Rebuilding my old analog console

[Greg]

Impressive. Both the 3D printed panel and the cycloids. Does the shape in video2 look the same on an oscilloscope? Are you using the G120 scanners for this?

[TheHermit]

Thank you so much Lasermaster for the encouragement. I've been racking my brain.

The biggest challenge is dealing with, how to balance the voltages when changing effects. For example, the AD633's drop the voltage so when you turn off the AM signal the image gets gigantic, or goes to zero depending on how they are hooked up. For the z-axis rotation when I turn off the rotation signal I have to ground one side or the image goes to zero. So, I'm trying to figure out how to wire all this stuff together so it interacts nicely.

If anyone has some advise on how to manage signal paths, that would be great.

Yes, very nice work and cycloids, Kevin.

.MP4 Workaround tip: With a Gmail account, you can upload your .mp4 video files onto Google Drive. Right click onto that file, click on "Sharing", "Share with anyone with the link", and "copy link". Then, when posting your message here, click the 'video' button on the above toolbar and paste the link into the popup dialog box. ;-)

Don't know whether any of this is relevant, but FWIW, I've also noticed interactions on my regulated 5V offset rail when tweaking XYRGB trimpots. But, figured that was normal behavior for parallel variable voltage dividers and once set, they're pretty stable, except for another unrelated image offset drift during warmup.
Then followed a problem with multiple Laser Synth PCBs sharing 5VDC from a powered USB hub. So, instead of powering from the computer's USB, now I use external 3 amp 12VDC power bricks individually, plus internal 12V > 5V regulated buck convertors.
Also upgraded the 5V 4010 cooling fans to quieter, more efficient 12V 5015 fans, with 12VDC still available for lumia projector devices.
Buck convertors are more efficient and run much cooler than LDO voltage regulators, which is a bonus considering how much heat my ADC/DAC CODEC IC generates. Plus the buck convertor effectively isolates the 12V fan noises from the synth's sensitive 5V, 3V3 digital, and +/- 9V analog audio circuits. That seems to have solved those woes.
I'd still be very interested in reading Lasermaster's suggestions.

Image #8 has a very nice intensity effect, smoothly increasing from the middle outwards to each point. Is that intensity modulation at the same Hz as the AM ramp, video banding, or the side effect of XY offset levels interacting with diode signals? 🤔
Roj

[Kevint]

Lasermaster1977: Thank you.here is the basic lay. I'll add more detailed schematics shortly

Greg: The 3D printer is a blast. I'll include some more pictures. I'm using Fusion CAD for my designs.
Not using the G120's at the moment. I got frustrated with the scan amp. For now, I took the cheap way out and bought some 25KPPS galvos with scan amp, power supply and mirrors for $68 and they work great.
My digital scope does have an XY feature. I'll see if it works to get an image.

TheHermit: thanks for the info. I think my issue is more basic and deals with how I should hookup and connect the various signals correctly. I will post some more specific examples of the issue shortly. Here is how video 8 was made:
Oscillator 1: 976Hz 7.6V
Oscillator 2: 959Hz 14.6V
Oscillator 1 VCA: both X&Y Ramp 67Hz 10V
Oscillator 2 VCA: both X&Y Inv Ramp 67Hz 10V
VCA 1 & VCA 2: 17Hz 8V
See photo for more details.

Video 8 Making:
Oscillator 1 circle
Add Oscillator 2 pattern and decrease to 0
Add Ramp 1 signal to get spiral
Add Oscillator 2
Add Laser diode modulation
Add Ramp 2 to XY VCA's
Play with levels

[laserist]

Thank you so much Lasermaster for the encouragement. I've been racking my brain.

The biggest challenge is dealing with, how to balance the voltages when changing effects. For example, the AD633's drop the voltage so when you turn off the AM signal the image gets gigantic, or goes to zero depending on how they are hooked up. For the z-axis rotation when I turn off the rotation signal I have to ground one side or the image goes to zero. So, I'm trying to figure out how to wire all this stuff together so it interacts nicely.

If anyone has some advise on how to manage signal paths, that would be great.

Nice work. Just from memory - I believe the Laser Images image rotator had a high impedance 10v signal always present that was swamped by the lower impedance rotation signal when present. Alternatively you could use an analog switch in a DPDT configuration, or a couple of 8 to 1 analog switches to give yourself lots of room...

[Kevint]

Thanks Laserist. I think that is what I am seeing.

Here is a drawing of how I am currently using the AD633 for a VCA.

I thought I could just pass a signal through the 633 if the AM signal was turned off. But in that scenario the output goes to 0V. However if I put -15V on the OpAmp AM input, I get a signal around the same voltage as the X1 voltage, but I'm not sure if that will cause future or upstream issues. If I disconnect the AM signal after the OpAmp the image blows up - over voltage.

I could use a switch (which I think is what you suggested) to bypass the 633 and use a gain stage so in both the bypass and non-bypass signals the images are equal in size. But that is a lot of switches. I was really hoping to just turn ON an AM signal and have it added without changing the image size too much.

Or is there a different VCA approach I should look at?

[lasermaster1977]

Thanks Laserist. I think that is what I am seeing.

Here is a drawing of how I am currently using the AD633 for a VCA.

I thought I could just pass a signal through the 633 if the AM signal was turned off. But in that scenario the output goes to 0V. However if I put -15V on the OpAmp AM input, I get a signal around the same voltage as the X1 voltage, but I'm not sure if that will cause future or upstream issues. If I disconnect the AM signal after the OpAmp the image blows up - over voltage.

I could use a switch (which I think is what you suggested) to bypass the 633 and use a gain stage so in both the bypass and non-bypass signals the images are equal in size. But that is a lot of switches. I was really hoping to just turn ON an AM signal and have it added without changing the image size too much.

Or is there a different VCA approach I should look at?

Maybe this insight will help you understand why what you seeing is happening.

I thought I could just pass a signal through the 633 if the AM signal was turned off.
As you've found out, nope is the answer. Here's why: The AD633, by design, is a 4-quadrant multiplier which means it will numerically multiply the differential voltage signal appearing across Y1+ and Y2- by the differential voltage that appears across X1+ and X1-.

In your implementation of the AD633, you have X1- and Y1- tied to GND (which is fine and dandy) and makes the AM signal voltage on X1+ multiply the Oscillator signal put on Y1+.
Therefore, when the AM signal amplitude on X1+ is zero volts or GND, 0 x Y1 voltage = 0.

In order for any output to appear on the pin 7 output, for any point in time, the voltages on X1+ and Y1+ must be > than or < than 0. At anytime, instantaneously or long term, if either of these inputs go to 0 volts the output on pin 7 will be zero.

So if you would like for the output of the AD633 to pass the Oscillator input signal without change there must be a positive or negative DC voltage on X1+.

As I stated earlier, there are many ways to accomplish this. One way is to have the input of X1+ start out as a DC offset voltage that makes the Y1+ full scale and of whichever voltage polarity you choose. Then, implement a cross mixing opamp circuit that provides this DC voltage and as a slider or rotary pot is change it starts "subtracting" a small AM signal from the DC signal. As the pot wiper is moved to its extreme, the DC voltage gets reduced to 0v as the AM voltage's pk-to-pk value goes to maximum but with a DC offset half of what was started with.

Another way is to use differential voltage waveforms for the X1-X2 and Y1-Y2 inputs. This method would keep the resulting output waveform peaks centered about 0 volts and independent changes in their respective amplitudes ( with and/or without DC offset changes) would control the final AD633 output signal results.

Thoughts, suggestions?

[TheHermit]

TheHermit: Here is how video 8 was made:
Oscillator 1: 976Hz 7.6V
Oscillator 2: 959Hz 14.6V
Oscillator 1 VCA: both X&Y Ramp 67Hz 10V
Oscillator 2 VCA: both X&Y Inv Ramp 67Hz 10V
VCA 1 & VCA 2: 17Hz 8V
See photo for more details.

Video 8 Making:
Oscillator 1 circle
Add Oscillator 2 pattern and decrease to 0
Add Ramp 1 signal to get spiral
Add Oscillator 2
Add Laser diode modulation
Add Ramp 2 to XY VCA's
Play with levels

Thank you for the detailed info and taking the time to video demo how video 8 was created.
Great reply, Kevin, and very much appreciated.
Hopefully Lasermaster1977 or others can help to resolve your voltage interactions issues.
BR
Roj

[Kevint]

TheHermit, Your welcome.

Greg, here is a scope image. This is sbout the best I could get with my scope. Also, here are a couple more 3D print images.

Lasermaster1977 and Laserist. With your advice, I think I have a game plan. More to follow once I test everything out. Also, your input got me to go read the AD633 datasheet. I know have a better understanding.

There is a tone of good circuit info on the eurorack synthesizer sites. Here is the schematic of the LFO circuit I'm using. That coupled with the offset circuit works really well.

[lasermaster1977]

TheHermit, Your welcome.

Greg, here is a scope image. This is sbout the best I could get with my scope. Also, here are a couple more 3D print images.

Lasermaster1977 and Laserist. With your advice, I think I have a game plan. More to follow once I test everything out. Also, your input got me to go read the AD633 datasheet. I know have a better understanding.

There is a tone of good circuit info on the eurorack synthesizer sites. Here is the schematic of the LFO circuit I'm using. That coupled with the offset circuit works really well.

Glad to hear your news, Kevin.

btw: I also love your 3D printed slider knobs and panel.

[Greg]

the Laser Images image rotator had a high impedance 10v signal always present that was swamped by the lower impedance rotation signal when present. Alternatively you could use an analog switch in a DPDT configuration, or a couple of 8 to 1 analog switches to give yourself lots of room...

For the curious, here's info in the older image rotator schematic that might be part of what is described.

Kevint: I have only been able to get the transformation shown in your video2 when using GS galvos open loop. But thanks to your work I think I now see the spiral is being amplitude modulated and rotation modulated by the same ramp signal.

Lore has it the LII version of this image class was a bit different because of the KQO clipper (FXCY callouts R K 6 7) circuit that flattened the outer loops against imposed amplitude limits. The GS galvos made the signal look amazing with loops compressed toward the outer edge of the spiral. But in my experiments it didn't work with modern galvos. Now I remember I think it was mentioned a while back that ramp shaping could get this effect too.