Prob Problem

[Stroutwoodworking@sbcglobal.net]

I have a problem with the Shopbot probe and I am wondering if anyone has any idea's.
I have customers come in my shop and want custom elipses cut out, so i use the 2d edge find routine to edge find there template and it never completes, what I mean is the probe never ends in the same place it starts and when i cut out the elipse the top of it bubbles out about a 1/4 of an inch so it doesn't match the template. I have tried this without using the shopbot probe and wrapped the edge of there template in tinfoil I used a straight piece of metal and I get the same result so it cannot be the probe. I am thinking it is the software. Any help would really be appreciated

[Brady Watson]

Ken,
Using a metal rod and ground to probe is dicey at best. Yes, it will work...No it is not 100% reliable. The software to run the probe is very basic and straight-forward. If I had to guess, I would say that it is an issue with the tape and the ground.

If you are going to be doing a lot of 2D scanning, invest in a real probe ~ especially if you use your bot for business. You'll thank yourself, save countless hours of frustration and wasted material and have a higher degree of repeatability and accuracy. The solid rod technique is a neat trick for hobbiests, but if you are running a business, you don't have time for unreliable methods.

The other option at the moment until you get a probe is to use the polyline probing routine. This will allow you to pull it into CAD or PartWizard, preview it and repair any discrepancies...and also check the dimensions of the scan BEFORE you cut anything.

-Brady

[bert (Unregistered Guest)]

isn't shopbot probe a real probe?

[garyb]

Yes it is Bert
I think Brady missed reading the first lines of your post.
Is't the probe a little overkill for a standard shape of an elipse?
As you stated, this was an elipse and you have the template why not measure it, draw in cadd or what ever you use and create a cut file, 5 mins and your done and cutting.

[gerald_d]

Brady, you are going on about hobbiests versus running a business.... Is that really necessary? There is no technical reason why the rod and tape/foil method would be less accurate - as Kenneth's post illustrates.

Anyway, to get this thread back on track, Kenneth has a problem in that his probe routine does not give him the true shape, irrespective of the type of probe that he is using.......

[mikejohn]

This doesn't help with the probe problem, but with Kenneths' particular problem, if you have even 50% of the elipse accurate, you can reproduce the entire elipse precisely in a CAD program.
(or am I again missing soemthing)
.....Mike

[mikejohn]

If you had a solid probe, and you were probing a metal original, would the result be any different than using a probe with a 'switch' tip?
If the answer is they are both equally accurate, then the only lack of precision will be in applying the conductive material (foil) to the un-conductive(wood) original.
.........Mike

[gerald_d]

Mike, we sometimes get old picture/mirror frames to reproduce, which are supposedly ellipses. But they are not pure mathematical ellipses and the CAD ellipse functions are useless for them. The customer wants their old glass to fit again - he or she (more often) don't want a maths lesson.

Probing a metal surface with a conductive tip works perfectly. Apparently the only fault with it is that it is too cheap.

Anyway, we are drifting the thread again....

[mikejohn]

Gerald, I do have a habit of looking at problems, assuming what is written is accurate.
An 'elipse' that isn't aa elipse (can't be shown as a mathematical formula) is an oval. Often the two are shown as being the same thing.
They are not.
....Mike

[gerald_d]

Ovals are also symmetrical about a centreline - some of the things carried in here don't have an obvious centreline. Maybe "rustic" is the word...

Here (http://www.talkshopbot.com/forum/messages/27/6100.html?1106119632) is a concept proposal for a dedicated edge finding probe.

[Brady Watson]

Actually...he must have edited his post. The 1st line of Kenneth's post did not specify the ShopBot probe, I believe that was edited and added after I posted. (that or I am cracking up...)

So disregard my post.

-Brady

[gerald_d]

You posted 4 hours after him, we can only edit within the first 30 minutes. The rod/foil is only mentioned near the end of his post. That confirms it.....you are cracking up.

[bert (Unregistered Guest)]

Elementary my dear Watson

[gerald_d]

Poor old Kenneth who started this thread must be wondering if anybody is really going to help him. This is what he said:

"I have a problem with the Shopbot probe and I am wondering if anyone has any idea's.
I have customers come in my shop and want custom elipses cut out, so i use the 2d edge find routine to edge find there template and it never completes, what I mean is the probe never ends in the same place it starts and when i cut out the elipse the top of it bubbles out about a 1/4 of an inch so it doesn't match the template. I have tried this without using the shopbot probe and wrapped the edge of there template in tinfoil I used a straight piece of metal and I get the same result so it cannot be the probe. I am thinking it is the software. Any help would really be appreciated."

Well, can the probe routine experts help the man out?

[bill.young]

Kenneth,

I think the metal contact method is actually the best for edge finding because there's no movement in the probe tip before the contact triggers...it records the point right when it makes contact whereas any method with a moving tip will move beyond the contact point a little before it triggers. You can also easily change the size of the rod to match the bit that you'll be cutting with and won't have to worry about offsets.

That doesn't help with your problem, though. Here are a few things to check off the top of my head.

* Could the pattern be moving a little?

* Is the tool losing position? If you send it back to 0,0 does it go where it's supposed to?

* Is the pattern connected to the same input switch as a z-zero plate? If so could the plate be vibrating when the tool moves and causing the plate to trigger?

* What version of the ShopBot softwware are you using and what routines...the DOS routines or the Copy Machine?

I'm on the road this week and have intermittent internet access, but will try to keep up with your answers.

Bill

[mikejohn]

Bill
What about this (http://www.semsupplies.com/Copper%20Tape.html) product to fix to your original?
Might also work if you were trying to 'slice' a 3D original?
Its very thin, I wonder if you could poke it into narrow spaces?
.....Mike

[Brady Watson]

Well there you have it...I am cracking up! Sorry!

(been running the bot 18+ hrs a day for the past 3 weeks...)

-Brady

[bill.young]

Hey Mike,

That looks like it would work fine. The tricky part in using metal contact is connecting the wire that goes to the input switch connection in the control box to the pattern. You also need to be sure that the foil (the electrical connection really ) is continuous all the way around the pattern, with no breaks.

Bill

[gerald_d]

Mike, if you want decent tapes, you can look across the pond to the old Minnesota Mining & Manufacturing Co., today known simply as 3M. Here (http://products3.3m.com/catalog/us/en001/manufacturing_industry/specialty_tapes/node_GSST0Z1YWLbe/root_GST1T4S9TCgv/vroot_GSNYTMLW46ge/gvel_60R9LNG52Lgl/theme_us_specialtytapes_3_0/command_AbcPageHandler/output_html) is some info on metal tapes. This (http://products3.3m.com/catalog/us/en001/manufacturing_industry/specialty_tapes/node_GSWYTJX0NFbe/root_GST1T4S9TCgv/vroot_GSNYTMLW46ge/gvel_60R9LNG52Lgl/theme_us_specialtytapes_3_0/command_AbcPageHandler/output_html) lead tape is meant for conforming to uneven surfaces.

[stroutwoodworking@sbcglobal.net]

Bill

Could the pattern be moving a little?

As far as i can tell it hasn't moved at all. I do have it screwed to the table so i would have to say it is not.

* Is the tool losing position? If you send it back to 0,0 does it go where it's supposed to?

Not sure on that one but I will check

* Is the pattern connected to the same input switch as a z-zero plate? If so could the plate be vibrating when the tool moves and causing the plate to trigger?

Yes it is hooked to the same input switch. I also thought it was the zero plate trigering and completly removed the zero plate from the wago connector to rule it out and it still does it.

* What version of the ShopBot softwware are you using and what routines...the DOS routines or the Copy Machine

I am using 3.3.5, I have also tried 3.3.4, with the same result, I am using copy machine.

Everyone else it is not a true ellipse if it were i would do it based on a mathematical formula but i cannot.

[bill.young]

Hey Kenneth,

I would try updating to the newest software...3.3.18 is the latest. Can't guarantee that it will help but there were some issues with input switch triggering in older versions causing lost postion, and I can't remember if 3.3.5 was before or after the fix.

If that doesn't help how about sending me an email and we can dig a little deeper.

Bill

[stroutwoodworking@sbcglobal.net]

Thanks Bill I will give it a try.

[Dave Rawn (Unregistered Guest)]

I am interested in the Shopbot but would like to talk to some in person about it and watch it work
If anyone in southern minneosta has one and would be willing to talk to me and give a demo please let me know

[gerald_d]

Oh well, Dave seems to be in the wrong thread, but when reading the whole thread again, it did seem to drift a tad.

[mikejohn]

I think there needs to be more discipline in this Forum regarding thread drift.

[fleinbach]

I'm sure you could make a very nice hickory paddle with the Shopbot.

[gerald_d]

But Frank, would it be square?

*Runs away, dodging assorted objects being thrown by Frank*

[gerald_d]

Here we use paddles for stirring paint. I was disciplined with a cane:

3837

[fleinbach]

My teachers prefered a paddle with holes drilled in it. Not square holes, but round ones. Didn't you know Gerald nothing I make is square!

[gerald_d]

Why have holes in a paddle? Were they for spikes in earlier times?

My first caning in high school was for de-facing my Latin textbook. Its title was "First steps in Latin" and I changed that to "First steps in Eating". Mr Jurgens gave me 3_of_the_best for that. My revenge was to put pepper on his false teeth that he kept in his desk drawer.....

Latin....Amo, Amas, Amat. Amabo, amabit, amabemus, amabunt.....or something like that.

[fleinbach]

There is a big difference between a paddle with holes and one without. Actually the one with holes is probably similar to caning. The one without holes is a bit milder. Putting holes in the paddle created a much sharper sting. It would actually raise a welt. Of course this was back in the 50s. I don't think they allow this anymore. Of course it may be different where you are.

[Mayo]

One could probe into the difference between identically sized paddles, one having holes in it and the other without holes. The one without holes may offer more wind resistance, thus requiring more effort to produce the same amount of sting as the paddle with holes and less wind resistance.

I imagine this wind resistance to be measureable but negligible unless you are talking about some really huge paddles.

I will make an educated guess and say that paddles can be as mild or severe as the person using them wants them to be. Both paddles will sting and or produce welts if used with enough force.

Less surface area contact (one with holes, or a cane) would most likely sting more given the same amount of force, than more surface area contact (one without holes) since the force is spread over a larger area.

Now are we talking wood or polycarbonate?

[mikejohn]

Gentlemen please!

I was using definition number 5 below, not number 4.

dis·ci·pline
1. Training expected to produce a specific character or pattern of behavior, especially training that produces moral or mental improvement.
2. Controlled behavior resulting from disciplinary training; self-control.
3.
1. Control obtained by enforcing compliance or order.
2. A systematic method to obtain obedience: a military discipline.
3. A state of order based on submission to rules and authority: a teacher who demanded discipline in the classroom.
4. Punishment intended to correct or train.
5. A set of rules or methods, as those regulating the practice of a church or monastic order.or Forum
6. A branch of knowledge or teaching.


I am learning things which, at my age I'm not sure it's good for me to know.

Gerald, are those legs yours, or did you hire them for the photo?

................Mike

[gerald_d]

Mike, I think those pins were borrowed off a snooker table.

Since my old man was a garage owner, I was responsible for providing the "anti-dote" for a caning....... a 1 foot square piece of truck wheel inner tube, cut on the outer curve for the best anatomical fit, with corners radiused to about 2" (less visible under the clothes) and punched with 1/4" holes at 1" centers for the correct sound effect. Teachers were not allowed to cane on the spur of the moment, they had to consult the headmaster first - those few minutes were vital for applying the anti-dote. Yes, my early high school years (end-60's) were in the inch system. For the record, all forms of corporal punishment were banned here by the end of the 80's.

Growing up in a garage had its downside too - the old man had ready access to V-belts and carburettor hose for applying some discipline. Tell me, would you hit that sweet little boy in my profile pic?

[wishiwasabot]

I am having a problem with the s_prbdxf.sbp file. When I run the file it gives me this error:

Error in - OPEN &outname FOR OUTPUT AS #1

I changed the code from:

INPUT " Type an eight-character or less file name now -->" &myname
&myname = "&myname"
&outname = "C:\SBPARTS\"&myname".dxf"

To:

INPUT " Type an eight-character or less file name now -->" &myname
&myname = "&myname"
&outname = "C:\SBPARTS\xxx.dxf"



Now it moves into the piece and gives me this error:

Undefined variable SQR (sqr ((&dx * &dx) + (&dy * &dl)); 0)


Do I need to have any other program on my computer to run this file ( basic, etc...) S_prb3d3.sbp runs fine (but I still need a good way use point cloud data)

Thanks.

[Brady Watson]

Richard,
Use the probe file that gives you a 3D SBP file. Then download the latest version of the SB3 control software & in it you will see a probe to surface conversion tool. Use this to export it as a 3D DXF.

Recently I used the 3D DXF probing routine...and it died about 45hrs into a scan. I then scanned with the SBP output one and had no problems whatsoever. The SBP to surface converter is a really nice addition to the new control software, even if you still use DOS to control the bot.

-Brady

[bill.young]

Hey Richard,

It looks like you've got 2 problems

1) I'd bet you're running the DOS edgefinding routine with the Windows software...that's what's causing the first problem which you fixed by disabling the ability to input a filename. The Windows software combined all the 2d and 3d probing routines into the Copy Machine Virtual Tool (you'll find it in the TOOLS menu), making probing much easier. I'd recommend giving it a try.

2) somehow one of the variables got changed in your file, giving the second error. If you look about halfway through the file you'll see the line that gave you the error. It should be...

&seglen = sqr ((&dx * &dx) + (&dy * &dy))

...but somehow your second "&dy" got changed to "&dl", giving the "undefined variable" error. Changing it back to "&dy" should solve that problem.

Bill

[gus]

Dave Rawn,

You may have been lost but I will reply to your question anyway. Even if I should be talking about probe problems here. Don't ya know.

There is a Camp ShopBot comming up in the Mpls the 21st of this month. You may want to check it out.