November 23, 2013

Building a Magnetic Coupled Thruster

This is a Magnetic Coupled Thruster that I was trying to build may years ago.  I start with a TLDR = While it worked… it didn’t work well enough.  The magnets I used were either not strong enough or I didn’t use enough.  At some point I’ll go back and add more but honestly this project was a long time ago so who knows when I’ll get back around to it.  All that said this is a long very detail build page that I did mainly for myself so if I ever wanted to build another one of these so I can look back and remember how I did things. This is just a build page… I do not show it working or being teted If your still interested at this point …. read on.

This is the basic design I will be working from and is subject to change through out the build.  (yes I drew a lot of my plans in MS Paint back in the day) The basic principle as with most magnetic coupled thrusters is…. An inner rotor that holds one set of magnets is attached to the motor and is sealed inside the main thruster housing, the prop is attached to another rotor which holds a second set of magnets on the outside of the housing.  This allows the prop to be coupled to the motor via magnetic force and eliminates any type of shaft seal.  Basically there should be no way for water to reach the motor because it is totally sealed to the outside water.  It is possible for the cap o-rings to fail but there is less chance of this happening over a shaft seal.  The unique thing about this design is the fact that the whole outside rotor spins with the prop.  I can’t take credit for this concept though, I saw this along time ago on a set of drawings by Thijs Struijs and just changed a few things here and there.  The other unique option I added was to thread the motor mount so the spacing between the magnets can be easily adjusted.  I want to magnetic pull to be strong enough to turn the prop but I also want it weak enough to break away if the prop should become tangled.  This way the motor should never be able to stall and draw to many amps.

These are the parts I have purchased all ready.  The housing and rotor parts will be turned from a solid piece of 2″ PVC Rod. I choose PVC because it is cheap and easy to machine.  The motor is a Brushless Turnigy Multistar 2216-800Kv 14Pole Multi-Rotor Outrunner and I am using a HobbyKing Brushless 30A Car Electronic Speed Controller w/ Reverser for control. Not knowing much about Brushless motors and controllers yet I bought a Hobby King Programming Card too just encase it needs reprogramming. The Magnets are 1/4″ x 3/8″ Neodymium Magnets.

I’ve never used brushless motors before so I want to make sure I’m not going to burn anything up before I machine up the housings.  I bought a few of these Vex controllers years ago that should be perfect for testing out the new controllers.

I also bought one of these hexTronik DT700 Brushless Outrunner 700kv motors because it all ready has a 4mm shaft.  I want to test out running a brushless exposed to the water and this one should be the easiest to do that with because I can just trim down the shaft and screw on one of the many 4mm thread props I all ready have.  (this is a thread for another day though)

I did a quick test setup today just to try out the motor. I’m definitely going to have to use that program card because it needs to complete stop and I need to release the stick before going in reverse.  I think this has something to do with a brake feature. It also is not as powerful in reverse as forward either.  Hopefully it’s something I can program into the ESC I really haven’t looked into the card yet.

I started by cutting of a 4″ piece of the 2″ PVC Rod on my bandsaw.

Next that gets centered in a 4″ 4 jaw chuck on the lathe.

Then a few quick cuts to clean up the face.

After that I use a few different drills to make a hole for my boring bar.

I then bored a hole in the piece leaving about a 1/4″ for wall thickness. Everything is pretty much done using the TLAR method (That Looks About Right) at this point.

I had to put a step on the inside edge to allow for a little more room for my internal threading tool as it isn’t quite long enough to thread as deep as I would have liked.

Next I mount up the internal threading tool in the tool holder and hope for the best. (I haven’t cut threads in a long time.)

I chose to go with a 18tpi thread for the motor mount. This is done but adding the right gear ratio to the lathe.

After the first pass I check the cut with a thread gauge to make sure everything is setup correctly.

They need a little clean up but here are the internal threads for the adjustable mounter mount.

Next I turn down the outer edge for the threads for the rear cap.

So far this is what I have for the rear section of the thruster housing.

At this point I’ve removed more material than I have left, but that’s machining for you….

Next it is onto the motor mount. I cut another chuck of PVC and again after centering it in the lathe it is faced.

Next I actually have to measure (no TLAR here) what will be the minor diameter of the threads for the mount. It’s about 1.465″ as you can see.

Knowing the thread depth of 18TPI is about .07″ I had to turn down the work piece to 1.535 for the major diameter before I can start the threading process.

Next using the external threading tool again I thread the piece to match the housing.

I quick test fit to make sure everything is correct before I remove the work piece from the lathe.

Next I measure the original mount to determine the correct drill size (25/64″) to use to match the center hole…

and drill the corresponding hole in the center of the piece. This will only be used to mark the mounting holes in the new mount.

Next it was over to the mill to cut a relief channel for the motor wires.

Then using a 25/64″ transfer punch to center the mount onto the work piece I use another punch to mark the center of each mounting hole.

I then mount the piece in the mill and using a center pointer I line up the first hole.

Next I use a center drill to start the hole. This will keep the drill bit from walking and also put a bevel on the hole face for the screw.

After that I used a 1/8″ bit for the through hole. I think the screws are actually metric so this will give me a little play when mounting the motor. I then repeat this process for the other three holes.

I also drill two 3/16″ holes in the mount, these will be used to screw it into the housing and later adjust the motor spacing.

One last trip back over to the lathe and I use a parting tool to cut off the mount from the work piece.

Here is the mount so far, it still needs a little clean up but its complete for the most part.

I’m using a pair of needle nose pliers to test fit the mount in the housing, I will probably make a tool for this later on. I need to buy some longer screws before I can mount the motor.

So I bought some longer screws and now have the motor secured to the mount, next its on to the internal rotor.

I haven’t bored out the front of the housing yet so I use the back (minus the internal threads to measure for the rotor diameter. The rotor will end up being about 1.53″.

After measuring my magnets I come up with a measurement of .45″ for the thickness of the rotor and mark that on my work piece.

Here I have turned down the piece for the correct diameter and depth.

I then had to measure the front step of the motor where the stock propeller adapter would mount.

I turn down the front of the work piece to duplicate that little step. (This will center the adapter while I mark the mounting holes.)

I don’t have a transfer punch small enough for these holes so I use a little pointy thing to mark the holes as best I can.

I have to turn the piece around and face the other end so it sits flat in the mill for the next step. (I should have done this to the workpiece at the start but I forgot.)

Next it was over to the mill and once again using my center pointer I align the piece for the holes.

I used a 5/64″ drill bit to drill through for the screws. My holes look a little off which is the result of not having the right transfer punch. (remember kids always use the right tool for the right job)

Next I motor the piece back in the lathe and clean off the little step I made before. Then using the cross slide I drag the tool across the face making a line, then rotate the chuck 90 degrees and do it again. These marks will be for the placement of the magnets.

After that I roughly mark how much of the inter rotor I can remove. This will lighten the rotor a bit and allow me to use the stock mounting bolts.

After measuring my screw length and knowing my rotor will be about .45″ thick I calculate that I need to bore out center of the rotor to about a depth of about .305″. I am starting the hole with a plunge cut using a 1/2″ end mill so I roughly mark the depth on the end mill with a marker. (being this is only a prototype I’m back to the TLAR method for most non critical things.)

Using my boring bar I then finish out removing excess material from the center of the rotor.

After that I measure whats left with my calipers (not shown) divide by by half and the using my tool bit again mark the center point for each magnet.

Back over to the mill I center each hole as before and drill through first with a 5/32″ drill bit (these holes will be used to transfer the magnet holes to the outer rotor later on.) I then finish the holes for the magnets with a 1/4″ end mill so they have a flat bottom.

Here the holes for the magnets have all be drilled. I’ll probably have to clean them up a bit because I wasn’t as critical with the depths as I should have been.

A quick trip back to the lathe to cut off the rotor with a part off tool.

Next the piece is rechucked and I drill a 23/64″ hole for the step on the motor.

Here we have the finished rotor mounted to the motor. I still have to glue in the magnets and give it a spin up to see how balanced it is.

The next step was to bore out the front of the housing just wide enough to fit the rotor.

A quick test fit to make sure there is enough clearance for it to spin without any interference.

I then screwed the motor half way into the threads (to allow for adjustment either way) and I measure the depth that I need to trim the housing.

With the piece back into the lathe I once again use the part tool to trim off the excess piece of the housing. The housing came out much shorted than I originally intended but this was all because I couldn’t cut the internal threads as deep into the housing as I wanted to.

After I had the housing trimmed down to size I turned the end down and threaded it for the rotor cap.

Another quick test fit of the motor and rotor and then it’s on to making the rotor cap.

I got a few more parts in today, we have a piece of brass (that will be used for through hull connectors), a nylon thrust bearing, a few Acetal Ball Bearings (went with these because they are much cheaper than stainless steal ones), A couple of o-rings, and a 4mm bolt. Next it is onto making the rotor cap….

I first measure the minimum depth I need to cut the matching threads in the cap.

I transfer this mark to my end mill and plunge cut the hole.

Next I measure the major diameter and subtract the thread depth to get my minor diameter.

I bored out the minor diameter to about 1.728″ and then cut the internal threads. (not shown)

Another quick test fit of the housing to make sure everything is copacetic.

I then measure the internal depth of the cap to figure out how much needs to be trimmed off the outer side.

I’m going to go with a .100″ thickness for the magnet spacing. (best guess here, I can always trim it down later if I need to.) I scribe a line on my work piece and its back into the lathe to turn down.

I used a piece of left over scrap from the housing to act as a spacer in the lathe. This will also help to keep the threaded section from being crushed by the chuck.

After that I turn down the cap to about .75″….

and then I trim the length down to about .75″ also.

Next I drilled and tapped the cap to 8/32″ to fit the shaft.

I also cut a recess for the thrust bearing washer.

For the shaft I turned down a piece of 3/16″ stainless steel rod to .164″ to match the 8/32″ threads I cut in the cap.

The treads of the shaft are being cut with a die by hand. I use the end of the drill chuck to keep the die holder straight while I start the threads.

To cut the threads I hold the Die holder and just spin the chuck by hand.

Here is the completed shaft.

I quick test fit of the bearings shows me I’m going to need a bigger outside washer on the thrust bearing, the way it is now the rotor would rest against the top bearing and shaft and the thrust bearing wouldn’t be doing much. I’ll have to turn down a larger washer so the rotor rest against that point and then the thrust bearing will take that load as the magnets will pull the rotor toward the housing.

Onto the prop rotor. I start with another chuck of PVC, face it and drill a 23/64″ hole in the end.

Using a 23/64″ Transfer punch to center the other rotor I transfer the magnet locations using a 1/4″ punch. (I later noticed the rotor is backwards from what it should be but the holes should still be close enough for the prototype.)

Now that I have my magnet locations I can start boring the piece out to mate with the cap.

I start by measuring the depth needed for the bearings.

I first drilled a 31/64″ hole and then bored that out to .50″ to a depth of 1.45″ to fit both bearings.

Next I measure for the new thrust bearing washer I made. (I didn’t show making this)

Then I bored out .620″ to a depth of .890″

Next I bored out to clear the step on the cap.

I tested out the thrust bearing by spinning the piece in the lathe and just holding the cap against it. It seemed to work well at this stage.

Next I transferred the work piece over to the lathe and plunged cut the magnet holes.

After that it was back over to the lathe to drill and tap for the 4mm screw for the prop.

Next I had to setup the angle on the compound slide to turn the cone shape.

After a bunch of cutting this is how it looks. I was off a bit with my measurements so I’ll have to clean the step up at the end later on.

I quick test fit reveals it’s a tad bit wobbly but it spins very freely. Next I have to glue the magnets in and go from there.

…at this point I’ve glued in the magnets and after a few quick spin up tests everything seems like its working pretty good so far. It’s a tad bit wobbly but better than I expected. Next it’s on to the end cap.

I first face off the back of the end cap and knock off the corner a bit to reduce the sharp edge.

Flipping the piece in the chuck I start a hole with an end mill so I can fit my boring bar in there.

I then bored out a section to almost the end of the cap to make room for the motors wires. I left the end cap a little longer than I needed because this is my last piece of material and if a screw up the threads I can just cut them off and try again without having to order more stock and redo the entire cap.

Next I measure my major diameter of the threads on the housing….

….after subtracting for the depth of the threads I finish boring out the cap to around 1.73″.

Using my internal threading bit I give the cap its threads and hope for the best.

A quick test fit and it seems I actually measure right.

Here is the almost finished end cap next to the housing.

At this point the thruster is pretty much complete with the exception of the exits for the motor wires.

Next it is on to the through hull connectors. I start with a piece of .25″ brass and turn the front down to .20″

After that using a 12-24 die I thread the piece.

Then a .1406 hole gets drilled in the end. These are where the motors connectors will plug in.

Using the parting tool the piece is then cut off.

The piece is then rechucked in the lathe and I turn down a small step and then drill a .1094″ hole in the end. The power wires will connect here and the step is for a bit of heat shrink tubing.

Here are the three finished connectors and a quick test to make sure the motor plug fits securely.

After that I drill and trap three holes in the end cap for the connectors.

The connectors get a bit of Teflon tape and are started by hand.

Next I protect the connector with a piece of soda can and I tighten them with a pair of pliers.

So far everything is working out as I had planned.

Next I soldered the wires right to the connectors. I was going to use a screw on fitting here but to speed things up I went this route for now. (I was a little worried that the heat from soldering would melt the housing but I was quick about it so I think its fine.

After the wires were connected I added a little heat shrink tubing just to pretty it up. (I’m not really worried about water proofing the connections.)

Before screwing on the cap I had to pre-twist the motor wires in the opposite direction so when screwing on the actual cap the wires don’t twist up and pull free.

I originally planed on using o-rings to seal the housing but for now I just use more Teflon tape on the threads. (at this point I’m not really concerned about it leaking)

Well here it is all ready get wet and start some testing.

---

Conclusion

I got this in the test tub today and while it kind of worked it needs some tweaking. It spins right up but before it can reach full throttle the magnets break away from each other. I might try adjusting the gap and see if that helps but it might just need more magnets. I didn’t bother testing the thrust or amps draw because I only was trying to see if the coupler worked. Anyway I’m going to try the gap adjustment first and then go from there…..

Some day I’ll get back into this idea…..

 

---

All Information, Pictures, and Material is copyright © 2022 by Stephen Thone and may not be used for any personal or commercial purposes without the consent of the author. All rights reserved. The Author makes no guarantees or warranties as to the accuracy or completeness of, or results to be obtained from accessing and using the Information herein.