Welcome to the SeeMeCNC Community

I'm attaching photo of first-print of calibration block. It started out 'ok', first layer is a bit too thick, then seems 'ok' for a few layers, but then layers started to get offset - angling toward between the X and Y towers. I am did calibration, got x y z and center all equal distance, can slide paper below head same on all points. Using 0.91 on the repetier upload to arduino.

It isn't like the X Y is going offset on an angle, it goes up fairly straight, then all of sudden shift over and keeps printing shifted and then later shift again.

This is on a Rostock Max, just completed the build. Please help, really want to get going on printing real files.

Thanks Jim

According to a previous post by Polygonhell the leaning implies a loose belt/pulley or a skipping stepper, it will be leaning towards or away from the offending tower when it's printing.

Eaglezsoar
Thanks for the reply. I looked up a thread on "Maybe Belt / Motor Skipping" and i don't think it is the same problem. I don't hear or detect any belt teeth slipping Or if it is, i am not sure other then listening for it, how to detect it. I ran a test G-Code script that was in that thread for running the head all around, up and down, etc, and never heard anything that could be perceived as a skip or slip.

The part does lean away from the Z axis, and that belt was tighter than the other two. I thought it might be too tight, because it was very taunt. I loosened it, not made it loose, but just loosened it to make it same as the other 2. But, it made no difference. Additional prints seem to look exactly the same, they shift over at the same places, same amounts.

Any ideas of where to go from here. I am running .91 repetier with the standard settings found in Gene B's 2nd edition manual.

Possibly a loose connection where it is loosing steps.

I've also seen this when the motor currents where too low -- specifically after a firmware upgrade. In Configuration.h, search for CURRENT where you'll see something like (175,175,175,225,0). Try chaning the 175's to 195 and see if that helps.

If the gear on the stepper is slipping, you won't necessarily hear it happen. That being said, dpmacri's suggestion is a good first step.

g.

did you clamp the glass down to the Onyx ??

You can see a clip in the background...

g.

Hey everyone, thanks for the help. Other things in LIFE came up and I couldn't keep on this till now.

Update: I tried to increase current, set to 195,195,195,200 the next print was 'better' there were no longer major shifts, but, now each layer is slightly shifted about the same, so the block still leans. When I saw that I thought maybe the towers weren't sq or ? But, I rechecked and they are all sq to the onyx / glass just as they were to base. I rechecked their heights and they are all the same, plus their distance from the base cut out.
Since the increase in current helped, I thought, what the heck and I increased it another 20 then it really started going OFF. Only did 1 pass, not sure if that's exactly if it was due to just increase of current or not, i'm going to go back to 175 and see what happens again.

One thing is, I do here a click, like something is jumping. I've heard it all along, and suspected a belt jumping, but, when i tried to listen to see which tower, I could have sworn it was coming from the extruder - I even ran a dry run and it didn't make the click, so I was sure it had to do with retracting or something and ignored it for now.

But, after stepping up the current again, now it clicks more often, and the head is going way off. At one point gouging into the plate and tearing the blue tape, so i powered off the unit.

So, it has 'something' to do with slipping, skipping, but I haven't pinned it down yet.

Gene, I checked the gears on the steppers, and both set screws are tight on all, with one in the flat side of the shaft, so i can't image the gears are slipping on the shaft. I'm going to keep probing around, any more ideas would be helpful.

(Another problem, which may or may not be related, is when the block was printing offset, it seemed that there was too much plastic being fed, after it went up ~10 layers and up I could see the head was pushing down into the plastic as it printed, rather that being just above the plastic. When it starts off, it seems fine, so maybe the skips are throwing off the Z height ? Hopefully once i can pin this down I can get some useful prints out of it)

Are the belts at the center of the idlers at the top or is one of them off to the side of the idler rubbing on the side of the belt?

Check out this video: http://www.youtube.com/watch?v=5LCHGMivJeE

Are your Cheapskates adjusted like that? It almost sounds like there's too much friction in the system and that may account for it.

g.

UnLtdSoul wrote:(Another problem, which may or may not be related, is when the block was printing offset, it seemed that there was too much plastic being fed, after it went up ~10 layers and up I could see the head was pushing down into the plastic as it printed, rather that being just above the plastic. When it starts off, it seems fine, so maybe the skips are throwing off the Z height ? Hopefully once i can pin this down I can get some useful prints out of it)

UnLtd' I don't think this is related to your main issue. At least not shifting to one side. Post a pic of your newest print.
I just "finished" my build and am mostly doing test prints. Mine also did the too much extrusion. For this, you definitely want to go through this thread. From it, there was some (for lack of better words) Spill over. that is the some of the extrusions would fall over the side. but it would do this all the way around, the average of the build was straight up. From a pic of the new build, it should be straight forward to determine. In fact, provide some pics from different views

http://forum.seemecnc.com/viewtopic.php?f=54&t=1163

It will help you dial that in. Personal opinion, from a person with limited experience, you should chase down the shift first.

Cheers and best to you!

Chris

Their in the centers, just a hair to the side

Pictures...

Check your toothed pulley that mounts directly onto the motor shaft. Check that the set screws are tightened down to the shaft. I had this identical problem and spent forever trying to find the issue only to realize that all I had to do was tighten the set screw holding the pulley tight to the motor shaft.

UnLtdSoul wrote:Their in the centers, just a hair to the side

neurascenic wrote:Pictures...

With a hair in the picture as reference to show how off it is? Don't you trust him?

Punchy!

Trust isn't an issue, I just find that people have different ways of thinking about things that easily get lost in text. Especially for us nub's at something. it is easy to misinterpret the "goings on" of things


Pictures are a little more concise.

Oh, and i am just voyeuristic... so there!

UPDATE: Sorry Neurascenic, no pictures on the belt position because I ran other tests and am narrowing things down -- a little

I stand up a battery on the platform (after it draws perimeter line) to where the extruder plate just touches the battery when it prints the outer perimeter of the box, in line of the direction that the parts have been leaning toward.

When I print in DRY RUN with no plastic - the clearance between the extruder plate and the batter remains the same, from bottom to top and no clicking is heard

BUT, when i print with plastic
I do hear clicks now and then, indicating 'something' is slipping.

When I hear the clicks, that's when the extruder is shifting over, after the click then the battery will move over as the outside perimeter is printed, because the extruder has shifted.

BUT, I checked and checked - the Clicking is coming from the EXTRUDER Stepper - not one of the Towers.
And, the shifting is always Away from the Z axis.

I have checked all the gears - i had lock tight on the threads, so they were really hard to break loose, but, I unthreaded them and then retightened them - especially the ones on the flat part of the stepper shaft. They were and again are on Tight, they aren't slipping on the shaft. I haven't perceived the belt slipping on the gear threads, although I will run a few more tests on it, but that isn't where the clicking sound is coming from.
But, 'something' has to be slipping in relation to the Axis - somewhere.

Since the Clicking sound is from the filament extruder, and is when the extruder is backing off, or retracting, i think to go up to the next layer, i am suspecting this is more electrical / electronics related then mechanical (or the mechanical is causing an electrical interference).

If it is related to the filament extruder retraction, then I can think of 2 paths where the Z axis is slipping :
1) would be that the plastic is not retracting easily, and when the extruder tries retracting it, it is Physically pulling up on the extruder platform so hard that it is causing one of the tower steppers (Z axis) to slip on the belt.
2) there is an electrical problem - where either there is a weak electrical connection where the Z axis stepper is functioning normally most of the time, but, when there is extra current draw from the extruder retraction it is enough to sometimes deprive the Z axis of enough current that it can't step at that time.

If it were scenario 1) then i would think i would hear a click on the Z axis as the belt would be jumping a tooth at that time, OR if the gear wasn't tight on the shaft, it could be slipping at that time. But, I don't hear a click on any of the tower steppers at that time, and I've totally taken out and re-tightened the gear set screws, i am 99.99999% sure it is not a slipping gear (hey, there is always that small chance the threads are damaged and the set screw gets tight, but not against the shaft, but, I just don't think that is happening, but again, i'll recheck it again).

If it's scenario 2), then there might be a number of causes and places to look.
It could be a weak electrical connection on the Z stepper connector
It could be a defective Z stepper motor
It could be that the plastic is putting up too much resistance to retracting that it is drawing excessive current drain, causing the Z axis to not get enough current at that time.
It could be a bad Rambo board / stepper driver
It could be a weak / bad Power Supply
It could be a bad wire connection from the power supply

Please help me out if you can think of other things to add to the IT Could Be list, and other things to check.
Also, HOW to i go about checking these things? Especially the steppers?

I have a 'slow' LCD DMV meter that checks amperage, but, that would mean disconnecting the wires and checking current inline, but, it is slow to measure, so i think it would be useless. I don't have a scope, but, I have a number of mini aduino boards and I have a arduino sketch and program that outputs the A/D inputs on the arduino to a GUI windows interface that allows me to have a 4 (or 6) channel scope . 0-5 volt without using any resisters to cut down the voltage. It will take a few hours to set it up, but, I am thinking of using it to see what the voltages to the steppers looks like, only if i have to,
Or is there an easier way to check steppers / etc?

In EEProm the Ext1 Steps per mm = 92.65
Attached is screen shot of the Slic3r extruder settings

I noticed that in the manual controls on repetier it has retract set to 100 mm is that correct?

A couple thoughts:

1) There have been problems reported where the end-stop wires receive interference from the stepper wires and the firmware things an axis is against an end-stop. Since the problem happens when the extruder motor is running, see if the extruder motor wires are touching one of the end-stop wires (or very close to them).
2) Your test of dry-run vs. actual run was a good one. Have you tried doing an "actual run" but without any filament in? That way you can test your theory on whether or not the skipping of the extruder is causing added force on one of the other steppers. The the run without filament still has a problem, then it's something electrical, not mechanical.
3) Sometimes the stepper motor mosfets overheat which causes them to shutdown briefly (basically causing a skip on that axis). Since the extruder mosfet is close to the Z-axis mosfet, I'm wondering if the extra heat from the extruder mosfet is causing the z-axis mosfet to overheat. Try opening the door and running a fan to blow cooler air over the board. FYI, the mosfets are near the bottom of the board where the motor stepper wires plug in. They're little square chips and you can buy heatsinks for them to help with cooling.

dpmacri wrote:A couple thoughts:

1) There have been problems reported where the end-stop wires receive interference from the stepper wires and the firmware things an axis is against an end-stop. Since the problem happens when the extruder motor is running, see if the extruder motor wires are touching one of the end-stop wires (or very close to them).
2) Your test of dry-run vs. actual run was a good one. Have you tried doing an "actual run" but without any filament in? That way you can test your theory on whether or not the skipping of the extruder is causing added force on one of the other steppers. The the run without filament still has a problem, then it's something electrical, not mechanical.
3) Sometimes the stepper motor mosfets overheat which causes them to shutdown briefly (basically causing a skip on that axis). Since the extruder mosfet is close to the Z-axis mosfet, I'm wondering if the extra heat from the extruder mosfet is causing the z-axis mosfet to overheat. Try opening the door and running a fan to blow cooler air over the board. FYI, the mosfets are near the bottom of the board where the motor stepper wires plug in. They're little square chips and you can buy heatsinks for them to help with cooling.

I agree with all you have said. Out of all of them the most frequently reported problem is your #1. If the end stop wires are run anywhere near or in the same channel with a stepper wire they will pick up the induced EMF from the stepper wire. Again, this is the most frequently reported problem, the second most frequently problem is your #3 where the mosfets overheat. A fan blowing on the Rambo card in the area you pointed out along with heatsinks is a good idea.

NEW UPDATE OFFSET RESOLVED

Sorry its taking me long time to get back. But, I've been getting somewhere (And it doesn't look like my posts are displaying in right order in the thread - at least not on my system)

dpmacri wrote:A couple thoughts:

1) There have been problems reported where the end-stop wires receive interference from the stepper wires and the firmware things an axis is against an end-stop. Since the problem happens when the extruder motor is running, see if the extruder motor wires are touching one of the end-stop wires (or very close to them).
...

Yea, We ran both the extruder motor and Z axis home sw down the same tower. That was the problem with the offset issue. I've rerouted the Z axis sw down the Y axis along with the Y axis sw.

But, I don't understand the logic - why would interference on the home sw cause the prints to offset? If the program detected the sw was hit, then wouldn't it be thinking the head is way up and try to bring it all the way down to where it's printing on that axis? Why just go offset at that point? And, the other question is, why is the program even looking for the Home sw when it's printing? Seems like it would only need to look for home after it gets a 'home' command and starts running the head up. Then it would seem that the program could just ignore the home sw while printing. Does anyone know the logic of why the home sw is being monitored during print?

There are tweaking issues i now need to address :

Here is photo of the new block (I scaled the block to 75% to save on plastic.) The block is definitly more straight, not shifting to one side as before. But, the block is slighter fatter at the base and the bulges back out toward the mid to top. The edges on the holes are not sharp, and at the top there are those rabbit ears on the corners - it prints like that, each layer has those ears, from close to the base on up.
Should i post these issues in a separate thread?

Thanks to everyone who helped out with the offset issue. Maybe Gene should make a note for the next edition of the manual to NOT run the extruder stepper wires in the same tower as an axis sw. When i had an analog input to an offboard arduino hooked up to the sw on the Rambo side (I was monitoring for glitches, didn't work), but it suppressed the noise and i got the same print as shown. But, i tried to put additional pull up resister on the sw input, and that didn't help. It must have been the capacitance helped. Maybe a small cap would have worked, to filter out the EMF, but i didn't have various sizes to try it out. Too large a cap would delay the detection of the sw, so i just re-routed the wires, and got the print shown (did several, they came out the same).

I'm not 100% certain why the firmware checks the end stop switches continuously, but I *think* it's a protection mechanism that's been built in for cartesion printers that use both min & max end-stop switches. That way the firmware won't keep running steppers if they hit one of the end-stops (min or max). And I'm also guessing that it doesn't screw up the position because the SW is just checking whether or not an end-stop is pressed. It doesn't adjust any values based on that.

Regarding your rabbit ears, I think it means you're extruding too much filament. If you haven't already, try doing the extruder calibration to make sure the extruder moves X mm when you tell it to move X mm. And make sure to accurately measure the diameter of your filament in a few places and average those to tell the slicer what it is.

dpmari
I haven't done a filament calibration yet, but have measured the filament a number of times (average 1.67).

But, it seems like 'something' is getting adjusted when the home sw was glitching, because it definitely shifted the head over. When it was printing, before i re-routed the wires, I hit the sw manually a few times and it started shifting over more each time. I'll try it again now that it isn't shifting and see if false home sw hits cause it too. If so, then, yes, the program is making some sort of adjustments when it detects the sw.

I think what is happening, is;

the switches are probably on an interrupt pin on the microcontroller for the reason dpmarci mentioned. This way the software doesn't have to actually look for such an event. it is kind of hard wired into the IC, and once the interrupt is triggered, the firmware handles it in a specified way.

as far as the rest of it goes, this makes sense. on a delta, when homing, and one of the LS is tripped, the other motors have to run.
where, on the tower that the LS is tripped on, the firmware stops that motor.


Makes me want to do some dead head prints and try it out.

When not homing, the switches are just a safety all it does when they open is stop moving the axis, this results in the shift.
The firmware just assumes the head is always where it should be.
There is a flag in configuration.h that disables testing the switch while printing and a lot of people run with them disabled (me included). On Cartesian printers what happens is when you hit the end with no checking is the motor stalls, it just buzzes a bit, and ruins the print. The issue on a delta printer is that if you are printing something very tall, and you do hit the top it will destroy the belt because with the long belt run the stepper has enough torque to eat the belt, rather than stopping.