FB3 hookup - help!!!

[Lasershow]

Ok - I did some more tests. Unfortunately, the results aren't very conclusive...

First, Gary - I checked to see if there was any potential difference between the isolated ground of the Scanner amp PSU and earth ground. There was no difference.

However, as Bill mentioned earlier, since just about every device in the projector is grounded to it's case, which is bolted to the optical table, which is already at ground potential, I suppose it's not surprising that I didn't get any reading other than zero. (I didn't disconnect the scanner amps from the PSU, so my guess is that they were grounded through the scanners to the optical plate.)

Just for grins, I connected a ground wire from pin 25 to the ground on the scanner amp power supply. Then I removed the ground from the scanner input. No change in scanned images. (I looked at the test pattern, as well as the quadrature square wave pattern with first the X axis collapsed, and then the Y axis collapsed.)

Then I connected pin 25 back to the input connector on the scanner amp, and left the other ground wire from pin 25 connected to the power supply. Still no change.

Finally, I removed the ground wire going to the scanner amp power supply and left pin 25 connected to the scanner amp input. Still no change.

I should note that in my projector, the red, green and blue modulation signals are connected with the positive leads going to the R+, G+ and B+ pins, and the negative leads are connected to pin 25.

If I get *really* ambitious, I might yank those negative leads loose from pin 25 and connect them to the R-, G-, and B- pins, just to see if it makes any difference. But I don't see that happening any time soon. (I've got another big project in the works at the moment, and I still haven't done my taxes yet!)

Adam

Hello Adam

The results of their tests, show that it doesn't have differences of potentials of the two grounds and then they don't create the circulation of electric current and then the ground loop doesn't induce noises, because his equipment was projected and built appropriately and consequently the Images of the projector, are projected without distortions and then for his projector and installation, including power supplies, scanners, scanners amps, leads with quality and good gauges, etc... it don't have any interference and YES you should leave connected the ground of the DB25 ILDA Signal to the ground of scanners amps for you protect it in case there is the need of some escape of current unwanted.

Then, the ground loop is not a problem and it does not cause troubles when all of the following thing are true:
I) - None of the wires in the loop carry any current
II) - The loop is not exposed to external changing magnetic fields
III) - There is no radio frequency interference nearby

Conclusion: Nobody can generalize and to affirm that we should not connect the "ground" the DB25 ILDA signal to the "ground" of scanner amps", because each wiring is a different situation, specifies and particular and therefore we needed first to test and to verify the final performance for later decide in connecting or not.

Best Regards

Carlos

[Pangolin]

Hi guys,

If you have a dual-axis amp (as maybe Lasershow may have), then, almost by definition, there won't be a difference between the two axis because... it's all one amplifier. But on these Chinese systems where they are implemented as two, separate, single-axis amps, we definitely see a difference, especially since the Chinese often supply power supplies and wiring with their systems, neither of which are very good...

Also, I will point out two excerpts from the manual that Lasershow seems to covet so much. You will see wording "can cause problems which are hard to diagnose", and even more important "This reduces crosstalk between the two channels caused by high ground currents". The latter implies that there will be "crosstalk between the two channels" no matter what.

If you do it my way, the crosstalk won't matter because it will be cancelled out by the differential receiver of the channels (with no common ground connection). If you do it another way, then, as Cambridge alludes... your results may vary, depending on the thickness of the wire used for ground, and other factors I mentioned in a previous post.

And there is one more factor. The difference will be much more noticable on amps which use single-ended scanner drive, rather than H-bridge drive, because with H-bridge drive there is almost zero ground current (at least in theory).

In other words... "your milage may vary"...

As for "generalization", we can certainly generally say that a single-point grounding scheme is preferred, and if you don't believe me, just consult the internet... A single-point grounding scheme will always give you consistent and superior results, no matter how thick your wiring is, or what kind of power supply or output system your scanner amp has. If you're a betting man, do it my way .

But if you don't want to follow my recommendations, then that's OK. Not everyone has to have perfect images .

Bill

[Lasershow]

Also, I will point out two excerpts from the manual that Lasershow seems to covet so much. You will see wording "can cause problems which are hard to diagnose", and even more important "This reduces crosstalk between the two channels caused by high ground currents". The latter implies that there will be "crosstalk between the two channels" no matter what.

I don't covet the manual, I usually try to follow the manufacturer's orientations, mainly of renowned companies and the excerpt "Can cause problems which are hard to diagnose", it refers the contact resistance in the crimp caused by dirty or oxidized pins or wire, or by a worn or improperly used tool, or by excessive stress on the wire and "This reduces crosstalk between the two channels caused by high ground currents", it refers the Power Connections and it is informing to connect a short, heavy ground wire between the two boards to avoid strong coupling between channels and the CT information are appropriate.

As for "generalization", we can certainly generally say that a single-point grounding scheme is preferred, and if you don't believe me, just consult the internet... A single-point grounding scheme will always give you consistent and superior results, no matter how thick your wiring is, or what kind of power supply or output system your scanner amp has.

It is evident whenever, "ground loops" there are in an installation, resulting in interferences/noises, will be necessary eliminate them and there are four basic methods to dealing with grounds: single point, multiple point, floating, and telescoping shield, each has specific advantages in different types of installation.

Carlos

[Pangolin]

The excerpts I took from the Cambridge manual are entirely relevant for ground situations, just as they are for bad crimps and connections. But if you don't understand, that's fine. Really.

For what it's worth, Pangolin is also "world renowned", and I personally have consulted for Cambridge on numerous projects, including on the 6800 and 6215 scanner projects and related amps.

And for what it's worth, I think the recommendations that you are following come from the very same process that produced what were the same recomendations in our own documentation (thanks for pointing it out DZ) -- basically it was a careless and quickly-done documentation. I'll get confirmation from Cambridge on this point. Maybe if it comes from them you will respect it more?

As an aside, from your attitude (and yes, I can often tell who is using what software from the attitude expressed) I imagine that you must not be using Pangolin software, and thus, half jokingly I would say that you don't deserve the great images that a single-point grounding scheme could provide. But this is of course merely a jovial comment. I have helped out lots of Laserfreak and PL members to get the best images they can, regardless of whether or not they are using Pangolin software. Ultimately, and through trickle-down theory, everyone wins if everyone has good looking images. That's why I am so passionate about helping people -- even you Carlos .


Best regards,

William Benner

[Lasershow]

The excerpts I took from the Cambridge manual are entirely relevant for ground situations, just as they are for bad crimps and connections. But if you don't understand, that's fine. Really.

I understood and I only explained with details the two excerpts, mentioned by you.

For what it's worth, Pangolin is also "world renowned", and I personally have consulted for Cambridge on numerous projects, including on the 6800 and 6215 scanner projects and related amps.

I know that Pangolin is also a renowned Company and is evident that is important your comment.

And for what it's worth, I think the recommendations that you are following come from the very same process that produced what were the same recomendations in our own documentation (thanks for pointing it out DZ) -- basically it was a careless and quickly-done documentation. I'll get confirmation from Cambridge on this point. Maybe if it comes from them you will respect it more?

As far as I know, the Cambridge Manuals are revised and corrected, because that is printed in the manuals and I usually respect the manufacturers' information and when I don't agree, I try to elucidate my doubts and you inform me, please if CT changes the recommendations on this point.

As an aside, from your attitude (and yes, I can often tell who is using what software from the attitude expressed) I imagine that you must not be using Pangolin software, and thus, half jokingly I would say that you don't deserve the great images that a single-point grounding scheme could provide. But this is of course merely a jovial comment. I have helped out lots of Laserfreak and PL members to get the best images they can, regardless of whether or not they are using Pangolin software. Ultimately, and through trickle-down theory, everyone wins if everyone has good looking images. That's why I am so passionate about helping people -- even you Carlos .

It is evident that I want great images and I also like of helping people a lot and I am happy in knowing that you can help me, I thank you very in advance and any doubts that I can have, I will consult you.

Best Regards

Carlos

[Pangolin]

Hi Carlos,

Thanks for the comments. I genuinely appreciate your attitude. And I am not trying to be an asshole, although it might come across that way in a low-bandwidth medium like email or forum text.

Now lets get down to business. I have had an email exchange about this topic with Mike Thanos at Cambridge. I am not sure what Mike's current title is at Cambridge (he used to be called "Engineering Manager", but he might have a more prestigious title now), but he is basically my counterpart there. Mike has been involved with Cambridge certainly since they began making scanners around 20 years ago, and before that, he worked at General Scanning. Although he is too modest to say so himself, I would consider him to be a master circuit designer. If memory serves, Mike actually wrote the original manuals for the 6800 and for the original Cambridge CB6580, and his wording has probably been propagated by whoever it is that is writing Cambridge manuals today.

I discussed this situation with Mike, along with my current "single point grounding scheme" recommendations. And, for space reasons, while I don't want to re-print the entire email exchange, I will point out a few relevant points in his response. Mike wrote:

> You are right on every point. We call this "coupling" and it works
> exactly as you've described. In fact, there are a number of ways to
> solve this. One way is to keep everything differential. But as you
> know, some people do use oscilloscopes to view signals. As soon as
> they do this, they add a ground loop.

Mike went on to show me the very same "Signal connections" PDF file that you had attached to your earlier post. Somehow I completely overlooked that PDF, but actually the answer is sitting right there in that document.

If you look at the diagram, it shows the shield of the cables going into pin 2 of the amp. But on the diagram, pin 2 is labeled as "N/C" which means "no connection". Now, here's the funny part. Pin 2 is not "no connection" it is the signal ground.

It was the intention to communicate that you DON'T connect anything to the signal ground of the scanner amp, although that gets a bit lost in the diagram PDF you have. Cambridge is in the process of updating that diagram as we speak, and will remove the wire going into pin 2, showing that the shield only protects the wires themselves, but should not make an actual physical connection to the scanner amp itself.

Also, I have taken another look at the picture of the printed wording from the Cambridge manual you have in your earlier post. This APPEARS to have some conflicting information, but it really doesn't. Unfortunately, the words on that page only make COMPLETE sense when they are interpreted in the context of another diagram in the Cambridge manual called "Power connections". This other diagram shows two separate, single-voltage, power supplies used, not one dual-voltage power supply as is commonly distributed with Chinese scanners. Cambridge chose to show a particular connection methodology whereby the two separate power supplies are connected to the two separate scanner amps, but I personally don't like that connection methodology very much, and it just about requires this separate heavy-gauge wire to be used.

Incidentally, the wire Cambridge discusses there really isn't related to this matter. It's there for power reasons, not for signal reasons. Nevertheless, we will see if Cambridge updates this document as well, to show that, for clarity purposes, the power supply connections could be made in a simpler way with ground connections meeting at the power supply, rather than using this separate heavy gauge wire.

Whew!! A lot of words for just one simple thing. Although this subject has been contentious and controversial (sheesh) I am glad that it has come up because it gives us a chance to review AND CORRECT our own documentation. (Thanks for pointing out these bugaboos DZ!)

And with all of this having been written, my recommendation stands. Don't connect anything to the "signal ground" of the scanner amps, and connect DB25 pin 25 only to some common, single point in the projector, preferably at the power supply.

Best regards,

William Benner

[Lasershow]

Hi Carlos,

Thanks for the comments. I genuinely appreciate your attitude. And I am not trying to be an asshole, although it might come across that way in a low-bandwidth medium like email or forum text.

My comments are with the intention of helping and you don't worry, because you are not being an asshole.

Now lets get down to business. I have had an email exchange about this topic with Mike Thanos at Cambridge. I am not sure what Mike's current title is at Cambridge (he used to be called "Engineering Manager", but he might have a more prestigious title now), but he is basically my counterpart there. Mike has been involved with Cambridge certainly since they began making scanners around 20 years ago, and before that, he worked at General Scanning. Although he is too modest to say so himself, I would consider him to be a master circuit designer. If memory serves, Mike actually wrote the original manuals for the 6800 and for the original Cambridge CB6580, and his wording has probably been propagated by whoever it is that is writing Cambridge manuals today.
I discussed this situation with Mike, along with my current "single point grounding scheme" recommendations. And, for space reasons, while I don't want to re-print the entire email exchange, I will point out a few relevant points in his response. Mike wrote:

Mike Thanos is VP of Engineering and there in Cambridge also has the Engineer Mike Danos and if you can, send the entire email exchange of Thanos for my e-mail.

> You are right on every point. We call this "coupling" and it works
> exactly as you've described. In fact, there are a number of ways to
> solve this. One way is to keep everything differential. But as you
> know, some people do use oscilloscopes to view signals. As soon as
> they do this, they add a ground loop.

If I understood correctly, Thanos says that to solve the ground loop, one way is to keep everything differential signal, in other words, the input configuration "Analog Differential", using + input, - input and maybe (????) ± input return GND.

Mike went on to show me the very same "Signal connections" PDF file that you had attached to your earlier post. Somehow I completely overlooked that PDF, but actually the answer is sitting right there in that document.
If you look at the diagram, it shows the shield of the cables going into pin 2 of the amp. But on the diagram, pin 2 is labeled as "N/C" which means "no connection". Now, here's the funny part. Pin 2 is not "no connection" it is the signal ground.

I saw this in the diagram and I think N/C means "No Charge", i.e. no electric charge, because it is ground, please confirm this with Thanos.

It was the intention to communicate that you DON'T connect anything to the signal ground of the scanner amp, although that gets a bit lost in the diagram PDF you have. Cambridge is in the process of updating that diagram as we speak, and will remove the wire going into pin 2, showing that the shield only protects the wires themselves, but should not make an actual physical connection to the scanner amp itself.

If it was really this the writer's intention, the diagram is very confused, and therefore it is wrong and it needs to be updated, please confirm with Thanos and inform any modification.

Also, I have taken another look at the picture of the printed wording from the Cambridge manual you have in your earlier post. This APPEARS to have some conflicting information, but it really doesn't. Unfortunately, the words on that page only make COMPLETE sense when they are interpreted in the context of another diagram in the Cambridge manual called "Power connections". This other diagram shows two separate, single-voltage, power supplies used, not one dual-voltage power supply as is commonly distributed with Chinese scanners. Cambridge chose to show a particular connection methodology whereby the two separate power supplies are connected to the two separate scanner amps, but I personally don't like that connection methodology very much, and it just about requires this separate heavy-gauge wire to be used.

The Diagram "Power Connections" is below and this other diagram shows two separate, single-voltage, power supplies and this CT particular connection methodology with two separate power supplies is called "Bridge Connections", because the power supplies are connected in series, but it doesn't have any relation with the diagram " Signal Connections", however it recommends to connect heavy wire between both grounds to avoid strong coupling between channels caused by high ground currents to reduce the Crosstalk.
Do you think then should not have two separate power supplies and yes a dual-voltage power supply as is usually distributed with Chinese scanners?

Incidentally, the wire Cambridge discusses there really isn't related to this matter. It's there for power reasons, not for signal reasons. Nevertheless, we will see if Cambridge updates this document as well, to show that, for clarity purposes, the power supply connections could be made in a simpler way with ground connections meeting at the power supply, rather than using this separate heavy gauge wire.
Whew!! A lot of words for just one simple thing. Although this subject has been contentious and controversial (sheesh) I am glad that it has come up because it gives us a chance to review AND CORRECT our own documentation. (Thanks for pointing out these bugaboos DZ!)

I think really very important we seriously discuss this subject and other important ones and to contact the manufacturers to update the points discussed by the members of the PL Forum, logically when we arrived in a more appropriate conclusion and with the manufacturers' agreement.

And with all of this having been written, my recommendation stands. Don't connect anything to the "signal ground" of the scanner amps, and connect DB25 pin 25 only to some common, single point in the projector, preferably at the power supply.

I will send shortly for your e-mail a Schematic Diagram with Input Power and Command Input Wiring (after I to draw) of my Micromax 671 series with 6215HB scanners for you to verify if it is everything correct the connections, OK?


Best Regards

Carlos

[Pangolin]

Hi Carlos,

You are confusing things a bit. Unfortunately, your confusion is exacerbated by the Cambridge diagrams...

And yes, it is Mike Thanos. I have known him for the past 20 years, and we have worked together on Cambridge-related projects. I don't need to "confirm" anything with him, because the facts are what they are. And any such further "confirmation" would only be a waste of everybody's time, and I am not interested in wasting their time. They are in the process of updating the "Signal connections" diagram as I already indicated, and once it is completed I will post it here.

For what it's worth, NC definitely means "no connection" or "not connected", and I am sure many others on PL will back me up on this one... If you don't believe me, look it up on the internet...

And yes, Mike is in full agreement with me. Only power connections should be "common" between scanner amps. The signals should be differential, just as I have been saying throughout this post. And no "signal ground" should be used.

This thick wire in the Cambridge diagrams is not something I like very much and I was amazed when I saw it in the diagram. At best, we can call it an "appendix". And we get a clue that this is an "appendix" by the fact that -- there is no real physical connector for it. On some boards you use a "spade lug", on others you solder a wire between this point and that point, etc.

You don't need the thick wire, if you simply "meet the power supply" at some central point, and then feed everything off of the centralized power supply connections. That's the cleanest and most understandable way to do. BUT, I admit that it does take more wire. The Cambridge method of tying power supplies directly to the amps basically requires less wire. Maybe that's why they did it that way. But there is no real central meeting place. The thick wire tries to create a happy medium, but... I just don't like it.

I won't forward Mike's email to you in its entirety -- not because I am ashamed of anything, but because I don't go forwarding the emails of others without their prior written permission willy-nilly. If you are interested, I will, however, forward the email I sent to Mike, so that you can see my points (all of which have been previously posted here), and see that I have not distored any facts. With my entire email and Mike's comment that "yes, Bill you are absolutely right on all points", hopefully that will give you a comfort level.

If you send me a diagram through email, I would be happy to give my comments, just as I am always happy to help out others here...

Nevertheless, all of this is really simple math guys... Only the power supply should be "common" among the scanner amps. On Chinese systems, this is natural, because they give you a single power supply that gives both voltages. Then you feed the scanner amp purely differentially (whether you are driving it that way or not) and connect DB25 pin 25 to the central grounding place, either on the power supply (in the case of a single power supply system) or at the power supply meeting place -- i.e. the central point. That's why it's called a STAR grounding schem, because all kinds of wires are meeting in a central point.

I do hope we can put this subject to bed now...

Bill

[Lasershow]

I do hope we can put this subject to bed now...

Ok, Bill we will put this subject now to bed, but it can be wakened up..... and as soon as I to draw the Diagram, I will send for you.

All the Best

Carlos