Notes from 1/19/13 Build Day


#1

Just my notes from the build day. If anyone sees something that I got wrong please let me know!

1) There are a couple parts in the Kit (mine were missing) that are intended for connecting power to the cape (which in turn, powers the beagle) and for connecting the lights. They need to be soldered on. Dave's ROV has the power that comes directly from the batteries simply spliced with wires going to the cape and the motors.



2) After doing the wiring and when sealing the electronics tube, the epoxy used is different than what was shown for sealing the batteries. This epoxy completely seals the wiring hole that is empty on one end cap, and then seals around the wiring in the other end cap. Dave used "Loctite E-90FL Hysol". The wiring harness feeds through the groove left from the last portion of the end gap and is epoxied on that side as well. Turns our this is well covered here)




3) Talked to Jim a lot about power fluctuation of the current design. He has it all laid out on a bench at the moment. He is thinking it might be good to add a capacitor to help moderate from power loss. If so such a thing could be directly soldered on to the cape.

4) The intent of having both a beagle and Arduino running in the ROV. The Beagle is for communication and video capture. The Arduino is for all of the Analog capture from sensors and for control of the motors. The Beagle is a bit expensive and apparently the Analog libraries are not as robust as the Arduino today. The Beagle could be swapped out for a Raspberry Pi for instance. The beagle and Arduino communicate through a serial interface. There was some concerns from one system powering 3V and the other 5V components, but it could be managed.

5) Just go ahead and epoxy on the LED lights.

6) The small blue part that the servo plugs in to on the end caps is epoxied in the inside with the same Hysol brand mentioned above. Leaves an opening for the syringe part.




7) The analog ports are all full on the current setup. So things like analog moisture sensors would need to probably run directly against the beagle analog inputs today. Conceptually, the idea is that it would be cheap to add another Arduino and expand the analog ports that way.


#2

Also, a real world example of what an end cap that is too loose looks like.


#3

Thanks Brian, this is very valuable. I had just used the glue gun on the lights but will put some epoxy now as well based on your advice, Eric had also said something to that effect in the video. I am about to epoxy my wire pass through for the caps and this is encouraging thanks again.


#4

At point 1 above, what parts where you missing? My kit had two blue plugs like this (not yet soldered)

From your posting I understand that the battery power terminates at left hand plug (power in) Does the cape in turn power the beagle via the connections in the two socket banks?

BeagleBone cape detail


#5

Yes on both accounts. -B


#6

Let's wait for Eric to chime in on the LEDs. Now I'm thinking he said something like "use the tape that comes with the ESCs for the LEDs" or something like that. Of course, epoxying them is one permanent solution.


#7

Well, I tested the tape that comes with the ESCs. I've tried compressing the tape to give a little bit more clearance, those small black components stand 0.5mm proud of the two light shrouds.

The whole E-chassis still slides into the tube. There is a a very small gap between the black components on the LED array and the acrylic tube wall. I'm happy enough to continue. I suspect that that gap may be enough to let light into the lens that I will consider using a couple of layers of neoprene or similar to add some thickness to the light shrouds.

(edit) sorry about the crappy shot. I've got an SLR macro rig handy, but the shot would have taken half an hour!

LEDs