Robbe Seawolf V2

I have been making a dsrv for the back of the sub, it's a rough first draft.

I had the sub out for its maiden voyage at the Largs model boat club open day and things didn't quite go to plan, I forgot to connect the control rod to the front planes, and I managed to pull a tube off one of the pumps, so when I tried to dive the water sloshed forward and hit the receiver and I lost RC, I had also forgotten to set the failsafe again as I had rebound the receiver while troubleshooting forgetting that it also reset the failsafe. So when it lost signal I was trying to dive the sub, so it ended up doing circles scraping along the bottom of the pond and I had to wade in to get it.

The reason I like the Largs pond is that the bottom is smooth, and it's only up to my knees, so there was no chance of losing the sub if things went wrong.

I didn't want to click "Like". if you see what I mean!

Heck, we have all been there. Why is everything underwater so difficult?

It will all be better next time. Stick with it.

Lovely pond, by the way.

David

I have plenty of plans for this sub, I am going to put the 2 water proof servos in the front along with a water jet bow thruster, camera in the mast, and some more work on the flight controller dive system. Its been tested in the bath again since it failed at the pond and its all working fine. I plan on doing that once the dark nights start, as I'm currently working on my solar rover.


I have been writing code to convert robot vacuum cleaner lidar scanners for use with mavlink protocol, so it can be used with flight controllers along with some other experimental radar and AI camera systems. it's a bit of a time crunch to get it working as the dark wet nights are coming. its basically made from 3 reprogrammed hoverboards with 3d printed tyres bolted to a plastic crate. It's a test bed for the electronics that are for a long range autonomous boat as it's much easier to get it all tested on land before I put it in the water.




https://www.youtube.com/watch?v=qBKi5ORgOj8

quote again lol

I have rebuilt the sub after its last trip out to the bottom of largs pond, hopefully to stop it happening again i have implimented a few fixes:

1. I have secured all tubes inside the sub, the main reason it sank was that one of the tubes got snagged by the WTC as i was closing up the sub, this pulled the tube off the pump causing it to fill the wtc instead of the bladder. so now it shouldnt snag when closing up the sub.

2, I have replaced the peristaltic main ballast pump with a geared pump, its just much faster to dive and surface, even overvolting the other pump took minutes to go under. https://www.ebay.co.uk/itm/353689843234

3, the flight controller has been totally reprogrammed with custom mixing so i can static dive while keeping the sub level.

4, failsafes have been set so it will blow the ballast if rc signal is lost.

5, I have connected the sub trim pump to the gimbal output so it will try and keep it level without using the pitch output.

once im sure it will work reliably then I will look at upgrades.  fpv camera, bow thruster, pressure sensor for inav and depth sounder.

I have the thruster and sonar to install, now that I have the software functioning I can finally get around to cutting holes. I am still not sure where I'm going to install the thruster, I would like it at the front but routing the wiring would become very difficult, fitting it at the back would be simpler but it would be less effective. The sonar will get installed on the rear of the sub at the bottom, there isn't really anywhere else it can go.

I have been giving the sub more thought on how to manage its ballast. So currently it has a main ballast bladder ran with a geared pump and a trim ballast bladder with a tiny 5v peristaltic pump at the back of the WTC for levelling the sub. The main ballast can be manually controlled but it's eventually going to be controlled using external water pressure for depth control and the trim pump can be manually controlled, but I'm going to connect it into the inav stabilised output using logic switches so it can trim the sub automatically.

After its issue last time when i couldn't get it to resurface, I have been thinking of adding an emergency ballast system, basically there is some free space outside the WTC at either end of the sub so I'm thinking I could add a bladder at each end that I can fill with air from the main WTC if it has issues with the main ballast.

I got the sub out today to see what it needed for the summer. I'm going to forget about the pumps the next time out and just get some time on the water before trying to dive.

I also left the 3s li-ion connected, and even tho it had a BMS it was totally dead when I plugged it into the charger so that will need replaced too.

Pumps, and ballast need some work, the trim pump isn't moving and I'm still looking at the barometer depth hold, so I'm just going to try dynamic diving next time I try to dive it.

When trying to cut a cable tie I managed to nick the boot for the rudder connecting rod, and trying to find large enough boots wasn't easy, I ended up having to modify one to make it fit.

To top it all off I stripped the thread on the bolt that holds the sub together, replacing that is going to be interesting as its in very securely. To be fair it was well-worn already, there was a smooth section on it when I got it so probably half the thread was gripping.


So a few issues before I can take it out again.

It's that time of year, so to get the sub ready for the open day I have replaced the battery, ballast pumps and the waterproof boot I managed to cut with a knife. It's just had its bath test and it all looks ok. so I really hope that I get more than 30 seconds this time as thats about as long as I got last time before it sank. im just going to be trying to dynamic dive it I think and il save the pumps for another day.

these little pumps are great and very cheap.

https://www.aliexpress.com/item/1005005933241656.html

well, things didn't go to plan, it was almost a total repeat of last year, as soon as it went under it just continued at speed to do circles until it hit the side of the pond and popped up again, once it surfaced I got control again but it somehow broke off the rudder linkage and that was the end of it.

After I got home I started an autopsy on the submarine, first thing was that there was no water at all in the WTC, so that ruled out water hitting the electronics as I thought happened last time. so I started testing the receiver failsafe, and when I gave it some throttle and switched off the transmitter it just kept going holding the last inputs. very bizarre, so I checked the transmitter settings and the failsafe had been set but it was acting like it wasn't. so I changed it to "no pulse" failsafe and repeated the test and got the same result, I think there is something wrong with the Radiomaster R900 receiver failsafe mechanism where it just holds the inputs rather than going to failsafe positions.


In hindsight, I totally forgot about the relay remote, i could have possibly shut it off with that and since I was just dynamic diving it would have popped back up.


https://youtu.be/AMzkQ-VsZD8

Some pictures from before it went wrong

I have ordered a new receiver, it's 500mw ELRS but i will be running a firmware called mLRS on it, I don't plan on running it aat 500mw, I suspect it would overheat above 100mw in an enclosed tube, but I wanted the extra for testing. I have been thinking, the pond is right on the shore so I think there could be salt in it from the wind blowing spray from the beach and thats why it's struggling to keep signal in only 12 inches of water and less than 20 meters away. The transmitter was at 500mw and as bad as the r9 system is it should have been able to do that.
https://www.aliexpress.com/item/1005005452264439.html


I think im going to test the water salt content next time im at the pond to make sure its not a mini dead sea, if it is then I might have to go with a radio mast.

Yes, you may well be right. Pools by the sea could well have a conductivity problem. Some of my observations on swimming pool water here:

https://www.theassociationofmodelsubmariners.com/t1783-433-458-mhz-openlrs-in-a-swimming-pool?highlight=swimming


Conductivity meters (from China) are not too expensive.

David

Geofrancis, what transmitter/transmitter module are you using with the expresslrs/mlrs receiver?

Its a frsky r9m module. mLRS isn't compatible with the esp32 based transmitter modules at the moment.

after watching the video back, I just realized that I had it in full manual mode. so it wasn't trying to auto-level, thats why it did the nosedive and popped up the first time I tried to dive, it should have stayed pretty much level if I had acro mode enabled on the flight controller. I need to make a checklist for next time, I had 5 boats on the same controller so some switch positions were set from the other boat.

Im thinking about just building the conductivity sensor into the sub and connecting it to one of the ADC inputs on the flight controller. This rain sensor gives a 0-vcc signal depending on the conductivity. I can use a cheap EC meter to calibrate since I don't need it to be very accurate, it just has to be good enough to show if the water is going to be a problem or not and I can set my radio to play a warning alarm. To probe the outside I'm just going to use a pair of m2 stainless bolts going through the rear bulkhead as contacts.

https://www.ebay.co.uk/itm/194436974772

I use them for bilge pump sensors in a couple of my boats.

I have been testing these little pumps and as good as they were at basic tests, they have some issues.

https://www.aliexpress.com/item/1005005933241656.html

I fitted a pair to my tug to replace the washer pumps as maneuvering thrusters and I noticed some bubbles coming out. I thought it was just some air getting sucked in the intake but even after holding it down there was still bubbles coming from somewhere. after dissecting one of the pumps I found that there was no shaft seal between the impeller and the drive gear. it was just a stainless rod in a plastic shaft with a trace of grease. Im sure if I just add more silicone grease on the inside it should be airtight for the pressures we are at. I have also ordered some rubber seals for the shaft to see if that will help.

When I was at the pond I got a chance to test the sonar on my tug, Its based on the sr04 that I was using before but it's specifically designed for use in water now so all the units are correctly scaled. I'm going to add one to the sub to avoid the bottom
https://www.aliexpress.us/item/1005005245293992.html

I have been trying to get everything connected to the flight controller but the omnibus f4 pro board I am using just doesn't have enough outputs. I thought 5 would be enough but with a motor, rudder, forward planes, rear planes and now a dive pump and trimp pump it just wasnt enough. and now im not using a reciver with servo outputs i cant connect anything directly to it and I also wanted to add some other stuff so it was becoming a problem.

so I have ordered a Speedybee F405 mini Wing controller its basically the same chipset and software but it's a much newer design with 10 servo outputs powered by an internal 4A BEC and native support for Express LRS style receivers that use a UART rather than an inverted SBUS input. all of that is going to make wiring much simpler.

another benefit is that it has a built-in camera switcher so I could have 2 cameras connected and switch between them. so one on the periscope and one on the sail.

A potential issue I discovered was that placing ballast in the nose and tail was a bad idea. with all that weight hanging off the ends its causing a lot of stress on the end caps od the WTC. I have removed it and I have lead strips to line the inside of the WTC to keep the weight in the center of the buoyancy.

I ordered one of these cheap EC modules, im just going to use the board with a pair of m2 stainless bolts protuding from the hull used as probes, im still debating if i should fit it to the sub or to my tug, the tug already has a sensor suit so it would just be plugging another one in. there isnt an easy way to interface unsupported sensors on inav.

https://www.aliexpress.com/item/1005001698342637.html?src=google

I have been updating my periscope idea to a digital system using a micro esp32-s3 sense camera module that's almost as small as the analogue cameras I was going to use. The only downside is it's on 2.4 so it's going to lag with other RC systems on around it. I'm going to print a more hydrodynamic fairing for it and for waterproofing its just going to get dipped in epoxy to seal it.