Shoot holes in my estop idea

[telmnstr]

Sorry I'm so late to this party...

2 Caveats:

1) You need 2 separate attenuation methods that are both activated by your e-stop. However, interrupting the AC power supply to the projector is not an approved attenuation method. You are allowed to interrupt the DC supply to the laser diode drivers, however. Other options include a shutter, and crowbarring the modulation inputs to ground.

2) The e-stop circuit needs a separate "start" switch that is separate from the mushroom switch, such that even when all interlock conditions have been satisfied and the e-stop switch has been reset, the circuit will remain in the shutdown state until commanded to restart by the operator. (This prevents unintentional projector restarts following a momentary loss of power, or momentary connectivity loss through the cabling of the interlock circuit.)

Number 2 is the tough one. To be compliant, you must be able to unplug any cable (control, data, power, e-stop), anywhere between front of house and the projector, and the projector must immediately shut off, *AND*, when the cable is plugged back in, the projector must remain in the shutdown state until the "start" button is pressed again.

Essentially, this means you need a latching relay inside each projector.

Adam

PS: If you are only doing this for a class 3B laser, you can ignore #2, but if you're going to use this on any class 4 lasers, then #2 applies.

Hmmm. Do the current projectors being sold that meet all these requirements really meet all these requirements?

What kind of reaction time does it require to satisfy the eStop stuff? Challenge/response over ethernet valid? Might as well move it all to ESP32 or something run it on the same switches driving the FB4s.

[buffo]

Hmmm. Do the current projectors being sold that meet all these requirements really meet all these requirements?

I can only speak for X-Laser projectors, as I have done *lots* of pre-show checks with those to verify proper operation of the interlock system.

For those projectors, yes, they really do meet all these requirements. Their remote E-stops have a separate run button that must be pressed each time you want to "start" a projector. If you unplug any cable, the projector will shut down and stay shut down even after you plug the cable back in.

While I've had the opportunity to play with other brands of projectors at various get-togethers over the years, I don't recall specifically testing the class IV reset on any of them. (Well, with the exception of a few projector brands that were clearly NOT compliant {cough} LASWERWOLD {cough} as they were lacking in several areas, of which the class IV reset was just one of many deficiencies that were plainly obvious.)

What kind of reaction time does it require to satisfy the eStop stuff?

Not sure there is a published specification, to be honest. I would guess that anything on the order of a tenth of a second or faster should suffice. (Especially since the limiting factor may well be the speed of the mechanical shutter, or the speed of the mechanical relay.) Of course, a solid-state relay would be much faster, but I've seen plenty of certified projectors that use mechanical relays (and even some that used Ledex rotary solenoids for shutters).

Challenge/response over ethernet valid?

So long as you preserve the functionality of the class IV reset (total shutdown on loss of communication, and no auto-restart upon resumption of communication unless you receive a separate operator command to restart), I don't see any reason why you couldn't do E-stop over Ethernet. Figure that if you don't get a response within some short window (10 milliseconds? 50 milliseconds? something in that range...), you trigger an e-stop condition.

That being said, I'm pretty sure the X-Laser solution is hard-wired, even though they use Ethernet cable to connect everything. I seem to recall that they only use 4 wires in the ethernet cable for data, which leaves them with 4 wires for their interlock circuit. Although I also remember them talking about how they use the Mercury controller to implement E-stop on all projectors that have Mercury built-in, so maybe they are planning to move away from the hard-wired solution as they incorporate Mercury into more of their product line...

Might as well move it all to ESP32 or something run it on the same switches driving the FB4s.

Yeah, like using the Mercury controller described above.

Adam

[Robin]

Lots of great info in this thread around e-stopping in a way that makes CDRH happy. I'll throw in a shameless plug for Stanwax's ILDA breakout boards which have facilities to easily include an e-stop. Looking at the the stanwax catalogue, it appears that there is one available specifically for FB4's: https://shop.stanwaxlaser.co.uk/ilda...-fb4-632-p.asp

I used stanwax ILDA breakout boards the last time I built a couple of PJ's and including an e-stop was relatively simple: https://www.flickr.com/photos/raytracing/7406362364/

Whatever route you go down, make the e-stop button as large and temptingly shiny as possible (ya know, fer safety!). Here's some human machine interface pr0n: https://eao.com/

[telmnstr]

Buffo - yea the X-Laser stuff is pretty cool, as I remember they use an ethernet switch IC in projector to split for Mercury, FB4 I'd guess and daisy chain to next projector. Then they used unused wires (100mbps) for e-stop.

Robin - oo nice! I use 3 FB4s (externally.) I wonder if the Stanwax board shunting the output + projector power cut is enough? I see the Stanwax board has the output shunting but I assume you would need either a mech shutter or DC PSU drop?


Going the ethernet route, there are these boards which appear to now be $10,
https://www.cnx-software.com/2020/11...with-ethernet/
And me and ChatGPT could probably make a program to handle doing something like DC power cut (3 x relays) and maybe crowbar the modulation. I would do a SPI OLED screen or something for status/IP and a SPI LCD on the pendant so it could show status. But if Stanwax PCB is less resistance I am all about making things happen easier.

[buffo]

I wonder if the Stanwax board shunting the output + projector power cut is enough?

The latest version of the Stanwax board crowbars the modulation lines to ground upon interlock, and also opens the main relay (which is typically used to switch the DC power to the diode drivers). When integrated in this manner, the Stanwax board is fully compliant with the requirements of a class IV reset. Just remember to use the relay to switch the DC power, not the AC power, and you are golden.

Alternately, you could use the relay to control a mechanical shutter. But you can also use the shutter signal on the output of the Stanwax board to control a shutter, since that signal will also be crowbarred to ground upon interlock.

If the Stanwax PCB is less resistance I am all about making things happen easier.

The main reason I incorporated the Stanwax board in my product report (as opposed to designing my own circuit) is that I was informed by Casey Stack that the CDRH was already familiar with the Stanwax board and they very much approved of its use. That meant they would be far less likely to ask a bunch of follow up questions about my interlock scheme.

Rob has a nice cut sheet that explains the features of the board, and I included that document in my product report, along with a block diagram that showed how it would be integrated with the projector's internal components. It's been almost a decade since I filed that first report, and they've never asked me for any further information.

Adam