What the heck?

[grahamshere]

I feel like saying more than heck but I've settled down a bit . I was doing a nice intarsia for my first grand son of a lion and giraffe , I did the ruff cut yesterday and that took over 3 hrs, then today I started the finish cut which is an 8 hr job, well the machine was running for 1.5 hrs and I went to have dinner and came back to a nightmare. seems my x axis missed or something and the router was pushing through the wood work. it ended up looking like some one was playing X's & O's all over it, Dam I'm mad. took all I had not to take the sledge hammer to it .That should be a bat and a cnc machine lol. I don"t know what happen. Like I said it ran fine yesterday and the day before, I tested it and tested it to make sure there where no missing steps. Now I run the x axis and its missing steps. Any body have any ideas what to look for it would make me a happy man again. Graham

[wyldesyde007]

I was getting missing steps from an underpowered power supply, I upped my amperage and my missing steps went away.are you running mach3? I have had lags before because of to many processes running on my pc, i.e. virus software auto windows stuff like optimization.

[grahamshere]

Thank for the reply wyldesyde, As far as I know I have no programs running in the back ground,and I make sure the anti virus is shut off. its a fairly empty comp. I am running mach3 . I have the xylotex system and the power supply came with it, 35v and I have the verf set as high as it will go, I don't understand it at all, it go's for hrs then bang it,s all over but the crying. I don't think it's resonance or it would lose steps right away wouldn't it? This will get costly when I'm going into full production of my intasia products, after spending hrs cutting each piece out by scroll saw only to have it destroyed when the machine fails. Graham

[trevis1st]

Do you normally turn the computer off when a job is complete?
If so what is the longest time the computer has stayed on without problems?
How old is the computer? (yes the do give out after a few years)
Is there a dust filter for the computer?
Is there any remote controlled (radio wave) devices near your system?

[grahamshere]

Hey Trevis, Ya I usually shut off the computer, I only use it for the machine, the comp is only a few months old, I do have a filter on the comp and there is no radio waves, I think it has something to do with the drivers just don't know what, the motors do not get too hot and there is a fan on the drivers, it's only happening to my x drive. Graham

[trevis1st]

I've worked for a company that bought 30 identical computers with identical software. Good brand and nothing on them to make you think they should be any different. It seemed like each had it's own quarks. The one I used simply died after about 3 months of use. It does sound like the controller. Maybe a solder point that wasn't quite right, a connection slightly loose, or maybe some dust and static had it's fun.

[grahamshere]

do you think sawdust could cause that? I do have a filter on the box but still get dust in there, I don't think there would be loose connection or it would be happening all the time, at least I think it would. How can you tell if dust causes this, I could clean it and it runs fine for a few hrs then come back its doing it again, guess thats why you pay 100 grand for a top of the line machine. Graham

[trevis1st]

Well what it could be is thermal expansion on the controller. Metal and plastics have different thermal expansion rates. If there is a solder point that is bad it may be touching enough when the board is cool. But when it heats up after hours of use it could have a gap that drops that axis until that part of the circuit cools enough to closes the gap. I guess a way to test this is to give it a line pattern to follow for a few hours to see if it comes back again. Trouble is tracking exactly what solder point or component on the board that is bad.

[grahamshere]

Thanks mate for the advice, Ill run a test like you say, probably slow down the x axis, give it a good cleaning recheck the verf and when I have enough money im going for those gecko g540's and see if that makes a difference, if you have any other advice fire away. Graham

[MrWild]

Well, it may be a simple fix. When you tested and tested, and when you started the job, did you by chance use the computer to change the way you looked at the screen, or used the mouse at all? I ask because if you have a screen saver running, the little bits of use kept it at bay. When you left to eat dinner, the computer wasn't getting little inputs and decided to throw on the screen saver or other cost saving device(s) when it sensed no one was around.

If this isn't the case, try slowing your feed down. Running a program without cutting anything has a lot less stres on the system than cutting.

[grahamshere]

Hi Mrwild,Thanks for replying. I have all those power saver devices turned off, I did have that problem when I first built the machine and learned a fast lesson on turning every thing off. Turning the feed down will be the first thing ill do, Im sure thats the problem, just dose not make sense though, why would it cut for over 3 hrs then the next day 1.5 and that happens, wouldn't it be affected shortly after running the code?

[xcayba]

Lately, my biggest issue with losing steps has been associated with getting chips stuck in my linear bearings. My machine has roller ball bearings on a tube. The chips get on there, then bind up where the bearing runs, causing the axis to totally lock up. To clear it, I power down the Xylotex controller and manually run the axis back and forth a few times (its a belt drive system). I need a better dust collection setup to fix this issue. Perhaps you have a similar issue.

I'm also a little concerned to hear that you have your vref maxed out. I don't think the Xylotex likes that too much. I know that you can adjust the pots on the board beyond the safe limit that will fry the board too. If you are passing more current to your steppers than they can handle, an overheating condition may be existing too. The ambient air temperature along with the extra over-heating may play a role in the inconsistency.