Gearing/Drive question

[wcarrothers]

This is my first post...

I've gone though and done my mill conversion using Jeff's plans. That went really well and works Great in my opinion..

I'm working on the design for my 4x8 router mill now.. But I have a question about gearing...

I plan on using ball screws for the Z and Y axis. Was thinking I would gear the motors similarly to the way the mill was done (ie a .2 pitch ball screw then make the motor pully half the size of the screw pullie. So 10 turns by the motors per inch I think it is)

Plan on using 1100 oz servos from camtronics (by the way)

And a rack drive on the x axis.. The question is .... to get a rack drive to produce 10 turns of the motor per inch it will take a pretty large pullie on the rack. (figure a 1" pitch diameter pinion gear will go 3.14" per rev it would take a large difference between the drive pullie and motor pullie to do 10 turns per inch..

How important is it that the motors on the different axis run at the same ratios? Guess it isn't that important since they all will move a different amount of weight and thus to really syncronize they would all have to be geared different

Any suggestions of rack/pinion ratios would probably help me.

Also what about direct driving the racks and screws? I've seen people do it with stepers but not with servos so much. Should I do it that way? I kinda figured gearing them the same as the mill plans was a good idea to reduce the stress on teh motor..

b.

[ger21]

Originally Posted by wcarrothers How important is it that the motors on the different axis run at the same ratios? Guess it isn't that important since they all will move a different amount of weight and thus to really syncronize they would all have to be geared different

Not that important at all, really.

Imo, the proper way to gear your router is to pick a target rapid speed, say 500-700ipm, and gear to reach that speed near the top rpm of your servos. If you just pick an arbitrary gearing, you may end up limited in your speed.

I personally wouldn't want any slower cutting speed than 300ipm and double that for rapids, with a 4x8 router. Anything else will be painfully slow once the initial amazement wears off.

Your spindle may be the limiting factor in regards to cutting speed, but you can always take multiple light cuts at high speeds if the spindle is underpowered

[JerryFlyGuy]

Gerry's right, the programing will take care of differences in gearing, I would recommend looking at two more specific area's on your design when specing the rack/pinion. One is resolution. You might as well try to get it close to the same as the other axis. The axis w/ the worst resolution is your "weakest link" so to speak. The other thing is if you use a 1100 oz stepper, your going to have trouble getting a 1" pinion which will take the torque of a 1100 oz stepper. Esp if there is a significant reduction on it. Bostongear.com has lots of info in their Open Gearing Cataloge of how much torque etc..

If need be we can get you some more help on which to pick and use. Make sure you use the 20deg gearing as its better for Linear motion over the 14.5deg

Hope that helps..

Jerry

[wcarrothers]

Think I'll be looking at a 1500rpm max from my servo drive with the power supply I'm using (acording to dave at cam tronics)...

Mcmaster car only has 14.5 deg angle pinions and racks so being pointed in the right direction (boston gear) would be of help..

For some reason over the years I've lost my better mathmatic ability and of course hopefully can get some balpark help here anyway. Sounds like that should not be a problem



b.

[JerryFlyGuy]

Ok, I'll jump at it

The Base level webpage at Boston Gear is here you can download the entire cataloge, or go here for the seperate chapters to download.

This is a direct link to a pdf which tells how to select gearing as well as has all the torque ratings for different gearing [ a must read]

Lots of good info on this site. If your designing it via CAD either 2 or 3d, they have all their products modeled so you can download them in DXF/DWG/IGES etc.. The nice thing is.. you don't have to buy from them [ I think they are a bit pricy] just get the spec's you need and then buy locally to you.. [look up Bearing supply in the phone book] most likely you'll get a cheaper price locally, I know I did.. and you save lots on shipping etc.. The local product [ a Martin Gear] was exactly the same as the Boston stuff [ I don't know of any standards in Gearing but either there is, or they all just make them the same I don't know]

What is your continuous torque on the servo's? I assum 1100 is peak? What size output shaft do the servo's have? I'd be looking for some precision planetary gearing over belts, Ebay usually has lots.. lately they have been CHEAP too.. Look for Parker-Bayside, Thompson, or just "Planetary Gear reducers NEMA" on there.. I get emails pretty much every day of new listings.. You shouldn't have to pay more that $100/each for them.. Don't get more than about 10arc/min backlash and you should be fine.. unless your looking for SUPER accuracy.. [ 10 arc/min is like 0.05deg or something.. >memory there<]

Hope that helps.. Ask away w/ any specific's hopefully I or someone can help [I don't concider myself an expert, but I've designed and spec'd out mine.. so I have found some of the "got ya's" like torque ratings etc..]

Jerry

[JerryFlyGuy]

One thing I was going to mention was make sure your gear drive is pressure mated to the rack [ spring loaded or via gas shock] your pinion is going to try and push itself out of the rack and you need to keep some good force-pounds pressure to keep it in there.. it will elliminate any backlash between the rack and pinion also.. once you know the gearing your using I can get you the pounds/force number to use..

Jerry

[wcarrothers]

Cool.. I was thinking of planitarys but was also thinking belts to keep things formilure with what I did on the mill conversion. Of course it all comes down to cost as with most of us.. I will find the shaft size out (still waiting camtronics to get the motors in again) and start watching ebay for reducers. have to decide if I'll do belt of planet final drives. I think how i'll be setting up the rack it would be better as a belt system due to space. But the gantry could be direct by planetary.

b.

[JerryFlyGuy]

I think you'll find that Planety reducers take up alot less space than belts [ even just a 3:1 is pretty large. These reducers are about 3-4" cubed [ 3 x 3 x 3], very compact. If you can get them for $100 each.. you'll be a fair ways ahead of belts [ at least belts that are rated for the loads your gonna be using --like an XL or bigger]

I actually designed my system for belts the first few go-arounds.. to get the 5:1 reduction that I wanted it took 4 pulleys,4 bearings and alot of plates and what not to get it all mounted and held together, yet light.. it was fairly compact.., still its no match for the planetary reducer.. not even close.. and o/a cost for the belts was about 2.5x more.. and you had to adjust the tension etc.. yuck.. compared to a point and shoot planetary..


I've attached a couple PDF's of the belt reducers..

Jerry

[wcarrothers]

Ya your convincing for sure... add one more expence to the list I guess.... Darn..

b./

[wcarrothers]

Well been thinking of the targets Ger 21 said to do I'm wondering..

If most people run their screws direct drive with the 1100oz servos or steppers. Since I est the max rpm of the motors with the power supply I'm using is going to be ~1500 rpm and the screw is I think 5 turns per inch I'd see my Y move around 300inch per min? If I geared it down by 1/2 or something with a belt I'd see 150 or so max...

Just wondering if the motor has enough grunt to move smoothly with out gearing down. be of course easier to direct drive but want to be ready on the other hand to gear it down if I have to change the design..

Figure I will probably go 2:1 on the Z axis since I don't think it will have to move as fast as x and y since it will only have maybe 12-18" travel anyway and I will want the motor might some day be lifting a small gear head mill up and down.(becides the design already has it lifting itself and the router up and down)

Still wondering about how x will end up. Hopefully will fina a low cost planetary. 10:1 with maybe a pinion with 3.5inch of travel per rev would put x around 525 inch per min? Seems a little faster then y might end up... Course could go down to a 3" size pinion and hit 450 which is closer to the y target.. And of course x will be draging more weight.

Hard to eyeball all this for sure.

Maybe I will setup for a belt final drive on the Y even if I start out with a 1 to 1 ratio so I can change it up or down easier then re-working..

Also I think the 1500 rpm target might be a little on the high side for max rpm of the motors since my power supply was on the lower side of the voltage output when I assembled it..

Got to keep my eyes on the targets some how so my initial initial amazement lasts..

Any suggestions?

[JerryFlyGuy]

Gerry is correct in that you need to set a target and then design towards it. It does matter, however, what your system is to be used for. I think its common to look at a router and/or a plasma cutter and think.. speed is the need.. and then accuracy, followed by cutting loads. When it comes to mill's however I think the proper order would be, cutting forces, accuracy and then speed. Not that speed is less important on a mill, but if you can't put out the needed lb-force to do the cut, speed is moot. I would sit down and decide what force level you want and then design towards that. If you want 150lb's force you would balance the force requirment w/ the need for accuracy. Some people might argue that accuracy is even more important than the cutting force, and for some this could very well be true, if all depends on what you want to do with it.

If I was to sit down and design a mill from a clean pc of paper this is the order in which I'd do it:
1) Decide what it is that I'm milling [ ie, steel, alum, Ti, copper.. etc..]
2) Decide what size cutters and spindle size [ how much power are you going to have available]
3) Decide or calculate a max cutting force needed from above.
4) Take that number and add 50% to it [ unless your VERY comfortable w/ the number you've come up w/, in that case only add 10-25%]
5) Take your servo motor and find out what type of continuous load it will put out.
6) Calculate the needed reduction to get to the force you want.
7) Using the calculated reduction #, check your resolution.
8) Using the calcualted reduction #, calculate your acceleration and max speeds.
9) Is the mill resolution w/in your spec's? If yes continue. If no, decide if a higher or lower encoder count is needed or economical.
10) Is the mill rapid and max cutting speeds w/in your spec's? If yes, great your done! IF no, decide if you need a larger motor or can live w/ the lower rapid's, revise accordingly and re-do from step #4.

Note again.. some might do those in other orders, depending on what they think is most important. It might sound like a long trek to find your answer, in reality, w/ a spread sheet you can go through and spec this stuff out in a matter of minutes..

One last pc of advice, its worth it to get it right on paper before you ever start cutting metal [or wood] for your machine. Once you have all the item's nailed down.. stick to the plan. If you have to revise your plan, make sure to go back and check through all your numbers to see if your new changes effect something else.

Also,if you don't know what the cutting forces are and aren't sure how to calculate them [ ie a wood router, routing wood] just do some simple tests using a fish scale and a marked length, you'll soon get an educated guess. Use a torque wrench to check how much power is required on a mill [ while milling a pc of steel or whatever..]
You don't have to be terribly accuracte w/ your tests [ the more accurate you are, the more confidence you'll have in your numbers however] you just need to get a ball park number.


Jerry [hopping off my soap box now ]

[wcarrothers]

All cool suggestions Jerry..

I think for the most part my 4x8 will be working wood. I don't really see it doing much metal work since I don't do very large stuff, getting the small mill head is out of budget and I already converted my harbor freight mill to cnc to do metal stuff..

I'm also curious about the 20deg pressure angle stuff for pinion drives.. Mcmaster car sells 14.5 deg and of coruse it would be easier to buy their stuff (since I will probably also get my bearings through them and stuff as well) rather then going to bearing supplys inc localy or something else to do 20.

How important is that deg angle I know it means more face contact for the gear probably but is it important enought to be mandatory?

b.