Analog modulation connections

[DZ]

Common mode rejection ratio

Sounds like a personal problem to me...


Guess I'll have to log back in after the fun and see what the hell you're talking about...

[Dr Laser]

I cud help if I cud ridd

[DZ]

With respect to what DZ pointed out -- the earlier recommendations made in our help file, I have only four words: "Hey man, live and learn...". OK, five words .

After all, if I am full of bullshit, why is it that you will find discussions on the internet about "single-point grounding scheme" and "ground bounce"?

As I wrote in a previous post, you can follow my recommendations or not. It won't hurt my feelings or my reputation one way or another. And doing the experiments is part of the fun and life experiences!

Well Bill, I guess then there is no point in further discussing this with you. Fact is, I'm not going to rewire my projector simply because Cambridge says so. I'll rewire it, if there is a problem that warrants my time rewiring it. If you still think that my projector produces "crap images", then so be it. Half my problem here, is that I did follow your recommendation. I've always appreciated and respected your advice Bill, but I want to see proof on this one. I contest that, if using proper gauge wiring and very short runs, as in the case of my projector, the longest wire run is about 16 inches with most others being about 6 inches, then there is no difference electrically. I can say that I know of at least one person here that has done a test of removing the pin 25 ground connection going to the amps and there was no apparent change in image quality. A test that I will try myself when I have time later this week.

[buffo]

DZ: It may be that on our projectors, where we have short lengths of heavy gauge wire and beefy power supplies, this problem doesn't really manifest itself.

However, Clandestiny has already pointed out that he's run into this problem in the past on some of his equipment. (Typically though, his projectors are much larger, with longer runs of wire both inside the projector and to the controller.)

It's true that I pulled the ground connection on my unit last night and didn't notice any difference. However, the power supplies that I'm using to drive my scanners are *seriously* overbuilt. (They're each capable of delivering 24 volts at just over 5.5 amps continuous, which is more than double the max current draw for a set of DT-40's) Then too, the longest groun wire in my projector is about 10 inches long. (Most are 6 inches or less.) So that may be a factor as well.

Here's an experiment that Bill suggested: Load the quadrature square wave pattern and collapse the Y axis to zero. You should have a horizontal line with two bright dots at the ends. Look for any vertical "noise" in that horizontal line. Now put the Y axis at full size and collapse the X axis. You'll have a vertical line with two bright dots at the ends. Look for horizontal noise in the vertical line.

I performed this experiment last night both with the ground wire conencted and disconnected. I didn't notice any change. (I had no noise in all cases.) What does that mean? Probably that the ground lift / ground loop phenomenon is so minute in my projector that it doesn't affect the image. And Bill even said that if you don't see any difference in the image after you do the test, then don't bother re-wiring everything.

I'm going to look into the issue some more, because I'm almost positive that the ground on my scanner amp power supply is actually isolated from earth ground. If that's the case, then I still have a single-point ground scheme for the amps, except it's on the signal input rather than the PSU side! (oops) My next test will be to ground that scanner psu to earth ground, and then see what difference I see when pin 25 is connected to the input.

Adam

[carmangary]

Don't forget that the scanner amps are already connected to the power supply's ground via the power connector. It makes no sense to run a wire from the signal connectors ground pin to the power supplies. I think the most useful purpose for those pins would be for connecting to your signal cable shield. Just run a wire from Pin 25 of the ILDA connector to the P/S ground and that's all you need. If you have dual power supplies for you scanner amps then just run a wire to one of them and tie the two P/S grounds together with a nice guage of wire. If your ground fluctuates when you have a hard movement then perhaps your P/S sucks.

[buffo]

Don't forget that the scanner amps are already connected to the power supply's ground via the power connector. It makes no sense to run a wire from the signal connectors ground pin to the power supplies.

Gary, you misunderstood me. I'm not going to connect the signal input ground to the power supply. I'm going to connect pin 25 on the ilda connector to the DC output ground on the power supply, and disconnect it from the signal input ground, like Bill suggested in his post above.

Right now I don't really know what potential the ground on my scanner amp PSU is at. The outputs of those power supplies are completely isolated from the inputs. So it could be floating at just about any potential.

In fact, I suspect that right now, the entire mess is actually grounded through the input connector - through pin 25, and all the way back to the pangolin board.

However, Bill made another good point. All my lasers are mounted on a metal plate. The scanners are also mounted on that same metal plate. And the outside of the scanner case is at ground potential. Likewise, I'm pretty sure the metal cases of my lasers are all common with the ground (or negative) side of those PSU's, as well as common with the negative blanking leads... So everything is already connected in one great big ground loop inside my projector! Not sure if this is really an issue or not though. (For sure it's not affecting my image quality. Not that I've been able to notice, anyway.)

If you have dual power supplies for you scanner amps then just run a wire to one of them and tie the two P/S grounds together with a nice guage of wire. If your ground fluctuates when you have a hard movement then perhaps your P/S sucks.

I *don't* have any ground fluctuation, but I think it's because my power supplies are exceptionally robust. As for the ground on the output of the PSU's, like I mentioned above, it's completely isolated from earth ground. But I'm thinking that I want to go ahead and ground it now...

Adam

[carmangary]

Sorry for the confusion. My reply was just to the general conversation and not specifically to your last post.

[buffo]

Ahh.. Gotcha

I'm really curious as to how this will pan out though. I've got a couple different wiring schemes I want to try.

Actually, I suspect that my projector is so small, and the wiring runs so short, that I won't be able to notice a difference no matter how I hook it up. But I want to experiment anyway, just for the hell of it.

Adam

[Pangolin]

Well Bill, I guess then there is no point in further discussing this with you. Fact is, I'm not going to rewire my projector simply because Cambridge says so. I'll rewire it, if there is a problem that warrants my time rewiring it.

Exactly. Well put!

I am not saying that people should drop everything they are doing and re-wire their projector if they are happy with the images they have. Mostly trying to give advice to people who are putting together new projectors, or experiencing problems with their current projectors.

I would say about a good 30% of the improvement in the images comes from this trick of "single point grounding" for the scanners that people have been sending me. BUT, I will also say a few other things. These scanners, so far, have univerally included a power supply with "factory" wiring. The wire gauge is really pretty small -- something on the order of 20 gauge wire.

Let me see if I can illustrate something verbally. I plan on making a picture to show it, but maybe I can do it verbally as well.

Picture this. When a 6800-like scanner (all Chinese scanners) undergoes hard acceleration, as it does with a squarewave input or "large seek", the scanner will be receiving 4 amps. If the scanner amp does hot have H-bridge drive (no Chinese amps do), then this 4 amps will be conducted between the power supply and the "ground" connection of the scanner amp.

Now, lets say that the ground wire that goes from the power supply to the scanner amp has 1/4 ohm resistance. I think that's reasonable given the gauge and length of the wire. (Note that for people with a digital meter, this resistance would be imperceptible -- it would look like zero ohms). Well, with 4 amps going through this wire, it means that the ground is LIFTED on this amp by 1 volt!! during hard accelerations.

Now, lets say that you connect the scanner amp's ground to "somewhere else" -- like, another scanner amp, or to the DB25. Well, what you are effectively doing is "sharing" the ground lift. This will produce some movement in the opposite axis, and can produce color effects, etc. Basically, whatever you connect that scanner amp's ground to, will experience this "lift". If the grounds are shared, it's like a big party. Every connection will share in the lift.

Now, if you wire as I suggest, what happens is this ground lift still happens, but it happens on a "local" basis. It means -- only the axis undergoing hard acceleration experiences the ground lift. If the INPUT of the scanner amp is wired differentially, then this ground lift will be ignored.

Probably these words are not as good as a picture, but hopefully you can visualize it, and visualize that ground being lifted by 1 volt! can cause problems.

If you still think that my projector produces "crap images", then so be it.

HAH! As much as I hate to admit it, I guess "Lasershow" was right about generalizations .

Dave, as far as I am concerned, your projector produces images among the best I have seen. How about that . BUT, as Adam has alluded, this could be due to you using thicker wire and perhaps shorter runs. Note that the ground lift still exists, but to a lesser extent. Lets say your wire is 10 times as good as that being sent to me for a tune-up. Well, then you experience a 1/10 volt lift instead of a 1 volt lift. I guess a question is -- at what point does it become visible?

Still and yet, the best way to wire a projector (or any electronic circuit) is with a single-point grounding scheme. That's one generalization I'll stand by.

And Dave, what you don't know is -- how much better can your images be? Maybe it was wrong to say "crap images", universally. But I can universally say "better images with a single-point grounding scheme". How much better? Well, as I said in another post -- "your milage may vary"...

Bill