Shark does unexplicable and inconsistant actions

[pjmayer]

Bob,

Ed is right that when you load SD card, the entire TAP file is put on the SD card. Your computer then just becomes a display, allowing you to start, e-stop, pause, etc. It shows you what g-code line is executing, but isn't really involved (the Shark should just carry on if the computer crashes or goes to sleep).

I always leave the USP cable connected when I run. I don't think the FRO slider would cause any of the trouble you're seeing. It really just scales down the feed rate written into the program. I do have problems with the jog buttons if I put the FRO slider all the way down (25%), but that's the only bug I'm aware of in the latest version of the control software.

I regularly run programs with more than 3000 lines, and haven't encountered the error message you mentioned. I've read of others who have problems with the SD card, but I'm still using the card that came with my Shark. I suggest calls to both Rocker and Nextwave (I've actually had more luck emailing NextWave).

PJ

[Tim Owens]

Bob,
you need to do a consistant set of steps so we can figure out what might be happening. You say you are using the Windows 7 and XP why are you switching computers? Are you using the same version of the controller on both machines?
I would suggest using the 7 since the XP can be flakey talking to the USB. Also are you hitting the pause button rather then the E stop then trying to run the program again with a jog? You state that the machine "goes all crazy" but we need some more info like, is it moving your program over still going? Does it look like it is running an older program you just ran? Have you checked the couplers to the motors to make sure you are not having something slip?

Thanks

Tim

[BobKrattli]

Tim; Thanks for your response, we are back to using Windows 7, we just tried the Xp to see if we would get the same problem, and yes we did. No I have not checked the couplers on the motors, but I will. I don't see this as the problem, since I can run the same code file, and it will happen again at the same place. As an example Saturday we were trying to run a test set of house numbers with a street name below. I think the number was 9854, 4 inches tall, ( I always start in the center of the part) I think we had 4700 files in the g code. It started on the 5, came down and cut the 2nd half of the street name,back up to the 4, did the 8, down to the frist half of the street name and up to the 9. Just as it was on the last past of cleaning out the 9, it went crazy and started the 9 again, on top of the orginal 9. It had about a inch to go on clean out, the second 9 was about 1 1/2 inch tall and it cut about .0625 deeper. When this happens I hit E-Stop. If I hit 000 the unit will go back to almost center, X and Z are good but Y is a little off, or X and Y are good but Z goes about .125 down into the part. We ran this same file several times, between runs we changed Sd cards, reformated the Sd card, changed the USB cable, tried a differant port in the computer, and checked again for a loose wire both inside the control box and the XYZ connectors. During this time is when we got the screen "NCpodgcodecomplier.exe Has Stopped Working", this happened several times, to get back up we did a system shut down. We ran the same file and got the same problem at the same place. Sunday I made a differant file with the same number, but no street name, had 2700 files-Ran fine. It just seems that if I have over 3000 files I may have a problem. It is not getting a old file, most times the problem is on the last number cut, once in a while it will be the second last number, but it always tries to cut a smaller number of the one just cut, when the problem occers. Sorry for the long response, I am not that good on computers, so if this gets over my head I may need to get my son-in-law in on this, but I do appreciate your help and look forward to your input. Bob

[Tim Owens]

Bob, I have not seen this issue before. There should not be an issue with the amount of lines. I would say provide us with the controller and a program tap file that you have the issue with and we will check it out here to see if we have the problem. But to be on the safe side we will just swap out the controllers whatever the result.

[thoelscher]

Just to be clear, we aren't unplugging the USB cable from the laptop when running the Shark. We understand it needs the power coming from the laptop. Bob is correct that we can see the lines of code progressing through the control panel while the Shark is cutting. It appears to me that when we get to about 3000 to 4000 lines of code (or commands) is when the problem begins. I've looked at the G code and the Shark simply isn't listening to the code. We've tried multiple memory cards and USB cords and USB ports. The error message Bob is describing had no rhyme or reason. I could restart the laptop, unplug the USB, try loading different G codes that worked previously, etc. and couldn't seem to figure out why it was doing it.

[spinningwood]

Have you tried using a smaller capacity SD card? Like 2GB instead of 4.

I don't know what file system the controller uses, but I do know that it had problems when I was using filenames longer than 8 characters for scan probe output. That reminded me of the days when filenames were limited to 8 characters (not counting extension) and disk size was limited to 2GB.

The fact that you hit the same problem at the same exact spot (if I am understanding you) and there is not a mechanical problem (junk on the lead screws or a chunk of wood slipped under the table blocking the gantry) sure sounds to me like it is a problem with the file system on the card. I've run programs with 50,000+ lines of gcode on a 2GB card and haven't come close to exceeding the capacity. Maybe a 4GB card won't work.

A 2GB SD card costs around $5 - $7. Might be worth a try.

Another thing to consider is power saving features. Make sure you have it all turned off. Screen savers, disk drives,displays, everything. Disable all power saving features. Especially on a laptop. Check the advance features and make sure USB isn't powering down. Although the computer isn't actively running the shark, spurious signals sent over the usb port as the machine powers the ports up or down can cause the shark to do unexpected things. I learned that one the hard way.

What anti-virus stuff do you have running? Consider diabling that and see if that makes the problem go away.

Or you can send your controller and a problem *.tap file to Tim and see if he can duplicate the problem. Mabye send him the PC as well....

Good luck.

Ed

[BobKrattli]

Thanks for all the input, but frist things frist, Ed , I did read the instruction manual twice before we started, and I know my son-in law also read it, I was thinking of the 150 page VCarve Pro Reference Manual is when I started to go crossways. The couplers to the motors are good. I think the Power Saving and anti-virus features are off, but will double check. Before I bother Tim with replacing the controller and things, I will get a 2-GB SD card and try it. Thanks again, you have no idea how I appreciate your input Bob

[spinningwood]

Bob - One other thought. Try a different brand of SD card. Changing brands solved some problems with a friends digital camera a while back. You never know.

Ed

[leonard1234]

Ed, Bob if I may interact here I have had my CNC for a month now and still have not ran a part yet. I was having problems with update firmware constantly and also getting the same error message gcode stop working. I did get in touch with Tim and his request returned the 2 boxes. The snet me new or repaired boxes and not works fine at this point. The problem I am having is when I goge to zero as I set in Vcarve all XYZ with Z set at paper width from my stock. I load G codes turn router on hit run from SD card the router goes back up and not down into the stock. What qam I doing wrong?

[spinningwood]

Leonard - I started a new thread and posted a response there. Poor Bob has enough problems without having his thread hijacked.


Ed