epilog problem I could really use some help

[Jonathan Noble]

Hello folks I have a problem with my machine that’s really cutting into production. I was hoping somewhere within the hive mind known as sawmill creek engravers forum I could find an answer. Here’s all I know and if anyone could help with a possible solution it would be really really fantastic.


I’ve talked with tech support 3 or 4 times so far and continue to have the problem I will continue working with them and I’ve spent many hours reading old posts and searching the forum for answer with no luck yet.


The machine is an epilog legend helix 36ext


when I use the machine two things happen


the x axis randomly jumps location ie. if I’m cutting a row of circles somewhere along the line it just cuts over a random amount and continues cutting as if it hadn’t which means now everything is out of alignment. This happens erratically every day and often enough that the machine is barely usable.


also occasionally the carriage head will take off out of no where at a frightening speed and smash against the side of the machine.


Sometimes these two things happen in conjunction sometimes separately.


What I have done already:


I have cleaned the linear encoder and reader as per tech instructions. (though this machine gets x axis location from rotary encoder in axis motor)
I flipped the cable connecting linear encoder
I then replaced that cable and linear encoder with new ones
I replaced the x axis motor
I tried flipping the cable that runs from the x axis motor to mother board
I’ve run the machine with only itself on the circuit and no other electronics on around it and through a power cleaner with also shows voltage and amperage and it is getting the correct voltage
the erratic behaviour continues seemingly completely at random.
It has also happened occasionally when the machine was just starting and no computer was connected to it.


often i have to end cutting for the day and then it’ll work again the following day until it becomes erratic after a short period of cutting and I have to stop again.


If anyone has any suggestions it would really help me out as this has been ongoing and a very frustrating experience.

[Chris J Anderson]

Not sure if your laser has the same features, or if you are printing from a stored job in the laser but...

our Epilog Helix does random wierd stuff sometimes if we try to run a job that has been saved into memory on the laser.

It never does this if we print a new job to the laser normally.

So we dont use the on board memory any more, its way too risky.

Hope you find the issue, I know how frustrating it can be.

Cheers,
chris

[Mike Audleman]

The skipping I had a similar thing happening with a stepper project I was builting off an arduino. Ended up being a loose setscrew on the drive pulley which allowed it to slip a part revolution until it was cinch up again.

Not familiar with the Eplog internal mechanics but does it use a timing belt? If so, check for missing teeth on both the belt and the pulleys. And check for possible debris in or on the belt(s). Also check the set screws where the drive pulley mounts on the motor. Is it slipping?

But given the other weird take off and run issue, sounds like a possible memory issue, controller board or motor driver. Maybe a cold or cracked solder joint on the motor driver?

[Kev Williams]

My Gravo LS900 has developed a similar problem, which started when I tried to run the machine using the new driver that came with my Gravo 7 software. It started when trying to run some vector passes. The machine would either go someplace then move at about 1 inch per day, or it would do the Whheeeeee-SLAM thing. My problem is that even though I don't use that driver any more, whenever I send a job to the laser using Gravostyle, no matter what it is, once that's done, then the machine refuses to run a vector pass. Rasters fine, but won't vector... It either does the creepy-crawl thing, or the whee-slam thing. And it won't vector from Gravo whatsoever, ever. But after running a raster job from Gravo, it won't vector using Corel either. I then have to delete the printer, then re-install it and reload the driver. Then all is well

The funny thing is, I can run the machine indefinitely from Corel without issue. It's only when it receives a job from Gravostyle does it mess up.

My rep and his crew at NH CS say the firmware upgrade from the Gravo 7 corrupted my controller software, and short of replacing a board in the controller (about $1300) I'm stuck with the problem. Been living with it for over a year, and it's second nature now to just unload and reload the driver after running a job from Gravo...

Not much help, sorry!

[Jonathan Noble]

Thanks for the input so far! I never print from memory so it's definitely not that. I will check for cold/cracked solder connections and have checked the pulley's and other mechanical do dads as much as I can not being an expert in that sort of thing I may have missed something I will look that over again.

[David Somers]

Kev,

So if I am following you correctly. You have a Gravo machine that you use Gravo supplied software with. The company believes that their upgraded Gravo software and its associated Gravo firmware update toasted your Gravo hardware and that their way of dealing with it is to say that life is tough, man up? Seriously? For some reason that grates on me. ?

Sorry Jonathan. That had nothing useful to do with your problem. That was just me saying "whaaaaaaaa???!!!!" <grin>

Dave

[Michael Hunter]

My Epilog did something similar when the X motor failed.

[Bill George]

FYI My son does building automation control. He had re-written the control program and then down loaded to a Flash (USB) drive and then installed in the controller. At random times it would crash. They checked everything wires, connections, replaced parts and so on.
Finally a call to mfg, who suggested another software reload. Turned out the old Flash drive had corrupted the file, not enough so it would not load or run... but crash at random times.

You might also check to see if you are getting interference from RF or radio signals or EMI from welding machines, or other machines in the area putting a spike on the power going into the machine. You can not see these on a voltmeter or amp meter.

[Dan Hintz]

Turned out the old Flash drive had corrupted the file, not enough so it would not load or run... but crash at random times.

Sounds like the company did a really poor job at programming their firmware loader... any proper loader should verify the code before and after loading.

[Bill George]

Sounds like the company did a really poor job at programming their firmware loader... any proper loader should verify the code before and after loading.

Dan that brings up an interesting Question. How many times can a USB or flash drive be written to, and what happens to the old data?

I know on a "real" hard drive when you erase, your just erasing the Index to the file, and the file could be scattered all over the hard drive. The more the file gets fragmented the slower the access to that file. On hard drives you can either de-fragment the device or if you really don't care about the data it can be re-formatted.
My son just copied the program file to the USB drive, and when it loading into the controller not sure if there is any bit checking or not?

[Jamie Bourland]

I don't know if you can have circuits or circuit board exposed while laser is running but I have narrowed down search for defective circuitry in computers using a Freeze Spray. Doesn't really allow me to fix anything but it shows me what to replace, like entire motherboard.

http://www.mcmelectronics.com/produc...1550-/58-17235

I like to keep Contact Cleaner spray on hand as well. One type is mainly for grease and such, the other is for oxides (corrosion.)
http://electronics.mcmelectronics.co...leaner%20spray
When I am away from my classroom for Xmas vacation, enough corrosion builds up on my monitor cable connectors that when I return the signal can't get through; no video.
The anti-oxide spray (25%) fixes this problem.

[Dan Hintz]

Dan that brings up an interesting Question. How many times can a USB or flash drive be written to, and what happens to the old data?

I know on a "real" hard drive when you erase, your just erasing the Index to the file, and the file could be scattered all over the hard drive. The more the file gets fragmented the slower the access to that file. On hard drives you can either de-fragment the device or if you really don't care about the data it can be re-formatted.
My son just copied the program file to the USB drive, and when it loading into the controller not sure if there is any bit checking or not?

Depends upon the quality of the flash chip, as well as the technology (NAND, NOR, etc.)... but it is typically rated in millions of write cycles. Wear leveling (either via software or built-in hardware) can extend the apparent lifetime significantly. Even high-quality chips have bad sectors, but the bad sector table hides this from the user. Wear leveling tends to leave old files on the flash, just "hidden", similar to a spinning platter drive. There should be CRC/bit-checking done by the flash hardware as well as any driver software.

Fragmenting mean very little on a flash drive... access speed is so significant, and there is no need to take into account physical parameters like how long it will take for the platter with the desired data to spin back into position. But the limited write lifetime of flash is also why you don't (and shouldn't) defragment flash drives. For purists who feel the need to defragment, it's best to copy all data off of the flash, wipe the flash, then write it all back again.

[Mike Audleman]

Dan that brings up an interesting Question. How many times can a USB or flash drive be written to, and what happens to the old data?

Old data moves to Miami where it wears speedos that are too small and blue blocker sunglasses that are way to big. You can usually spot it driving 10 under the speed limit in the fast lane with a blinker thats been going for at least 5 miles.

In all reality, flash memory has a life span. It does wear out. The memory can only be written to a set number of times before failure. This is one reason its not a bad idea to leave files on it and not constantly erase them and only erase them when its actually needed. If you constantly erase them, it reuses the same locations time after time and the same directory entries. This wears certain areas more than the rest of the drive.

I leave files on mine until it begins to get full. Then I quick format it and let start all over again.

[Dan Hintz]

This is one reason its not a bad idea to leave files on it and not constantly erase them and only erase them when its actually needed. If you constantly erase them, it reuses the same locations time after time and the same directory entries. This wears certain areas more than the rest of the drive.

I leave files on mine until it begins to get full. Then I quick format it and let start all over again.

This is incorrect... the control hardware for flash drives have wear leveling built in for all but the crappiest of drives.

[Mike Audleman]

This is incorrect... the control hardware for flash drives have wear leveling built in for all but the crappiest of drives.

Not all but some do, yes. Higher quality ones do. But this is not a guarantee and I am unaware of any way to tell if yours does or does not have wear leveling firmware short of looking at the firmware. At least no way the common user could tell. I don't have the article handy but someone did a review, 2013 or maybe 2012, where it showed less than half did bitwise wear leveling on the chips.

Thing is, its easy just to leave the files on the drive and only delete when full. It costs nothing and distributes the writes across all the cells. Then you know.