[onepimpinsummer]

ok, I've been lurking(reading) for months now and thought it time to post. I'm probably a little nutz, but I'm buildling a large(4x8x1.5) 5 axis cnc router for milling wood and foam molds mostly. Most of the machine will be welded steel and some aluminum. I'm going to use servos. The gantry will be heavy(150lbs maybe because of the large z axis). Will two 850/oz on the x, a single on the y and single on the z should be enough power? I will probably put a counterweight or spring on the z to help with the weight. This machine is fairly large, hence the heavy gantry. I don't need lightning speed, but don't want a total slug either. I'm going to use rack and pinions. With the RPM the servos run at(approz 4000rpm), I think it wise to build transmissions to slow them down. Maybe 5:1 so I can get 800rpm to the 1 inch pinion. That should run me about 2500 IPM of travel at full throttle right? Those calculations sound right? 800rpm x 3.14(circumference of 1 inch pinion)= 2512 inch per minute. Thats fast for such a heavy gantry I guess. Maybe gear down to 8 or 10 to 1 and get closer to 1000 or 1500ipm? Sound like a good plan? slower/faster? bigger servos? ideas?

If I build transmissions(don't know where to buy decent/economic tranny's), should I use timing belts or gears? The belts would be easier to change ratio later if I wanted to. any suggestions or input greatly appreciated.

[Splint]

If I build transmissions(don't know where to buy decent/economic tranny's), should I use timing belts or gears? The belts would be easier to change ratio later if I wanted to. any suggestions or input greatly appreciated.

I would suggest you go with belt and pully gearing. You can buy precision reduction gearsets but they are very expensive and you cant mix and match parts to get the ratios like you could with a belt and pully system. You'd be best to do that as you'll need to keep backlash to a minimum because gantry type five axis machines typically have difficlty holding tollerence (compared to a 3 axis machine) because of the long z axis and the presence of the 4th and 5th axis which is an opportunty for error.

[wcarrothers1]

Personally I don't see 5:1 being enough of a reduction to a 1" diameter pinion. I'm running 10:1 + a 18drive to 24 tooth drive belt final to my rack with 1" pinions and I can get 800+ IPM out of my gantry.

5:1 would be pretty hard on the servo in my machine (and my gantry is easily >then 150lbs) and I have 10" of Z..
also 5:1 would not yeald very good accuracy as the servo would not spin very fast.

After building I'm pretty sure I would not be a fan of my machine if I did rack for the Z axis as when your servos fault out with out a support mechanism (ie spring or something) your tool would fall easier with rack then with say ball screw drive due to gravity..

I got my 10:1 thompson gear drive off a guy on ebay for 70-80 bucks as I recall. he had others listed that didn't match what I needed but after a few emails he had exactly what I needed..

Be sure you get the right input shaft size for what you need. For my trans the input coupler/gear would be around 200.00 new from thompson if it were not right to begin with.

Belt reduction from those ratios (ie 10:1 or more) would be a pain to maintain compaired to a planetary if you can find one at the right price. I would have easily spent more then 80 bucks on belts bearings and pullies to do it that way.. Heck I spent about that much on the 2 pullies and belt that made the final drive as it's now built..

b.


b.

[JerryFlyGuy]

I've also gone the planetary route, I don't have them running yet but they are sure alot more compact than the belt reduction's I'd planned.

Also, plan for twice as much reduction as you need, reason being is you can always go to a larger gear [twice as large] as I can garrentee that @ 850oz-in you'll exceed the rating of a 1" pinion. You'd be better off w/ a 10-1 reduction and a 2" pinion. Even this might not be large enough.

Bostongear.com has some great spec's for these issues. Just look under their 'open gearing' heading.

I'd recommend a 12DP or less gearing solution, also stick to the 20deg, not the 14.5

Also as wcarrothers1 say's plan for a ball screw on the Z axis, the rack/pinio idea works great for the x/y axis however.

Also plan to press the pinion into the rack, w/ a 850 oz-in servo w/ a 10-1 reduction a 12DP rack, w/ a 2" pinion you'll get 194lb's force [ the pinion trying to push itself out of the rack] I'd recommend a couple gas shock's to overcome that, 150lbs each @ a min.


HTH
Jerry [another large mill builder]

[wcarrothers1]

I'm runnign 1100oz servos through my 10:1 then I just got 14.5deg rack./pinions from Mcmaster. They seem to work well so far but agree the 20deg's are better.



My rack/pinoins are hard mounted (ie no gas shock or spring) holding them in to or against the rack. Any inconsistancy is taken up by the flex of the 1/2" shaft that sticks ~1" from the carrier bearings from the side of my gantry. Although my rails and rack are pretty much dead on streight so I don' treallly see a need making the shaft's able to spring although if a piece of something gets gbetween the gear teeth and rack then maybe that would be a reason..



Rack drive is cool for some things though compaired to screw..

for example I like being able to manually move my gantry with the power off or rather with out having to power up my system which is something I can 't really do with Y or Z power off.

b.

[JerryFlyGuy]

Ya, I thought about making mine solid but liked the ability to adjust for wear as it moves as well as adjust for any slight inconsistancy's in the rack. I'd be a bit worried about the shaft bending. In your case, assuming it's a 2" pinion, the force is only 17.8 lb's of drive out force, which is peanuts for a 1/2" shaft overhung 1"

Now, I have a confession to make, I transposed a decimal place on the previous calculation instead of 194lb's it will be 19.4 lb's.

Sorry about that.

The end result is you could do it either way, your going to have more backlash issues with a solid mounted system over a floating one, but that all depends on your resolution requirments.

One other nice thing about having a floating connection as wcarrothers1 has said, you can move which-ever axis you want manually. I'm not sure if this is a good or bad idea, as afterwards you don't have a resolved position, but it might be usefull in some case's. W/ a floating connection, just use a air cylinder instead of a gas shock. This could be controled by a solinoid/relay and controled right from your console if need be.

Lots of possibilities.

Jerry

[onepimpinsummer]

thanks guys. I actually wasn't planning on using the rack and pinion for the z for exactly the reasons you mentioned. Only for the x,y. The one inch pinion idea was just thrown in there. Probably a good idea to go with a little larger pinion. I was thinking about it last night after I posted, and after reading your input here, I am going to take the advice and go with much more reduction. Bigger servos too if I can find them for a decent price.

[wcarrothers1]

I just eyeballed a lot of the stuff with my build.. came out pretty good. I did the 1" pinion because I didn't want the ratio to go back the otherway. Did the ball screw drive on Y because I didn't want to have to hang another motor (ie more weight) out on the gantry which was already getting heavy. (my 1100oz's are 12+ lbs each) So that was another reason I didn't do rack on Y becides the trans also can double the weight of the motor..hehe

Although I would love to be rack in order to manualy move the motor power off.

When I say manually move the motor I'm saying move the gantry out of the way for loading or unloading or something like that. No other reason to have manual movement.

b.

[turmite]

Onepimpinsummer

Do you have your 4th and 5th axis designed yet. If not I have leads for you on that as well.

I have run a Digitial Tool rack and pinion machine that was set up for dual z axis all running racks and never have had a problem with either the rack or the heads not staying up when not powered. Now with your setup and the extra weight that will likely be a problem for you. My machine uses BishopWisecarver dual vee rails and the pinion is held captive into the rack by the rail system. Kinda need and again, never a problem with that either.

Mike

[JerryFlyGuy]

Mike got any pictures? I've been tossing around several idea's for building my own 4/5 head. Are how are you ensuring there is no backlash in the rack interface on the ring gear?

Jerry [ I love to see how other people do things, prompts idea's of my own ]

[Rance]

...Do you have your 4th and 5th axis designed yet. If not I have leads for you on that as well...

turmite,

If you are referring to a horizontal rotary axis (like a lathe) then I'd be interested too. Thanks.

[turmite]

Well guys on the a axis that lays along the x I designed it and had a friend with a machine shop build it. It is a simple transmission based on timing belts and gears. The problem was when I finally got it working there was not enough holding torque and my part would twist out of the way when the cutter made contact. Part of this was a design error and part was the simple fact I designed the 4th axis to save on high dollar wood, which took center of mass considerably off to one side.

The b/c head combination can be found from a couple of suppliers. One is Donny from Whiteriver who makes them like Rab made his for his 5 axis using the sat drives. He can provide some pretty heafty motors for the two heads as well.

The other is Colin Doughty who lives in New Zealand and has designed a nice system where you can buy one head at a time if that is all you need. You can find both sites here by searching for 5 axis and looking for the name hodarthestrange (Donny) or searching for a link for Colin. If you can't find it post back and I'll try to find them for you.

Mike

[smoker]

Hi, I am in the design fase of my machine which is similar to this one.
I see that you recommend gas shocks pushing the pinion with a force of 150 lbf against the rack. I do not completely understand this.

So, in rest (no movement) the pinion is pushed against the rack with that force?
Is that not very hard on the gearbox bearings?

Sorry for this stupid question maybe but I am not a mechanical engineer...

David

[JerryFlyGuy]

Nope, not a stupid question

Gearing, in its nature, causes a force which tries to push the gears appart [when they are in use] This force is just part of the nature of how gears work. It has to do with the tooth angles and all that jazz [ 20deg or 14.5Deg tooth angle creat's a vector which puts out a force which tries to drive the two gears appart] This is easily calculatable. Now, some people just fix the gear drive so that the pinion cannot move [only rotate] and rely on the flexture of the overhung shaft to flex as the gearing is used. This will work, however over time it would need to be adjusted due to part wearing and ever increasing backlash. On my machine I've put the planetary reducer and the drive motor [which is bolted to the top of the planetary] on a pivoting mount. This mount is then pre-loaded w/ a 130lb gas shock to keep the pinion pressed into the rack. Will it damage the planetary reducer? Maybe, it all depends on what your reducer is rated at. I know mine will take the loading because it's rated for over twice the pressure that I'm applying [however I can't remember what the finite number is as its been a while since I looked at this part of my design] You can also adjust the mechanic's of the floating mnt to keep the pressure to an acceptable level by just adjusting the arm at which the shock is mounted to the floating/pivoting mount.

A good example of possible loading would be here http://www.baysidemotion.com/Web/Pro...0?OpenDocument

In my case a Nema34 Gear head could take several hundred pound radial load before it gets to its rated load. I'm not using a Parker but the basics will be similar.

HTH

Jerry [I'm no engineer either ]

[smoker]

Hi Jerry,

I see better now with your very good explanation.

However some minor questions just to make sure I have it correct.

The force by a gas spring is progressive, so in rest how much pressure is there from the pinion to the rack? the full 139 lbs?
Then when the motion and acceleration is applied, there will be even more pressure?

If there is that much pressure between rack and pinion, does this not take down a lot of the effeciency of the drive system? Do you then need bigger motors or more gearing down?

David

[JerryFlyGuy]

I don't actually know at what point the pressure rating for my shock's is listed. Whether its at full compression, 50% or at 1%. I went a little high for this reason. In reality I think I could have gotten away with a shock that was 25% of the pressure rating of these ones but whatever. I guess thats just designing for you, there are always compromises. If you want a perfect gear mate, it has to be pressure loaded, you pay however with some level of reduced efficency. If you choose to just fix/mnt the pinion, you may end up with accuracy issues, or flexture/fatigue issues with the drive shaft later in the life of the machine.

I'm sure there will be some loss of efficency as there are two forces fighting it out in the process of the gear turning. Again, I don't know what that will be. Each of my steppers is 'supposed' to put out 400lb's force at their stall torque so I'm sure the added stress shouldn't be too big an issue. The other thing is pre-mature wear. Being as I have a larger force level pushing the pinion and rack together, I know it will increase the rate of wearing on the two. I still expect that the life of the gearing is going to be measured in years and could well still exceed the life of the machine. I'll know more about these questions here in the next few days. I'm in the wiring stage right now and hope to have a machine which is under computer control at some level, by this weekend.

Jerry

[smoker]

THX! Jerry

BTW I following your wiring post with great interest as well!

[JerryFlyGuy]

No problem, Also its worth of note that I'm no expert in these things. I do, do industrial design for a living but I work with a team, so we're always bouncing idea's off each other. There are times when I suggest something and they look at me like I'm an idiot

I should be posting some more on that thread here shortly. I've been working on that schematic and will post it when I get a bit farther along. The basic wiring stuff is pretty straight forward. I'm working to get my 'non-electrical' head around wiring up a mod-io, Digispeed and a Sj200 VFD. The rest is pretty simple

Jerry