ESC's ( thrusters) and Servo Not Working - LED's are

electronics

#1

After a full day of troubleshooting, we have managed to get the ROV control systems from nothing, to communication with the computer and a successful upgrade of the Firmware to the latest version of the 2.5 BETA. We now have full control of the LED's in the cockpit for Brightness, and the webcam image is coming though, however the ESC's seem to have no power coming to them at all, and the Servo Motor for the camera is also in the same position. We are currently using a benchtop power supply for testing ( our batteries have not arrived) and when we turn the switches on the ESC's to the on position the Power supply starts flickering and saying that there is a short circuit, and when this happens the Web interface freezes and we lose contact with the ROV.

if the ESC's are in the off position the cockpit seems to be sending signal to them to move., but neither the thrusters nor the Servo are reacting.

Any help in this matter would be greatly appreciated!

Nathan & Arnaud


#2

Hi Nathan:

Let's hit the camera tilt servo first. Is the camera tilt servo plugged into connector slot D9 or D11? The first 2.5 beta used slot D9, the latest changed that to D11 in order to work around some Arduino issues.

I think the Dozuki instructions have now been updated to reflect this.

Make sure that the camera tilt servo is plugged into the D11 slot, then let us know how things are going.

-Walt


#3

Hi Walt,

I changed the Servo to D11 and no change in response, bowever, i did some testing with a multimeter, and i'm not getting a voltage reading between the BEC and GND pins on any of the servo Outputs, but I am getting signal coming to the SIG pin. It seems like theres a problem somewhere along the line where I'm not getting power to that part of the board. I got to my lab about an hour ago ( I live in Atlantic Canada and this OpenROV is to be used for a geography Research Project), and i'll be here all day testing. I suspect if i can put 5v to the "BEC Power" pins on the board I might be able to solve part of my issue.

Nathan

Nathan


#4

Hi Nathan:

Good problem description- here's what the issue with the servo is:

Normally the servo gets powered by the battery eliminator circuit (BEC) that exists in each of the 3 ESCs. They feed +5V power back onto the BEC bus, which then powers the servo. If you have all the ESCs turned off, since you're having issues with them, then there is no +5V power being generated for the servo.

The "BEC Power" pads exist to allow somebody to use servos with this Controller Board in a design that doesn't have any ESCs, and therefore doesn't have a BEC to provide power to that buss. Apply 5V to those pads, and you should have a working servo.

Once you verify that the servo is working, we can turn to the ESCs, and once you have at least one of those working, you won't need to apply power to the "BEC Power" pads any longer.

Walt


#5

Hi Walt,

So, A lot has happened today.

First of all: the Servo and ESC's now work! The initial issues were due to my Bench top Power Supply that I was using for testing ( our batteries are not in yet) was in the wrong setting, as soon as I changed it to parallel mode they all worked!

That said, we are now in a slightly different position, first of all, the camera doesn't want to start up ( i.e. send data to the Cockpit) when the ESC's are in the "ON" position, I need to turn them off, power up the board and cockpit, and then turn on the ESCs, then it works.

And there is also another problem. Our Beaglebone has suddenly stopped working, and refuses to boot. the Power LED comes on, but the USer LED's do not. We are in the process of trying to re-flash the Microcontroller with the base software from Beaglebone, and then try again from scratch in an attempt to bring the Beaglebone back to life.

If you have any suggestions or help in any of these issues that would be great! Thanks for your help so far, it's been quite useful!

Nathan


#6

One of the issues I had early on with the camera not booting was that it wasn’t getting enough initial power to kick over. This was fixed by just swapping out batteries.

I don’t have anyone else other than the forums helping me along with this, so this may be dumbed down advice for you, but when you do calibrate the esc’s make sure you remove the camera servo jumper. If you don’t the esc’s go insane. This may be dumbed down advice or something you already know, but it took me a weekend and a fresh look at the ROV to figure it out.

Good luck


#7

I have some expertise here at my disposal, but im more so supposed to be the SME for the people from Geography for whom I'm helping assemble this ROV.

My biggest frustration currently is that we don't have our batteries yet. We are in Canada, and its not easy to get the 3.7volt Lithium Ions into the country. they should be here sometime next week. because of that were just using a benchtop supply. Initially power was my issue was well, prior to me being able to get the ESC to start up at all. I had to put my benchtop supply in parallel.

when it comes to calibrating the ESC's, people have mentioned this numerous times on this forum, and I ( to my knowledge) never completed this step. I don't know if it happened when I was trying to solve some other issues we had and was bench testing just the controller board with the ESC's on it, but I haven't had any problems. all of my thrusters respond correctly to the inputs as I give them from the software. what should happen when I calibrate them? I don't know if that's been added to the assembly instructions since I last looked at them or not.

Thanks for your help on this! it's nice to know someone somewhere else has reproduced the same situation as me, means that its likely not something I managed to break something, and more like there is a simpler fix out there somewhere!

Nathan


#8

Calibrating the ESC's wasnt in the instructions I had for the 2.5. I came across that on the software release page of Github.

https://github.com/OpenROV/openrov-software/releases

but I assume you have already found this.

In my experience before I calibrated the ESC's, when I turned on the ROV some of the motors would engage intermittently. Calibrating them killed this problem and helped them know where zero input was.

Batteries were also a pain for me. I had never seen this size of battery, but thankfully Amazon had them.

Cannon


#9

Yes, Amazon is currently the best place to find them.

Here are some links to ones on sale right now:
http://www.amazon.com/5000mAh-TrustFire-Rechargeable-Li-ion-Battery/dp/B00ADJH20E/ref=sr_1_1?ie=UTF8&qid=1381776689&sr=8-1&keywords=26650+battery
http://www.amazon.com/Mltifunctional-26650-Battery-Charger-Rechargeable/dp/B00ADJ4478/ref=sr_1_2?ie=UTF8&qid=1381776689&sr=8-2&keywords=26650+battery

#10

Hi Nathan and Cannon:

I've been away from the forum for a couple weeks, and am trying to catch up now. I can sympathize with the challenges you're having with the power supply.

One thing you can try if you think you're having bootup power supply issues is to turn the ESCs off while sorting out the other problems. The ESCs in the kit draw a truly massive current spike when they are first turned on, for I think maybe 10 usec or so. I'm guessing it's a bug in the ESC firmware (probably the output MOSFETs are all turned on at the same time) but I don't know for sure. Anyway, this current spike can't really even be handled properly by the lithium cells, and it kicks off the 5V switcher for a short period of time. The reason the output of the 5V switcher has such a huge capacitor on it is to hold up the 5V bus while the ESCs sort themselves out, and the 5V switcher comes back online. In time we will likely switch to a different ESC that doesn't have this "feature", but we've got bigger fish to fry right now, so that task is on the back burner.

If the external power supply you're using can't source as much current as the Lithium cells do, then it may be that the cap on the output of the 5V switcher can't hold up the +5V bus for the length of time that your external power supply needs to recover. If that happens, the Arduino tries to reset everything (including the ESCs), and the whole process starts all over again.

Nathan, If you still have issues finding lithium cells in Canada, you could modify the unit to run on 8 or 10 NiMh C-cells, with 4 on each side, or 5 on each side if you cut yourself some longer battery tubes. The wiring harness would then need to be changed to hook the battery tubes in series. We want to test this configuration here at the lab sometime, but again it just hasn't bubbled up to the top of the priority list yet.

Anyway, I hope this explanation at least gives you some background as to what might be happening. Please keep us posted as to how things progress.

-Walt


#11

Walt,

Thanks for your help! From reading your post I can say that the issue is likely due to power, and once we get our lithium cells in from what you're saying we shouldn't have any problems. Our cells are on their way, finally in the country (they had to come shipped from china) and should be here early this week! I'll make sure a post with some pictures will be up once we have a successful test!

Nathan


#12

a cheap solution to provide enough power (before recieving the batteries) could be to use a ATX powersupply from a computer.

these have ALOT of amps available on the 12v output.

the power supplies has 12V on the yellow wire going to the various devices.

with the power supply standalone, it can be turned on by jumping a wire between the green wire and a black wire on the 20-24pin atx connector.

guide: http://www.instructables.com/id/How-to-power-up-an-ATX-Power-Supply...

if you want a permanent bench supply you could follow a guide like this and have various voltages available for testing of electronics.

http://www.wikihow.com/Convert-a-Computer-ATX-Power-Supply-to-a-Lab...


#13

I am running a total of 10 (5 on each side) 1.2v 5000mAh Rechargeable NiMH cells. We cut longer tubes and as you say, wired the tubes in series. I opted for 5 per side because of the lower voltage (1.2 vs 1.5) of the rechargeable NiMH cells. I'm using rechargeable due to the high draw of the motors and as was explained to me, non-rechargable have too much resistance - plus I'm taking the ROV out every day or so, which would be way too many disposable batteries.


#14

Hi Laura-

I was interested to see in your videos the changes that you had made from the stock ROV, but curious as to your rationale for some of them.

Here at OpenROV HQ we've been using rechargeable 26650 LiFePO4 batteries for the last 9 months or so, and are very pleased with their performance. Compared to NiMH, they have better energy density, so for a decent battery pack capacity you won't need to use flotation on the ROV to get the buoyancy right.

If you decide to stick with NiMH for future builds, here's a couple of thoughts:

- 12V is not needed to run the ROV. The recommended LiFePO4 battery configuration is 9.6V. If you use 4 NiMH batteries per side, that would match it at 9.6V. You'd have less energy available, but also less need for flotation.

- Since the NiMH cells are all placed in series, the total vehicle current draw goes through a single set of wires to the electronics tube. I would consider upgrading the wire for the battery pack from the 20ga supplied with the kit to 18ga.

- Similarly, when you bring the power into the vehicle through the endcap and the DB-25 connector, make sure that you bridge the power across both sets of power pins in the DB-25 connector. The power pins in the DB-25 connector, and the associated blocking diodes on the controller board, are not sized for continuous full-power operation from a single channel. With the Lithium batteries, the current gets split between two separate channels.

Keep posting your videos. They're great fun to watch!

-W