Our team is looking at a method to measure the distance our telescoping arm has extended. We have some laser distance sensors which would work for our purposes (specifically the Panasonic FX-102). The idea is to mount the laser on the main frame of the robot and point it along the arm at our grabber to measure its extension. The laser would never point outside of the robot and the only way the beam could hit something else is if that something moved close to the arm and broke the beam. According to R203 d, examples of items that will violate this rule include exposed lasers other than Class I. My question is this: does our setup count as an exposed laser? Will this break the rule?
I have a feeling it would count as an exposed laser but it would be great if it didn’t!
Sounds like a great question for the official Q&A to get and official answer rather than unofficial answer here on CD.
Thanks Chuck, will do!
R203d says only class I are allowed, but I agree, it is worth asking.
If it ends up not getting approved ik the RI3D Redux team used this Time of Flight Sensor to measure their elevator extension.
You should definitely ask the Q&A question, but I wouldn’t count on the answer being what you want. The likelihood is that this will be considered an exposed laser, since it would have to pass the beam through the atmosphere outside of the robot’s elements, even if it’s still within the frame perimeter. If you could run it up the inside of the arm, for instance, where it would be entirely contained within a solid structure, that might be okay. But if there’s even a chance that it could go beyond the robot, they’re likely to rule against it. I’d suggest that you begin exploring other ways of doing the job to be on the safe side.
Like Chuck mentioned, the official answer can only come from the Q&A (or your event’s inspectors if the Q&A doesn’t comment on a specific design) - though I think I can try and explain what I think the answer will be and why.
tl;dr - I think the way you described your system would meet the definition of an “exposed” laser and not be legal, but I couldn’t rule out different ways of using the same sensor legally. You would just need to do a bit of research and make the case to the LRI at your event.
As you pointed out, R203d lists “exposed” Class I lasers as being a violation of the rule. So the question becomes, what counts as exposed?
For some background, any company that is looking to sell a higher class laser as a lower class “laser product” has to certify that the integrated product complies with the Code of Federal Regulations sections that define the different laser classes (the relevant section would be 21 CFR 1040.10, linked here). I’ll copy the definitions of a Class I laser product and a Class II laser product:
(5) Class I laser product means any laser product that does not permit access during the operation to levels of laser radiation in excess of the accessible emission limits contained in table I of paragraph (d) of this section. 1
(7) Class II laser product means any laser product that permits human access during operation to levels of visible laser radiation in excess of the accessible emission limits contained in table II-A, but does not permit human access during operation to levels of laser radiation in excess of the accessible emission limits contained in table II of paragraph (d) of this section.
Note that the definition of a Class II laser product references “human access” to the laser radiation. Human access is also defined:
(15) Human access means the capacity to intercept laser or collateral radiation by any part of the human body. For laser products that contain Class IIIb or IV levels of laser radiation, “human access” also means access to laser radiation that can be reflected directly by any single introduced flat surface from the interior of the product through any opening in the protective housing of the product.
We don’t have to deal with the extra Class IIIb or IV requirements, but we do need to make sure no part of any person can intercept a beam at Class II levels if we were trying to consider a robot as a Class I “product”. This would include work being done in the pits, where hands (and heads…and eyes…) have the potential to be anywhere in the robot that isn’t specifically guarded against them.
Like @Strategic mentioned, if you kept the beam fully enclosed a tube and sealed the ends off you might be able to make a case, but the final judgement there would belong to your local LRI. Pointed through the air at an end effector, though, probably won’t fly.
Sorry for the long and boring post, hope I explained things in a way that made some kind of sense. Looking forward to seeing what you end up with!
Pending the answer from the Q&A, I would generally consider “exposed” to be something that is available external to the device in question. A COTS CD drive would be enclosed, while a distance sensor like this would not be.
I don’t recall a question like this being asked in the past, so I hope you can get a good solution!
It never hurts to ask. But I do think you’re stuck at class 1. Have you guys looked into linear encoders? Those I feel would be more accurate for your use.
I was thinking a string pot.
Or there might be a Class 1 laser you can use.
There’s no chance this doesn’t count as an exposed laser. What’s stopping you from using the encoder built into brushless motors, or using a hex bore encoder?
We are currently using that. But it’s relative positioning. Have something absolute is way better. I would guess the laser falls more on the absolute side then relative.
How accurate do you think an absolute sensor would be here? I’m not sure how the sensor you linked works but it kind of looks like it is purely an on/off output? Most distance sensors are accurate to within inches… the motor encoder will be accurate to within a fraction of an inch.
If you want an absolute sensor, running 2 absolute encoders on slightly different gear rations may give you what you want. Or 1 absolute encoder on a big gear reduction from the main drive.
Correct me if I’m wrong but wouldn’t a string potentiometer be better in this case?
Motor encoders on the telescoping arms are in a big gear box. Spins multiple times before it is fully extended or in. So let’s say you turn the robot off and on. It won’t remember it’s position accurately vs absolute will.
You (usually) wouldnt cut the power unless:
- End of match
- Emergency stop
- Maintenance
When the robot is first powered on, what should the length be?
When would it NOT be in this position? Why?
Is it possible to/what is stopping you from “resetting” the robot before the power is on?
I usually prefer to have a manual zero button and to zero both on power on and on autonomous init. That way for practice you rely on the first two, and at comp it’ll always work with the 3rd.
Yup, i intended to lead the conversation towards that, without trying to offer this as an option.
Was just in a hypothetical moment. Disconnection in a match sometimes what ever.
If it’s disconnected to the point the RIO won’t detect the encoder, assume the arm ain’t gonna be used the rest of the match.
Ditto for the motor controller.