PROXIMITY SWITCH

[larryk]

WERE SHOULD I PUT THE TARGETS 0,0 / -.25,-.25 ANY HELP WOULD BE GREAT! LARRY

[rgbrown@itexas.net]

Larry,

I don't believe it makes much difference where the proximity switches are. I think they should be 'approached' from the same direction each time to be moat accurate. But, other than that consideration, I doubt it matters much. Near the "0, 0" makes a lot of practical sense.

Ron Brown - rgbrown@itexas.net (mailto:rgbrown@itexas.net)

When looking for a reason why things go wrong,
never rule out sheer stupidity.

[billp]

Larry, Ron is right, there is no specific place you HAVE to put them, but you might want to consider the kind of cuts you'll be making. I gave myself a -3.75 range in the X axis because I sometimes go "off" the table edge to use the machine as a panel saw and I like to be able to use a large bit when I level the table as well.It wasn't so easy to do that on the Y axis, and I think I have about a -.75 figured in there..Bill P.

[Matthew WB Inc]

We have a major problem with the registration of items over multiple, repetetive cuts. I think I know what the problem is, but havn't a clue as to the resolution.

We cut 16 panel sides with 3 rotolock holes each. By the end of the run, we were over 1/8" off.
We cut a 4X8' panel of 1.5" singfoam into 30 letters. We had to make two passes because of the depth of material and were 1/32-1/16" off on one side or the other with the #2 cut.

The ShopBot has a tolerence of .02". The router will slow down, just a bit, while cutting- .01" is not very much- and this is not at all reflected in the coputer because the computer is ESTIMATING the distance traveled and has no way of even knowing the error is building. Our screen says that we are at "X" and we turn around and see that we are at "X+4"
We cut with a very slow travel time- .2, use as sharp bits as we can and run the router at the two highest speeds. We ought to be vaporizing the MDF. Well we are that- we just can't stack the pieces so the mate exactly.
I think that it is very easy to overestimate the machine and it does best cutting pieces that are unique within an assemblage- 1 of 30 pcs. to go into a unit, not a run of 30 Interchangable pcs.
I know that it would be a lot to ask of a program to communicate with the cutting tool, but if it did, we might be able to handle this registration problem better.
Any Ideas???

[larryk]

Thanks, I see what u are taking about that it dose not make much diffence I did not know that u can change the value in the home program. Did not know how it works. Thanks alot ! Larry

[Mayo]

Just some things to think about for Mathew:

If your material moves when cutting, it will mess up the cutting distances. (Obvious, I know)

If you are cutting at only .2 speed with your router going at the fastest or almost the fastest speed it can go, I would think your bit is getting very hot. Heat makes metal expand. Enough to account for an error of 1/16th? maybe not entirely, but perhaps the heat is also helping to create runout or perhaps it's making the bit more flexible once it's hot.

You should be able to cut signfoam at a fairly quick speed - much quicker than MDF. As long as it's held securely to the table you shouldn't have a problem. I have seen posts here by other people who cut MDF at 1.0" per second. I have cut signfoam at 1.0" per second and I'm sure I could cut it faster. I guess it depends what you're trying to cut but .2 just seems painfully slow.

Can you look at the letters that you cut, and the leftover scrap, to determine if it's always cutting too far in a particular direction? Is it always leaving 1/32 extra in the X direction or the Y direction? Or is it both?

Check your ramp speeds. I'm not sure that it would matter if you're cutting at such a slow speed, but normally the ramp speeds should be about 60% of the cutting speed.

[gerald_d]

A bad ground connection to our router spindle is the only thing that has caused drifting zeroes on our PRT.

[beacon14]

Matthew
Just my own experience to share: we noticed that the cuts were getting "sloppy", similar to the inaccuracies that you report. I figured out it was a combination of tool spindle runout, and looseness in the shopbot carriages where the wheels grind away minute amouts of the rails (we have the older, PR machine). We also discovered that the wheels on our Z-axis were no longer snug. The result was that the bit would "pull" itself into the workpiece as it was cutting as much as 1/16". Tightening all the wheels to their tracks and replacing router bearings made all the difference. Even with our older machine, a second pass typically falls exactly over the first pass. Multiple parts cutout of a sheet or sheets line up extremely well.
While it maybe likely to overestimate a machine that is not tuned up properly, I find it more likely to underestimate the abilities when it is running right.
I find that the .02 that Shopbot claims is very conservative - our limit switch repeats within .002 over 100 tries, and the error does not accumulate. We have even engraved plaques using .02 diameter bits, cutting 1/4" tall letters, to match plaques that were laser-cut.
Take a close look at the machine set-up, see if there is any rattle in the carriages, grab the spindle (power off, of course) and see if you can wiggle it at all. The machine may be more capable than you think

[Ted Hall, ShopBot Tools]

Hi Matthew:

There are a couple of threads here. This response is to Matthew. I'm pretty certain we should be able to improve the reliability of cuts in your situation, but we probably need to talk through a few things based on your particular model ShopBot and application. If you can give Tech Support at ShopBot a call, we'd be glad to try and help.

In general, while various factors can add up to problematic errors in location, these problems should -not- accumulate over cutting. The only time you would get a cummulative error that the computer is not aware of would be when steps are 'missed' and thus lost. Otherwise, the computer will be just as good at knowing where the tool is in the early parts of a cut as at the end. Steps should not be lost if the tool is working correctly. I expect we will need to consider other possibilities to explain your registration problem.

[davidbeede]

I got proximity switches at the Jamboree - when I went to install them I see the hardware is for the rail arrangement on the newer PRT machines. Any suggestions for mounting on a PR machine? I figure I need to rig a bracket and drill for it, but if anyone has "been there done that" and has tips I'm open.

I'm guessing I need to connect them to P1.6 on our PR control box. Since the Z zero plate went to P1.4.
Thanks,
David Beede