DT-40 Pro + FB3 Kit having grounding issue

[DZ]

DZ, I suggest you take up the inquiry with Bill or VaThInK.

For what reason? I seem to understand it quite well. If you have a differential signal from a DAC to an amp, you need +X, ground and -X. 3 lines per channel its that simple... And despite what some folks think, -X and -Y should not be used as "grounds" for single ended amps...

I think leaving the V- floating would cause bother.

I'm not quite sure I understand how floating V- has become part of this thread, did I miss something? This is something completely different and has nothing to do with sending a signal to an amp.

The Flashback has single-ended outputs

I wasn't aware that the Flashback had single-ended outputs. I think that you can just tie the -X and -Y inputs on the amp to ground and it should work fine. I had a set of DT-40pro's for a while and I think thats what I did when I was running single ended from LD2K. Either that or I just left -X floating, I'd try with it floating first and if that looks bad, just ground it.

[buffo]

Bill Benner talks about the FB3's single ended outputs in this thread. He also mentions that the newer versions of the board (coming out soon) will have differential ouputs.

I agree that you don't want to use the X- and Y- terminals as ground if you're running single ended... (That's what caused the problem at SELEM, remember? My projector was wired that way and when I hooked it up to your QM-2000, everything got skewed sideways.) I've since corrected my wiring, BTW.

I think the Doc is referring to the X- and Y- pins being open when he talks about floating V-. But I agree, it shouldn't matter if you're running single ended signals.

VaThink;

The scanner amp is supposed to be able to work with either a single ended or a differential signal. It's quite common to connect a single-ended controller to an amp that supports differential signals as well, because there are so many single-ended controllers on the market. When doing this, you are only supposed to conenct to the positive signal input and the ground pin on the amp.

Getting these connections wrong will cause the image to be skewed, which will force you to apply the geometric corrections you mentioned.

However, I think the crazy jumping and jittering you're seeing is something else. Maybe it's noise in the ground (analogous to ground-loop problems with audio gear) as the Doc suggested, or maybe it's something else.

Adam

[DZ]

(That's what caused the problem at SELEM, remember?

I think there was more to it than that. Weren't you trying to use my cable on your projector? My cable is not physically wired for differential. So you would have had +X and ground with -X floating. Which sort of supports my thinking that -X and -Y may need to be grounded.

[buffo]

No, I had the splitter plugged directly into the back of your QM-2000 board, and then I ran my cable from it to my projector. So I was taking the X+ and X- signals only and applying them to the single-ended input of my DT-40's.

Your cable came off the other end of the splitter and went to your DT-40's, where you had the X+ and ground signals connected.

What we found was that my projector started displaying everything at a skewed angle. When I unplugged your projector cable, the problem didn't go away. That's when I remembered that I was wired differently than you were. (Remember that at the time I was running the QM-32, which pre-dates the current ILDA standard for the DB-25 connector.) Thus, I was using the X+ and X- pins to feed the projector, even though the X- was actually grounded at the QM-32. The pins swapped places in the QM-32 to ILDA jumper box that you saw hanging on the end of the ribbon cable coming out of my QM-32 card.

It really drove home the point that you made a while back in another post - namely that no matter if it "works" or not, you don't violate the standard wiring, because if you do you loose the inter-operability that the standard created. (If I had been wired correctly we could have run both of our projectors off your QM-2000 board and performed the "head-to-head" test much easier...)

Adam

[The_Doctor]

Got to wonder how the issue even arose, if the current FB3 and the DT40 inputs are both single ended. Seems there's a need to look closely at what is there in this case or all bets are off.

One clue that's been bugging me because it doesn;t fit is that VaThInK said he needed geometry correction to stop a square appearing as a trapezium. I know he didn't mean the modulation, those apparent distortions are just time based illusions due to the relative closeness of frame rate and noise mod rate. So I'm wondering what else he needed to correct for, and whether it was caused by an analog or digital means. We might be able to rule it as irrelevant but I think not just yet.

Edit:
I think an RS-232 breakout box might be a useful ILDA connection diagnostic and patch tool. Uses DB25, allows rewiring. Not a lot of call for those anymore, so often cheap on eBay.

[VaThInK]

I think my DT-40 Pro have differential inputs. Although I'm not certainly sure, but there are 4 input pins on each amps, marked +, G, G, -.

I have attached some pictures I took during my preliminary testing. Those are before I did the geometry correction, analogue modulation was turned off and Bill's temporary fix applied. Those test pattern looked a bit off because I was scanning them full size. If I remember correctly I think it was at around 25° to 30°.

PS: I have edited my previous post and for anyone who are curious, I also attached a picture of how my FB3 is wired to the amps.

[buffo]

Doc;

He's using the new DT-40 pro's, which support both differential and single-ended inputs. Only the original DT-40's (non-pro) had just the single-ended connectors.

VaThInk;
Have you tried hooking it up like this: (see pic)

Adam

[carmangary]

I don't have DT40s but my scanners do have differential inputs (+ G G -). If using single ended I hook X or Y to + and then connect G, G, and - all together and to gnd from my dac. Now, with a +-5v input, I will only get 1/2 deflection. So, in order to get full deflection I have to modify my DAC to output +- 10v or adjust the input scale on my dacs.

Now, if I connect only to + and G and leave the other G and - unconnected, my image looks OK but it is only 1/4 deflection.

[The_Doctor]

In completely other business, I had cause to see the word 'squegging'. Look it up, it might be relevant. VaThInK, what are your cables like? If you can, try shorter, better screened ones. If it makes a difference, you'll have something else to go on.

Basically, the idea I have is that instead of two high frequency noise sources with a slow detuning beat, you might just have one, driving your amps nonlinear to some extent. That modulation looks nonlinear enough to be a squegg. Also, the nonlinearity could explain your trapzoid error, unless that was a purely optical thing based on the shape of your screen. (It might be that even if there is a cause of nonlinearity, it might not be able to cause both effects on its own).

[VaThInK]

Buffo: I have thought of running my setup that way, but then since this is my first proper closed loop scanners and I'm pretty much a newbie in this area, I have convinced myself to refrain in trying anything that I'm not sure of. I'm afraid that I would ended up with a pair of broken scanners. Plus, the fact that Bill who I consider to be an expert in this area told me to wire them that way with the ground floating. If your suggestion is safe enough, I'm willing to try it out. I'll give an update when I have done so.

carmangary: Your last sentence seems to indicate the kind of result that I should expect if I tried Buffo's suggestion.

The_Doctor: Hmm, I don't know for sure, but I don't think it's relevant to my problem. Anyway, the cables from the scanner's amps to ILDA male connector are about 38cm long shielded. My FB3 is attached with a 26cm ribbon cable commonly used for computer data interface (floppy, PATA, etc). However I doubt that cable length or type is actually the culprit here as Bill didn't face the same issue as I am despite our setup is nearly identical, except he was using his own PSU to power the amps as I have mentioned earlier. I'm kinda confused now of what should I do next.

Does anyone know the ILDA specification when it comes to wiring differential amps to single ended DAC? Based on the response so far, it seems that there are a number of ways to achieve this and each have slightly different results. Is there actually such standard specified by ILDA in regards to this?

PS: On a side note, how do you actually measure the projection angle and Kpps rating of scanners? Is 2(1/tan(opposite/adjacent)) the right formula to find the angle? I have no idea how to measure the Kpps .