Mag coupling version 3.0 vid (updated new video 1/25/15)

[a_shorething]

Making progress on my magnetic coupler idea.


Here was version 1.0: the proof of concept:


And here is today's test (version 2.0):



It uses an 1100 gph bilge pump motor, some 3D printed parts and fits inside 1.5" sched. 40 PVC. My motor controller is just an Arduino shield for testing and I already cooked 1 of them so I haven't cranked it up to full power yet, but I am sending it almost half power and it seems to have pretty good torque. The motor doesn't seem to heat up at all (mostly because the board is the weakest link I assume). I'm just sending 3-4 vdc at less than 1 amp so far, but it's working!

I plan to play with the size, pitch, and number of prop blades in order to find the best option. Once it's done, the plan is to have a completely self-contained and sealed thruster design that you can pretty much just attach to your frame wherever you like.

Total cost for the sealed unit: under $50.

My plan is to build a test stand and see what kind of numbers it's putting out and compare different configurations of thruster blade.

It's coming along nicely though...

What do you think?

[FJM]

Looking good!

Are you thinking of some material (Teflon or something) as a bearing surface between the PVC and the prop? Would either the PVC or the prop melt due to friction? I guess the water would cool it though.

Really interested in how well this works.

Fred

[olegodo]

Great progress!
Looking forward to seeing test results

[Zaibach]

Cool beans

Any concern about the motor overheating? Seems like the bilge pumps are at least partially cooled by the water or at least they seem to burn out a lot faster without it.

[a_shorething]

Looking good!

Are you thinking of some material (Teflon or something) as a bearing surface between the PVC and the prop? Would either the PVC or the prop melt due to friction? I guess the water would cool it though.

Really interested in how well this works.

Fred

Fred,
The thruster doesn't really have that much friction when it's dry and I'm thinking in the water it will be even less. I have sanded the inner surface of the prop and the outer surface of the PVC with 800 grit emery paper and it feels like glass. Any friction will probably only be from an 'out-of-round' situation and I don't think that will be a problem with my process.

Thanks Ole and Zaibach,

Cooling doesn't really seem like it's going to be an issue, at least not in my initial tests. We'll see about later.

The prop is going to be designed to have the right pitch and surface area to provide a good combination of propulsion and control. Too aggressive in the pitch department and I'll stall the motor or if I'm lucky and there is enough power in the motor to handle it, the ROV would lurch when the thruster engages even a little. If the pitch is too shallow, it will be working too hard all the time (like driving in first gear).

Since I can adjust the amount of voltage that will be sent to each one in the software (using the map function in Arduino if that's the board I go with) I just need to find the sweet spots of where I get all the right variables to work together without overheating or causing jerky motions.

I was really happy to find that I can stall the motor using only 4 neodynium magnets (two in the inner hub and 2 in the thruster hub). They may even be more than I need since they easily stall the motor when I stop the thruster. If it turns out I could use smaller (cheaper) ones that would be great but these are only a little over $3 each at this size and that means 4 of them is still less than the cost of the bilge pump and keeps the whole assembly under $50 which was my initial price target.

This will take many iterations and lots and lots of plastic. But I think I'm up to the task.

Step 1 will be thrust test in a tank or ideally in the river by here (so there's no eddy currents set up by the thruster to confuse the results). With this I plan to 'bracket' my parameters. The most aggressive I think I should start with is the 3 blade 45 degree prop you saw. The next one is a 65 degree one I have with a thinner outer ring. That one will more easily spin but who knows if it will provide any thrust? I think the sweet spot will be somewhere in between these but maybe not.

Step 2: Then I'll figure out if more blades helps or not. I set up a 'template' in my AutoCad program so I can adjust the pitch and number of blades now that I have the length of them and thickness of the hub worked out (I think). I just enter it in and then print. (and four hours later) I have another prop to test.

Step 3: I will need to go with a different motor controller and the one I really wanted to try is backordered, so that will take a while. I'm using the Adafruit motor shield v2.0 but it only handles I think 2Amp max per output. I've doubled up on two of the outputs in an effort to double the max amps it can handle but I've already cooked one of them as well as my previous version 1.2 shield so I think I need to go with a 'real' motor controller instead of messing around with the low amperage ones that work as an Arduino shield.

This whole thing could work with brushless motors and ESCs as well, I just think it sounds cooler when you can build a magnetically coupled ROV thruster from a bilge pump, some PVC and some 3D printed parts (plus 4 magnets).

I just got a digital scale so if I can get a test jig set up to rapidly change out the thruster blades I'll post the results of that soon.

(PS- I will work for PLA ) I go through a LOT of black 3mm PLA and I can't prototype without it. If anyone is interested in some printed parts that they can test and/or use I'd be willing to print and send them for cost + a supply of plastic I can use for prototyping.

[FJM]

A bearing might also help stop wear, but your testing will determine whether you need it.

I noticed that in the second version you've stripped off the housing of the bilge pump. Does that mean that what you've basically got is a brushed DC motor? If so, using any DC motor would bring your costs down. Or have you already factored that in?

With your template, you could set up a nice little sideline printing custom props!

Fred

[a_shorething]

A bearing might also help stop wear, but your testing will determine whether you need it.

I noticed that in the second version you've stripped off the housing of the bilge pump. Does that mean that what you've basically got is a brushed DC motor? If so, using any DC motor would bring your costs down. Or have you already factored that in?

With your template, you could set up a nice little sideline printing custom props!

Fred

You're correct about the bilge pump essentially being just a brushed DC motor at that point. It comes out of an 1100 gph bilge pump though so it's got some power and is cheap (I think like $15?). Most of the motors I've seen with that kind of power online are twice that or more, which is one reason I'm trying to stick with this one.

Who knows if this thing will work and if not, this kind of prop won't really have a market.

This prop design is so basic it won't work for any other application I can think of. The cross-section of the prop is not optimized or 'contoured'. It's just a basic wing cross section with the same angle forward and back, and it probably requires the nozzle to be connected as I have it here because of the limitations of 3D printing.

If I can get the mag-coupler working though.... who knows?

[FJM]

Are there any clues on the motor as to its specs? I have no idea how a motor from a bilge pump compares to some of the hobby motors out there. I'm looking at other motors but have no idea whether they will be grossly underpowered.

As for the price, Bass Pro lists a Rule 1000 GPH pump at $55 online, excluding shipping. I don't know whether this is typical, or whether I'm looking at the wrong sources. The price is the main reason I've been looking at the hobby motors.

As for the sideline, I was assuming that the mag coupling would work and that you would be selling for ROV-purposes. I think having the ability to seal up the motor is a big selling point for mag coupling and it always helps being one of the first to market.

[a_shorething]

Are there any clues on the motor as to its specs? I have no idea how a motor from a bilge pump compares to some of the hobby motors out there. I'm looking at other motors but have no idea whether they will be grossly underpowered.

As for the price, Bass Pro lists a Rule 1000 GPH pump at $55 online, excluding shipping. I don't know whether this is typical, or whether I'm looking at the wrong sources. The price is the main reason I've been looking at the hobby motors.

As for the sideline, I was assuming that the mag coupling would work and that you would be selling for ROV-purposes. I think having the ability to seal up the motor is a big selling point for mag coupling and it always helps being one of the first to market.

I didn't see any markings on the motor itself. I got a bunch of them for $15 from this guy on eBay http://www.ebay.com/itm/NEW-PROPUMP-110 ... 43&vxp=mtr

(no affiliation). I know it's a cheap knock off of a Rule pump. I'm OK with that.

That's why I wanted to try this because I think most of the hobby motors are at least $20 each and the ones that are this large typically draw 10 amps or more. This 1100gph motor is rated at 3 amps to run, 5amp max (requires 5 amp fuse).

I may end up switching to a different motor type if that the way the research leads me, but I'm going to stick with this path until I find a reason not to.

And thanks for the vote of confidence on this working. I think the initial tests are very positive, hopefully there isn't some fatal flaw I haven't anticipated...

[thegadgetguy]

Cool beans

Any concern about the motor overheating? Seems like the bilge pumps are at least partially cooled by the water or at least they seem to burn out a lot faster without it.

I believe it is just the lip seals that needs to be kept cool. In normal operation in the bilge of a boat, the entire motor wouldn't necessarily be underwater anyway.

a_shorething - Nice job so far. I think we will all be very interested to see where you go with this. Keep up the good work.