Brownouts and motor power

propulsion

#1

Hi everyone,

Problem: I'm having the usual problem of communication blackout when putting more power into motors. I am flying using a keyboard on throttle setting *one*, yet the communication is still very fragile and drops occasionally. I tried a stress test by attaching ROV by a piece of wire and driving forward - after 4s voltage goes down on the cockpit and I can't steer the ROV (even at slightly less than 11V). If I'm lucky this lasts only a few seconds before I regain the control and the video is still transmitted. If I'm less lucky, then I need to reset the rov (USB) or even open-close the battery tubes. The soft freezes tend to be less related to the motor load, the hard ones are. Very reproducible, even on fully charged batteries.

ROV version: 2.6
Batteries: TrustFire 26650

Is there anything I can do to limit power provided to the motors? I was thinking about reprogramming the ESCs with something other than "Rock Crawler". Does it make sense?

Problem above aside - I just did open water testing last week and it turned out quite positively. Other major problem I had was fogging, but I guess I haven't yet put enough desiccants inside.

Cheers!
Paweł

292-20140727_081049EFFECTS.jpg (1.26 MB)

#2

Hi Pawel. You seem to be having problems very similar to mine. I eventually found that most of them could be solved by software modifications. Look at my comment and Walts reply here:

http://openrov.com/forum/topics/rov-work-on-land-but-not-i-sea-water?commentId=6365107%3AComment%3A93601

In particular the software modification #243 was important. Without this, a simple right or left turn ("yaw") movement would just ignore the thrust level setting and use maximum power. This would trip the TrustFire batteries every time, causing a hard freeze. You can check your rovpilot.js code to see if #243 has been implemented: https://github.com/OpenROV/openrov-software/pull/243/files. (The important thing is to have a muliplication of positions.yaw by this.power. The 1.5 factor actually makes the tripping threshold lower.)

FT


#3

Thanks for the pointers! I'll play with the software then and report back within a week or so.


#4

This issues of brown outs has been discussed a number of times with various solutions proposed and or implemented (change battery types / software warnings or limits / settings to limit motors) and not trying to tell anyone to suck eggs but

As a thought what is the possibility of a UPS type device for the Beaglebone to cushion or prevent these brown outs??

Potentially a supercapacitor on the 5V rail which could power the Beaglebone for say about 10 seconds?? I understand it may need a resister to prevent too great a current on the inrush charging but this may overcome several of this style issue

Thoughts


#5

I know, the last post is rather old, but I wanted to comment on that.

We had the same issue with our ROV 2.6 but with total blackouts when turning twice (one for each battery tube to die). Saline environment seemed to increase the issue. Therefore we potted the motors with epoxy and changed the Arduino code regarding the thrust factor while doing yaw motion. That solved the problem and we had no break down since then.

An UPS for the BBB would be cool but most likely not work since the batteries need to be disconnected and reconnected to overcome the safety circuit shutdown. I am actually not sure but I would guess this is not possible to do when being submerged and without physically opening the tubes. Other than that, this wouldn’t help with the overpowering issues since you’d need to do it after every second turn…