Messages

241

General Mach Discussion / Re: Connecting an MPG to Mach3

« on: December 24, 2008, 11:51:02 AM »

Great !!!
Glad to hear it. You may want to check the two logic levels. The high logic level voltage on the pin should be over 3v and the low should be below - say 1v else you may get flakey operation. Lower the resistor value if the high logic voltage is below 3v. And as previously mentioned it will be a balancing act to maintain a low logic level that insn't too high.


Sage

242

General Mach Discussion / Re: Connecting an MPG to Mach3

« on: December 23, 2008, 09:24:28 AM »

Rogerbsst:
4.7k (Written as 4k7 in the industry so the decimal point will not be misinterpreted - mostly a problem on lousy paper prints) is what I suggested. But it sort of depends on what they have as a pull down. Pull-downs are a bit of a pain (and quite un-necessary if the circuit is designed properly) since they always form a divider with whatever else you put on the line to pull it up.
 The trick is to get a pull-up resistor value that gives to at least 3v on the line when the circuit tries to go high. You may have to try a few values.
 What value do they have as a pull-down? Is there a circuit diagram of the output available to look at?

Sage

243

General Mach Discussion / Re: Connecting an MPG to Mach3

« on: December 22, 2008, 08:32:22 AM »

It seems he has lost his logic levels in the transition through the BOB.
I'm not familiar with the model of BOB quoted above.  Maybe the outputs of the BOB which are connected to the parallel port inputs require a pull up to restore the logic levels. They would be required if the bOB had open collector drivers on the output.
I'm not sure if parallel port inputs have internal pullups to satisfy the requirement.
It wouldn't hurt to try a pullup of 4k7 or so between the parallel port input (along with the BOB output) to +5v to see if the logic levels return.


Sage

244

General Mach Discussion / Re: Connecting an MPG to Mach3

« on: December 21, 2008, 07:53:19 PM »

Is a pullup resistor required on the parallel port input pins?

Sage

245

LazyCam (Beta) / Re: Pocketing not working

« on: December 17, 2008, 08:46:46 AM »

I will reiterate that the main purpose for posting the problem I was having with LCam was to document the problem so it can be corrected and so others can see it and read a possible solution.
 It appears Art is probably aware of all the quirks and when (if) he gets around to recoding the program those quirks will be gone and, as is usually the case with code, a new bunch will arise  .
 I guess we just have to sit back and see what happens.
 For now I'm actually quite happy with the program. Each quirk I've found has been worked around (with the help of others here - thanks) and I keep going forward. I mostly use it to generate the math of the curves and then build on that code manually. Most times I don't even bother with setting depth-of-cut because the hundreds of lines of repetitive code generated by LCam are more sensibly handled with subroutines which I write myself.
 Since we worked around the latest pocketing issue I'm happy again.

One note on what Art has posted in his replies though:
 I'm not sure why he goes to such great lengths to FIX bad drawings in LCam, such as lines that don't join etc etc. and thereby implementing things like the connection tolerance. Second guessing the human just opens up the file to mis-interpretation and quirky action.
 IMHO if he can detect a problem then the program should just indicate to the user there is a problem in plain english (not in typical Windows cryptic fashion) and let the human fix it in the originating program.  I'm sure the program could even point out the co-ordinates of the offending issues like "lines not joined at X2.234 Y5.234 Do you want to connect or Cancel.
Mach3 does it now when arc end points don't end up where they should be. It doesn't try to fix it. It just lets you know and refuse to go any further.

Enough said.

Sage

246

LazyCam (Beta) / Re: Pocketing not working

« on: December 16, 2008, 02:56:43 PM »

OK. I guess I got everyone  annoyed now. Sorry.   
Thanks for your help 
I guess I was happy a few posts back when I chose a 3/16" cutter and ALL of the profiles pocketed without issue.
I think we are all getting frustrated with something we have no power to fix.
The problem and symptoms are documented well enough here (the real purpose of startng this thread) so that the programmer can figure it out. Assuming there is some mechanism to advise him of that.

Thanks Again

Sage

247

LazyCam (Beta) / Re: Pocketing not working

« on: December 16, 2008, 08:16:25 AM »

RICH:
I guess we have another set of variables now. I notice you have switched to zig-zag pocket fill. How does your process react to the original circular fill?
Also, there is little point in having a lead-in on an inside pocket (IMHO). I'm not sure how you started on that.
Although it is nice to figure out what works at all, it would be good to stick to a set of standard tests. We started with a circular pocket with no lead-in. How can we get that to work? Please test that scenerio with your process. Does changing the cutter still have the same effect? I'm getting a bit lost when the whole ballgame is changing.
 Trying everything has some merit in that you might find something that always works but the investigation process doen't seem to scientific so far.

To bad the programmer who created this monster can't chime in here. Some insight from the prorammer would go a long way.
 Anyone know how to contact that person??



Sage

248

LazyCam (Beta) / Re: Pocketing not working

« on: December 15, 2008, 03:40:44 PM »

I'm going to have to absorb and try what you just told me, but this is just way too much to have to do.
That one corner should have been filleted like the others. I guess I missed it.

Curious to find out if that one corner was the cause of all the grief. Is the same corner on the opposite end also un-filleted.

Not trusting LCam too much I'd be afraid scaling and un-scaling might screw up the drawing. I'd have to export the g-code and convert it back to DXF to check it thoroughly. I have a program that will do that.

How much do you scale it up?

It depends right?   I know, whatever works right.



Sage

249

LazyCam (Beta) / Re: Pocketing not working

« on: December 15, 2008, 09:29:18 AM »

Ok. Lets call it CrazyCam. I got it to work. 
Just after I posted above a light went on. Why was I using a 1/16 cutter when the maximum possible is 3/16 for the .093 radius corners? 

Drawing from docltf's process:

I tried a 3/16 cutter and voila the corners pocketed. Yet they still won't pocket with a smaller 1/16 cutter. That's just plain weird 

So what kind of work around "rule" do you derive from this. Try every possible cutter?

Too wierd for me 

But I guess I  can't complain, the solution was found - again.

Please try it yourselves with the file posted in my last message (above) and see if it was some fluke that it worked this time.

(crossing my fingers)

Sage

250

LazyCam (Beta) / Re: Pocketing not working

« on: December 15, 2008, 09:08:46 AM »

docltf:
I don't have anything newer than the code in message 25 but I've attached it again here anyway.
I don't expect (or want) someone to write new code for this. It was just another example of LousyCam going wrong (apparently) and also an exercise to figure out a work around. Having said that I need to get a piece someting like this one done sometime soon. I'll mess with it some more to see if I can gather any clues.
 I can see your plan of attack varying some of the parameters to see what happens. I use that method myself - it's a good one. I didn't mean to be critical. Just pointing out that sometime trial and error parameters and reality don't jive.

Material Aluminum. Thickness of part is 1/2" for outer profile.  I'll probbaly modify the drawing to add an offset to mill outer profile as well.
The rectangular and corner pockets are 0.15" deep.
Cutter - whatever willl accommodate the .093 radius - (3/16dia). I set 1/16 in LazyCam. Maybe I should have tried 3/16.
Speed and feed to suit the cutter and a mill/drill machine.

Sage