

LED1 Red 5mm
A: 1.62 V
K: 0.00 V

LED2 Red 5mm
A: 3.21 V
K: 1.62 V

LED 3 Red 5mm
A: 2.63 V
K:1.17 V

D1 1n914
A: 1.13
K: 0.66 V

Thanks,

Dusty

[dustfilledhobo]

I got the oscillator working, a resistor that should have connected to pin 2 on the TL062 was connected to pin 3, once this was fixed it oscillated.  Thanks for your help.

Dusty

[DougH]

Cool! 

Hope you have better luck with it than I did. 

[dustfilledhobo]

So, I finally tested my ring frobnicator, and the vibrato sounds awesome, but the ring mod function seems to do nothing to the guitar signal.

I have read on here that you can hear the LFO tone, and some people have been able to get rid of it almost completely, but my problem seems a little different. 

When I am in ring mode I can hear the clean, unaltered guitar signal and the LFO pitch, they seem to not be mixing at all.  I can change the pitch of the LFO tone, but the clean guitar never changes in this mode.

Thanks for any help in advance.

Dusty

[soggybag]

In the ring mod the LFO should speed up to audio levels. With mine I can hear a faint whine from oscillator when in the ring mode. When in the ring mode it's like an audio rate tremolo. If you have a scope you should probe the oscillator and see if the rate increases when you switch to ring mode.

[dustfilledhobo]

In the ring mod the LFO should speed up to audio levels. With mine I can hear a faint whine from oscillator when in the ring mode. When in the ring mode it's like an audio rate tremolo. If you have a scope you should probe the oscillator and see if the rate increases when you switch to ring mode.

When you write "probe the oscillator" where in the circuit do you mean exactly?  If you could tell me an exact node to place the prob that would help me. Thanks.

Dusty

[dustfilledhobo]

I ended up placing a 1M in parallel with R6(10M), bringing the total resistance down to just under 1M.  Now the trim pot (tr2/tr1) can adjust the volume of the modulation, but I cannot get it to be completely silent.  Has anyone been able to get the modulation completely silent?  Would placing a small resistor in series with R6 and using a multi-turn (lower value) trim pot allow me to get the modulation silent?  Thanks.

Dusty

[soggybag]

Glad to hear that you got it working. I haven't played with this for a while.

My comments about ground refer to the original schematic. Here everything marked Vb on R.G.s schem is shown as ground. The actual ground is shown as 0V.

The 13600 is similar to the 3080. But it's 14 pins and has two OTA's. It's sort of a double 3080.

If you have a scope I'd try probing pin 1 of the TL062. This is where the CV wave form comes out. You could also use the audio probe. In the ring mode you should here an audio tone. On the scope you could see the low frequency wave and see it change. When you switch to ring mode it should increase by 10 or a 100 or something.

If the ring mode isn't working. I'd be sure to check all of the connections around C5 and C6. When the switch is closed you are adding C6 to the system. Since they are in parallel their values add to create a larger value cap (.1 + .002). If isn't working when the switch is openit could be that C5 is not connected correctly. Which cause the oscillator not to oscillate.