15kpps scanner signal, what are the values?

[lasermaster1977]

My dad uses G124. Any idea what the proper signals for those are? I want to say they are single ended and not looking for a differential signal.

This is very helpful information because while building my analog console I have been thinking about things incorrectly.

The G124's were open loop amps run on + and - 15 to 18 volts. G100 series scanners required a max of 1 amp for maximum deflection so the power supply needed had to provide at least 2 amps per scanner for head room. Input voltage range was set by the input potentiometer and could range easily from 1v peak-to-peak to (if my recollection serves me right) to 10v peak-to-peak.

[lasermaster1977]

I've built examples of this circuit before and played with it several times over the years, so I know it works.

Adam, sorry to pop-in on this discussion. I agree your 3 op-amp circuit would work fine, but you could also achieve the same results with only two op-amps. You have 3 inverting, unity-gain op-amp circuits when you only need 2 to create the differential output. Why not just take the output of the first amp, feed it to the inverting input of the second like you are doing, but also use it as the X- output, then take the output of the second op-amp for the X+? You save 2 resistors and 1/4 of the TL084 allowing you to use one TL084 quad package for X and Y channels, 1 and 2 for X's diff outputs and 3 and 4 for Y's diff outputs. This works whether the source input is either 0-5v needing the 2.5v offset or a bi-polar input that requires no offset adjustment. Just a thought.

[buffo]

I agree your 3 op-amp circuit would work fine, but you could also achieve the same results with only two op-amps.

I posted that circuit as an example, because of the discussion that Andrew and I were having about how to bias the op-amp, but I should have clarified that the 3 op-amp circuit was not actually designed to do what we had originally been discussing. In fact, the circuit I posted had another purpose altogether. It's actually one stage of a buffered ILDA splitter that would split a single ILDA input to multiple ILDA output ports. (Much like the DZ ILDA splitter does.)

Thus the signal output from that first op-amp (the differential receiver) would then be fed to multiple copies of the latter part of that circuit, including separate X/Y gain and offset pots and invert switches for each output port. This allows you to adjust each output separately, making setup much easier in a multi-projector show. But for simplicity's sake only a single iteration of the circuit is show on the schematic.

Anyway, that's why you have the unity gain differential receiver at the head end.

Adam

PS: No need to apologize, btw. This has been a good discussion! I know I learned some new things from this thread, and I'm sure others have as well.

[mixedgas]

It all depends where the manufacturer sets up the position sensor gain resistors and the servo gain pot. Some of them do not follow the rules. Some of them have NEVER even read the ILDA standards docs. Some of them have ever even read the 6850 manual to know what they copied.
~
So it depends if your manufacturer calibrates the amplifier for 0.1 volt per degree or 0.3335 volts per degree or whatever... In the olden days the spec was 0.5V per Degree Optical at Test Point One (position sensor TP) on the Cambridge Galvo amps. The actual input voltage number depended on your ordering either a 40, 60 or 80 degree swing.
~
I've seen amps need the full swing of the ILDA signals and others set where the angle slider at 10% maxes out the galvo.
~
+/-5 driving a single input of a differential input should get you about half angle scanning if you followed the rules. Keep in mind who tunes your amps at the factory and how they do it makes a massive difference in performance and standards compliance. For many of these production line tuners all they have been taught is " Make It look like This" :-{
Many of them probably make only the specified minimum wage in their district.
~
See attached. A page from the 6850 tuning manual. Which at the time would have been one of the guidelines for the designers of the ILDA standard.
~
Yes, there is a buffer built into the original ILDA conventions used so you should be able to swing at least half a scan field with DC positioning pots.

~
The ILDA standard does NOT tell you how to scale the signal inside your projector!
~

Steve

[lasermaster1977]

So we all know that the ILDA standard for X/Y signals is "+/- 5 volt differential". But what does that mean? It means that you have two wires labeled X+ and X-. Each of these wires (or pins, if you will) can be at any voltage between -5 volts and + 5 volts. So if you take the *difference* between the X+ and X- wires, you end up with the following maximum possible values:

Voltage on X+ Voltage on X- Difference
+5 volts -5 volts +10 volts
-5 volts +5 volts -10 volts

As you can see, the total swing between these two maximums is +10 volts to -10 volts, or a total peak-to-peak voltage of 20 volts.

Adam, now this really confuses me so maybe I've misunderstood the ILDA standard for XY. According to the ILDA Standard the absolute maximum difference between the X+ and X- outputs (given the ILDA input standard is +5v and -5v bi-polar) is 10v peak-to-peak regardless of polarity when measured across these two differential outputs, and when converted to a single-ended signal yields + and - 5v signal swings with respect to ground. Am I missing the math somewhere on how 20 volts peak-to-peak comes in, and I really want to understand this.

Thanks

[buffo]

According to the ILDA Standard the absolute maximum difference between the X+ and X- outputs (given the ILDA input standard is +5v and -5v bi-polar) is 10v peak-to-peak

All of this ^^^ is correct.

regardless of polarity when measured across these two differential outputs

This is the part that is confusing you. "Regardless of polarity" all by itself would be incorrect. But "regardless of polarity when measured across these two inputs" is correct, in that the absolute maximum *magnitude* of the voltage between the two signal pins would be 10 volts.

The problem, of course, is that polarity *IS* regarded. The galvo amps can tell the difference between a +10 volt difference and a -10 volt difference between the two inputs.

To review: if the X+ signal is at +5 volts, and the X- signal is at -5 volts, then you have +10 volts between them. However, the opposite case is also possible... If X+ signal is at -5 volts, and the X- signal is at +5 volts, now you have -10 volts between them.

Thus, the "total swing" between these two edge cases is +10 volts to -10 volts, or a maximum of 20 volts peak-to-peak on the OUTPUT of the differential receiver. (Remembering that a differential receiver doesn't care what "ground" is with regard to the input... All it does is compare the voltages on the inverting and non-inverting input and produces an output equal to the difference between those two inputs.)

And yeah, you can easily bias the receiver such that the maximum -10 volt signal output becomes your ground reference, giving you a zero to +20 volt signal output. But even if you don't do that, you still will have a maximum swing of 20 volts between the -10 and +10 edge cases. It will just be centered around ground.

If you look at my 3 op-amp circuit, you'll notice that the gain of the middle op-amp is much less than unity... That's because we need to reduce that 20 volt max signal from the output of the differential receiver stage!

Adam

PS: I should also point out that this post and my previous one above are both referring to the schematic labeled "ILDA X and Y Size and Position Adjustment" that I posted. I just realized that even earlier in the thread I also posted a different schematic (Labeled "0-5 volt single ended to ILDA differential") which is slightly different and WILL NOT WORK. That is actually what started the discussion between Andrew and me. But that other circuit is flawed, so please disregard it. (I'm leaving it in the thread because we refer to it later in the discussion, but that circuit is definitely wrong.) Don't want to confuse anyone by referring to the wrong circuit...

[lasermaster1977]

All of this ^^^ is correct.

Thanks for the quick response. I wasn't referring to any of your schematics and I should have so stated. So let me clarify that I am referring to a unity gain single-ended input to balanced driver output, such as:



This is a portion of a schematic image uploaded by mixedgas depicting a unity gain balanced line driver. So this is my point of reference in my posts on this topic. Got to run for now, but will follow up on this soon.

[buffo]

I am referring to a unity gain single-ended input to balanced driver output

Yep. You'll notice that the firs op-amp in your circuit has no feedback resistor. It's configured as a voltage follower. Whatever voltage you have on the non-inverting input of that first op-amp will be present at the output. Notice also that there is only 1 input connection to your circuit.

In contrast, a differential receiver needs two inputs, and in order to get unity gain you need to make sure all the resistors are equal. Example:




As long as R1 through R4 are all the same, the output voltage will be exactly equal to the difference between the inverting and the non-inverting inputs. This is very advantageous for noise reduction. By using differential signalling, any noise that might end up getting into the wires leading to the receiver will be present on BOTH inputs in equal amounts. So when you subtract them to get the difference between the two inputs, any noise is automatically cancelled out. (The example on the right is not something you would actually build - it's there just to illustrate how the circuit works for those common-mode noise signals.)

This is a portion of a schematic image uploaded by mixedgas depicting a unity gain balanced line driver.

Makes sense. And yes, a circuit designed similarly to the one you posted will work for the original purposes mentioned in the first post in this thread.

However, my circuit needed the differential receiver front-end because it needed to capture the differential output from a controller, allow for gain and DC offset to be adjusted, and then a new differential output created. That's why there are 3 op-amps in my circuit.

Adam

[lasermaster1977]

Yep. You'll notice that the firs op-amp in your circuit has no feedback resistor. It's configured as a voltage follower. Whatever voltage you have on the non-inverting input of that first op-amp will be present at the output. Notice also that there is only 1 input connection to your circuit.

In contrast, a differential receiver needs two inputs, and in order to get unity gain you need to make sure all the resistors are equal. Example:




As long as R1 through R4 are all the same, the output voltage will be exactly equal to the difference between the inverting and the non-inverting inputs. This is very advantageous for noise reduction. By using differential signalling, any noise that might end up getting into the wires leading to the receiver will be present on BOTH inputs in equal amounts. So when you subtract them to get the difference between the two inputs, any noise is automatically cancelled out. (The example on the right is not something you would actually build - it's there just to illustrate how the circuit works for those common-mode noise signals.)



Makes sense. And yes, a circuit designed similarly to the one you posted will work for the original purposes mentioned in the first post in this thread.

However, my circuit needed the differential receiver front-end because it needed to capture the differential output from a controller, allow for gain and DC offset to be adjusted, and then a new differential output created. That's why there are 3 op-amps in my circuit.

Adam

Very familiar with op-amp circuitry and variations, just confirming ILDA's balanced line driving for XY and RGB. I do like your implementation for being able to adjust gain and offset in the differential signal chain for a given projector in a multi-projector chain. My other point was about your showing within this discussion thread that there was 20v p-p signal out of an ILDA differential output. My point is that it is not possible to measure this 20v p-p value at the diff. output, and that at best it is a conceptual idea, since empirical maximum measurements across the ILDA + and - outputs is never more than either -10v or +10v due to the way a balanced line driver works, or when measured across either output to power supply common it is never more than +5v on one output and -5v on the other (or visa-versa) yielding a full signal swing of 10v p-p maximum.

I'm taking a TTL signal from an Apple IIe or IIgs and using it for only "beam ON/OFF" for each RGB channel, buffering it with a 74HCT244 octel chip emerging as a very clean 0-5v pulse, then scaling down by half yielding a 0-2.5v pulse then sending this through the basic voltage-follower/unity gain inverter balanced line driver circuit. The output of differential "+" when measured (looked at with a high-falluten scope) to common is a 0 to +2.5v signal, and from the differential "-" when measured to common is a 0 to -2.5v signal, and when measures across "+" and "-" outputs is a +2.5v to -2.5v signal. Further, this differential output signal when converted to a single-ended output yields a 0 to +5v signal with respect to common. So I should be good using this method for driving RGB channels ON/OFF, if a given and supposed ILDA compliant projector, is working already properly when driven with Quick Show 3.0 or 4.0 image signals through Pangolin's FB3QS USB DAC or with LSX v20180416 images using an EthernetDream2 DAC, right? (I hope, I hope)

[buffo]

Very familiar with op-amp circuitry and variations

Fair enough. Sorry if I was being overly didactic.

My other point was about your showing within this discussion thread that there was 20v p-p signal out of an ILDA differential output.

That's not actually how I explained it. I'll admit that I used the phrase "20 volts peak-to-peak", but I also clarified exactly where that number came from. Specifically, I said you get 20 volts peak-to-peak at the output of the differential receiver. I never meant to imply that you could put an oscilloscope at the output of an ILDA controller and measure 20 volts across the X+ and X- pins. (OK, if you had a dual channel 'scope you *could* eventually determine that you'd get a 20-volt max swing, but you'd need to interpret the traces and do the math in your head...)

This is a topic that has come up multiple times in various threads in the past. Most people understand that "+/- 5 volt differential" means you will never exceed 10 volts between the two pins at the source of the differential signal, but they forget that both +10 and -10 are possible, so at the receiver end the total swing can be up to 20 volts peak-to-peak.

My point is that it is not possible to measure this 20v p-p value at the diff. output

Completely agree. You need a differential receiver to measure it.

I'm taking a TTL signal from an Apple IIe or IIgs and using it for only "beam ON/OFF" for each RGB channel

Woah! Wait just a second... Alden? That you? If so, then I've got egg on my face! When I suggested to Neil that you check this thread out, my intention was for you and Andrew to discuss things... He's way, *way* better at this than I am!

The output of differential "+" when measured (looked at with a high-falluten scope) to common is a 0 to +2.5v signal, and from the differential "-" when measured to common is a 0 to -2.5v signal, and when measures across "+" and "-" outputs is a +2.5v to -2.5v signal.

That will work just fine. However, unlike the circuit you're describing, the ILDA X and Y position signals we've been talking about in this thread are not limited by zero. They can swing positive and negative, which is why you get the wider voltage range at the output of the receiver. But yeah, for color signals the plan you have will work provided that both outputs are prevented from crossing the zero point.

This differential output signal when converted to a single-ended output yields a 0 to +5v signal with respect to common. So I should be good using this method for driving RGB channels ON/OFF, if a given and supposed ILDA compliant projector is working already properly when driven with Quick Show 3.0 or 4.0 image signals through Pangolin's FB3QS USB DAC or with LSX v20180416 images using an EthernetDream2 DAC, right? (I hope, I hope)

Absolutely. It should work on any ILDA-compatible laser projector.

Interesting side note though: Many modern controllers no longer implement differential signalling for color, in direct contradiction of the ILDA standard. Since nearly all laser drivers use single-ended zero to +5 volts for color modulation, companies started deviating from the published ILDA spec and began using single-ended signaling for color. In fact, I have several QM2000 boards in my collection that actually have the negative color signal pins internally connected to pin 25 (ILDA ground). But they work just fine with all my projectors.

Adam