Dome rotation motor slipping problem..

[Jannis]

I have installed the motor and roof now, with a 5:1 reduction gear. I have also managed to get the sidereal speed pretty much OK, but i've run into a slipping issue...

While i can slew the roof quickly to a new location without any issues, the issue is when leaving it at sidereal speed. It appears that while the friction was enough for slewing, it's slipping after a while on sidereal speed. I would normally assume it was the other way around, but i guess i was wrong...

Now this probably leaves me with 3 options:
1) Get better friction somehow. Better "rubber" for the motor wheel, or get a friction coating on the steel ring, or both? I also don't know how this rubber will work in the winter, as after an hour in the freezer it got rock hard, so i think i need a better option for it regardless (optionally i can install a heating element on it, keeping it at 20-30c to keep it soft).
2) Abandon the current design and weld on a chain (bike chain for example) on the entire inside of the ring and install a gear on the motor instead of a rubber wheel.
3) Abandon my current idea of sidereal tracking, and change it into a slaving function, even though i finally got the tracking right...

I would naturally prefer option 1, but does anyone have any idea on  how to increase the friction? I tested it last night and it worked well for about an hour, then it was stuck spinning. I'm retesting again now after cleaning the steel ring with acetone in case it was any dirt left on it, it's only been 30 min so far but it's working for now. I'm shooting a time-lapse of it now to better spot the problem, as sidereal speed is simply to slow to see anything. In fact, it took me about 30 minutes yesterday to notice it was stuck...

Does anyone have any other thoughts on how to solve it easily, or any idea how to get better friction?

The motor system:

 

[sloz1664]

That's not rubber you are using it looks like reinforced hose and it will slip.. As a simple experiment, cut a piece of 120 grit emery cloth the same length as the cicumference of your "reinforced hose" and glue it in position. When the glue has set try the sidereal test again. If it works then you have a solution.

Steve 

[Jannis]

Thank you!

Yes, you're right, it's a reinforced hose, but i thought it was still rubber though? Or are they made of different materials?

It's funny you should mention using 120 grit cloth to test, cause i just got back in from testing exactly that. Well, not exactly like you said though, as i tried to fasten some anti-skid tape to the steel ring instead, but didn't have much success. Again it works perfectly at slewing speed, but at sidereal speed it just starts spinning again.
I tried it both taped around the steel ring, and just underneath, but it simply ripped it apart at sidereal. I then tested with another stronger tape, and it worked OK for the first 15 minutes before that got stuck too.

Do you think it will be any difference if i glue it on the rubber hose instead or the metal ring? I was thinking i would get a better grip if it was attached to the steel ring instead, but it appears that while i get enough grip now, the tape can't handle it.

I'm worried if i attach the sand paper to the drive wheel instead, it might cause irreparable damage to the steel ring before i notice anything, should it slip again.

Edit: I have a can of Plasti-Dip, I wonder if maybe this would be any better, assuming it's tough enough?

 

 

[sloz1664]

Hi Jannis

The reinforced hose is nylon which unfortunately has a slippy surface. I see now how you have tried to use emery tape to try and cure the slippage, which unfortunately didn't work. The best non slip material that is available and cheap and could be suitable for your application are tyres. Or more specifically bicycle tyres. You could purchase the lightweight racing tyres which have the smallest diameter. These cut, stretched out and glued on the underside of the tubular ring, the reinforced hose also replaced with a piece of tyre. You would then have two mating surfaces which are made of materials which are designed not to slip.

Steve

[Rusted]

You have only point contact via the hose drive. The axes cross at your drive point.

Instead of adding tyres I'd suggest the skinniest racing size inner tubes. 25x700C?

As a keen cyclist I had loads of old tubes at one time. No doubt others do too.

Or ask the local bike shop for some secondhand racing tubes?

It is quicker on a ride to replace a tube than to mend a puncture.

So over the years they build up in the bike shed.

The inner tubes can be easily slit and glued in place.

Rubber hose is still available. Gas welding kit or fire extinguishers?

An alternative is to make a V-pulley which fits your exact needs.

It will offer better contact on the tube buy jamming the tube in the V.

You can [carefully] turn a plywood pulley on an electric drill if you don't have a lathe.

Or rotate your pulley against a suitable router bit if you have a router. Drill plus router?

I've made large pulleys from plywood using a drill and sharpened screwdriver cutting tool resting on a support block.

And I still have most of my own limbs. 

[Peter Drew]

My first thoughts were to change the drive "wheel" profile as suggested by Rusted to improve the contact area or to strongly spring load the drive support wheels to take up any variations in loading as the unit rotates. We use an adjustable "pinch" wheel to help with friction to drive our very large dome.   

[RayD]

I'd go with spring loading the drive wheel. It's highly unlikely your ring is totally flat, and you may find that it is just raising a tiny bit at a specific point, which in sidereal is causing an issue, but at slew is being skipped over.

I'd change the hose for rubber also.

[Jannis]

Thank you very much for your advices guys!

I have now removed the anti-slip tape again and will first of all try to get the nylon hose replaced by some real rubber and see if that alone works.
The most easy option would be to use racer bike tubes as they might be small enough to just stretch over the current drive wheel - but are the bicycle tubes strong enough, and are they not made to have minimum friction for easy installation/replacement?

I don't think there would be space to glue a racer bicycle tyre to the underside of the metal ring, but it should be space to glue on a bike hose to create that rubber<->rubber contact.

I also thought about making the drive wheel U or V shaped to give larger traction area, but i'm not quite sure how to get that done, but i will look more into it.

A spring loaded system won't be possible due to there is no space above the metal ring, however the drive wheel is intentionally slightly oversized for that spesific reason, so that it takes most of the weight off the 2 closest wheels. There is probably about 25-30KG resting on the drive wheel.

I also learned it the hard way today that adding strong sandpaper to the metal ring and holding it in place with strips was not a good idea, as the roof actually got pushed off the wheels when it hit the strip. So i have now attached 3 extra guide wheels to make sure that sliding off simply will now be possible ever again. This howevere, made it even a bit heavier to rotate, but at least now i know the roof will stay on the wheels...

 

[Rusted]

A spring can be fitted almost anywhere. You just need to rethink the support system to isolate it.

Change it from being fixed to the structure and made into a vertically sliding, or even a pivoted unit.

A V-groove 'pinch' wheel will provide immediate and greater benefits over your hose drive. IMHO.

The V-groove will have a different drive diameter to your present, slippery hose, of course.

I wonder if a steel pulley might not offer higher friction than an aluminium one against your steel rail?

Similar metals have much higher friction between contact surfaces than mixed metals.

Do you have access to a lathe? Or a local school or college with a metalwork department?

BTW: Normal [rubber] cycle inner tubes are very grippy but I wouldn't bother getting involved until you've tried a V-pulley.

The rubber is not really designed for high local pressures and creeping friction drives.

It will probably tear itself to shreds.

[Jannis]

Thank you for all the help so far guys, and also to Xplode for your PM with great info and tips!

I have so far tried to use Plasti Dip to rubber coat a thinner rubber hose(real rubber this time, but too thin), and while it indeed had enough friction and didn't slip at all, it ripped apart at sidereal speed - even with ~10-12 thick layers.

Because of that failed attempt I've been checking up options for having a custom made wheel with a U profile to match the steel tube and have it coated with high friction rubber designed for this use - but i have yet to get a response back from the company on the price and time needed to make it.

In the meantime, i might have found a more usable option. I found a piece of solid rubber designed for resting for example engines on to absorb vibrations. This is seemingly very tough rubber and designed to not not let go of the metal. It also appear to have decent friction, and is just about soft enough to get a good grip without twisting and tearing.

I sanded it down to a rough U profile so it now have 3-4x the gripping surface compared to a "flat" wheel. Since it only had a short M8 bolt i gave a flexible connector a go as i needed M12.

I was skeptical at first with regards to how much torque this flexible joint would handle - and i guess i had reason to...

Well, so, plan B... I sanded a bit of the M12 rod down to roughly 9mm and threaded it with M8 1.25 threads and used a bolt connector to solve the issue. To make sure it would stay put i also filled the connector with high strength epoxy glue before attaching it all.

I run a test with slewing, and it did not slip even one bit. I'm currently running a sidereal test, but naturally it will take time before i know if it works or not - however it's been running for ~90 minutes apparently without slipping som far!  
I'm thinking that if this still slips at some point, but the rubber can withstand it, my easiest option would probably be to attach another motor on the opposite side so that it will rotate more easily.

 

[Xplode]

Very nice solution, lets hope it will work when it gets colder too

[sloz1664]

That's what I /like about this hobby. Never give in attitude. Hope this works without the pain of fitting a new motor

Steve.

[Jannis]

Thanks guys!

Well, it appears i wasn't that lucky though. ~2:15 in it started slipping again...
It's an improvement though, as with the 3 added support wheels pushing from the inside the nylon hose was unable to even run for a few minutes, and even started slipping at slewing speeds, where it would previously work for 30-60 min.  I will start with a cleaning of the steel ring again and run a retest though, as i have been lubricating a bit with silicone grease, and i don't know if maybe there was some spilled..

I can still sand down a bit more to get more traction area, and i probably don't need the 3 support wheels pushing so hard on the ring. I will work more on it, but i think i will also order the gears for the 2nd motor just to be sure. If it could run for two hours with just 1 motor then it should probably not be any problems at all with two. More work and cost involved though, as well as waiting time for the parts.

[Peter Drew]

Looking again at your white nylon ring support bearings, it seems they are running with plane bores on fixed steel bolts. This could cause more rolling friction force if so and this would be improved greatly with ballrace bearings added if possible. The lateral support units are probably ok as they are, the plate they are fitted to looks reasonably flexible.  

[Jannis]

You are correct there Peter, the nylon wheels are with just a center bolt. I have used silicone grease on them though, and i think that helped a bit. I agree ball bearings would help even more, but I'm not sure how i could attach it properly. I do have a set of brand new skateboard wheel bearings available though, maybe i can make them fit somehow. I'd need more bearings in the end though, but 4 wheels with bearings would help to start with.

The bracket for the lateral support wheels are a bit flexible to compensate for any unevenness, but it's quite stiff. I will try to make it slightly less tight and see if that changes anything.

A 2nd gear set, pulley belt, and motor driver have been ordered now anyway just in case, but what a joy the 15-45 day shipping from china is...

[Rusted]

I would not rely on the tubular rail remaining pristine or even dry.

Even condensation will greatly reduce friction.

Silicone lubrication and a friction drive is not a marriage made in heaven.  ;-)

[Jannis]

32 minutes ago, Rusted said:

Silicone lubrication and a friction drive is not a marriage made in heaven.  ;-)

Indeed! There is no more risk of silicone grease spilling now and the metal appears to be clean, but i will give it one round of acetone just to make 100% sure. I don't yet know if condensation will be any issue, but i'll keep an eye on it. 
It may have enough friction as it is now after some slight modifications, but some more testing will be needed. It run fine for 4 hours last evening until i had to call it a night, and the rotation speed was only about 2-3% off (i will improve that later, but it's a great starting point i think).

[sloz1664]

Good to see you are making progress.

Steve

 

[Jannis]

Ok, so progress have been made..

I noticed an increased resistance when rotating as the silicone lube started to wear off, and it reached a point where it got so unstable i was afraid to rotate it too fast and with too much acceleration.
Naturally this also caused it to slip even more easily and something just had to be done. Luckily the 2nd gear system just arrived from china and i already had a 2nd motor and a spare motor driver by hand.

I've now replaced one of the 5 guide wheels with another drive wheel on the opposite side of the first motor to give a much more even rotation force on the roof. This means the ring is now resting on 4 guide wheels and 2 drive wheels. I have also now installed 4 lateral support wheels to increase stability (and safety?), especially in strong winds.

First test was like pure night and day. The stability and precision was fantastic compared to before, and at no point was the roof stuttering due to too high friction. If i wanted to move it just 1mm in one direction, then i could easily do that now. With 2 motors there was also no more "backlash" (can i call it that?) where it needed to move certain amount on one side before it starts moving on the other.

I also upgraded my hand controller with an Arduino nano to get more precise pulse frequency. I'm currently running the first "long run test" now to verify speed and confirm it's not slipping.
While frequency change is no longer just a push of a button, the Arduino is still easy to access for re-programming should the frequency need to be changed - though i can't see why it should be needed once i find the correct frequency.

[sloz1664]

Excellent news. I certainly know how frustrating this is at the time. But doesn't it feel great when it starts to work correctly.

Steve

[Jannis]

Indeed it does!

I'm sitting outside giving it a first real try now, and so far it seems OK. Also doing my first RGB test of the night sky, so crossing my fingers.
At least i still see stars and not the roof inside the observatory after 2 hours, so that's a good sign!

[Rusted]

Great discussion, problem solving, real progress and excellent images all make for excellent reference material.

Well done, Jannis! 

I think those long, skewed screws will have to be replaced though.

Push longer screws through some alloy pipe to act as straight and rigid stand-offs?

Old aerials [scrap TV and radio antennae] provide lots of good aluminium pipe in a number of diameters.

[Jannis]

Thanks!

Yes, those long screws will be replaced. I just needed a quick solution to stop the bracket from flexing and screwed them in at different angles on purpose. It probably wasn't needed anyway though.