Need some help! My first RGB projector project (expecting approx. 2W system)

[taggalucci]

Layout proof-of-concept:


I'll spread this out a bit more on the 440mm x 400mm plate allowing space for the shutter and dust barriers.


A couple of rough beam combining shots:




There seems to be a fair amount of red being reflected off that dichro. There's also a little bit of blue transmission through the other dichro. I'm not sure if that's right

Just to sum up my current list of items I need to work through:

• I need to read the safety standard document to work out what each safety component needs to conform to, e.g. is a keyed kill-switch required (an integrated one where a key is required to reset the switch to "open" after it has been used; does the key switch on the projector need to work in such a way that it cannot be left on without a key in the lock; emission indicator and turn-on delay may not be required.
• I'm a little confused by Adam's interlock circuit . I understand the diagram, it's just that I'm not sure about where every switch goes physically and the function of each switch from a use-case perspective. I'm sure I can figure this out with a little more attention to it. One thing I'm most confused by is how to integrate the interlock circuit with the 3 separate laser drivers.
• A/C supply and internal distribution. Will I need internal power filters/conditioners?
• ILDA wiring integration. Done this before, but not with RGB.
• I am leaning towards a simple solenoid shutter. I need to design the mechanism and find a suitable beam stop. Have considered using a small anodised aluminium heat sink attached to the end of a "push" solenoid.
• Need to find a smaller PSU for the CB6580 scanner amps as mentioned here.. http://www.photonlexicon.com/forums/...0-scanner-amps
• Once I have the 2 shelves designed, I need to design a suitable housing. I have a few ideas regarding cooling and airflow, filers on the inlets and outlets, shock mounts, removable panels and suspension rig.

Sure was great to see the 3 powered up today

[danielbriggs]

There seems to be a fair amount of red being reflected off that dichro. There's also a little bit of blue transmission through the other dichro. I'm not sure if that's right

Looking good.
A little bit of transmission is normal, they are not 100% efficient. It would be best to get a power meter to measure exactly how much of the beam you're loosing, then you can express that as a percentage and see how it compares to the manufactures specs.
As for the red: Try to get the most out of it by angle tuning. i.e. play about with the angle of the red module and dichro angle, as sometimes there peak efficiency is not at 45 deg.

What make is the dichro?

[taggalucci]

Thanks Dan

It would be best to get a power meter to measure exactly how much of the beam you're loosing, then you can express that as a percentage and see how it compares to the manufactures specs.

Yes, I wish I had a power meter, I'm not sure I should spend the money right now. It's on the list though.

As for the red: Try to get the most out of it by angle tuning. i.e. play about with the angle of the red module and dichro angle, as sometimes there peak efficiency is not at 45 deg.

I'll have a play with that, thanks for the suggestion.

What make is the dichro?

I got these from lasershowparts.com so I'm not sure of the manufacturer. Any recommendations on alternatives?

[Insanity]

Thanks Mark for your information and advice



I've laid all the components out, considering all of the points raised in this thread. I will be drawing this out soon to verify, as well as probably building a prototype using some spare 6mm aluminium sheet (which I have lying around).

I'm proposing a 2 shelf design; upper shelf for the PSUs and drivers, lower shelf with just the lasers and optics. Each shelf with a base plate size of 440mm x 400mm. I expect the housing will have an internal frame with sheet metal panels. The base would most likely be supported by this frame (on the outer edges at least).

Given these dimensions and proposed design, do you consider 8mm would be sufficient? I agree and would like to keep the weight as low as possible.

I think 8mm will be plenty at that size....

Great looking work with that, I would like to do something similar. It doesn't look airtight though, is that/isn't that desirable? I'm assuming a design which allows a removable panel for cleaning and optics adjustment.

Thanks It isn't airtight by any means... It's basically a 2 sided perspex box with a lid screwed on.. The front and back of the box are open, but the edges of the side panels fit snug up to the front and back of the projector casing. I could use silicone sealant to make it really good, but as it is, it keeps things pretty clean. Had the projector built nearly a year now and it has done a lot of shows in clubs and *very* dusty tents at festivals. I have only needed to clean the optics once so far, so I am pretty pleased with the results. Like I said, if I made it again I would do a two box design...

Yes, perhaps for projector number 3 (or a later rebuild of this one). Until then, 635nm it is

Great shots and perfectly demonstrates the difference. I'm no optics expert, but I do wonder about reverse telescopes, using a beam expander in reverse and even an aperture/iris to cut out the fringe. Probably just some crazy thoughts and a lack of knowledge on my part I have seen a lot of info on the web about beam expanders, but nothing about beam "reducers".

640nm is still quite pricey but those prices seem to be slowly dropping. You will be fine with your 635nm if you want to do mainly beams. I used mine for a good few years and loved it... When it comes to graphics, beam profiles begin to matter more though... Most of the fringe you see in my pic are down to the divergence of the laser as opposed to initial beam size.. Like you, I am no expert on optics, so I will let someone better qualified answer your telescope/expander question...

What are the specs of each of the lasers you've used in this projector?

The lasers in my projector are:

650mw @ 640nm
850mw @ 532nm
300mw @ 473nm
500mw @ 445nm

My green has a few issues with stability and has a pretty ropey beam, so I will replace it very soon....

And here they all are, before they were all housed....


Mark

[buffo]

Looking great so far! You are definitely on the right track.

is a keyed kill-switch required (an integrated one where a key is required to reset the switch to "open" after it has been used)

The key switch is separate from the kill switch. Both are required. The kill switch does not have to have a key, but the key switch does. And yes, the key must *not* be removable in the on position.

As for the kill switch, it is designed to be a fast acting switch that will stop laser output. Thus, a normally-closed, push to open switch is an obvious choice. However, the kill switch must not automatically reset itself. This is why some people choose to use a latching kill switch (the big red mushroom switches that have to be twisted in order to be reset). The "twist to reset" feature satisfies the "no automatic reset" requirement in most cases. But there is an exception to this protection that many people ignore, and it is this exception that caused me to use a different arrangement.

In my case, I decided to implement a separate run button in the circuit. Thus, you press run to start the projector, and then if any interlock condition occurs - including pressing the kill (or "scram") switch - the projector will shut down and stay shut down until the run button is pressed again.

The reason I consider such an arrangement to be inherently safer than using only a mushroom-type kill switch is that my circuit offers protection against an automatic projector re-start if you unplug the ILDA cable and then plug it back in. Using just the mushroom switch does not. (Think about that - some girl in high heels steps on your ILDA cable and breaks the interlock circuit. You go over to see what's wrong with the unit. Then she steps away and the wire ends inside the cable touch again, and your projector comes back on all by itself without any command from the operator. Zap!)

emission indicator and turn-on delay may not be required.

These are requirements under US (CDRH) rules. Not sure about Australian standards though. They may not be required under ANSI 2211.1; you'll need to check with someone else to be sure.

I'm a little confused by Adam's interlock circuit . I understand the diagram, it's just that I'm not sure about where every switch goes physically and the function of each switch from a use-case perspective.

My circuit is perhaps overly conservative from a safety standpoint. Every control on the projector is duplicated on the remote connector. (The components that are housed in the remote connector are the ones enclosed in the box in the lower left portion of the drawing labeled "remote kill switch box".) Thus there is a key switch on the projector and also one on the remote box. Turning either key switch to OFF will shut down the projector. (Both must be on for it to function.)

There is a kill (or "scram") switch on the projector and also one on the remote box. Pressing either one will shut down the projector. Likewise, there is a run switch on the projector and on the remote box, and pressing either switch will start up the projector (assuming all interlocks are met).

Note that my schematic doesn't show everything. For one, there is a time-delay circuit downstream of the main relay for the circuit that sends power to the lasers. (The shutter signal is connected directly to the shutter driver with no delay, however.) Also, there are several indicating lights on the projector that are not shown in this drawing. (But the remote emission light on the remote box is shown.) And finally, understand that you could add extra interlocks (such as a case-intrusion interlock, or a mercury switch tip-over interlock) simply by inserting a normally-closed, open on interlock switch in series with the ILDA interlock loop. (Pins 4 & 17 on the ILDA cable.)

The key to understanding my circuit is to understand how the first sets of relay contacts are used to latch the relay in the closed position. Here's a basic explanation: Q1 supplies 12 volts to everything, assuming all interlocks are met. If Q1 is not conducting, your interlocks are not satisfied.

Pressing the run button applies 12 volts to the relay coil to pull it closed (the local projector run button applies power directly to the coil, while the remote run button turns on Q2, which applies power to the coil. This eliminates the need for the relatively high coil current to have to traverse the length of the cable to the remote box and back, which would otherwise cause serious voltage drop in the line.) Once the main relay is closed, you can let go of the run button and the coil will remain energized because power is now flowing through the first set of closed contacts to the coil.

Thus if you interrupt this supply of 12 volt power to the coil at any point, the relay will open and REMAIN open. In practice, this can be done by interrupting the supply of 12 volts to the collector of Q1 (by using the projector-mounted key switch or the projector mounted scram switch), or by removing the voltage from the base of Q1 (by using the key switch on the remote box, the scram switch on the remote box, or by interrupting the ILDA interlock loop) which causes Q1 to stop conducting, again removing power to the relay. The only way to close the relay again is to push one of the run buttons again (assuming that all Interlocks are satisfied and Q1 is conducting), which essentially shorts around the now open contacts to apply 12 volt power to the coil again.

The 330 ohm resistor is sized to limit the base-emitter current in Q1 and Q2 to an acceptable range, while the 10K ohm resistors from the base to ground on Q1 and Q2 are there to pull the bases low to ensure Q1 and Q2 stop conducting immediately upon a loss of voltage applied to the bases.

One thing I'm most confused by is how to integrate the interlock circuit with the 3 separate laser drivers.

Assuming that all three laser drivers run off 5 volts, you replace the three separate DC power supplies with one large 5 volt power supply. Then you use the main relay in my schematic to switch +5V power to the drivers all at once.

If you have drivers with different voltage requirements (say, 2 of them run off 5 volts and one runs on 12 volts), then use a 4 pole relay and run the 12 volt power for the last laser through the 4th pole on the relay.

A/C supply and internal distribution. Will I need internal power filters/conditioners?

Probably not. I don't, anyway.

ILDA wiring integration. Done this before, but not with RGB.

It's no different with RGB. The X+/X- and Y+/Y- signals go to the respective scanner amps. (Don't connect anything to the signal input ground pin between the + and - inputs though.)

The R+/R- signals go to the red laser driver's modulation inputs. Same for G+/G- and B+/B-... Connect them to their respective laser driver's modulation inputs.

All DC power supply grounds (negative DC terminals) get connected together, and those get tied to ILDA pin 25.

The ILDA interlock loop goes out on pin 4 and comes back on pin 17. (These pins should be connected to each other inside your DAC at the controller end of the cable.) Your projector sends signals (a few 10's of milliamps at 12 volts DC, if you're using my schematic) down these wires to verify that the cable is plugged in at both ends. If the cable becomes unplugged, the projector should shut down.

That's about it with regard to the ILDA wiring. Use one of DZ's break-out boards and the wiring is a snap. The whole idea behind the ILDA wiring standard is that you can take *any* ILDA-compatible controller and connect it to *any* ILDA-compatible projector and they will function correctly.

I am leaning towards a simple solenoid shutter. I need to design the mechanism and find a suitable beam stop. Have considered using a small anodised aluminium heat sink attached to the end of a "push" solenoid.

Nothing wrong with simple. Just be sure you have a spring on it to force it to the closed position when power is removed. (You can't rely on gravity - it must be spring-return.) Also, solenoids can be noisy. But if you can live with it, they will work fine. (Be sure to select a solenoid that is rated for 100% duty cycle. Not all of them can withstand being energized continuously for long periods...)

Need to find a smaller PSU for the CB6580 scanner amps

Send a PM to Dave and see if one of the ScanPro 50 power supplies would work. The voltage is right, but they might not be able to source enough current to drive those amps. (I'm not all that familiar with the different Cambridge amps, so I can't say how much is enough...)

Alternately, you could go shopping for a pair of single-ended power supplies with floating outputs and place them in series to make your own dual-ended power supply. Just make sure you get the voltage and current ratings right. There are plenty of US suppliers I could recommend, but shipping would be killer. Perhaps someone in Australia can recommend a local electronics supplier...

ADam

[Doc]

but I do wonder about reverse telescopes, using a beam expander in reverse and even an aperture/iris to cut out the fringe. Probably just some crazy thoughts and a lack of knowledge on my part I have seen a lot of info on the web about beam expanders, but nothing about beam "reducers".

If you halve the beam diameter you double the divergence and as 635s tend to have large divergence to start with this is a no go.

[dave]

I got these from lasershowparts.com so I'm not sure of the manufacturer. Any recommendations on alternatives?

They were the new laser-wave dichros, so the losses should be minimal.

Maybe I can drop past sometime to meter your lasers and check the losses of your dichros with the fieldmaxII or gentec meters we have here..?

[taggalucci]

Thanks Mark, very informative Great looking colours you have in those 4 lasers BTW.

You will be fine with your 635nm if you want to do mainly beams. I used mine for a good few years and loved it... When it comes to graphics, beam profiles begin to matter more though...

When you say graphics do you mean the projected kind where the audience is looking at the projected image, say in a water curtain, or scanned animations projected towards and over the audience?

Looking great so far! You are definitely on the right track.

Thanks Adam and thank you also for that detailed and extremely helpful description of the interlock circuit. This is very helpful as I was keen from the outset to have safety an intrinsic part of the operational design. Sharing your design with me (and the forum) has allowed me to cross check some of my design assumptions (and saved me quite a few dollars no doubt)!

I am planning on adding another switch to the remote switch box to allow the shutter-open signal to be overridden for the purpose of allowing the lasers to warm whilst shuttered. I'll have to give the wiring of this some thought.

Assuming that all three laser drivers run off 5 volts, you replace the three separate DC power supplies with one large 5 volt power supply. Then you use the main relay in my schematic to switch +5V power to the drivers all at once.

If you have drivers with different voltage requirements (say, 2 of them run off 5 volts and one runs on 12 volts), then use a 4 pole relay and run the 12 volt power for the last laser through the 4th pole on the relay.

The blue laser has a lab-style PSU and driver, and unless I were to replace this or take it apart (neither of which I'm that keen on) I would need to switch at 240v. The two CNI laser drivers require 5v @ 5A each. So I assume I would either need a relay that can handle switching at 240v, or a second 240v relay triggered by the 5v signal.

It's no different with RGB. The X+/X- and Y+/Y- signals go to the respective scanner amps. (Don't connect anything to the signal input ground pin between the + and - inputs though.)

On the CB6580 there are 3 wires for the X/Y signals, the +, - and ground. Are you indicating that this ground should not be connected to the ILDA pin 25?

All DC power supply grounds (negative DC terminals) get connected together, and those get tied to ILDA pin 25.

I'm going to draw all the components out in a circuit diagram with my interpretation of all the advice given here. I want to be really careful not to short anything - especially considering I may be needing to link two SMPSs in serial to make a dual ended supply. I'm thinking which of the 3 terminals would be considered the PSU ground in this scenario?

That's about it with regard to the ILDA wiring. Use one of DZ's break-out boards and the wiring is a snap. The whole idea behind the ILDA wiring standard is that you can take *any* ILDA-compatible controller and connect it to *any* ILDA-compatible projector and they will function correctly.

I've used the ILDA connector scheme on my little greenie too (utilising a lot less pins of course)!

Just be sure you have a spring on it to force it to the closed position when power is removed. (You can't rely on gravity - it must be spring-return.) Also, solenoids can be noisy. But if you can live with it, they will work fine. (Be sure to select a solenoid that is rated for 100% duty cycle. Not all of them can withstand being energized continuously for long periods...)

I bought one today and will be retro-fitting a spring. It will need to work horizontally. The man in the electronics shop wasn't sure about the duty cycle, and the manual's not clear. I have 7 days to return it if need be.

I'm not all that familiar with the different Cambridge amps, so I can't say how much is enough...

I'm emailing Scott Hicks at Cambridge Technology, who has provided me with some info. I'm just checking a couple of things with him and then I should be good to go. I'll post my findings.

Alternately, you could go shopping for a pair of single-ended power supplies with floating outputs and place them in series to make your own dual-ended power supply. Just make sure you get the voltage and current ratings right. There are plenty of US suppliers I could recommend, but shipping would be killer. Perhaps someone in Australia can recommend a local electronics supplier...

I know only of Jaycar and RS Components here. Jaycar are in the city and walking distance from work, but only have a limited range. RS are bigger, but way out of town. RS's online site shows stock mainly in the US or UK, so I don't know how that helps me! Their website is also hard to use, just try searching for a specific SMPS like the one I'm after!

They were the new laser-wave dichros, so the losses should be minimal.

Yeah, there seemed to be plenty of power reaching the galvos, the leakage just seemed quite apparent in the smoke.

Maybe I can drop past sometime to meter your lasers and check the losses of your dichros with the fieldmaxII or gentec meters we have here..?

Thanks Dave, I appreciate the offer, that would be very helpful I'll PM you...

[buffo]

The blue laser has a lab-style PSU and driver, and unless I were to replace this or take it apart (neither of which I'm that keen on) I would need to switch at 240v. The two CNI laser drivers require 5v @ 5A each. So I assume I would either need a relay that can handle switching at 240v, or a second 240v relay triggered by the 5v signal.

If you purchase a 4-pole relay that has contacts rated for 220V AC, it will also work for the 5 volt DC power to the other lasers. Just be sure you don't exceed the current rating for the contacts.

On the CB6580 there are 3 wires for the X/Y signals, the +, - and ground. Are you indicating that this ground should not be connected to the ILDA pin 25?

Correct. The X+/X- signals are differential. There is no ground connection needed. Furthermore, connecting the signal ground can actually induce distortion in the signal, because even a momentary shift in ground potential will be interpreted as a signal change. (And such momentary ground potential shifts can happen when you have a large change in the DC load inside the projector.)

I may be needing to link two SMPSs in serial to make a dual ended supply. I'm thinking which of the 3 terminals would be considered the PSU ground in this scenario?

If you do this, you must first be absolutely certain that the DC outputs are floating with respect to the AC ground. (That is, the DC outputs are completely isolated.) If that's that case, they you connect the - terminal from one PSU to the + terminal of the other, and this middle point becomes your new ground point that gets connected to ILDA pin 25. (Remember, with an isolated circuit, ground is wherever you decide to make it.) Now you'll have +24V from one PSU, and -24V from the other, with the ground in the middle where the two connect.

RS's online site shows stock mainly in the US or UK, so I don't know how that helps me! Their website is also hard to use, just try searching for a specific SMPS like the one I'm after!

Well, you could go with a US distributor like Mauser, Digikey, or allied electronics. But shipping to Australia might be killer. Then too, it never hurts to check on E-bay... Also, I like allelectronics.com for small stuff. They have really good deals, but don't always have exactly what you're looking for.

Might want to ask Dave where he buys his parts from...

Adam

[Insanity]

Thanks Mark, very informative Great looking colours you have in those 4 lasers BTW.

When you say graphics do you mean the projected kind where the audience is looking at the projected image, say in a water curtain, or scanned animations projected towards and over the audience?

Exactly..... Overhead atmospheric effects (waves, tunnels etc) will look just fine, you won't really notice the different sized beams at all, unless like me you are really fussy

When you project graphics and logos onto a screen, you may see a few red outlines around some of the colours, but even so they still look cool... If all your beams are the same size, everything looks spot on...

So don't worry, your 635nm will look great and very bright (Compared to say 650nm), but if like me you think, "hey what can I upgrade" then 640nm is a nice idea...

Cheers
Mark