Water/ Pressure Proofing
A brief introduction: Hello, my name is Zack and I am the TechShop liaison for the OpenROV team. TechShop has taken an interest in following this story and I have been charged with supporting the team with the tools and community available at TechShop in San Francisco.
A topic of debate is keeping your brushless motors working after repeated and possibly prolonged exposure to seawater. A post-dive cleaning with freshwater then WD-40 will get you fairly far, though it is fairly far from a long term solution. I tried three other methods over the weekend to see if they had any merit.
Silicone Mold Release Spray
Spraying the coils with liquid silicone will coat them in a water resistant silicone gel. This stuff is very difficult to remove even with solvents.
Silicone Conformal Coating
This liquid is commonly used to protect electronics and is resistant humidity. It hardens after application and is able to penetrate into the coils. Once hardened it forms a permanent coating.
This was, at first, going to be a control for the experiment. But I soon realized the merits of not caring. These motors are fairly inexpensive and frankly, my dear reader, replaceable.
All three methods were subjected to a solo cup filled with tap water and three packets of In-And-Out to-go salt--only the finest. The coatings were applied to some old DC motor rotors I had lying around that exhibited similar attributes to the motors we will be using for OpenROV models. I then left for a weekend of waiting in Memorial Day traffic. Upon my return I washed and wiped the motors to remove any rust that had accumulated on the magnets and snapped a picture.
From left to right,
Not Caring, Silicone Conformal Coating, Silicone Spray
As you can see, the "Rhett" method motor had a little corrosion on the magnets but nothing to stop the next dive. The conformal coated magnets actually accumulated a little rust, surprisingly, but copper coils were pretty well encapsulated--a result that could mean better performance underwater. The silicone spray fared the best against rust, showing few signs of build up.
Silicone spray is an excellent water repellent. Coupled with regular cleaning and re-coating, this could get you much further than simply not caring.
Another issue to contend with is that there is a possible drop in power of the motors as they operate underwater due to a short-circuiting effect between the coils. It is my next goal to test some methods for sealing them without compromising durability or performance. I will probably use the following methods: polyurethane sealant, epoxy, and marine vinyl-ester resin. Any suggestions would be appreciated.
This was an excellent experiment! The data you've gathered is exactly what is needed to develop the ideal way to handle propulsion for low cost ROVs. This is what open hardware is all about!
I look forward to seeing the results of your next tests relating to short prevention!
This is awesome, Zack!
I think you're right: Silicone Spray with a healthy cleaning regimen seems like the way to go - the way to get started at least. We can make changes if problems accumulate.
We use spray silicone extensively at my work. I've found the easiest way to clean it is with Dawn dish soap and warm water. (I am in no way connected with Dawn or whoever makes it) :)
Make sure you get a good rinse with warm water. It's best if you can do a final rinse with DI water to remove any city water deposits. Blow it dry with low pressure air or a hair dryer, then bake in an oven at low temp (130 F to 150 F) for a few hours after.
And just in case your housing ever fails, this process works on electronics also. We get a failure every now and again!
We used a similar process when the USS Sandlance almost sank at the pier in Charleston back in the early 90's.
That sounds pretty much like the plan at the moment. Thanks for the advice, John.
I grew up in Charleston. Cmdr. Dutch Shultz was my granddad--but we didn't talk about the subs much, he had retired by the mid-90s when the Sandlance flooded but I definitely recall the story.
Looks good, please be aware that if you are using brushed motors the silicon will form an abrasive layer on the commutator and you will see greatly accelerated brush wear.
- another way would be to keep the motors in a dry space and use a shaft seal (www.mikessubworks.com) to extend a shaft and prop to the wet side. A lot of RC sub guys have (www.subcommittee.com) tried the wet motor/wet servo approach and are usually rewarded after a few dives with a seized up motor and/or servo. Sure they're cheap, but who wants to mess around re-fitting motors all the time?
There are seals for push rod shafts too.
Good call Tim. It does make a lot of sense to have them inside. Eric came up with a pretty cool magnetically coupled drive system (built from mostly laser cut acrylic!) but the cost and complexity just didn't compete with the brushless approach. So far anyways. The push rod shafts are great, thanks for pointing them out.
Use flexseal and spray it on with the motor running. This works on the one working prototype.
This is good for a isolated saltwater test, add stray electrical currents, minerals and ions found in seawater, and a more complex set of materials for some galvanic action and things will happen much faster.
You're right, Chris. Zach and I have been working on some other solutions. A coating of marine resin seems to work, although we still need to do some tests on how effective it is as an insulator. We'll post test results.
Another good one is the epoxy coating that electricians use on bare wire. The mini rov's motors have that and seem to be doing good so far.
If you want to do a magnetically coupled drive system, I'd suggest borrowing a design from the pond pump makers. you can build a set of coils, then pot the whole thing to make it forever waterproof, and then have a magnetic "core" that is connected to the prop shafts.
For example: http://www.pondpetsusa.com/water_pumps_sub/pondmaster.html for a cutaway. the o-ring is to keep the water inside the pump, since the "cap" where the water inlets and outlets are connected to are removable.
A picture of the replaceable impeller may make more sense: http://www.marineandreef.com/Pond_Mag_and_Mag_Drive_9_5B_pump_Impellers_p/rdf12746.htm Note that the shaft must be connected on both ends to be stable. You could theoretically build a shaft that didn't need it, but it would have to be pretty stiff or have some sort of a slip "bearing" on each side of the magnet. the shaft on the impeller is actually stationary, and the little rubber end caps allow for some vibration without destroying the motor. these motors CANNOT be run dry; they rely on water to keep the magnet from rubbing on the shaft and/or sides of the motor housing.
These types of pumps can run for years @ 24/7 without maintenance.
Has anyone actually tested brushless motors in the real world? Followed by being flushed with fresh water and then lubricated after dives to see how long they will last?
Anthony, we've tested it several times and they've performed well. We even went a few days without silicon spray and freshwater rinsing. We did start to see some corrosion after that, though.
Thanks David, I am scratch building right now my own and have lots to do still. 99% of my dives will be in fresh water anyways. I sure am learning alot as I go and I haven't read or researched this much information since university : )
I am using 750Kv brushless motors for my build and they will be exposed to water. My next build will be oil filled housings and magnetic couplings. You can check it out on Home Built ROV.com
Sounds great! Keep us posted. Lots of folks are interested in the magnetically coupled drive strategy - excited to see some of the different methods on the forums.
Anyone tried potting the windings - in the bldc outrunners they don't spin so there's no problems with balance? Could use a thermally conductive potting epoxy and perhaps a vacuum to make sure there's no air bubbles.
We've tinkered around with potting the motors a bit, but it's the bearings that really need to be protected from corrosion. Tarnish on the stators doesn't really effect performance, however if the bearings corrode, it can freeze up the prop. We also have to insulate the solder junctions between the motor leads and magnet wire for the stators. This could be done with potting like you describe, but liquid electrical tape seems to do just fine.
There may be use in adding a little more protection to the magnets on the motor bell, since they are plated with only a small amount of nickle and the neodymium underneath is very susceptible to rust. This might be doable with a light layer of conformal coating, but I've never worried to about ti too much since the part is so easy and low cost to replace.
next page →