Motor required for spin casting, 10-100rpm

[tiviny]

I'm attempting to spin cast a parabola, and aim to hang my spinning vessel directly from the driveshaft of a motor to get it to spin as cleanly as possible.
Are there any PWM (12v ideally) 10-100rpm dc motors than could support 10kg thrust load?
I would strongly prefer to not use gear systems, but I suspect it may be ineviatable.
Also, could AC motors be used?

Thank you, Tom

 

[jim hardy]

my old standby - a Ford windshield wiper motor.

available from junkyard or roadside derelict.

 

[tiviny]

possible to make a pwm controller? forgive my ignorance

 

[jim hardy]

We're all ignorant just on different subjects. What's spin casting ?


"""possible to make a pwm controller?""



that is very do-able, though i never have..
here's a kit
https://www.amazon.com/dp/B006L7SWJ2/?tag=pfamazon01-20
i've used Ramsey kits before and been pleased.

and a DIY hobby article
http://www.dprg.org/tutorials/2005-11a/index.html

to experiment,
A car battery charger will run it, and on 6 volt setting it runs slow. Put a headlamp in series and it'll slow down ..

if you use a battery charger to power the PWM control add a huge capacitor in parallel.

 

[tiviny]

Can a windshield wiper motor be changed from ten to a hundred rpm? I'm guessing 5A PWM will be good enough for a wiper motor?

I want to get form a mold of a parabolic shape, and spinning liquids form parabolas. So, spinning plaster should set into a parabola. Tests look good, I just need to get it all smooth.

 

[jim hardy]

i used them for rotating fishing poles while the clear coat set, about four RPM. I had then a 4 amp battery charger and Variac adjustable transformer.

What's highest speed on your car windshield wiper? Time a few sweeps - seems to me it's at least a sweep per second.

I'll check my car today...

sounds like a neat project - what's end product? (if not a trade secret)

 

[tiviny]

Reflector for a spotlight. Still working out how to apply reflective materials.

 

[jim hardy]

sounds neat.

might ask on the math forum whether you'll get a parabola or a catenary, though...
just to be safe.

probably you already know.

 

[tiviny]

It will make a parabola. I know that at least, ha.
How does the motor perform at 4rpm? I didn't know wiper motors could do that.

 

[jim hardy]

well - i was just trying for real slow rotation, which it did very well.

as i recall it had good torque and that was with low voltage applied. all i had for a supply was my battery charger and Variac - this was like 1972.

They're just a DC motor with worm gear speed reduction.. If it has a shunt field (i think so) you can set field field current and apply PWM to armature for very stable speed and better torque.

 

[tiviny]

If it has a shunt field (i think so) you can set field field current and apply PWM to armature for very stable speed and better torque.

Im afraid I don't understand what you mean. Two inputs into the motor, one going through the PWM?
Also, is there some way to check if it is shunt field? I'll probably get two, so I can open one.

 

[jim hardy]

take a look at this introductory video..

http://www.wisc-online.com/objects/ViewObject.aspx?ID=IAU13708

indeed get a couple they're certainly inexpensive.

 

[sophiecentaur]

If this process lasts only a few minutes then I'd be inclined to think in terms of a variable speed electric drill, or even a pillar drill with the pulleys set to minimum speed. Perhaps a flexible / bungee coupling could help - or a further belt drive if the speed is too high. Either way, the messy electric / mechanical bit would be straightforward and you could get on with the interesting casting bit.
What material are you going to be casting?

 

[jim hardy]

that's a good thought Sophie.

i just assumed he was making a sort of permanent shop fixture.

anyhow - reason i logged back on is i just looked under hood of my 83 Ranger.
Windshield wiper motor has its earth(ground) wire of course,
and going into motor are two small black and one large white wire.
I expect large is armature and small ones are field. Field is brought out with both leads separate so motor can be reversed to "Park" the wipers, and control speed.

Last one i opened (decades ago) was a three brush design - has better speed regulation.

 

[tiviny]

Aim is to cast plaster of paris as bulk filler. if it doesn't set evenly then maybe a thin layer of epoxy resin on top. I can only assume this will be good enough.
I considered a drill but I will be using this more than once, and also plan to make a 2m wide rotating version where a decent motor will come in handy.
As regards the 'messy electric/mechanical bit' its going to have to be very messy anyway, I am going to have to make some sort of individual screw adjustment system for each of the container support struts for nice spinnage.

I assembled a 20A PWM kit today and got a wiper motor. The motor has red, black, yellow, brown and green wires. With R/B to +/-, touching yellow and brown makes the motor run for about 2 seconds, stop, then go for another 2 seconds in the same direction. No other combos seem to do anything, however I just touched y/b/g and its not working anymore so that's something, ha.
Maybe I might need an older-style motor?

 

[sophiecentaur]

How about a gramophone turntable? It may take some time to get up to speed but it will be steady and you can always tinker with the capstan diameter. 78rpm could be just what you wanted!

Just read the "2 metre"bit - so forget it haha.

If you plan to go so big then surely your main problem is with the turntable / suspension. A real crude solution could be a potter's wheel mechanism. I think the speed is controlled by sliding a belt on a tapered shaft.

I must say, I love the idea of producing a large parabaloid by spinning. You may find bubbles in the plaster are a problem because they will affect the density of the liquid mix at different levels and that would spoil the parabola shape, I think. It all depends on the accuracy you need.

 

[tiviny]

Bubbles are a big problem, they accumulate right at the very bottom, where it has to be the most precise. I am sure there's something better than paris plaster.

I built a potters wheel with a pulley attatched to a high rpm sewing machine motor. There were all sorts of problems with levelling, the thing has to spin invisibly to get a smooth surface. Thats why I am hoping hanging it will make this easier.
With hanging, tho, the vessel has to be constructed evenly, else I suspect it will spin with a wobble. I plan to get a potter to make me a container that is roughly a paraboloid shape (maybe I don't need the plaster, and could use resin right away . .?), as 'throwing' looks like it would give consistent thickness.

 

[jim hardy]

/ start babble/

maybe a centering bearing at bottom? an old bicycle cone & ball bearing could support a mld axial load. A boat trailer wheel would be very stout..

http://www.northerntool.com/shop/tools/product_200365090_200365090

http://www.northerntool.com/images/product/images/128006_lg.jpg

/ end babble.. /

 

[tiviny]

ideally there should be no axial load. I was looking at this thread. I think it needs to be supported from the top. 3am, sleep time.

 

[sophiecentaur]

I think you have to make the support (underneath) as massive and rigid as possible. Hanging your spinning drum from above is to invite instability. When speeding up, there is always the possibility of a wave forming in the mix and a serious wobble to develop.

I reckon a good solid (concrete) base with a spindle on two, well-spaced, bearings should allow the spinning to establish a good parabola even if the centre of the drum is not on the axis. A 'live' back axle from an old scrapyard vehicle would do fine.

On the subject of bubbles, why not use a very wet mix of mud plus cement, vibrate for some while, to let the bubbles out, and then spin it up gradually? If the water in the mix is allowed to drain out slowly, the shape can establish itself and then the cement can set. Possibly not a strong casting but it could easily take a coat of resin as a female mould for copying.

I reckon you need to overkill a bit on the mass and rigidity of your machine and be sure your investment is not wasted, were it to fail. Good luck with it.

 

[tiviny]

I'm not keen on the idea of a base, I think hanging will give the best stability all round.
Even once I get the spinning plate level it is near impossible to correctly place the container right in the middle. Even the slightest bit out is not good enough. Hanging the container (accurately!) eliminates both this problem, and the issue of having the container pointing at zenith. Even with no sections of material in the container to aid the plaster in spinning up, waves do not occur, so I assume the same will be so hanging.

Do you mean dirt-mud with cement? I'm pretty sure I just realized, however, that the easiest thing to do would be to just do a final layer of watery resin on top of the plaster, as this needs to be SMOOTH, and the grains in the plaster are too rough for a good reflecting surface.

 

[jim hardy]

i just noticed your post about the windshield wiper motor.

""The motor has red, black, yellow, brown and green wires. With R/B to +/-, touching yellow and brown makes the motor run for about 2 seconds, stop, then go for another 2 seconds in the same direction.""

It sounds like your motor has internal electronics. Can you remove a cover and get at wires going to brushes and field? Was it junkyard or can you post a datasheet ?

Fords were the same from about 1960 to at least mid-80's. I never messed with any others.

 

[tiviny]

Its just from a junkyard. I will try to disassemble it now.

 

[tiviny]

ok photo attatched. yellow and brown appear to go into the gear housing.

 

[tiviny]

ok I took apart the gear section, a photo is attatched. Its pretty simple, 3 contacts touching tracks on the rotating gear, changing how the motor operates as it spins. I am going to muck around with this for a bit.

 

[jim hardy]

oh my gosh - a permanent magnet field!

i expect a third brush opposite the red wire

power from that brush to one opposite should run the motor.
bypasssing switches entirely.
With that permanent magnet field your PWM will control speed.

Probably prudent to place a rectifier across the motor terminals. this is called a "flywheel" diode purpose is to absorb motor's inductive kick.

here's what i expected:
http://www.freepatentsonline.com/3497788.pdf

 

[jim hardy]

and here's how they were used in Model A Ford days and at least into mid eighties.
http://en.wikipedia.org/wiki/File:Three-brush_dynamo,_circuit_(Autocar_Handbook,_13th_ed,_1935).jpg

with permanent magnet field instead of a winding , disconnect third brush. i don't know how they used it in your motor.. is there a semiconductor or integrated circuit on the assembly?

 

[tiviny]

It doesn't look like it.
There is indeed a third brush opposite red.
Hooking up the wires that come straight out of the motor, red and green, works a treat.
About that rectifier, the PWM kit came with a 100nF MKT capacitor to be connected across the motor terminals. I forgot this bit and ran it for less than a minute, hope i didnt do too much damage. Do I place the rectifier inside the case, bridging the red brush with its opposite? What if it gets bumped and gets caught in the mechanism?

 

[tiviny]

as you can see in attatched, red and green are next to each other. i only applied power through these contacts and it still worked well. But ideally I want to only use red and black, correct?

there are no semiconductors or ICs anywhere.

 

[jim hardy]

I think motor will run better with power applied to physically opposite brushes.
Of course polarity determines direction of rotation.


I'd mount flywheel rectifier outside not inside..
It will be reverse biased in normal operation so cathode goes to positive..
get a six amp or so, ... unless there's already one built into your PWM which is likely if it's made for motor control.


the capacitor just cuts down on RF from arcing at the brushes so may or may not affect your PWM if absent.
Probably running it without cap hasn't hurt a thing.

How's the torque feel to you? Your machine is bigger that i thought, still those wiper motors are stout.

old jim

 

[jim hardy]

Red and black, aye !

 

[jim hardy]

great photos by the way.

How do you post them?

 

[tiviny]

Torque is good. It really only needs to overcome air resistance in the end. How does my motor being bigger affect torque?

There is enough space to make the capacitor fit easily. I put 4 of these capacitors on the board, and the last one is supposed to go across the motor terminals.

I just use the attach file option to post photos.

While I am here i might as well ask, could you think of anything that would attatch to the shaft thread? Its only meant for a worm gear, I not sure how to secure anything to it.

cheers jim

 

[jim hardy]

I think you'll want to attach to the shaft coming out of nylon gear. In the automobile the windshield wiper is driven by an arm held on by the nut.
The worm drive screw on main motor shaft is pretty high speed, that's why i envisioned instead using the same output shaft as automaker did, after worm reduction..
Worm gear drive is pretty smooth and gives high torque at about the speed you need.
I see from your photo that shaft has flats on it where wiper drive arm was attached..
One of those shaft extenders for electric motor should grab it (unless it's metric), or epoxy a friction wheel to it.


"..motor being bigger affect torque? " i don't follow.. torque will be in proportion to motor current.

The two empirical equations for any DC motor are:
Voltage across motor = (K * flux) * rpm
where flux is field, in your case from permanent magnet so relatively constant
and K is proportionality constant for your specific armature.

so if you measure voltage at any speed you know product (K * flux)
and
Torque = 7.04 * same(K * flux) * Ia , where Ia is motor current and torque is ft-lbs.
Are you in a metric country?

Keep us posted.. This is interesting...!

 

[sophiecentaur]

I'm not keen on the idea of a base, I think hanging will give the best stability all round.
Even once I get the spinning plate level it is near impossible to correctly place the container right in the middle. Even the slightest bit out is not good enough. Hanging the container (accurately!) eliminates both this problem, and the issue of having the container pointing at zenith. Even with no sections of material in the container to aid the plaster in spinning up, waves do not occur, so I assume the same will be so hanging.

Do you mean dirt-mud with cement? I'm pretty sure I just realized, however, that the easiest thing to do would be to just do a final layer of watery resin on top of the plaster, as this needs to be SMOOTH, and the grains in the plaster are too rough for a good reflecting surface.

If you hang the container, loosely, and you don't get the centre of mass right on axis there is nothing to stop the whole thing precessing and going wild. The thing needs to be constrained so it won't do that. If you use a firm, massive base then the accuracy you need for mass distribution is much less. In fact, the only 'accurate' centring you need is for the geometry of the paraboloid. That would be very easy to arrange as long as the container is a good circle. Let's face it, a lathe will give you a good circular shape, whatever the shape or position of the workpiece. You don't need the workpiece to be 'balanced' if the chuck holds it firmly enough. I can't think anyone would try to 'turn' a chair leg by hanging it from a rotating bungee support (for these reasons).
I understand that you may have had problems mounting your spinning plate satisfactorily but how would this be easier with it hanging? You still need to be able to adjust the position of the centre plus two tilt angles. A baseplate with diagonal slots can be used for side to side adjustment and three screws can give tilt adjustment. It may sound over the top; it's not a trivial thing you're proposing. But all this could be a lot easier than other methods of producing a paraboloid. What is the required accuracy of the shape? Is it for an optical, microwave or sound application? Will you be measuring the paraboloid as you spin it up or will you rely on the speed to define the focal length?