Z AXIX DRIFT

[mark holm]

Has anyone experienced anything like this?

At the beginning of a Shopbot session I'll zero the z axis with the zero plate & program. The S.B. runs fine initially with the depth of cut being dead accurate. After finishing one piece, it may take an hour or so before I can load & run the next piece. Upon running the next piece, it becomes apparent that the depth of cut has increased.

Is it possible that gravity has pulled the z axis lower during the 1 hour break between pieces? I did not power the unit down.

[bjwat@comcast.net]

Mark,
A few possibilities:

1. Your spoilboard has warped due to humidity and settling. Mine was dead on a year ago....just resurfaced because of the same issue.

2. Z-plunge speed is too high and Z stepper is losing steps when it plunges into the material. I like to plunge into the material at .2 to .5 to go easy on my bits.

3. PC to control box timing is off. Sometimes my machine gets wacky and loses steps because it doesn't have a good sync with the PC. I use the UP command to manually get a number that I know is in the ballpark. For my machine, 59 is about where the timing number is. Your number will vary. When you first run the DOS software you will see the [Timing ck...] remark as it initializes. Look for the number.

Most likely...it's one of the 1st 2 issues.

-Brady

[beacon14]

Brady,

Mark said his depth of cut has increased - to me that says the bit is now cutting deeper. If he was losing steps due to plunging too fast (like I used to!) his cutting depth would decrease.

Not sure about the spoilboard moving in an hour enough to notice a difference (but it's possible).

What about the bit slipping out of the collet. Are you sure the previous part finished up at the same depth that it started? Or is the difference strictly happening during rest periods? Not sure what else to suggest.

[jkforney]

What version of shopbot software are you using. Do you have a two z axis machine with one or more being a Columbo. There was an earlier version of the dos software that turned the power off to the motor during a z axis switch and the early springs weren't enough to hold the spindles staionary.

If none of the above is true, maybe a little more info about your machine setup.


John Forney

[gerald_d]

This topic is discussed here very often and it does seem to be a recurring problem.

We also experienced this problem until we put a lot of effort into grounding EVERYTHING electrically. Our router is plastic cased with a metal nose-piece carrying the main bearing - we found the main problem lay with the grounding of that nosepiece.

[mark holm]

Gentlemen,

Thanks for your comments. I offer in response the following:

-It’s doubtful that my spoilboard had changed by 1/16" (the offset between correct & too deep cut) over an hour. it was very recently re-surfaced
-I'll try backing off on my plunge speed, I've been plunging .25" per pass into phenolic resin @ 2ips w/a .25" bit
-I'm using the shopbot software that came with the unit in Nov. of '02. Is there an upgrade since then?
_I'm fairly certain that I've got a good ground. I have a direct wire to earth from the PC and the S.B. table, as suggested by the installation manual.

I may simply have to re-zero the Z axis between parts.

[gerald_d]

Mark, you use the term "earth" as opposed to "ground" which makes me think that you use the Queen's English. Would it be better if I drop my pseudo-American and speak in a language that we both understand better?


One earth wire, from the table only, assumes that all the grease-filled V-rollers are the electrical path. We made sure that we actually have a copper wire across all of the V-rollers. Yet that was not enough - the nose of our router was somehow generating a stray signal that would move the z-motor.

Realise that the wire to the Z-motor is the longest one - which makes it the longest aerial (antenna). This wire is strapped to the router's wire, and interference can happen between the two. Separating the z-motor wire from the rest by only 30mm (1.25") can also make a world of a difference.

[gerald_d]

Here are some related discussions on this forum:
http://www.talkshopbot.com/forum/messages/312/1790.html
http://www.talkshopbot.com/forum/messages/26/1654.html
http://www.talkshopbot.com/forum/messages/26/1258.html

[bjwat@comcast.net]

Mark,
I misread your post...

I would look at the spring on the Z axis. The constant-force springs (look like rolled up band of steel) sometimes get lazy...and can cause the Z to drop if it isn't retracting and seating properly. I have this on my Colombo and have purchased a coil spring upgrade for the Colombo from ShopBot. I am not sure if they offer one for the PC.

-Brady

[cncdr@comcast.net]

I have owned my shopbot for about 10 months. I recently noticed when I turn the power off the z axis drifts down. Is this ok or should the spring be adjusted, if so how do I go about it.
thanks. Don