CET FK zero issues (flat y-base moving around)

[artbot]

i have the used 6x10 fk512 for several month now and it's been a great machine. except for one thing that has me puzzled. my "zero" (aka flat y-base in the control panel) isn't consistent.

this has been happening since our install. but recently has gone cater-wonky and i find myself watching for it to pull up to the right spot every print. now tonight it's off more often than it's on target.
my flat y is set to 50. when it malfunctions, it passes up 50 and goes out to around 70, sometimes 60ish(?). i've scrolled through the config file and it reads "50" even after it malfunctions. so i don't think
i'm dealing with corruption issues.

any input of any kind would be greatly appreciated.

 

[Matt-Tastic]

i have the used 6x10 fk512 for several month now and it's been a great machine. except for one thing that has me puzzled. my "zero" (aka flat y-base in the control panel) isn't consistent.

this has been happening since our install. but recently has gone cater-wonky and i find myself watching for it to pull up to the right spot every print. now tonight it's off more often than it's on target.
my flat y is set to 50. when it malfunctions, it passes up 50 and goes out to around 70, sometimes 60ish(?). i've scrolled through the config file and it reads "50" even after it malfunctions. so i don't think
i'm dealing with corruption issues.

any input of any kind would be greatly appreciated.

Its a sensor issue most likely. the FK is older, so I haven't had any experience with it, but computers typically read data until they find the part they are looking for. in this case, the data would come from an encoder. if it is misreading the encoder it would find the wrong point to start at. the other option could be slop in the belt, but I can't remember if the y-axis is carriage movement or gantry movement. That is where I would check first.

 

[10sacer]

Matt is right on target with the encoder strip. Try wiping it or blowing it off with air gun. If not - I think I have an encoder strip for the printer around here if you need a new one. Pretty sure my new model uses a different strip now.

 

[artbot]

the plot thickens

of course this is a thread about y- position. i didn't get to it today because we were dealing with another issue. but i noticed something odd concerning the y- start position. i needed to run a test so i sent the print. i saw that it didn't
go to the "main" printing position (i have the main one marked with tape). in that it didn't run to the "main" position, i quickly shut off the vacuum, bumped it forward about 3/4" of an inch and ran the print, only to see what seemed to be a print missing the bottom 3/4" of the print. this really had me confused because i've been stopping prints so many times in order to not screw up the sheet. does the printer have some odd compensation? ...like it knows where y-0 is, prints to it but doesn't always start from the exact same y-home position? i'm going to run some tests tomorrow to see if i can get it to print from two different zeros and track if it prints high or in the same place. i may have no issue at all and just don't understand some of the firmware gremlins of this machine quite yet.

that said (and this is probably a new thread all together). we've got this issue that i figured i'd have solved right now. but ... i don't. the printer is "stepping" to the left about an 1/8" over every 3 feet. i figured it's the encoder (on the gantry) so i cleaned it. no result. i ran a special encoder test that i have to see if i could isolate the location on the strip that it was missing count but it just seems to be missing from the right coming back worse than on the left but no spot in particular. so we change the encoder reader, no effect. and today change out the encoder strip, no effect. it's as if we've got the exact same printer.

at first i figured, that maybe one servo on the worm drive on the right is pushing out the gantry to then pivot the gantry to an angle thus cutting off the approach of the head on the right. but our square test print when finished has a perfectly plumb top and bottom. so the gantry is running straight. but still we have the stepping. is there some odd left count vs right count that could cause an encoder to measure longer inches going right or vice versa? i'm baffled. i've attached an image illustrating (exaggerated) this issue. the read is the way that the print leans.

 

[Nameci]

The raster data is very important for accurate print.

I suggest for you to open up the shield for the raster strip. If you are facing the printer, on the right side end of the raster strip there is a white sticker. Do the carriage overrun this white sticker and hit the right limit switch?

Before starting a print, the raster reader will look for this white sticker to take a mechanical reference for its zero. If it misses the mechanical reference it would mess up your x-axis prints and you will not be able to print accurately.

Also another cause would be a loose block nut on one of the side lead screw. What will happen is that, the misregistration can be caused by the slant.

Kindly send me pm if you wanted more info.