Build Thread- need help with Z-axis setup

[zeed]

Could somebody explain what is the diference between this 2 setups of the Z axis carriages?
On firsth pic carriages are bolted and dont move while on the second they move with router.
Which is the better option?

pics attached.

[CarveOne]

Either way will work. I designed mine similar to the second method in order to lower the weight of the moving parts and to have the weight of the motor and rail assembly attached to the Y axis. I still ended up having to lift and lower 10 pounds of weight. All of the moving parts are aluminum carriages and plate material. I'm happy with doing it this way. There will be other opinions on this subject.

CarveOne

[PaulRowntree]

This is a timely thread, as I am wrestling with this right now; I would appreciate any and all comments on my own logic here.

With the second model the spindle is probably on a Z-offset dogleg to provide clearance from the stationary portions of the Z axis. With thin working stock, the spindle is always working at its maximum extension, and with the maximum twisting moment due to the large tip-to-bearing distance. In fact, the tip-to-bearing distance is always the same, and always large. However, with the first model, thin working stock can be raised on support blocks, thus reducing the tip-to-bearing distance, and (hopefully) increasing the mechanical stability. With thick stock being cut, the mechanical aspects appear identical to me.

Based on this, I was leaning towards the first model.

Does this make sense?
Cheers!

[ger21]

With thick stock being cut, the mechanical aspects appear identical to me.

If you cut all the way through, isn't the bottom of the cut the same, whether the material is thick or thin?

Bottom line is that the more Z travel you have, the more chance it'll have to flex, so it needs to be more rigid.

The first pic can give you more clearance, because the spindle can move farther below the supporting structure.

[stevespo]

I am using a Parker actuator for my Z axis. After 3+ years of use, I recently switched from setup 1 to setup 2.

I did not want to lose any rigidity, but I really needed to take advantage of the additional clearance that the 2nd style of mounting would give me. I wanted to be able to bring a longer tool up and over a tall part, and setup 2 allows me to do that. I also thought it might be helpful to avoid dragging 30+ lbs of actuator and servo motor up and down on every move.

So, far I haven't noticed any loss of accuracy or difference in cutting. This is mostly thinner materials with the spindle extended downward to reach the table. Raising the work, as Paul mentions, is a good option. I also think Gerry is right in that the biggest issue with flexing is the length of travel, which probably doesn't change much in either case.

Maybe someone wants to run some numbers to quantify the difference?

http://www.cnczone.com/forums/showpo...&postcount=103

Steve

[ger21]

but I really needed to take advantage of the additional clearance that the 2nd style of mounting would give me.

I was thinking the other way around, but I see what you're saying. I think either one could be built differently so that both gave the same clearances, but the construction of the gantry will play a part in that as well.

That's why when you design a machine, you start at the spindle and work from there.

[PaulRowntree]

Since this is important, let me try to explain my thinking more clearly.

Imagine a machine that has a Z range of 6". When the tip of the tool is at the table height (minimum of Z) then the distance from the tip of the tool to the middle of the Z axis bearing is going to be 8"-10", so lets say 9" for a working number.
With Z at this low position, I think that the flex will be about the same between the two models, because the tip-support distances are about the same, as Gerry states.

However, when the tip is raised above the minimum Z, the tool-tip to bearing distance of model 1 will decrease, which should make model 1 stiffer at high Z. Model 2 on the other hand will always have the same tool tip to bearing distance, so the flex of model 2 will not decrease with higher Z work.

Since we can always choose to work at higher Z with thin materials by raising the working surface, I thought that this would give the stiffness advantage to Model 1. As CarveOne points out however, there will often be less moving mass in model 2 because it does not move the tracking. It should still be possible to have the motor stationary though, in both models.

Comments?
Cheers!

[louieatienza]

I've pondered this also. It's been many moons since I've been in physics class however. So I'll try to convey my thoughts with the most basic principles I can recall...

It would seem to be that design 2 creates a longer moment from the tool tip relative to the bearing, thus it would require less torque to "deflect" the system in design 2 (since torque is force multiplied by distance, one can see that the shorter the distance from tool tip to bearing is, the more toque is needed to "deflect" the system.)

Another thing to consider: With design 2, usually the carraige is made taller than design 1 of same Z movement, which might also be a design consideration. With design 2, the ultimate effective length of travel along the Z ends up being the distance from the tool tip to bearing, since the bearing can only travel to the extent of the rail. With design 1 the rail can be made as long as needed for the required Z travel; this is most apparant with 5-axis gantry routers.

In regards to the mass: I would think one would want as much mass as close to the tool tip as possible and practical. Just my hunch...

[CarveOne]

Physics aside, sometimes you do something against all normal practices in order to get what you want. And then compensate for any unacceptable conditions. There are all kinds of trade-offs to consider.

Mine was set up for a 6" clearance to the table with the Z raised and a 5" or 6" bit in the router. That's why mine looks so odd. This is for cutting molds from foam or Renshape materials. I expected to have some flex in Renshape so I added side plates to compensate for the router plate hanging lower than it normally would. If I were designing another machine for flat sheets only it would have a lower gantry height and a different Z setup.

I'm a nuts and bolts guy. Make it work, then fix it.

CarveOne

[louieatienza]

Without a doubt, the materials you intend to machine ultimately determine the strength you need to "engineer" (I use this term loosely ) to.

Pf course, both designs have their merits and shortcomings, and that has to be weighed against the final use of the machine and the capabilities and means of the individual.

Don't think your design looks odd at all; it's infact the way a linear stage or actuator is designed, right?

I guess I should clarify... I don't think one design is better over the other; it all depends on one's needs, tooling, and supplies available.

[CarveOne]

The younger engineers right out of college quickly learn from the older engineers that you can spend weeks on calculations and modeling or you can develop a general plan and go build something in the lab. Take your choice. The older engineers have years of experience and no longer need to do all of the math unless they are unsure about something in particular. In the last 10 years of my telecom engineering development lab work the ever shorter development cycles didn't allow time for much of the fancy stuff.

CarveOne
(Who once was rumored to have said "Engineers are like oranges. They squeeze the juice out of you and then throw you away.")

[ahren]

Having done the CAD design for both of the systems you referenced above, I feel qualified to throw in my $0.02 .

The issue with z-axes in general is that they are inherently cantilevered -- you need to hang down below the gantry to reach your workpiece. The question is whether you get enough cantilevered distance from the length of your router + tool, or if you need substantially more travel. If your router hangs down most of the way, then the fixed motor (like option 2) makes good sense -- easy to assemble, not a lot of moving mass on the z. I wouldn't hang it below the gantry the way Glacial Wanderer did -- that just means you need to raise your gantry higher, but the idea is sound.

If you need more travel, then option 1 makes a lot more sense. For option 2 to give long travels, you need a long cantilever to get you down to the workpiece, and an equally a tall axis sticking out above your gantry (to allow for clearance in the "up" position). The path of force to the tool is a long one, making this prone to vibration, and you need a beefy plate for your cantilever to avoid flexing.

I'm working on a new plate to support an "option 1". I'll post a photo tonight.

Ahren
www.cncrouterparts.com