[ebangor]

Hi everyone. First post here, and I'm hoping someone can at least point me towards some good resources. I have a few questions:

First, is there a reliable formula for determining minimum screw diameter to move a given mass?

My work wants to build a rig to spray test panels. Although we usually don't paint anything much bigger than about 8" x 8", we do paint several at a time, and occasionally work with 2ft x 2ft panels. Given the sizes we're looking at, the initial thought was to make a frame that allows about 5ft of travel on both the horizontal and vertical axes. The spray gun itself we want to use is only about 1.5 lbs. Our intent is to construct as much as possible out of 8020. Including the weight of a charged paint hose, the entire setup shouldn't need to lift more than 12 lbs vertically. Precision location isn't as much of a factor as speed control is, since the speed of the gun over the surface has a direct relationship with out finished coating thickness. Speed maxes out around 400in/min horizontally, and we can go much slower vertically.

Given some of the solvents we use, I'm hesitant to go with a ball screw, simply because I'm not sure how well internal lubrication will hold up. The most common solvents in our paints are acetone and dimethyl ether. My thought process for going with a lead screw is that it's far easier to visually inspect and re-lubricate than a ball screw if the aerosolized solvents start to break down the lube. It's also part of the reason why I'd rather not use a belt system to drive anything.. I don't really have any experience with ball screws, however, so I'm not sure if this is a valid concern or not. Am I worried about nothing? Given the substantial size, would rack and pinion be better for horizontal movement?

Those two solvents also mean that the entire apparatus is going to be explosion proof, which I'm sure won't make sourcing motors or a PLC easy. But that's a problem for another day.

[hfjbuis]

Those two solvents also mean that the entire apparatus is going to be explosion proof, which I'm sure won't make sourcing motors or a PLC easy. But that's a problem for another day.

Depending on the zone you have to take more ore less effort to make it safe. Put all electronics and motors outside the cabinet. It is not difficult to to seal the hole for the shaft of a lead screw. I am not sure about limit switches but to be sure, they can be activated by a sealed shaft and placed outside the cabinet.

Given some of the solvents we use, I'm hesitant to go with a ball screw, simply because I'm not sure how well internal lubrication will hold up

A polished steel lead screw running in a bronze nut could do without lubrication. Covering the lead screw won't hurt. You could construct in a way the lead screws are fully covered by the table. A lead screw is easier to clean/maintain than a ball screw.

First, is there a reliable formula for determining minimum screw diameter to move a given mass?

There probably is a formula but your demands aren't high. I would go for a 20 mm lead screw. My small lathe has one and its length is about 40". It can handle a lot of load and still is accurate (0.02 mm)

[joeavaerage]

Hi,
un-supported length and axis speed will determine what size screw you need, given that the weight (momentum) of your axes is very modest.

The general rule is that is the screw travel is 5' then consider then max unsupported length to be 5' also. If, to achieve a given axis speed then the screw must rotate
at a given speed. For instance if you need 5m/min axis speed then with a 5mm pitch screw it must rotate 1000 rpm. To prevent the screw from whipping it will need
to be a certain diameter. If the unsupported length is reduced then the diameter of the screw can be reduced also. The required diameter is very sensitive to unsupported
length, thus it might cost $500 for 4' screws good to 1000 rpm but $1500 for 5' screws at the same 1000 rpm.

NSK publish data and design equations for their ballscrews and is a good resource.

I would consider ballscrews, they are just so much better, smoother and without lash than any trapezoidal screw ever made. Note that ballscrews are used in semiconductor
manufacturing where the gases are way, WAY, WAY more corrosive than 'mere' acetone and ether!! The suggestion that they could not be used in a painting application
is nonsense.

Rolled ballscrews from China are an economical way to approach a project like yours. They are good to about 50um per 300 mm, zero to vanishing small lash.
If you don't require the axis to move more than 2500mm/min (500 rpm with a 5mm pitch ballscrew) then 20mm diameter screws sound about right at 1.2- 1.5m
length. If you want to go faster, and/or especially longer length then 25mm or 30mm may be required.

Craig

[ebangor]

Hi,
The general rule is that is the screw travel is 5' then consider then max unsupported length to be 5' also.

Does this apply even with bearing at both ends? I know I don't have the numbers worked yet, but my initial idea was for the bottom end of the screw to be fixed and the top end still radially supported. Does that still get treated as though the top end of the screw is moving completely freely for speed considerations?

[joeavaerage]

Hi,
the NSK design formulas cover the situation where there is only one bearing and one end free. Naturally the possibility of the screw whipping is very much
exacerbated by not having both ends supported.

My rough and ready:

The general rule is that is the screw travel is 5' then consider then max unsupported length to be 5' also

actually applies where there is 5' of screw between supports, namely the ballnut and the further away bearing.
There again consult the NSK publication.....its a good resource.

Craig