Questions - Help - 3DOF AC Platform IT Begins -

Hello folks,
I'm at the stage of my build (See Projects - 3DOF AC Motion Platform IT Begins)
that I need some assistance and help from forum members.

OK - My 1st Question of many to come:
My Gearheads are all cleaned, gear lube installed - and due to the weather not allowing me to paint,
I want to start working out the pots and how they will interface with gearhead.

The above attached photo of my gearhead, along with some attached information as to what "me thinks"
will work, just as well. REASONING - My gearhead output shaft is 1 1/8" in diameter and it would take a very large toothed pulley to be able to drill out a hole that would fit the shaft..

Most Gearheads of this design, builders use an aluminum gear pulley, large one that fits on the motor shaft, and a smaller one that attaches it to the pot shaft - all coming together with a belt - the pot is usually attached to the base frame with an "L" bracket.

Info:
1. Crank arm length - 4" from center to center
2. Pot http://www.digikey.com/product-search/en?vendor=0&keywords=987 1387-nd
Digikey - Sensor Non-Contacting, 1 Turn, 1/4" shaft -
WILL the above work since it's only one turn - my shaft will only be turning about 160 degrees,
and I want a very reliable pot, LONG TERM use without intervention.
3. Off of the crankarm - using a 1/2" diameter bolt, LONG enough to attach Rod End and Connecting
rod to upper platform - The bolt will have to be long enough for the rod end to not come into contact
with pot - OR, could I assume that with setting the travel points that the arm WOULD stop before
hitting the pot?
4. IF this will work and I have some blessings here, I will have to drill the gearshaft 1/4" with carbide bit,
tap for a 1/4" threaded shaft to accommodate the splice/connector/sleeve (whatever we need to call it)
to attach the pot shaft.

I am asking for advice, thoughts, why it will work IF, or why it won't work BECAUSE - well crap, I just need you advice and help.

Thanks so much friends for hanging with this ol' guy...it has been a blessing.

Tom

Hi
The pot you have linked to has a 180 in the product code, thats 180 degree the same as the one I'm using

regards Terry

Hi Terry,
Thanks for your reply - I went to the data sheet and I'll need the #6127 (has a 1/4" shaft) - and it says "Single Turn" - does that mean it will measure the whole 360 or just 180 degrees?
AND, gosh forbid, it the gearmotor would start spinning, would it destroy the pot?

Thanks so much,

Tom

Gday Tom, my main expierence is with 12v DC operation, I connect my pots directly to the centre of the shaft as in the pictures below.

Pros - Pot rotation is the actual rotated position of the lever. ie lever moves 45 deg , pot has travelled 45 deg.
Con - If motor spins out of control (rare thing for the Jrk) the pot is broken very quickly.

Ok to fix the con in 12v gear Im going to use mechanical stops as the Jrks sense a major increase in amp draw but no movement so it actual stops the motor until the reverse direction is detected and they move on. Another way I though about was limit switchs way past the deg of movement I want but not too far that the pot breaks, so if a motor overshoots it clips the limit switch and kills the power to the driver. It should only reach these limits if it goes out of control.
Not sure if such a safety thing can be incorporated with the VFD's but would be a good safety point to have.

Hi
The 180 means the you get from about 0.4v to 4.9v over half a turn 180,
these pots will go 360+ full turn without damage, but you feedback voltage is 0v for the second half of the 180

regards Terry

eaorobbie & Terry,
THANK you for your information...
Based on what you have said - and the fact "anything" can happen and will - I'm going to go the standard route with pots and how they're mounted. FYI, my ac/gearheads produce 117 FT.LBS torque ea. and no doubt can tear a pot off in a heartbeat.

I think I have found a vendor here in US to furnish the "aluminum" timing pulleys. Other AC users/platform builders have put the large pulley (about 3" in diameter") on the gear shaft and mounted the smaller pulley on an "L" bracket, then attached the drive tape/belt on. The rotatations of the pot becomes about 10:1 ratio.

So fellows - please give me your input:
I will need a multi-turn pot?
1. Do they make a multiple turn magnetic contactless pot? IF so, what do I need to look for - nomenclature wise.
2. If a contactless/magnetic will not work as a multi-turn - please suggest what I need in voltage/resistance, whatever.

The reason I'm so hell-bent on contactless is, "in my mind", they will last much longer than the wiper type...just don't want to be changing pots every week.

As stated a hundred time by me - I am learning, but still very ignorant to the electronics and terminology.

Thanks so much,

Tom

The pulley idea is a good solution its what is used on the early CKAS units.
But a better solution if a controller would allow is the use of encoders, again contactless and well as long as led in simple photo emitters dont fail could last a very very long time, and can be done quiet cheaply too. If I could run an encoder on the Jrks I would.

In turn there are contactless rotary hall effects on the market that have not stops in them so if the motor did overshoot or spin out of control there is no risk of them breaking. Could be an alternative to a std Pot.

eaorobbie,
I "am" going to go the pulley route - safer in case gearhead shaft goes past the 180 degree - THEN I'm not concerned about tearing off the pot - Also - I can get the rod-end closer to the crank arm - "Just makes better sense to me"....Thanks for pointing these things out.

At this point, I am sure I will be going with the SimAxe board, Opto_Iso., PWM's as yobuddy had in the diagram he sent me...

Would you or others be so kind as to point me to the Contactless Rotary Hall Effects Pot that I do need??? Again, with the timing gears, for 180Degrees needed for my setup, the smaller timing gear on the pot will
make 5 turns to the 1/2 turn of gearhead/180degrees.

I do apologize for my ignorance, but as I stated in the beginning of my build thread, mechanical I'm pretty good, electronics, pretty bad - so help is really needed.

@cthiggin The hall effect pot that you have linked to above looks like it's almost perfect for your situation imho. You need to register 160 degrees and the pot is 180 degrees, which is 89% of its range already. Therefore no adjustment in the turn ratio is necessary. These pots don’t have limit stops on them so there is no way to twist them off in an over rotated condition although your lever still could knock on them (they are multi turn but only register over the 180 degrees and repeat after 360 degrees). As telfel stated if you go over or under the 180 degrees they will just not output a voltage until they are back in the registered range. Also they have a noiseless output and the only thing to wear out on them is the bushing the shaft runs on. I think they’re rated at one million turns if I recall correctly. If you spend about two or three times what those cost you can upgrade to ball bearing shafts, which I think, are rated at 5 million turns. Also they are 12bit output over the 180 degrees, which gives an accuracy of around 22 positions per degree (but your controller likely wont register that high). While a hole and a setscrew would probably be the best method to mount the pot shaft to the motor there are low-tech methods that can also work well. I call my method hot melt glue from a hot melt glue gun! To keep the body of the pot from rotating just get some bailing wire or something similar and wrap one turn around the pot and clamp on it with the pots nut. Then attach the two ends of the bailing wire to mounting points somewhere. If your rig goes haywire your pot should simply get knocked off ready to be glued back on.

Edit: Upon looking at the data sheet these are rated at 10 million turns. The ball bearing ones must have been 30-50 million.

BlazinH,
THANK You -
I am going to go the Timing Pulley/Belt Route - Due to the fact I can keep my rod end closer to the crank arm, which "me thinks" will provide more stability...I had concern that a 4" x 1/2" bolt may cause more stress - PLUS, I don't want to have to drill out and tap the gear shafts.

The Timing Pulley/Belt Route will rectify this problem, and as stated, I have finally found a source for the pulley's to go over a 1 1/8" gear-shaft.

From the great info you sent, we are assuming that I am going with the mechanical setup of attaching pot to gear shaft?????, Like initial photo at beginning of thread.

Please help "here": By going the timing gear/belt route, you will see in the two pics that "these builders" use a large pulley on gearshaft, a smaller one on the pot - which means we are going to turn the pot about 5 complete turns per 180 degrees of gearshaft movement - IF this is correct, then I would need ????? a different pot - the timing gear/belt builders seem to be using a wire wound multi-turn pot. I want a pot that will last longer and more precise, such as the magnetic contactless one I listed. I don't see how this would work considering I'll be turning the pot shaft about 5 times??

MAYBE this is an acceptable thought process: Go with the same size timing gears on both the gearhead shaft and on the pot - then we have a 1:1 ratio and the above pot would work great - DOES this seem acceptable to you folks??? Like I said, DUH, I need help in these areas.

Thanks for your information and help, most of all your patience.

Have a blessed one.

Tom

@cthiggin I am unaware of any digital type pots myself that are multi-turn. So if you are going to go through the hassle of a timing gear/belt route anyway, you will probably want to go with a high quality wire wound multi-turn pot or a mechanical encoder. One issue with a digital pot is that since all the output is over a short range, they are very sensitive to any movement. So any play or backlash added by your gear and belt will have a move adverse effect then when the range is spread out over several turns. Also the SimAxe only has a 10bit adc (analog to digital converter) so the digital pots input will be cut to 10bit also. You could go with a mechanical encoder too but your controller must have the means (inputs and code) to read it because it is outputs a pulse or count, not a voltage. But since an encoder doesn’t need an adc that does eliminate the 10bit bottleneck.

Ps. I guess I’m not getting why you can’t just hot melt the digital pot to the end of the shaft where the big gear goes and get rid of the gears and belt altogether. I prefer not to use high quality wire wound pots myself because they are as much or more expensive than the digital pot mentioned, still have noise on the output, and have friction that will wear them out sooner.

Edit: Or maybe you just prefer not to use southern engineering, lol!

BlazinH,
Now you not only "see" but "know" why I need you guys. Makes sense explaining gearhead and timing gear/belt backlash...

OK - Your "hot glue method". Your saying to hotmelt the shaft end of pot to the center line of gearhead shaft ??? then use a method of stabilizing the pot body, via a wire, bracket, or whatever works? (ME thought that you wanted me to hot-glue the ass-end of the pot to the shaft - couldn't figure that out since the terminals are on the back side - GEEZE)

Like my first pics in this thread - (We are eliminating the drilling of gearhead shaft/and tapping - This "would" also bring pot closer to gearhead shaft. IF this is the case, then my rod end shaft could be closer to gearhead shaft - but STILL provide enough clearance to miss the pot, just in case of over-run? AND, best of all, I could use the proposed contactless pot.....
Hopefully, I'm on the same page.......IF so, let me know and I'll proceed in ordering the pots.

PS - Keep it simple, stupid - really applies to me.

Thank you so very very much.

Tom

Yes, shaft end to shaft end. I have found though that the pot body needs to be able to "float" a little bit but still not twist. If there is any perpendicular movement between the motor shaft and the pot shaft, the pot needs to move with it otherwise it will fatigue the glue and it will eventually pop loose. But the bailing wire, for me at least, is strong enough not to twist with the shaft movement but still give a little so it elevates the stress on the glue joint. The bailing wire is attached at my frame, not to the motor itself so it must give a little. I am using ball bearing hall effect encoders at the moment though and there is virtually no resistance at all in the twisting of the encoder shaft. I do have some of the pots you linked to however and am planning on using them in my new build. But with it though I will be able to drill a hole in the lever to connect the pots shaft to but will still use the wire to hold the pots body in position and hot melt glue to hold the pots shaft in the hole.

Hello again friend BlazinH,
ME got it - I'll order the pots tomorrow.
The brain juices tells me I have seen a round plastic disk with a collar in the middle that would or could fit the gearshaft on one side, and then glue the pot on the other side....think I saw it at Home Depot. I can glue the disk to the shaft end, then the pot shaft, pressed into the center collar, hot glue, and that part is done. Secondly, I can make a bracket to affix the pot body to make it solid and stationary - I have two threaded holes already on top of my gearhead and I can easily fab a good bracket to secure. ?? Sound like a plan???

A ???? - Is there an indexing mark on the pots - I WILL know how my crankarm will move , 0 to 180 degrees, but if I glue the pot shaft out of the acceptable range, then I'm screwed ..
Let me know on this please...

I'm very excited to have a resolution to this....I sure didn't want to drill the gearshaft...nightmare, and tapping would be even worse.....PLUS Factor for me - being able to have the crankarm miss the pot in case of over-run....LOVE IT!

Thanks BlazinH for taking the time to educate me......it sure is a blessing to have knowledgeable folks here that will help in any way they can. Sure wish all of "we" Americans" had that attitude.

Tom

@cthiggin If the plastic disc you speak of will hold the pots shaft so it cannot shift sideways from the gears shaft at all then a hard mounting of the pots body may work. But it is imperative that if there is any sideways stress between the pots shaft and gears shaft when the motor turns, it is either eliminated or accounted for (for example if the pots shaft is not glued perfectly in the center of the gear head shaft it will wobble slightly). Without any give in the pots mounting point, all the wobble stress will be applied to the glue joint and it will fail at some point. One of the last things you want to happen is have one come loose in operation unless your controller tests for this and shuts down automatically or you have limit switches. How I mount mine with the bailing wire is form the wire in a half circle around the size of a soccer ball. Make a loop in the middle of the wire around your pots neck and tighten its nut (would be better though to have a piece of metal with a hole for the pot to go through and then attach the wire to it somehow). Now find a place to attach the ends of the bailing wire (maybe even still to your gearhead). You can bend and manipulate the wire to your needs. Just make sure that is strong enough when your done so the pots body can’t rotate with the gear shaft, but can still move slightly right, left, up, and down as necessary with the stress. Make since?

Also there is no index on the digital pots that I see anyway. You will have to use a voltmeter to find the correct position for them and make your own. However, if they are not correct after gluing the shaft all you have to do is loosen the nut holding the pots body, rotate it to the correct position, then tighten the nut again. Also, with the bailing wire setup I use, you can make minor adjustments to the center position by just bending the wire slightly one way or the other.

PS Your welcome for the help. I hope my education is not a bad one though as I'm sure some others here have different opinions.

Hello BlazinH,
First, your education is a great one. We all have ideas, some good, some bad - but I find that little bits of good information, assembled together, great things can happen.
I'm going to take a stroll to HomeDepot this weekend, and look isle by isle, (like a lady looks for a dress) - and find what I am looking for. Yes sir, I am aware that if the pot and disk are not dead center - well could be a terrible thing -

Please send me a photo of your bailing wire form - just need to get my head around it better.
After my post last night, it sank in that IF the pot index is out, loosen and rotate...DUH.
Thank you for you info and patience. Please stay in touch and hopefully you will send me a photo.

I will report back in a few days to update.

Have a blessed one,
Tom

I agree! I also like the old adage,"Think out of the box". Just because its not the "norm" doesn't mean its a bad idea. However, if it is a bad idea it can be quite painful in wasted time and money.

I used hot melt glue originally because I wanted to have a not so permanent connection in the case they needed to pop loose to avoid damage. But now that I have the kinks knocked out I have continued to use them because they continue to work good for me as is. There are, of course, many other methods of gluing so something like JB Weld may work for you too! Good luck in your hunt at Home Depot! They have gotten sooooooo much of my money with Lowes and Ace Hardware a close second!

BlazinH,
You talk about Home Depot getting your money - last year, 7 months of virtually ALL my spare time, had me remodeling our 3br and 2 bath upstairs - + painting our master bath - (had to take wallpaper off - geeze), new sinks, fixtures, granite counter-tops - AND this really IS the reason I'm starting my motion sim a year late.......12 grand and counting...

I'm looking forward to HD for my treasure hunt - going to be fun, "me thinks"... I'm continuing to do a lot of study on "our" method - just want to get it right the first time and for it to last a loooooooooong time.

I'm also considering Gorilla Glue - that crap is just superb for about anything and with the pot/shaft, won't be hardly any stress at all......
Please stay in touch and I'll will also. Please do get me a picture if you have time.

Thanks so much friend,
Tom

Hi folks,
Weather still too cold to spray my finished steel platform base. Have been busy cleaning motors, lubing gearheads, cutting, shaping etc. - you all know the drill - AND have my pots ordered..
so still moving right along.

Another question... My U-Joint is out of an automobile - the driveshaft yoke assembly... I started dis-assembly so I could really clean it, ready it for welding, then painting.
I moved the yoke assembly, both directions, (pitch, roll) and what I noticed was that the "ears" (I call them - they hold the caps and needle bearings) would NOT allow for the pivot of a total 90 degrees each way.

I put a 3' yardstick at centerline, moved the yoke both ways and I do have a lot of pitch and roll.

Over the years on the platform type sims, most of the builders show their u-joint and they lay completely down in each direction.

DO you folks think this may be an issue??

Have a blessed one and thank you.
Tom

If your uni joint has a min of about 25 deg to me is enough for most simulation modes, actually the more angle you induce the hard the motors will have to work to bring it back to centre. To me I would work out the angles you want the sim to be able to roll and pitch too and look for a uni that will comply to that.

Yes place the axis of the uni with the axis of the sim or you will find the motion wont roll through its centre properly and it all feels like it is off centre.