Z Axis Creep

[robin]

When cutting complex objects 3000+ lines the z axis will often lose its setting. For example, at the end of a cut, the z axis zero point will either be a little high, or a little low. It usually isn't more than a 1/16th or so but leaves the background uneven. I haven't seen the problem on smaller files.

I have duplicated the problem on two shopbots (older cable models) both with the ball screw updates. Am I the only one with this problem? Any one have any suggestions?

[bobgrim]

I also have this problem but mine is aways lower.
I just did a 40,000 line drawing and I had to stop the Sb 4 times to adjust the z axis. It drop as much as a 1\2 inch. I have the new ball and screw update.

Does anyone have any suggestion?bob

[phil]

I have also experienced this problem. Mine also drops around 1/2 inch or so.
I haven't determined the how and when of the cause of the problem. But my suspicion is that if you have a high speed setting for the z-axis and your ramps set too high that at some point the z axis tends to jam up because of debris piling up on the z axis screw, maybe? Or the z motor loses some info and

[engiplast@n-link.com]

I was having problems with the Z axis and if you listened closely you could hear a popping noise. It finally got so bad I had to take it all apart. I was tipped off to the problem by marking the screw, bearing race, and coupling. When the marks didn't line up after running for a while I knew where the trouble was. I couldn't believe what I found. The ballscrew did not have the end turned down to fit the bearing or the coupling. It was also a little small in diameter, so there was shim stock wrapped half way around the screw to take up the slack. This caused it to be off center. The ball screw is so hard the set screws for the bearing and the coupling won't bite into it and stay in position. The popping I was hearing was the set screws jumping over the threads on the ball screw. Depending on the position of the nut it would pop one way and them the other. The misalignment caused by the wobble in the bearing would cause it to jump one way when the axis was up high. A heavy load would cause it to jump the other way. This movement was in the bearing and in the coupling. I turned the end of the screw down to 1/2" and put in a new bearing and coupling. No problem since. It only took about 15 minutes to turn the end of the screw. I couldn't believe how it had been done. I am sure they must be spending that much time cutting shim stock and messing with it, so I don't see why it would cost more to do it right.

[bruce_clark]

Don,

Did you use a cast iron pillow block or another
stamped steel pillow block? I have thought about
do this also, but mine has not been too bad.

Did your Z axis accuracy improve?

Bruce Clark
bwclark@centuryinter.net (mailto:bwclark@centuryinter.net)

[joa]

Somehow this doesn't surprise me. I found so many things wrong when putting my SB together that I just ended up fixing them myself because they would have had to ship me a whole new machine (which I don't have time to wait for… if you want it done right…)

Every impression I get is that they are so focused on throughput (at $4k a piece mind you!) that they ignore quality.

Yaa, I'm bitter. The only reason I'm not irate is that the machine works fine once you get the parts drilled, cut, and assembled correctly. In defense of SB they do have a good product but its frustrating having to re-work about every piece to get it that way!

JH

(FWIW I have emailed SB two separate times regarding my difficulties but so far have not received any replies. Hmmm, wonder where they are- probably at trade shows selling more SBs!)

[bruce_clark]

Joa,

Sorry to hear about your troubles. What exactly
was wrong with your machine?

Bruce Clark
bwclark@centuryinter.net (mailto:bwclark@centuryinter.net)

[joa]

You name it... multiple holes drilled in the wrong places (for instance the holes in the Y-axis were off by 90 deg.), the insides of the control box hanging by their wires (the few nuts supposed to be holding them in vibrated loose apparently), motor cables with shorts in them, missing items, the list goes on.

Fortunately I was able to fix the problems at home faster than new parts could have been shipped.

Neverless I think the SB is a good unit and am having a ball cutting things.

[engiplast@n-link.com]

Bruce,
I used another stamped steel bearing because I had a job to do and couldn't find a cast one with the same center height right away. I would like to put in a better one, but I think it will take some modification of the mount. My accuracy is much better now and much smoother. I also threw away the aluminum channel for holding the router and the one on the back where the wheels are mounted. I put a solid alum. block on the back and attached the wheels by drilling and tapping. I machined a heavier channel for the router mount out of an alum. block. Made it a little longer so I could move the hose clamps to each end of the motor. Also machined "V" grooves in the back of it to center on the tubes. This keep it from shifting around. Did you see a lot of difference with the higher voltage supply? Mine is external, but 12 volts. I am now in the process of trying to square up the Z axis and probably am going to tack weld some of it when I get it right. I am going to leave it so I can shim it, for adjustment, where it mounts to the Y axis.
What rubber bumper, for holding pinion tension, are you talking about? Mine doesn't have any. Mine is a factory R &P unit.

Joa,
I sent a e-mail asking about the noise in the Z axis, among other things, some time ago and never received a reply either. I had some of the same problems as you assembling mine. Two motor cable 6 feet too short, etc. I have had alot of computer lockup problems and have had reasonable response from them about this. I am not sure it is fixed yet, however. I told them they needed to keep better records of what they are doing and who has what on their machine. They keep making changes on the fly, with no documentation, such as the gear boxes instead of the torque box, I think the new ones are coming with higher voltage supplies, etc. They told me I should change the current limiting resistors because of the higher voltage, but I only got a 12 volt supply. I wish we were advised of changes and had the opportunity to make them on our units. I got mine at the end of April, so I don't feel like it should be "last years model", especially since the price is the same.

[bruce_clark]

Don,

I think that when I "fix" my Z axis, I will see
what it takes to make a Dodge CI pillow block
work.

As for welding the Z axis, I personally like the
idea that I _COULD_ tilt it _IF_ I have had a use
that required it (like tilting the head of a
Bridgeport mill--you _could_ do it, but I have
never personally seen someone do it as it is
usually easier to tilt the peice) but it is still
nice to _know_ you could.

Did the 12V ps clear up your lock up problems?
Now, on my original cable drive machine, I went
from an external 24V ps to a 13.8V ps (I had to
use the 24V for another project) and did not see
much top speed difference on the X or Y but I did
see a slight difference in the Z axis speed. I
have not tried any other PS with the new upgrade.

Well, they were _SUPPOSED_ to ship it mounted on
the gearbox (I only had 1 of 3) but it just so
happened that my local hardware store had the
EXACT same ones (package of 12 for $2). Anyways,
email me your address and I will send you some.
The "L" bracket is suppose to push up against the
motor/gearbox via the rubber "bumper" to help keep
the pinion engaged in the rack. I wish I could
tell you which page it is on in the manual, but I
don't have the manual in front of me.

Bruce Clark
bwclark@centuryinter.net (mailto:bwclark@centuryinter.net)

[dale]

Bruce:
I found a use for tilting the head. Last year when I drove down to pick up your Shopbot (sorry mine), I mentioned to Gordon that one of the things I wanted to make was dollhouses. He suggested that if I tilted the head that I could give the impression of having siding on the house. I must say that it looks great. Send me your address again and I will send you some pics. When you gave me your address last year it was on the notebook and the drive has been formatted since then. There are way too many to e-mail.
Dale

[Ted Hall, ShopBot]

Comments on Several Concerns Expressed in this Thread:
------------------------------------------------------

1. Quality Issues --

We apologize here to Joa and a few others for difficulties experienced with assembly of their ShopBots owing to quality problems with parts, and with our failure to get him more timely support (we are also trying to make sure these problems are individually resolved).

While not offering it as an excuse, in the last few months we have been bringing a few new people on board at ShopBot and have learned the hard way the difficulty of transferring production and support tasks from the initial core of a couple of people who were intimately familiar with every little nuance of a ShopBot (since they developed it) to a more routine manufacturing process. We have made some mistakes in learning what the important details are for communicating and teaching production procedures and in recognizing at what points quality assessment vigilance is critical.

We believe that a moderate and sustainable rate of growth is important for us as a company, and just as important to insure that those who have invested in a ShopBot tool will see that tool become increasingly productive and valuable as new software and hardware features provide enhanced capabilities, ease of use, and reliability. The problems noted above are not a fault of any of the new crew here at ShopBot, in fact we believe we've hired several great new people. Rather, our glitches arise from the fact that we've had to learn how to train, how to effectively communicate methods and technical skills, and have needed to develop standard procedures and an organizational structure for production. We feel that after a difficult few months, we are on track.

Our primary goals here are to keep ShopBots simple and affordable, and to make them right. We will continue to work hard it.


2. That Z-axis Ball Screw ...

This is an instructive thread for us because we may not have fully appreciated the problems that people were experiencing.

First let me explain that what we have been trying to accomplish with the Z-axis, as with all ShopBot parts, is to keep the tool as much as possible something assembled with off-the-shelf and readily available parts. Thus, each component of the Z-axis is readily available (standard ball-screw and nut, standard BG pillow block, etc...). And, as noted in the comments here, there are no machined parts other than basic cutting and drilling.

The diameter of the screw is ~.016 less than the inside diameter of the pillow-block bearing. Our intention is to have placed the .008 shim stock exactly opposite (what we will call) the primary set screw. This screw is then set hard and fully into the -valley- of the thread and embedded in locktite, pressing the screw into the cradle of the shim. The secondary set screw is tightened moderately into position and intended only to further resist rotation. When we do this correctly, the screw should be perfectly centered in the pillow block and the primary set screw so well seated that no movement is possible.

As noted in this thread, this system may not provide as much redundancy, resistance to wear and tear, and assurance of positioning as is desirable. We will be evaluating several alternatives to this system and you will hear more about what we find out in the next few months. In the meantime, we will post some tuning and alignment details as a Technical Bulletin.


3. ShopBots are Changing ...

As I hope we have indicated throughout our documentation and web site, we work to continually improve the hardware and software components of ShopBots. We are committed to making these improvements available to all ShopBotters, and to making improvements that can be incorporated as upgrades on all tools. We provide these upgrades at our costs, and in many cases are able to provide directions on how users can make improvements themselves if they want. Our new and enhanced Version 2 software will be distributed free to all ShopBot owners ... and early trial releases of the software are available for downloading here. This is to say, we expect improvements and we intend to do them in a way that these improvements are available to everyone. We explicitly want to avoid any sort of built-in obsolescence to the tool. And that is part of the idea behind the erector-set construction of the tool. We plan to keep making enhancements to the tool that can be adapted to all tools, and we expect ShopBotters to make improvements and modifications that others will also want to share ... hopefully it's a system that is straightforward enough to be easily understood, upgraded, modified, adapted, adjusted, beat around, and just used.

A comment in the above thread notes that early rack & pinion customers had 'thrust boxes' that were built by ShopBot, while newer customers had out-sourced gearboxes on their motors. This is the case. We have not offered these gear-head motors as a formal upgrade, because we do not believe they offer any enhanced performance over the ShopBot-built thrust boxes. We switched to the gear-boxes as a manufacturing convenience because we felt too many of our limited resources were going to building the thrust boxes in house (and we were finally able to get gearboxes at a reasonable price). In an earlier posting on the forum, we described the two motor/gearing systems in detail and discussed the relative merits (and subtle trade-offs) of the two motor/gearing systems and elaborated why we do not emphasize the differences (see the 'gear' discussion under the Variations & Modifications section).

(One difference between the two systems is that the gearbox motors do not develop full torque at high speed without a 24+v power supply, thus we have been supplying the higher voltage supply with the gearbox motors.)


I hope I've have addressed some of the issues of this thread. We know talk is cheap. So, beyond providing information here ... we hope to be judged by what we are able to accomplish in the coming months and years, by how we are able to be responsive, and most importantly by the continuing enhancements in what your ShopBot is able to do for you ...

[engiplas@n-link.com]

Ted,
FIW there wasn't a Hint of ever being near a bottle of Locktite on my Z axis!
Don

[robin]

Ted,

Any update on the z axis problem?

Robin

[Ted Hall, ShopBot]

Hi Robin,

Thanks for the question.

I'm afraid I got distracted with the other Z-axis issues above. These are mostly related to mechanical issues (and assembly quality issues).

Your problem is different and it has been reported to us by a couple other users. The symptom is as you describe: a small error in the Z location that occurs during the cutting of a large file. In the case of this problem, all mechanical issue have been ruled out, by for example, marking the various parts in the system and being certain that nothing is slipping and no steps are lost.

We believed that the particular problem only occurs with cable tools, but you may have found something different. An error seems to be occurring during the Safe-Z pull up that occurs when the software pauses to load a new movement block while it is in the middle of a series of moves. When the tool pauses to load more of a file, usually between 2000 and 3000 moves, it pulls up and then repositions itself in the material after the move. We think the error happens at the pull up and the problem seems worse if the pull-up occurs in a circle.

The work-around for the moment for a large file is to use the 'SC, 2' Command (the SC 2, sets continuous moves off and then turns them back on ... the equivalent of doing an SC, 0 and then an SC, 1) in the file to force the beginning of a new movement block, before it is handled automatically -- putting an SC, 2 between letters in a sign, for example, puts the file reading at a logical place. [I realize this is a bit arcane and were working to make use of the SC Command totally unnecessary.]


We hope to get this problem solved shortly. Try the work-around for the moment. We will also shortly have an expanded memory version that will load a file (and movement block) up to 8M.

[Fred A. Coyle]

Reguarding Z moves by Fred Coyle
I have several large files and experienced Z plunges after batch loading. I now fix my files by going to line 2,000 or multiple thereof adding M3, , ,0 enter
SC,2 enter
M3, , ,0 enter
I really had the problem when altering X,Y,Z values but this fixed it and eliminated router "pause" marks in my pieces.

[john.forney@acc-net.com]

Fred
I just add sc,2 and get the same results. On all my carved signs I would get jogs when the program would stop and load new instructions. This work around, as per Ted Hall, has cured the problem.

John Forney

[joa]

OK, now I'm having problems with my ball screw Z axis. It's been getting progressively worse and is stalling when going both up and down (even at very slow speeds) and the screw has a very visible wobble to it at the top.

I took it apart and noticed that the bearing was loose on the shaft (no Locktite in the setscrew?) and the shims seemed to not really help. I'll weld build the end of the shaft and turn it to the proper diameter and hope this helps.

Anybody have suggestions for putting the balls back in the lower section? It ended up coming off the end and I caught most of the balls though lost a few. How many are there supposed to be anyway?

Anybody have any other fixes that help the Z-axis?

Thanks!

Joa
PIW E&M

[chandler@n-link.com]

Joa,
Just turn the end down to 1/2". You can get the exact bearing with a 1/2" bore from Mcmaster Carr. Order a new half of a Lovejoy coupling with a 1/2" bore while your at it. it will match up with the one on the motor. You can also order a new nut there also. It comes with a cardboard sleeve inside, so all you have to do is push the tube against the end of your screw and screw the nut on. If you have any questions send me a email.
Don

[rfpaul@hotmail.com]

Has anyone been using the Typesetter program in the new sb2b14 program, I'd can't seen to figure out where to insert the depths, when I zero out the tool it will cut with out lifting up between letters. Rick

[J.E.T]

I was not aware there was a sb2b14. The download page only shows a sb2b10. I will check again but, If you do have a 14 version, where did you get it and if your version is 14 why isn"t it availabe to everyone? I thought Shopbot was going to supply all owners with a bug free copy of the software when it came off the press. Can anyone supply an update on this issue?

[J.E.T]

I did check, and there it is sb2b14 to download. Will download today and check it out ---Thanks Shopbot---

[Ted Hall, ShopBot]

Hi Rick,

The plunge depths in 'Typesetter' are computed for you on the basis of the bit angle that you indicate and the height of the lettering you select. These depths are all 3D moves and depend on the letter.

The clearance height between letters is the height you indicate when the program asks for Z up height. Something like .2 is usually safe.

[donchapman]

I'm having z-axis problems. I just routed a 6'x2'sign using my usual method of hogging out (pocketing) the background and leaving the lettering standing about 3/8" proud from the background. (you can see what I'm doing under "model signs" under "ShopBots is use" at the ShopBot web site. The problem is that the first half of the routed background is as smooth as a baby's bottom, but then the z-axis went nuts and left lines of different depths throughout the background, ruining about $150 worth of SignFoam high density urethane. The depth difference is only about 1/32" to 1/16" but next to impossible to correct by sanding amidst all my lettering. I didn't notice how bad it was until after I cut the sign free from the blank, so it's too late to fix this piece since I've lost my precise xy-zero starting point. I looked at the z-axis and found vertical play which I reduced by tightening very loose set screws in a collar at the top of the z screw. I use a dust collector so I don't have any debris build up on my z screw. I hope that tightening the set screws fixes the problem. I'll try routing the sign again tomorrow. In the meantime, I'd appreciate any feedback from anyone who has experienced a similar problem.

[davidallen]

when you tighten the screws, be sure to use some thread locking compound (Loctite). vibrations from the router and steppers can loosen set screws over time.