My "MINIRAY" Rov.

[bigbadbob]

I like how you've installed an OSD that only displays what You want.

lol. yes, as standard the minimOSD shows a lot of stuff that's meant for RC planes and the screen gets cluttered.

Of course, you're right, the rov could have drifted backwards under my boat and still be facing the same way.
in that case it should show on my fishfinder as a huge fish.
Sonar returns from hollow objects stand out like fighting dogs.... erm.....tails?
USBL would be nice I must admit, but I don't think I'll go there.

[bigbadbob]

Yet another motherboard modification today-



after another MAX485 chip and two ATMEGA 168 processors died and another MiniOSD died too, I decided to use a different RS485 daughter board with more protection and a higher Vcc supply range, this one will cope with 30V.
I've shifted the OSD board up to the controller and will add 9v varistors across the video lines to hopefully stop these pesky (and expensive) transients.
I did find a bad earth connection that probably wasn't helping, and I decided to power my board from the 12v camera power feed via a seperate 7805 regulator instead of using the BEC from one of the ESC's as I had been.
As the ESC's are powered via the 12v-48v and 48v to 12v dc-dc converters I figured a 48v transient might be creeping through the BEC's own 7805 regulators.
my old analog scope is not up to displaying transients. I'd need a modern digital one.

I've also added an analog joystick to the controller since it finally arrived from China. It has a good feel to it and is very controllable.
Much nicer than the ps2 one. the spring tension increases progressively the further you move it from the center position which is nice.

[bigbadbob]

Another problem popped up today....
Powering the ROV arduino from the camera supply means I had to link the arduino ground and ESC grounds together inside the ROV,
this causes a ground loop and I get thruster interference on my video signal which is what I was trying to avoid by using separate
propulsion and electronics grounds.
I think the solution is to use opto-isolators between the Arduino outputs and the ESC pulse inputs.
I'll give it a try and see how it goes.

[Oddmar]

When the FPV guys have video interference they put a "sander-style" filter on the power line.

https://www.youtube.com/watch?v=xDUoxsMpfYQ

https://www.rcgroups.com/forums/showthr ... -LC-Filter

[bigbadbob]

Thanks for the hint Oddmar.
my old board had a similar filter on the camera power lines which helped a lot on that design.



I might go back to it if I find I need it after opto-isolating my ESC pulses.

Later the same day...... opto-isolators fitted... problem solved.
Hopefully I don't kill any more microprocessors now.

Doing some research on RC forums it appears the max7456 chip on the MinimOSD board is prone to voltage spikes from the BEC output of ESC's.
Apparently there is now a new version available that will handle these spikes but I'd rather keep them away from the rest of my electronics by isolation
rather than absorbing them and I've hopefully managed that now.
No failures yet..... hahaha... watch this space.

[bigbadbob]

I found a handy toolbox that MiniRay fits in nicely.



and also sorted out my pressure compensator.



The ROV is fed with CO2 from the surface down my pneumatic pipe in the umbilical.
there is a 1 bar relief valve in the bottom of the hull which will limit the internal pressure to 1 bar above water pressure.
As the ROV surfaces the excess pressure will be automatically vented to prevent over pressurising the hull.
I'll convert to nitrogen rather than CO2 to reduce fogging up the camera dome and I'll have a regulator on the bottle rather than the flow valve that's there at the moment.

now I need to figure out how to automatically pressurise the hull. it's manually controlled at the moment.
hmmm..... strokes beard in a thoughtful way.....

maybe one of these mounted on the outside of the hull, with the adjuster and spring removed so the water pressure acts on the regulator diaphragm and opens the valve when the water pressure is greater than the hull pressure.

[Oddmar]

No, because the air pressure is far greater than the water pressure. Unscrew the adjuster most of the way, if that's not enough, replace the spring with a weaker one.

[bigbadbob]

Cheers Oddmar.
I found that unscrewing it completely closed the valve so you're right, it needs the screw in a couple of turns to balance the upper and lower springs.
I also found it's a self relieving regulator so I don't need the blow off valve in the hull but it's in now so it can stay there.
This also means the water pressure acts on the diaphragm through the pressure relief vent which is what I need.
A quick dip in the tank shows tiny bubbles coming from the thrusters so it's doing the job of keeping the water out of the wiring.
I think these bubbles are hydrogen as my Nitrogen bottle contains 5% hydrogen for leak testing water pipes. hydrogen molecules being smaller than nitrogen ones, they get through the tiniest passage.
Hopefully it doesn't ignite due to sparks from my brushed axial thrusters.



I've sorted the on-screen data display too.



Which flags up a problem with my compensater..... My depth gauge will always read erroneously as the pressure sensor diaphragm dry side is open to the internal pressure of the hull. This did occur to me before but I had dismissed it, thinking the spring pressure would take care of it but of course i was wrong. so now I need to measure absolute pressure in the hull (15-87psi for 0-50m) and subtract the water pressure sensor reading to give me a depth. (hull pressure is approx 15 psi above water pressure due to the compensater.).
The problem now is that absolute pressure sensors have a narrow pressure range or poor resolution..... hmmmm...

[bigbadbob]

I've discovered a 100psi sealed gage pressure sensor made by Honeywell that will replace my depth sensor.
it's physically the same as an absolute sensor with a vacuum reference but is calibrated for an output of 0 psig in atmosphere. ie: 0m depth.
Mouser sell it here in the UK so I'll give it a go.

Honeywell part number- PX3AM1BH100PSAAX

In the mean time my existing sensor will be handy to show that my compensator is working as it shows pressure difference between hull and water.

[bigbadbob]

A few dips in the tank today to check thruster currents and adjust the shunt on my panel meter so that it shows just shy of full scale with all thusters at full throttle. so the meter gives me a rough idea of throttle but also indicates a fault if I'm drawing more than normal.
each is drawing about 6A at full chat but the vertical only needs 50% to dive at a reasonable speed and 10% to hover.
I still only have a one direction ESC on the vert so the rov is ballasted to be bouyant.
The rov electronics draw about 1A.
Now I'm just waiting for good weather and a high tide at a convenient time so I can throw it off the pier for a test dive.

P.S. the bubbles are nitrogen/hydrogen mix, I checked with a lighter when the rov was safely submerged.