Spectrum 3847 - Build Blog 2020

That would be the same camera, the other driver camera will be out the other side of the robot.
Ball tracking for auton, and loading bay tracking is for auto aligning to the loading bay while cycling.

Not sure yet, those are still to be designed. Last year our driver camera was mounted in a 3D printed mount attached to a 7/8" tube sticking above our cargo/hatch intake and it worked well. Here is last year’s camera mount file -

Opensight H264 streaming seems very promising in intial testing, haven’t run it on a real field clearly.

Plan is for both the drive views to be streamed back, hopefully we won’t need to see the limelight view.

Initial tests were just done on a Raspberry pi 4 - 2gb


Awesome, thank you! So you’re not doing vision processing on the other driver camera, it’s just for driving?

We don’t have plans for anything on the other camera, it’s on the same side as the limelight, so the limelight should take care of that. If for some reason the students on the team figure out a way to get better performance from a custom setup than the limelight we could switch to that but it’s not on the to do list at this point. Limelight has worked well for us for 2 seasons now (only really tracking anything in 19, just a driver camera in 18).

1 Like

Hi there, I’ve been doing quite a lot of work with opensight and it’s been working well however I see you don’t seem to be using the FindAngle function, every time I try using it, it seems to be crashing the script. Any ideas why this might be, are you having the same issue? I have attempted to contact a dev through the discord but they don’t seem to know what the problem is surprisingly.

What method/product do you find effective for adhering the thin PTFE sheet to your components?

Amazing that you can keep this blog so rich with up-to-date info in build season, while effectively running at least two in-house teams (competition bot and everybot). Thank you for the community efforts.
3245 remains ever-thankful to Spectrum for sharing their 2013 frisbee shooter design - we adapted it and it led to some great results for us!


I don’t think we have tried to use it, sorry we can’t help.

We haven’t done much yet, in testing we have used thin double sided tape. That will likely be the final solution as well, we did it in 2017 to keep rubber on our hood and it worked with out issues. The stuff we have in the shop is called QingLong Tape (we call it dragon tape), we got it on Amazon.

1 Like

You should try posting this question in an OpenSight thread (e.g. the latest release thread) or our Discord server.

This will be fixed in v0.2.1 which should be ready in approx. 30 minutes, sorry about that.

Day 25

Design Review # 4



What do you mean you don’t need the middle passage with a belt? Does this mean just the wheels will suffice?

We had a belt on the floor of the V funnel in our design, that no longer exists. The belts on the sides still exist.

Cool that looks really good… we tested for the same thing with similar results

1 Like

I believe you guys use dacron sail material for your bumpers. Have you done any testing to see if that material helps to prevent the power cells from getting sucked under the frame when you run into them while driving?

1 Like

No we haven’t, we are assuming a low slippery bumper is better than a higher grippier bumper. That assumption could be wrong we haven’t tested.

1 Like

Yes, it seems like the sail cloth might be better. They also make this dry lubricant that can be applied to sail cloth. I wonder if that would work even better.

Let me know what your results are after you test your bumpers. My team talked to me about making the switch to sailcloth after talking with your team at Houston last year. Maybe this is the year.


I just posted a new thread about this last night. Maybe that is where you got the thought from but if you haven’t seen it I would and if you do have results please post them there!

1 Like

How does your friction break work? I don’t see what the piston is pushing.

It just engages the teeth on the VP ratchet slice directly, it’s not really a friction brake, it’s an interference latch.

I’d Q&A if lubrication on bumper fabric is legal before recommending that teams do it.


I assume you are referring to R10. That is a good point. I did not really consider that this product would fall into that category. I will try to get a question off to the Q&A.

I’ve used this product in the past for some very different applications and it does not rub off in any way in those applications. I would want to test it on the bumper material and make sure it has the same result here. And I will ask in the Q&A about it.

Thanks for thinking of that.