gcode resetting my surface zero to Z-0.200"

Post Reply
ger21
Vectric Wizard
Posts: 1592
Joined: Sun Sep 16, 2007 2:59 pm
Model of CNC Machine: Custom DIY
Location: Lake St Clair, MI, USA
Contact:

Re: gcode resetting my surface zero to Z-0.200"

Post by ger21 »

I’ve looked at several machines and they don’t use a bearing on the bottom end, so I’m just going to let that 5 start be loose and free, but keep the npmount as the bottom Z axis stop. That makes things kinda self align.
If the nut is not perfectly aligned with the bearing mount at the top, it''ll bind. It can't "Self align", because the screw is constrained at the top.
Gerry - http://www.thecncwoodworker.com

User avatar
TReischl
Vectric Wizard
Posts: 4597
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 48X36X7 RP 2022 UCCNC Screenset
Location: Leland NC

Re: gcode resetting my surface zero to Z-0.200"

Post by TReischl »

Mr.Chips wrote:What holds the Z assembly when you power off?
Nothing but friction. Mine occasionally will slowly slide down until it bottoms out. Sometimes it doesn't.
"If you see a good fight, get in it." Dr. Vernon Johns

User avatar
TReischl
Vectric Wizard
Posts: 4597
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 48X36X7 RP 2022 UCCNC Screenset
Location: Leland NC

Re: gcode resetting my surface zero to Z-0.200"

Post by TReischl »

wmgeorge wrote:I have never heard of a having rack and piñon on the Z? All my machines have had screw on the Z and 2 had ball screws on X & Y, this last one is rack & piñon X & Y, ball on Z of course. There are large commercial machines with ball screws on all, instead of R&P on X&Y but they are not your common hobbyist machines :D
Now you have heard of R&P on a Z axis. Not mine, but they do exist.

When it comes to large commercial machines it depends on what type of machine is being discussed. For instance, virtually all of the large format laser and plasma cutters use rack and pinion for X/Y. Some of them use rack and pinion for Z. And yes, the Z axis on a laser has to be precise to maintain a precise focal distance. They also use a sensor on the Z that accounts for waviness in sheet material. That is not the case with milling type machines probably because of the reduced travel distance in X and Y precludes the needs for rapid speeds in excess of 2000 IPM and on an industrial machine using large tooling the screw provides much more torque than a R&P provides.

As I discussed in an earlier post there are issues with ball screws on larger machines. A two inch diameter ball screw will sag even if it is preloaded (stretched). That is not so much an issue if the machine has shorter travel distances.

Now here is one for you: In the late 70's I designed several large laser cutters that used a rotating ball nut to achieve much higher speeds. At the time no one was using R&P because of the inaccuracies associated with the racks. (the problem with the racks was that they did not move so an encoder had to be mounted to the back of the motor, that did not actually indicate where the machine had moved only that the motor had reached a programmed position). Another issue was moment of inertia on long ball screws and not being able to preload them sufficiently to preclude whipping. Those machines also included linear glass scales to provide a closed feed back loop for the control. On a large ball screw there is a certain amount of twist which causes problems in the servo loop gain. By spinning the nut and monitoring where the table was actually going solved those issues.

Once industry realized that R&P was a viable option to obtain high feed rates racks started being machined to close tolerances. Prior to that racks were more of a crude way of converting rotary motion to linear motion. The only way to get any precision was via things like glass scales.

BTW MrChips, do NOT run a steel pinion on a steel rack. What you want to have wear is the pinion, it is much cheaper and easier to replace.
"If you see a good fight, get in it." Dr. Vernon Johns

ger21
Vectric Wizard
Posts: 1592
Joined: Sun Sep 16, 2007 2:59 pm
Model of CNC Machine: Custom DIY
Location: Lake St Clair, MI, USA
Contact:

Re: gcode resetting my surface zero to Z-0.200"

Post by ger21 »

Our big router at work has a 15ft long ballscrew. It's stretched about .03", and has a spring loaded support in the center to keep it from sagging. It has a rotating nut, and the support pivots out of the way when the nut passes by.
Gerry - http://www.thecncwoodworker.com

ger21
Vectric Wizard
Posts: 1592
Joined: Sun Sep 16, 2007 2:59 pm
Model of CNC Machine: Custom DIY
Location: Lake St Clair, MI, USA
Contact:

Re: gcode resetting my surface zero to Z-0.200"

Post by ger21 »

BTW MrChips, do NOT run a steel pinion on a steel rack. What you want to have wear is the pinion, it is much cheaper and easier to replace.
I think these days, almost everyone uses steel pinions. But the racks are hardened, so it's still the pinion that wears.
Gerry - http://www.thecncwoodworker.com

User avatar
Leo
Vectric Wizard
Posts: 4083
Joined: Sat Jul 14, 2007 3:02 am
Model of CNC Machine: 1300 x 1300 x 254 Chinese Made
Location: East Freetown, Ma.
Contact:

Re: gcode resetting my surface zero to Z-0.200"

Post by Leo »

ger21 wrote:
BTW MrChips, do NOT run a steel pinion on a steel rack. What you want to have wear is the pinion, it is much cheaper and easier to replace.
I think these days, almost everyone uses steel pinions. But the racks are hardened, so it's still the pinion that wears.
My machine has steel pinions on "X" & "Y" and a nice Italian made heavy duty ball screw for the "Z" axis
Imagine the Possibilities of a Creative mind, combined with the functionality of CNC

User avatar
TReischl
Vectric Wizard
Posts: 4597
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 48X36X7 RP 2022 UCCNC Screenset
Location: Leland NC

Re: gcode resetting my surface zero to Z-0.200"

Post by TReischl »

I see you guys do not like my thoughts on not running a steel pinion on a steel rack.

If the rack is hardened, like Gerry states, then I am in agreement. HOWEVER. . . most of the gear racks that a hobbyist would purchase, like from CNC Router Parts, McMaster Carr, Boston Gear are not hardened.

The pinion can be replaced for about $10 and is easy to do. So I still vote for putting an aluminum pinion with a steel rack. I ran the heck out of my machine pretty much everyday for the last ten years and I have yet to replace any of the aluminum pinions. But hey, YMMV!

From McMaster:

"Steel and stainless steel gear teeth are not hardened so you can harden them to fit your application. Hardening a gear’s teeth increases its wear resistance"

I don't think MrChips is going to be sending his gear racks out to be hardened.

I get it on an industrial machine, but on a hobbyist machine? Nah.
"If you see a good fight, get in it." Dr. Vernon Johns

User avatar
TReischl
Vectric Wizard
Posts: 4597
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 48X36X7 RP 2022 UCCNC Screenset
Location: Leland NC

Re: gcode resetting my surface zero to Z-0.200"

Post by TReischl »

ger21 wrote:Our big router at work has a 15ft long ballscrew. It's stretched about .03", and has a spring loaded support in the center to keep it from sagging. It has a rotating nut, and the support pivots out of the way when the nut passes by.
I hear ya Gerry. Is that a moving bed machine rather than a gantry? For a long time "common knowledge" was that gantry style machines were not as accurate as moving bed machines. The question in my mind always was "ok, so they are not as accurate, but how much not as accurate?" Never did hear an answer to that one. I would think that they are not as "accurate" but I never had to run the calcs to figure that out so I guess I will never know.

When we built the rotating nut machines we also stretched the ball screw and then had to do all the compensation stuff in the control. It also required a big honking lower bed to be able to withstand the force of stretching the ball screw.
"If you see a good fight, get in it." Dr. Vernon Johns

User avatar
wmgeorge
Vectric Craftsman
Posts: 288
Joined: Sun Nov 23, 2008 6:12 pm
Model of CNC Machine: Workbee 1000x750 Mach 4 PMDX Controller

Re: gcode resetting my surface zero to Z-0.200"

Post by wmgeorge »

The "real" world in high speed commercial router machines is a lot different than it was say 20 years ago or back in the day. R&P and rotating ball screw is now a fact of life. Wear happens when you making money with your machine. The Income offsets the minor expense of replacing wear items. Most folks lube the R&P system at recommended intervals. Gerry (Ger21) has been doing this forever, and not as a hobbyist I would trust his opinion.

ger21
Vectric Wizard
Posts: 1592
Joined: Sun Sep 16, 2007 2:59 pm
Model of CNC Machine: Custom DIY
Location: Lake St Clair, MI, USA
Contact:

Re: gcode resetting my surface zero to Z-0.200"

Post by ger21 »

I don't have any experience with CNCRP rack and pinions, but I see people wearing out pinions and replacing them all the time. So maybe their steel pinions are softer steel.

I have a box of helical rack and pinion in my garage, waiting for some free time to put a machine together. I think it's been there 2 years now, and I still haven't even opened the box. :(

Ted, it's a moving, cantilevered gantry, open at the front.

Bill, the machine we bought in 1997 had a rotating ballnut too. The main difference between 20 years ago and now is that they go faster now. And helical rack and pinion is more common then it used to be on $100K+ routers.
Gerry - http://www.thecncwoodworker.com

ger21
Vectric Wizard
Posts: 1592
Joined: Sun Sep 16, 2007 2:59 pm
Model of CNC Machine: Custom DIY
Location: Lake St Clair, MI, USA
Contact:

Re: gcode resetting my surface zero to Z-0.200"

Post by ger21 »

Btw, Happy Easter. :P
Gerry - http://www.thecncwoodworker.com

User avatar
TReischl
Vectric Wizard
Posts: 4597
Joined: Thu Jan 18, 2007 6:04 pm
Model of CNC Machine: 8020 48X36X7 RP 2022 UCCNC Screenset
Location: Leland NC

Re: gcode resetting my surface zero to Z-0.200"

Post by TReischl »

wmgeorge wrote:The "real" world in high speed commercial router machines is a lot different than it was say 20 years ago or back in the day. R&P and rotating ball screw is now a fact of life. Wear happens when you making money with your machine. The Income offsets the minor expense of replacing wear items. Most folks lube the R&P system at recommended intervals. Gerry (Ger21) has been doing this forever, and not as a hobbyist I would trust his opinion.
I have straddled both worlds, the hobbyist and the commercial over the last 45 years. The problem with most hobby machines (mine included) lubing the rack and pinion is that they are not covered and most lubes will just cake up with dust in short order. In my particular case the racks are outside the machine cutting area behind the bed rails and are upside down (teeth pointing down) so they do not get all full of debris. I have never lubed them.

Dust covers for things like ways, balls screws and r&p add quite a bit of cost to a machine. Not just the cost of the covers but the fact that they eat up travel distance like crazy. So if you want travel distance of 8' you cannot just add the space the bearing trucks devour to the rails and call it a day. Nope, you have to add a whole bunch of feet to the ways, which also means the base has to be several feet longer. It all adds up pretty quickly. Most of us hobbyist guys are not willing to absorb that cost.

BTW, I think you meant "rotating ball nut" up above. I first decided to use it when I saw an article in Machine Design way back in the late 70's discussing how it sped up the machine, reduced vibrations, reduced the required size of the servo motor and a whole bunch of other good things. So it isn't like I thought the idea up. The key at the time was to use a GoodYear HTD belt which IIRC was fiberglass reinforced and did not stretch. It may have been reinforced with steel wire, not sure anymore.

It has been interesting to watch and participate in all the changes over the years. Who woulda thunk it would make for a great hobby in retirement? One of my old boss's who knew everything there was to know (just ask him) told me when we were designing the laser cutters that they would never cut steel with them so there was a very limited market. What foresight that man had! Ten years later he visited our booth at IMTS which was chock full of large laser cutters cutting, OMG!! STEEL! I was nice and did not remind him of his prediction.
"If you see a good fight, get in it." Dr. Vernon Johns

wb9tpg
Vectric Wizard
Posts: 456
Joined: Tue Sep 04, 2018 2:49 pm
Model of CNC Machine: Shapeoko 3 XL

Re: gcode resetting my surface zero to Z-0.200"

Post by wb9tpg »

I've been following this thread and it feels like I've been reading an 800 page murder mystery and never found out who the murderer was ......

I'm hoping you all figure it out
Gary Mitchell
Kentucky, USA

Mr.Chips
Vectric Craftsman
Posts: 235
Joined: Wed Sep 17, 2008 2:28 pm
Model of CNC Machine: MAKE DO
Location: GILBERT AZ

Re: gcode resetting my surface zero to Z-0.200"

Post by Mr.Chips »

I’M 90% SURE I FOUND THE PROBLEM, FOUND IT LATE LAST NIGHT. GOING TO THE GYM NOW.
WHEN I GET BACK ILL DOCUMENT WHAT I FOUND AND SEE IF YOU ALSO AGREE THEN WE CAN MAKE IT A 100% PROBLEM FOUND.

HAGER

ger21
Vectric Wizard
Posts: 1592
Joined: Sun Sep 16, 2007 2:59 pm
Model of CNC Machine: Custom DIY
Location: Lake St Clair, MI, USA
Contact:

Re: gcode resetting my surface zero to Z-0.200"

Post by ger21 »

Mechanical issue? :twisted:
Gerry - http://www.thecncwoodworker.com

Post Reply