[kbillan]

Guys- i started a new thread, mainly because the Z axis thread was HIJACKED.. and well.. i am about to give up.

I just tried to run another part-- and well... not even close to completion without failure.

I have checked sherline pulse..i have optimized computer... Am i just expecting too much from this thing? I mean- the thing i was trying to cut was two identical opposite parts (mirrored) and a small part in the middle. Trying to make a master to pull a mold from.

The thing ran the roughing pass just fine- as always.. it got about 3/8" up, of the 4' pattern and I came back down stairs and it was silly how far off it was. YET, it seemed to have x,y working fine.

I cant take it anymore-- there is about to be a blowout sale. I think i would have been better off sending someone wax and cad files and a few hundred dollars...


tail between legs....

[Stepper Monkey]

Don't give up quite yet. PM me and I will give you my number, maybe I can talk you through it, way easier, faster, and more interactive than posting.

If all else fails, one of us may be able to run this part for you if you are in a time crunch, even though that doesn't solve the larger problem!

[123CNC]

I don't know the background of the HIJACKED thread, just hope there were no box-cutters involved and no one got hurt.

Although you do sound a little traumatized.

I would highly recommend a systemic approach in your diagnostics and troubleshooting, rather than drawing straws.

Without all the details, I would assume the good roughing path is indicative of the system behaves well at times. At other times when left to its own devices, not visually observed, you find it outside playing instead of working.

This is characteristic of a 'failure to communicate.' AKA lost steps.

Some more generalities, mechanical error tends to be more repeatable and may occur near system limits. Electrical induced errors can be more sporadic and difficult to chase, EMI/RF. If you are running off a PC parallel port, than the BIOS/OS and control software may be a contributor.

Identify the error the best you can and then start investigating in the area of the most probable of the usual suspects.

Don't giveup, many many others have been down this path. Many here will happily point you in the right direction and get you past the obstacle.

$100, opening bid...

[MechanoMan]

No it's not too much to ask.

Hmm have you tried what "dead slow" z-axis might do?

How about the motors themselves? Try swapping the Z with one of the X or Y motors. One thing I just read up on was due to demagnetization issues (which can result from electrical drive problems) a stepper is not immune from damage or mfg defects. I kinda thought since they were sealed and maintenance-free and all that they'd be assumed good if they turned and weren't smoking but hmm maybe not.

You don't need to run wax in there at all. Or even turn on the motor. Just make up some XYZ point on the table, run the bit to it, write down the exact location. Run the program on empty space, come back in the morning or whatever, jog to your reference point and see if it's the exact same spot.

My motors did not "lose steps" slightly. They would either keep working perfectly or do a stall that would persist for the rest of the transit run so it could lose inches.

Now you gotta be careful because, bring up the Jog Speed menu (hit TAB). Mine defaulted to 20%. So I had set 40ipm in Motor Tuning but keyboard jogging was actually going only 8ipm. Then I ran code which tried to transit the table at 40ipm and it stalled.

You know I kinda wonder if I might not have a similar problem. I tried several "harmonic dampener" designs to increase speed. And if I jog back and forth, 40ipm seems to work yet running code it stalls on the first move if I start with the table far from zero.

That's kinda curious actually. Wait a sec. Why does running code make it stall differently? Someone said that specific ratios of XY speed do it but I'm kinda uncertain of that now that I know more about resonance, it seems "unlikely" that the torsional vibration on one leadscrew would affect the other. There *might* be something else going on here. I DID put an oscilloscope on the DeepGroove driver power supply rails and I didn't see it folding back due to current limiting or noise or anything. Software? Well, Mach3 somehow not able to drive the same way when running code as sitting idle and doing manual jogging?

Anyhow with 15 ipm on all axes I run ok. I agree from what people said the Geckos should have helped.

BTW, make up a z-limit switch to prevent the bit from plunging into the table. Learned that the hard way.

[kbillan]

I am going bananas over here.

I just threw together another part drawing. Much more simple.. ran it and it looked like it was doing good.. roughing went well. I left more stock than usual hoping for a perfect part.. roughing started.. and was going fine. got a little ways into it.. maybe 15min.. and i turned around from my other project.. and im not kidding, it took all my might to not put my foot through the monitor.

I am now wondering if this is a software issue. Yes, i am still running on board video- but cmon, that really make a difference in the grand scheme of things? I turned off tool path monitor, and i run on the settings (green) screen. there are NO other programs running... no screen saver, WIFI disabled... nothing. THe computer is only thinking about the mill (yah, right-- windows).

the machine jogs fine in every direction with the jog thing (tab).. motors set at 50ipm.. accells are 4 on X,Y and 40ipm/ 2 accell on Z.

I just turned off Sherline pulse. I was running the kernel speed at 45k as per the guy at geckodrive.. and i just backed down to 35.

Non of these changes seem to yeild ANYTHING. Nothing ever changes. I mean, how do you know how to fix something if nothing is ever different. LOL.

SO- perhaps-- its meshcam. Maybe this program is creating something wrong and Mach3 isnt jiving with it. OR- perhaps Mach3 is doing something wrong? Maybe if I send someone a drawing, they could port it to an .NC file for me and ill see if it runs from their program??

You know, when i was doing the toolpaths on my macbook in parallels, i was having far less issues.. sort of. maybe.. or maybe i was just having them slower because i was running the stock controller..

I suppose it still could be a computer issue- but really- its brand new, and although I hate PC (this just adds to it) I do know how to make them operate decently.. maybe changing something with the Parallel port??

and again, why is my driver test screen different than the one in the manual? mine says mach2...

I need something. I am too tired to write emails to both mach and meshcam at the moment.. but somethings gotta give here.

Stepper== thank you for the offer. It is 2am at the moment, however i would appreciate it. I dont want to waste your time though.. as i think this is a much bigger problem than a 20 min call would fix.

Ill be back on around 9-10am. Tomorrow is the day. I need some satisfaction here..LMAO. IF anyone feels like converting one of my drawings to code.. uhm, that would be cool [email protected] is my email.. and ill send it first thing in the morning..

[Stepper Monkey]

Just a hint, edit your email address out of there before the spambot crawlers find it and you get thousands of Viagra offers... Preferably before it hits Google in a few hours.

In the future just break it up in ways we can parse but the bots can't understand, like "kbillian, [at] verizon {dot} net"

[TOTALLYRC]

I know you are upset,and I hate to add to it, but you are admitting to running an out of spec computer.
One of the main computer requirements is to run a discrete video card. 32mb IIRC.
When the video calls for memory to write something to the screen and Mach3 needs to use memory, steps get missed, not all the time but sometimes. It sounds like it would be a cheap and easy fix.
I would venture a guess that an email to mach support will get you a similar answer.

Of course it could still be a complety different problem, but before you start to look for more causes, I would think that you would start with the basics, like using a computer that complies with the recomended specifications.

Mike

[kbillan]

No-- no worries... If you think thats what it is--- Ill play along.. heading out the door now. I just HOPE that i dont have the same problem this afternoon. I mean-- you have a point- what else could it be?


There is no argument-- if they say thats what it needs-- they must be correct.. Like i said- i just didnt see how something so insignificant (in my world) would cause such big problems...

Are there any suggestions on what i should get for a video card? I am diving into the manual now-- maybe it says in there..

[fretsman]

Keith, you never mentioned if you ever did the Windows XP optimization on this PC. Have you done this?

If you go to Device manager/right click then choose properties/Hardware tab/Device manager button/and then click on the plus sign next to "Computer" icon, does it say "ACPI Uniprocessor" or "Standard PC" ?

It should say Standard PC, if not, theres a good chance you'll be having issues.

Dave

[kbillan]

OMG-- it says ACPI Multi---

how do I change this to standard computer?

[kbillan]

found it.. OMG i will be so happy if this is the end of the nonsense..

[cnccustom]

mine says it too...how do you change it

[kbillan]

9. Set Computer to Standard PC NOT ACPI PC
NOTE: This is only necessary if you wish to remove ACPI AFTER install.
If you followed step 0, you don't need to do this.


1. Right Click My Computer and select Properties
2. Click Hardware tab.
3. Click Device Manager Button in the middle.
4. Double click "Computer".
5. Right click on Standard ACPI PC and choose Update Driver.
6. Choose "Install the software from a Specific Location (Advanced)"
7. Click Next.
8. Choose "Don't search. I will choose driver to install."
9. Click Next.
10. Choose "Standard PC" from the listing.
11. Click Next.
12. Click OK.

[kbillan]

Okay- it seems to be going well now. the part is about half done and it looks good. It is a rather squared part and the edges seem to be slightly more round than I want. Is this due to constant velocity being on? and if so, how do i turn that off, or turn it down.. I dont mind the part taking more time.. Actually, if this completes well, i am going to get a little more aggressive with accells and motor speeds.. as i dont think this was ever the issue (it was the computer)

[fretsman]

CNCCustom, I recommend you reading the Mach 3 manual and you also may want to look this over:

http://www.machsupport.com/downloads...timization.txt

Dave

[MechanoMan]

Ah ha!

I have "ACPI PC" set too... and switching to "Standard PC" seems to have made my inconsistency problem go away. Still can't get the dampener-fitted axis to run over 40ipm but that's not really a "need" right now just a "want".

The machine launched a bunch of messages, said there was a new PCI modem installed that needed a driver which couldn't be found, that part of the video's functionality wouldn't work with this system, yada yada. I don't need that stuff I just hope the conflicts won't come back to bite me.

[kbillan]

LMAO-- 3 birds with one stone.. NICE. Mine seems to be working well.. although i did get it to stall, but i was standing on it,, so to speak.

I am now looking to get more precision out of the machine. I want crisp. I turned the units down in CV mode and hopefully that will make everything proper.

There is some test i ran and the software told me to run the kernel at 25k, rather than 35k... I was TOLD to run 45k.. can someone explain what this setting actually does?

I also just bumped the pulse up from 2 to 4 in motor tuning...

We will see what happens... Just fired up an intricate part...

[123CNC]

25 kHz vs. 35 kHz vs. 45 kHz

Sets your controller base frequency. Lower is slower, higher is faster. Optimum is as fast as your slowest requirement.

What software did you run, a driver test?

[kbillan]

i did run a driver test== system excellent. Mach3.

[kbillan]

Was able to run some parts over the last couple days-- and i have them under silicon now- BUT- it is still missing. You can tell in the last parts i ran, there is a point where is jumps up about .010".. ill be able to work this out by hand.. but COME ON.. LMAO. why am i having these issues?

Last resort is the new vid card, which i may just go buy tomorrow.. but really- driver test says its excellent.. what on earth else could it be?