Lucas' SeaFox (now on version 2.2.2)
Lucas' SeaFox (now on version 2.2.2)
Ever since I saw the footage of Alvin and Jason Jr. exploring the Titanic, I KNEW I had to have my own ROV.
Version 1 (2010)
Due to cost and geographical considerations, my design goals for the V1 were:
- Proof-of-concept with basic ROV functionality
- 3 thrusters for forward/reverse/turn/dive/surface
- basic camera setup
- basic relay and joystick control
- ~50 foot design depth (MN has very shallow lakes so that covers most of them)
- ~100 ft tether to allow for horizontal movement
Despite Steve's extensive documentation on his website, I still had a number of problems to tackle as I only had access to basic hand tools, a drill, and a Dremel tool. The frame was fun because I only had access to quick-drying PVC cement. It ended up coming out "close enough" but not perfect. I'm a perfectionist so this drives me crazy - there's always the next frame!
While I was trying to make this first version as simple as possible, I also wanted to make preparations for future changes and additions (lights, gripper, etc). I also really wanted to have a waterproof but still easily accessible electronics bay (which rules out covering it in wax). The design I came up with was to arrange the relays in three rows of 4 (3 thrusters used for now, have relays ready for future expansion) and put them in a PVC pipe with the wires running through the threaded end cap.
As you can see in the photo, I quickly decided three rows wasn't going to fit so I removed one of them. I filled in the plug with epoxy and the tube would rotate around the electronics when the cap was screwed in.
I'll throw the pictures here of V1 for your viewing enjoyment:
I had a hard time trying to find an enclosure big enough to house the joysticks but also not obnoxiously huge. I was running out of time and realized a large 3-ring binder was just big enough to hold them. A little rough, but it worked!
Two big problems with V1. I used 2 inch PVC for the floats and this was just enough to get it slightly positively buoyant without any extra ballast. While it worked for testing, I would like a more comfortable margin for the future. Other big problem was the electronics housing. While a good idea in theory, the epoxy I used was not runny enough so I never even got close to sealing it up. I ended up mounting the tube with the wires coming out of the bottom as a primitive diving bell type design to keep the water out longer. After about 5 minutes it would fill with water, the relays would stop working, and the whole thing would sink to the bottom because of the lack of floatation. I decided to go for broke and fill this with wax but forgot to seal the relays so when I powered it up they clicked but the thrusters wouldn't turn on. Oh well, back to the drawing board.
The one thing that worked really well was my camera setup. I used Steve's PVC union design and it hasn't leaked at all yet! The camera is a color board camera from http://www.supercircuits.com. I've used many of their cameras for many different applications in the past and have never been disappointed! Their prices are usually very reasonable as well.
Version 2 (2011)
The things I wanted to fix for the next year were:
- floatation
- new electronics housing that doesn't leak
Floatation was easy, I just went with 3" PVC to replace the 2" stuff.
For electronics, I decided to just bite the bullet and go with relays in a project box covered in wax. I was too ambitious the first time and once the relays are working, you don't have to mess with them much anyway.
And some close up shots of V2
And a few shots in the water
Overall this setup worked pretty well except I left it sit in the sun too long one day and some of the wax melted out. I didn't notice this until later but enough had escaped that water was able to get into the relays.
Areas I want to work on for the future:
- Thrusters: I ended up running modded bilge pumps for thrusters but had a tough time getting props figured out. The setup wasn't ideal but I was on a time crunch to get it in the water and they were the only ones I could find that fit the motor shafts. They work but it could definitely use some more thrust.
- Electronic control system: I already have a Velleman 8023 kit built. I'm at a crossroads as to whether I should run that or go all out with some Ardunios, a PS2 or Wii controller, and proportional control. I have a background in IT and programming so I don't think it would be terribly hard to do it and it would be nice to ditch the huge relays. Even if I stick with the relays and joysticks, I'm going to need to come up with a more usable controller as well.
- Lights: I would like to get lighting figured out
- Tether: My tether is currently made up of a cat 5e, 18ga speaker wire, and a video wire. I would definitely like to shrink my tether if at all possible.
- Power: Going along with the tether issue and battery technology improving, I'm thinking I want to run on-board batteries.
If I run on-board batteries, I will be able to skip the speaker wire. I also recently discovered those video baluns that let you send composite video over 2 wires, which would allow me to literally run just a cat5 for a tether.
Version 1 (2010)
Due to cost and geographical considerations, my design goals for the V1 were:
- Proof-of-concept with basic ROV functionality
- 3 thrusters for forward/reverse/turn/dive/surface
- basic camera setup
- basic relay and joystick control
- ~50 foot design depth (MN has very shallow lakes so that covers most of them)
- ~100 ft tether to allow for horizontal movement
Despite Steve's extensive documentation on his website, I still had a number of problems to tackle as I only had access to basic hand tools, a drill, and a Dremel tool. The frame was fun because I only had access to quick-drying PVC cement. It ended up coming out "close enough" but not perfect. I'm a perfectionist so this drives me crazy - there's always the next frame!
While I was trying to make this first version as simple as possible, I also wanted to make preparations for future changes and additions (lights, gripper, etc). I also really wanted to have a waterproof but still easily accessible electronics bay (which rules out covering it in wax). The design I came up with was to arrange the relays in three rows of 4 (3 thrusters used for now, have relays ready for future expansion) and put them in a PVC pipe with the wires running through the threaded end cap.
As you can see in the photo, I quickly decided three rows wasn't going to fit so I removed one of them. I filled in the plug with epoxy and the tube would rotate around the electronics when the cap was screwed in.
I'll throw the pictures here of V1 for your viewing enjoyment:
I had a hard time trying to find an enclosure big enough to house the joysticks but also not obnoxiously huge. I was running out of time and realized a large 3-ring binder was just big enough to hold them. A little rough, but it worked!
Two big problems with V1. I used 2 inch PVC for the floats and this was just enough to get it slightly positively buoyant without any extra ballast. While it worked for testing, I would like a more comfortable margin for the future. Other big problem was the electronics housing. While a good idea in theory, the epoxy I used was not runny enough so I never even got close to sealing it up. I ended up mounting the tube with the wires coming out of the bottom as a primitive diving bell type design to keep the water out longer. After about 5 minutes it would fill with water, the relays would stop working, and the whole thing would sink to the bottom because of the lack of floatation. I decided to go for broke and fill this with wax but forgot to seal the relays so when I powered it up they clicked but the thrusters wouldn't turn on. Oh well, back to the drawing board.
The one thing that worked really well was my camera setup. I used Steve's PVC union design and it hasn't leaked at all yet! The camera is a color board camera from http://www.supercircuits.com. I've used many of their cameras for many different applications in the past and have never been disappointed! Their prices are usually very reasonable as well.
Version 2 (2011)
The things I wanted to fix for the next year were:
- floatation
- new electronics housing that doesn't leak
Floatation was easy, I just went with 3" PVC to replace the 2" stuff.
For electronics, I decided to just bite the bullet and go with relays in a project box covered in wax. I was too ambitious the first time and once the relays are working, you don't have to mess with them much anyway.
And some close up shots of V2
And a few shots in the water
Overall this setup worked pretty well except I left it sit in the sun too long one day and some of the wax melted out. I didn't notice this until later but enough had escaped that water was able to get into the relays.
Areas I want to work on for the future:
- Thrusters: I ended up running modded bilge pumps for thrusters but had a tough time getting props figured out. The setup wasn't ideal but I was on a time crunch to get it in the water and they were the only ones I could find that fit the motor shafts. They work but it could definitely use some more thrust.
- Electronic control system: I already have a Velleman 8023 kit built. I'm at a crossroads as to whether I should run that or go all out with some Ardunios, a PS2 or Wii controller, and proportional control. I have a background in IT and programming so I don't think it would be terribly hard to do it and it would be nice to ditch the huge relays. Even if I stick with the relays and joysticks, I'm going to need to come up with a more usable controller as well.
- Lights: I would like to get lighting figured out
- Tether: My tether is currently made up of a cat 5e, 18ga speaker wire, and a video wire. I would definitely like to shrink my tether if at all possible.
- Power: Going along with the tether issue and battery technology improving, I'm thinking I want to run on-board batteries.
If I run on-board batteries, I will be able to skip the speaker wire. I also recently discovered those video baluns that let you send composite video over 2 wires, which would allow me to literally run just a cat5 for a tether.
Last edited by Lucas on Oct 11th, 2020, 5:15 pm, edited 7 times in total.
Re: Lucas' SeaFox
Saved for future use
- bikerbones1968
- Posts: 374
- Joined: May 10th, 2012, 5:21 pm
- Location: Annapolis Valley Nova Scotia
- Contact:
Re: Lucas' SeaFox
Lucas, I absolutley LOVE your control box!!!!! Simply GENIUS
Can't wait for Version 2
Can't wait for Version 2
Re: Lucas' SeaFox
Version one looks great to me! nice job! Cant wait to see that tether shrink on version 2
Re: Lucas' SeaFox
Well it's been a few years since I updated this thread but I've been playing with my ROV off and on throughout that time. Naturally, my big plans have taken longer to come to fruition - partially from life getting in the way but also from having a heck of a time dealing with getting the relays-in-wax to work for more than a short time. My last big attempt (and highest degree of success so far) was last summer on Lake of the Woods. These pics are all from that adventure:
Wiring up the relay box (again!)
Prepping tether wires for feeding through poly rope
As I'm sure some of you have experienced, this is a massive PITA. In my case, my video wire has a BNC connector on the end that I didn't want to cut off so that made it even more fun
Finally working again!
Still rocking the BinderControllerTM as I still haven't found a better option
Test dive in 3-10ft:
Trying to retrieve a toy one of the kids dropped in the water
Teaser screencap from the GoPro video
This was somewhere in the 20-30ft depth range IIRC. Ended up with nearly an hour of footage - I'm terrible at editing but hopefully sometime soon I'll get it posted to Youtube.
Unfortunately something happened with the electronics again after this run - guessing one of the relays heated up enough to melt some wax and let water in, or something similar. I lost motivation but after visiting there again a couple weeks ago the bug hit me again! Think I'm done with this control method though, already dusted off the Arduinos and PS2 controller
More to come!
Wiring up the relay box (again!)
Prepping tether wires for feeding through poly rope
As I'm sure some of you have experienced, this is a massive PITA. In my case, my video wire has a BNC connector on the end that I didn't want to cut off so that made it even more fun
Finally working again!
Still rocking the BinderControllerTM as I still haven't found a better option
Test dive in 3-10ft:
Trying to retrieve a toy one of the kids dropped in the water
Teaser screencap from the GoPro video
This was somewhere in the 20-30ft depth range IIRC. Ended up with nearly an hour of footage - I'm terrible at editing but hopefully sometime soon I'll get it posted to Youtube.
Unfortunately something happened with the electronics again after this run - guessing one of the relays heated up enough to melt some wax and let water in, or something similar. I lost motivation but after visiting there again a couple weeks ago the bug hit me again! Think I'm done with this control method though, already dusted off the Arduinos and PS2 controller
More to come!
Last edited by Lucas on Jan 9th, 2019, 2:30 am, edited 1 time in total.
-
- Posts: 160
- Joined: Jan 3rd, 2014, 10:04 pm
- Location: Vancouver B.C. Canada
Re: Lucas' SeaFox
I've been working on mine for two and a half years and just got it wet last month. I see other peoples ideas and have to add them to mine.
I used an Otter Box for the relays/electronics and 2 Velleman PCBs for control(same as Steve T). I used a Pelican case and a $5 cutting board for mounting the joy sticks/switches.
I used an Otter Box for the relays/electronics and 2 Velleman PCBs for control(same as Steve T). I used a Pelican case and a $5 cutting board for mounting the joy sticks/switches.
- Attachments
-
- Controller.jpg (93.1 KiB) Viewed 29179 times
-
- ROV Dec 23 001.jpg (189.05 KiB) Viewed 29179 times
Re: Lucas' SeaFox
Dive footage from 2015 is finally uploaded:
https://youtu.be/L-zw6FOZA9A
https://youtu.be/CZFUeZR8JZ8
Also working on a new build - stay tuned!
https://youtu.be/L-zw6FOZA9A
https://youtu.be/CZFUeZR8JZ8
Also working on a new build - stay tuned!
Re: Lucas' SeaFox
Ok it's been awhile since I've updated this thread but I've made a TON of progress! In 2017 I decided it was time to stop messing around and set my sights a little higher. Also, Blue Robotics has been making some high quality parts at attainable prices so I incorporated some of them into this build.
The new goals included:
Sticking with PVC for the main frame - primarily for cost and simplicity
Ended up going to an 80/20 aluminum T-slot system for mounting everything else https://www.8020.net/
And to that, my polycarbonate solution for holding the WTCs in the appropriate position.
Thrusters
Went with the popular NTM 28-36 750kv brushless motors. The hardest challenge here was connecting the Graupner props (4mm threaded) to the motors (4mm smooth) as apparently no one makes an adapter for those sizes (virtually every other combination is available however). After lots of searching and hairpulling, I found these threaded rod adapters:
https://www.thorlabs.com/newgrouppage9. ... up_id=1745
After consulting with my fabrication friend, we decided the threaded end could be drilled out and another set screw added for gripping flats machined into the motor shaft.
Control System
As I wanted proportional control, I decided to go with a PS2 controller and Arduino Uno on the surface talking over serial to an Arduino Mega on the ROV commanding the ESCs. Ryan's (KR2_Diving) thread was a huge resource to getting a lot of this figured out. I tested with the RS-485 chips and while it worked well enough, I decided to pick up the BR Fathom-S tether board set as for the price it has a compelling set of features and simplified this area.
A small interface board was soldered up for all of the ESC and servo connections
From a coding perspective, I borrowed heavily from http://www.techmonkeybusiness.com/opensource-rov.html for the basics and KR2_Diving's thread (and Bindo's comments in it) to point me in the right direction for coding up the vectored thrust logic. Thanks to all of you for that, it was hugely helpful!
Here's the vectored thrust code snippet:
Spent a lot of days with the lake teasing me while getting that figured out:
Here you can see the interface board I built for topside to keep everything kinda organized:
Pics of it all together
Here's an overview video:
https://www.youtube.com/watch?v=-JK1sZ_Onoo
First day in the water
Overall it worked well, except for some greater than expected water ingestion.
Here's some dive videos:
https://www.youtube.com/watch?v=kmwI3FrjR6g
https://www.youtube.com/watch?v=PgUwTe_qwT0
As you can see, even with the thruster power turned down quite a ways, it's still very twitchy. Taming the controls as well as fixing the leaks into the WTC were my primary focus going forward. I've managed to deal with both of those issues, and then some! More info in an upcoming post.
The new goals included:
- New frame with a more modular mounting system
Fully-proportional control
Vectored thrust
Utilize the BR WTC system for housing the electronics
On-board battery to shrink the tether
Sticking with PVC for the main frame - primarily for cost and simplicity
Ended up going to an 80/20 aluminum T-slot system for mounting everything else https://www.8020.net/
And to that, my polycarbonate solution for holding the WTCs in the appropriate position.
Thrusters
Went with the popular NTM 28-36 750kv brushless motors. The hardest challenge here was connecting the Graupner props (4mm threaded) to the motors (4mm smooth) as apparently no one makes an adapter for those sizes (virtually every other combination is available however). After lots of searching and hairpulling, I found these threaded rod adapters:
https://www.thorlabs.com/newgrouppage9. ... up_id=1745
After consulting with my fabrication friend, we decided the threaded end could be drilled out and another set screw added for gripping flats machined into the motor shaft.
Control System
As I wanted proportional control, I decided to go with a PS2 controller and Arduino Uno on the surface talking over serial to an Arduino Mega on the ROV commanding the ESCs. Ryan's (KR2_Diving) thread was a huge resource to getting a lot of this figured out. I tested with the RS-485 chips and while it worked well enough, I decided to pick up the BR Fathom-S tether board set as for the price it has a compelling set of features and simplified this area.
A small interface board was soldered up for all of the ESC and servo connections
From a coding perspective, I borrowed heavily from http://www.techmonkeybusiness.com/opensource-rov.html for the basics and KR2_Diving's thread (and Bindo's comments in it) to point me in the right direction for coding up the vectored thrust logic. Thanks to all of you for that, it was hugely helpful!
Here's the vectored thrust code snippet:
Code: Select all
//Take PS2 joystick values and mix to give vectored output
txdata.upLraw = -((128-UpVal)-128);
txdata.upRraw = -((128-UpVal)-128);
txdata.FHLraw = (ForwardVal + RollLeftVal + YawLeftVal - 256);
txdata.FHRraw = (ForwardVal - RollLeftVal - YawLeftVal + 256);
txdata.RHLraw = (ForwardVal + RollLeftVal - YawLeftVal);
txdata.RHRraw = (ForwardVal - RollLeftVal + YawLeftVal);
//print total values for debugging
pwmSum = (txdata.upLraw + txdata.upRraw + txdata.FHLraw + txdata.FHRraw + txdata.RHLraw + txdata.RHRraw);
Serial.print(pwmSum);
//Scale the values to be in range for ESCs (note: the PWM limits can be adjusted to adjust total thruster output. 1500 is neutral/off)
txdata.upLraw=map(txdata.upLraw,0,256,1300,1700);
txdata.upRraw=map(txdata.upRraw,256,0,1300,1700);
txdata.FHLraw=map(txdata.FHLraw,256,0,1300,1700);
txdata.FHRraw=map(txdata.FHRraw,256,0,1300,1700);
txdata.RHLraw=map(txdata.RHLraw,256,0,1300,1700);
txdata.RHRraw=map(txdata.RHRraw,256,0,1300,1700);
Here you can see the interface board I built for topside to keep everything kinda organized:
Pics of it all together
Here's an overview video:
https://www.youtube.com/watch?v=-JK1sZ_Onoo
First day in the water
Overall it worked well, except for some greater than expected water ingestion.
Here's some dive videos:
https://www.youtube.com/watch?v=kmwI3FrjR6g
https://www.youtube.com/watch?v=PgUwTe_qwT0
As you can see, even with the thruster power turned down quite a ways, it's still very twitchy. Taming the controls as well as fixing the leaks into the WTC were my primary focus going forward. I've managed to deal with both of those issues, and then some! More info in an upcoming post.
Re: Lucas' SeaFox
Nice work and great improvements... it cool to go back and read this thread to see how far you've come.
I was thinking of finally getting around to trying some of the Bluerobotics stuff.... why do you think your WTC leaked?
How nice is that tether? that has to be a huge improvement.
Again Great work!
-Steve
I was thinking of finally getting around to trying some of the Bluerobotics stuff.... why do you think your WTC leaked?
How nice is that tether? that has to be a huge improvement.
Again Great work!
-Steve
Re: Lucas' SeaFox
Thanks for the kind words, Steve! Speaking of how far I've come, I was going through some old papers the other day and found this drawing I made when I was very young:
As you can see, this has been a long-running passion of mine!
In regards to the WTC leak - I did identify the culprit and it wasn't the WTC itself > https://youtu.be/yGuRsIjz4Rk
The issue was the thruster cables had water ingestion going through the sheathing and into the WTC. This was an issue I expected at greater depths, but not < 20 ft! I did come up with a potting solution for the cables that did fix this, more info on that in an upcoming post.
The tether is a HUGE step up over any of the DIY solutions I've tried. HIGHLY recommend it if you can spare the $$$. While it seems pretty stiff in the air, I haven't had a single issue with it in the water with snags, pulling on the ROV, etc. Also with the high strength rating, I'm confident in being able to retrieve the ROV if there is a power or buoyancy issue.
As you can see, this has been a long-running passion of mine!
In regards to the WTC leak - I did identify the culprit and it wasn't the WTC itself > https://youtu.be/yGuRsIjz4Rk
The issue was the thruster cables had water ingestion going through the sheathing and into the WTC. This was an issue I expected at greater depths, but not < 20 ft! I did come up with a potting solution for the cables that did fix this, more info on that in an upcoming post.
The tether is a HUGE step up over any of the DIY solutions I've tried. HIGHLY recommend it if you can spare the $$$. While it seems pretty stiff in the air, I haven't had a single issue with it in the water with snags, pulling on the ROV, etc. Also with the high strength rating, I'm confident in being able to retrieve the ROV if there is a power or buoyancy issue.