Greetings, long time lurker, but first time poster here. Love this forum, and the wonderful ideas I get from here so I figured I should start chipping in. I made some sealed miniature Hall pots for a modified Thrustmaster Cougar I made for my mechpit, which I will be posting about shortly. I'm also a sailplane pilot, and have some sailplane cockpit stuff to share later too. I figured I'd start with the pots though, since it is potentially useful to more people than either my mechpit or my glider pit contributions will be.

Sealed Hall pots. I wanted Hall pots that would:
1. not be easy to disturb once installed
2. be easy to install and calibrate
3. compact
4. cheap
5. repeatable
6. do not collect steel shavings

All compasses pointed to sealed units. After pricing them in the wild, I decided a different approach would be needed, one that fit my budget, which is close to nothing. The tried and true bic pen method it is, (for anyone not in the know: http://www.simpits.org/geneb/?p=299 ) only I decided to bring it up to date. Below is what I have come up with. The pots work great and were well worth the time it took to make them.


The Ubiquitus A1302...

Currently in the classic bic pen arrangement. A1302's necessitate that I use two of the 5x5x1mm N52 square magnets I procured for this, but the sealed units use the A1301, which are identical in appearance but only need one magnet on each side since they are twice as sensitive. Only using one magnet on each side is important for the sealed pots I made because they need to fit with enough clearance to freely rotate within the footprint of the 1/2" OD of the little bearings. I was worried they would be susceptible to interference, but short of sticking a big magnet right up to them, they are dead quiet. Also completely unaffected by the presence of ferrous metals, much to my surprise.

On the bench...

The little square next to the bearing is one of the magnets I'm using. Once I established the 1301's work fine with the single magnets in the bic pen arrangement, it was time to ditch that giant bearing and make something a little more elegant.


Size *does matter

...sorry, but it's unfortunately true. In this case smaller is better, and there is no need for a 200kgf bearing for this anyhow. The large bearing is a skateboard bearing: 5/16 x 7/8 x 1/4 vs the smaller bearing, which are found at hobby shops: 5/16 x 1/2 x 5/32.

Exploded view of the new pots

A refined version of the now classic bic pen method. These were not hard to make, and required no special tools, just time and care. My original test bearing is bottom right, and the grid is 1/2" squares for scale. The little bearings have the same bore (5/16"), but are significantly smaller. No need for a 200kgf bearing on these!

(QTY) Parts [size]-price(s): *-$...
(1) Alegro A1301EUA-T -$1.50
(2) 5/16 x 1/2 x 5/32 bearings -$1.20($2.40)
(2) 5x5x1mm N52 neodymium magnets -$.50($1.00)
(1) bic pen [~1/2"L] -$...
(1) K&S Engineering brass tube [17/32 x ~3/4"] -$.30
(1) mylar disc [.030 x 1/2"D] -$...
(1) Magic Sculpt epoxy clay -$...
(3) 22ga wires/pigtail/etc -$...
(1) 9/16" brass jam nut [drilled to 5/16" bore]

*too small to calculate and/or can easily be substituted for other material

Upon closer inspection

You can see the spiraling of the threads I cut into the bic pen. I used a 1/4-20 tap and held the pen body in my hand. The 1/4-20 thread makes them very easy to mount directly to moving bits or to linkages, and is very easy to adjust/refine/tune/change at a later date.

Semi exlpoded

The thing that looks like a nut is not a nut, but some hex stock I bored to 5/16", since that's the magic pressfit size for mating with the bic pen, same as the bearings. It gives me something to grab onto for indexing them or mounting them, as well as preventing the pen segment from splitting when I screw a 1/4-20 set screw into it.

One of these things is not like the other

It has female threads for it to mate onto an axis that terminates with a segment of threaded lamprod. I had to use a tiny dab of superglue to keep this one happy.

Hello hello hello...

Ok, not echo-cavernous, but pretty empty.

3 Pin SIP, FTW!

Discs are cut out of mylar plastic, using a 1/2" hole punch. Easy peasy. Poke some holes and your good to go.

5mm is juuuuust right.

This puts the sensor head dead centered with the magnets when using the bearings I did.

Bend prongs thusly

...but make marks to keep orientation of wire positions. If you make multiple of these, I highly recommend bending them all in the same relative orientation in relation to the chip, to keep form hosing yourself once soldering, a trap I somehow avoided this time.

Relief notches

at the bottom rim to allow the wires to pass. Note the 1/4-20 threaded stud sticking out of the pot on the left. I recommend using stainless (non magnetic) set screws for this, but this is just an aluminum threaded segment.

Hemostats are a must!

Not only to hold your chip, but to act as a heatsink while soldering. There is no danger in shorting the pins together at this phase, but overheating them will kill them.



Careful now!

It's really easy to disturb the wires when getting it into the recess, so be patient and work gently.

Wired for sound


aaaaand potted

Magic Sculpt works great to pot the bottoms. Easy stuff to work with, cleans up with water, dries in a few hours and a nice dielectric to boot.

I call them ducks

The brackets are just bent aluminum stock, 1/2" x 1/16". After making a few mangled ones, I was able to efficiently make some fairly clean ones. They securely hold the pots, but allow easy repositioning if needed.

Mounted X-axis...

This pot is directly mounted in the azimuth potion on my modified Thrustmaster Cougar I made for piloting mechs. The stick moves in pitch/azimuth instead of pitch/roll, and is very intuitive to aim with since it controls the upper body of a mech which moves like a tank turret.

Mounted Y-axis

Current shot shows crappy wire connecting my test arms. New arms much cleaner with less holes, and now uses miniature ball linkage made for R/C gear to transmit motion and looks much nicer, but the subject today was sealed pots. The stick is it's own thing...

Very neat work! And I am eagerly awaiting to see what seems a single centering spring gimbal in the background of that picture, I have wanted to replace the gimbal in my Cougar for some time now, though I was thinking about a way to fit a couple of cams in there to get rid of the center detent.

I am eagerly awaiting to see what seems a single centering spring gimbal in the background of that picture

Thanks Brandano, hate to disappoint, but the spring on the gimbal is actually not for centering but rather provides friction on a greased mylar rub. Friction on both axes are accomplished the same way and are adjustable via a nut. my post on the stick they're for will explain it better, but here's a shot of the gimbal with it's spring loaded rubs.



This gimbal was made for mech piloting (MechWarriorOnline specifically, but not limited to) although tank driving with it might be fun with it too since the upper body of a mech moves like a tank turret. In use, the stick moves in pitch as normal, but instead of roll it has azimuth (twist). There are no centering forces, nor center detents. I use it with absolute inputs, and when I let go of my stick my mech (or tank...) stays in that position.

As a pilot, I cringe of imagining flying with this thing. Intuitive for mech/tank driving, a total kluge for aircraft...

Now that is just neater than hell. Great job, LocNar!

Where did you find the bearings you're using? I've only seen them at about $6 each.

Do the bearings need to be "drifted" into the brass sleeve, or are they glued into place?

I love the idea of tapping the interior of the pen! You could also use the threaded portion of a 1/4-20 nylon bolt if you can't find aluminum all-thread.

I'm gonna have to build me some of these.

g.

Many thanks Gene, your projects are a treasure trove of inspiration to me so it's much appreciated.

Where did you find the bearings you're using? I've only seen them at about $6 each.

There are likely more sources at this price, but I bought a 10 pack of them from this ebay seller for $8+$4 shipping. They are commonly found in R/C rars and such, so tend to be sold cheaper in bulk through those outlets rather than buying from a bearing supplier.

Do the bearings need to be "drifted" into the brass sleeve, or are they glued into place?

The bearings have a sliding fit in the brass tubing (17/32" OD,0.503"ID, 0.014" Wall), but the clearance was nominal, especially after using a regular tube cutter. As an overkill measure on these, I chucked them into a lathe collet and squeezed them in around the bearing at that stage, but glue would suffice here since the loads on these things is nil. If you had tubing that was slightly too tight, a .500" hand ream would make for a nice fit as well. I wanted to stay away from any need to press-fit the bearing/pen body assemblies though, to insure the magnets stay put. I also put tiny dobs of cabosil thickened epoxy in the fillets where the magnets meet the pen body, to ensure the magnets don't ever jostle loose in there. Super glue dries fast, but is brittle...

As to threading the pen body, were I not already stuck with 1/4-20 threads elsewhere in the project, I would have rather used 1/4-28; shallower cut but more bite overall. The aluminum threaded section was just cut from a screw I had sitting on the bench, but nylon would work just as well. If I had one handy, I would have used a socket head set screw. Makes it real easy to install it into the pen body for final assy, and also leaves a convenient place to grab hold of the armature with the magnets to jog it around for any kind of adjustment down the road once dialing it in. It's actually not even necessary for it to be non ferrous, but I couldn't bring myself to put any steel other than the bearings on it. I played around with metal chunks and magnets too while the pots were working (wired to TM Cougar guts and TARGET device analyzer running), and the presence of steel seemed to have zero effect, and even stray magnets had surprisingly small effects, although noticeable for sure.

-Loc Nar

Loc Nar,

Thanks for the detailed tutorial. Well done. One question: Why mylar? Is there a specific reason or could one use
any other sort of stiff plastic or silicon or even cork? Could you advise on alternatives?

Again, thanks for your contribution.

Regards,

Flying Colander

Very nice. Might have to revive my old Suncom THrottle restoration after reading this

Loc Nar - Thanks for the info. I printed the thread and I'll get some made when I get some spare time.

I used two .25" x .25" x .125" Neodymium magnets on my build and didn't need to stack them up, so the A1302 is still a viable part I suspect.

Flying Colander: I suspect he used Mylar because that's what he had on hand. You could also use a circle cut from a 3x5 index card or even a circle cut from 1/16" acrylic. Any insulating support structure would work - it just to hold the part in place.

g.

The plastic discs don't need to be mylar, it's mostly a place holder. I make composite aircraft parts, and mylar is a common material used for this. Considered a 'consumable', it's available for dirt cheap at composite supply places. It has a 600deg melting point, excellent dielectric properties and mechanically very strong. I also use it as rubs between sliding metal components if I'm trying to save space, and I also use it for insulating barriers. I suspect most any plastic will suffice, although I don't suspect cork would hold up well after all the handling and poking.

As to the magnets, .25x.25x.125 magnets will be too large. The ones I used were 5x5x1mm, and since I used A1301 sensors (01 is twice as sensitive) I'm able to use single magnets on each side which leaves plenty of clearance. My bench test with the big bearing had an A1302, which necessitated that I use them doubled up. Stacked up to 2mm, even the little magnets I used will not clear the .503" ID of the tube while mounted on an unmodded bic pen body, but it's close so filing the flats they are bonded to would give the clearance needed and would allow the 1302 to be used. Since my magnets are so small, I bought N52 grade, the highest commercially available strength.

A1301's were harder to find but Newark Electronics has plenty of stock still. My magnets came from Supermagnetman.

-Loc

Gotcha. I've got some .125" x .062 round ones I should try.

g.

Loc Nar and G.B.,

Thanks for the mylar clarification and magnet advice.


Cheers,

Flying Colander

About mylar -I buy this stuff by the square yard and have no shortage of this material so anyone wants some to play with some for this or other projects, I have no problem sending anyone an envelope's worth. For reference, a 4x9" piece is enough >100 of these pots...

-Loc