First print NEED ADVICE

Quote
Ralphy
The prints z axis error is about twice. A 2" tall obeject prints out 2 15/16"

The picture shows 3- 15/16"

Change from:
#define DEFAULT_AXIS_STEPS_PER_UNIT {80,80,4000,90}
to:
#define DEFAULT_AXIS_STEPS_PER_UNIT {80,80,2031,90}

Save the firmware and upload it to the arduino.

Reprint. It should now be 2" given the information you gave.

That's really cool - the printer has placed the filament with twice the vertical displacement that was intended, yet you still have a nearly intact print. Good keepsake

well it always printed kinda loose vertically, with the new setting its a remarkable improvement. Im really grateful for everyones help. Check out the pic of before and after...i also stepped the temp up to192 on the pla.

Edited 1 time(s). Last edit at 05/04/2015 06:23PM by Ralphy.

If that print came out of 2031 instead of a 4000 range number and your printer have M5 threaded rods, even though it looks great it's not micro stepping right and you might be loosing quality.

Check your ramps board under the Z axis driver for all the jumpers, you should have 3 for full micro stepping.

Quote
ggherbaz
If that print came out of 2031 instead of a 4000 range number and your printer have M5 threaded rods, even though it looks great it's not micro stepping right and you might be loosing quality.

Check your ramps board under the Z axis driver for all the jumpers, you should have 3 for full micro stepping.

Yes that is the 2031 setting, I am currently doing a print and cant pull the z driver to confirm the three jumpers are in but I recall installing all of them just like in the instructions said. The rods appear to be the M5 but maybe the thread pitch might be different??

See attached for the jumper location instructions i followed

The problems is this: a 1.8 degrees stepper takes 360/1.8=200 steps per revolution, if you micro step you got 16 micro steps per full step so 200x16=3200 micro steps for a full turn. A full turn of an M5 screw is 0.8mm so for 1mm you need 4000 micro steps. A 2031 is most likely half stepping at best.

You might have no problems since most generic layers can be achieve with half stepping, but it might end causing you problems when you get to set ABL.

Quote
ggherbaz
The problems is this: a 1.8 degrees stepper takes 360/1.8=200 steps per revolution, if you micro step you got 16 micro steps per full step so 200x16=3200 micro steps for a full turn. A full turn of an M5 screw is 0.8mm so for 1mm you need 4000 micro steps. A 2031 is most likely half stepping at best.

So I confirmed I put in all 3 jumpers on the board, thats not the issue. Why is it that at 4000 is printed everything bad and twice the height.

Do I have bad software, driver, voltage calibration on the driver?

I'd like to get the full potential out of the printer and so far reducing to 2031 is the only way to acheive an acurate print.

How can the value be increased to the 4000 and without it printing at double the height?

Quote
ggherbaz
You might have no problems since most generic layers can be achieve with half stepping, but it might end causing you problems when you get to set ABL.

Whats ABL?

Hi Ralphy,

ABL is auto bed levelling where the printer uses a probe to measure the height of the bed at various points and then compensates for any tilt by moving the head in Z during each layer. It's very convenient and makes great sense for delta style printers where all three axis are in motion all the time. I'm not so sure it's the best approach for printers like the Prusa i3 which were built with the assumption that during printing Z would move steadily upward and could therefore use a relatively crude Z axis implementation.

Getting back to the mysterious doubling of Z, being out by a factor of 2 suggests that it's running with 8x stepping instead of 16x. Given that it's recommended to use Z layer heights that are a multiple of full steps, that's not going to be a problem (unless you use the auto bed levelling as ggherbaz pointed out), so I wouldn't worry too much. Long term I'm sure you'd like to get to the bottom of it. Given that you've checked the jumpers the most likely explanation is a bad solder joint on the path between the jumper and the chip's IO pin. You could try following the path and checking each solder joint, or you could order a spare ramps board to swap in (I guess the problem could also be on the arduino board, so you might need one of each).

Jim

Jim,

That info is much appreciated. The z axis always acted funny, for example when i try to manually move it via the Repetier controls any more than 1mm at a time it makes a weird nasty noise like "errrrrrrrr" almost as if the motors are stuck and after a couple seconds it stops trying without any effective movement. So lifting the z axis was a pain because you have to do it in 1mm increments one by one.

I have reached out to Folgers support for replacement boards or advice on how to correct it. So far the advice here has generated better results.

I will pull the boards pins and inspect everything. Voltage calibration on the z axis driver cant be the issue right?

Quote
JamesK
Hi Ralphy,

ABL is auto bed levelling where the printer uses a probe to measure the height of the bed at various points and then compensates for any tilt by moving the head in Z during each layer. It's very convenient and makes great sense for delta style printers where all three axis are in motion all the time. I'm not so sure it's the best approach for printers like the Prusa i3 which were built with the assumption that during printing Z would move steadily upward and could therefore use a relatively crude Z axis implementation.

Getting back to the mysterious doubling of Z, being out by a factor of 2 suggests that it's running with 8x stepping instead of 16x. Given that it's recommended to use Z layer heights that are a multiple of full steps, that's not going to be a problem (unless you use the auto bed levelling as ggherbaz pointed out), so I wouldn't worry too much. Long term I'm sure you'd like to get to the bottom of it. Given that you've checked the jumpers the most likely explanation is a bad solder joint on the path between the jumper and the chip's IO pin. You could try following the path and checking each solder joint, or you could order a spare ramps board to swap in (I guess the problem could also be on the arduino board, so you might need one of each).

Jim

That doesn't sound good, and definitely needs fixing. Assuming that the Z axis is free moving with the threaded rods out, checking the Vref has to be top priority.

Info here: [reprap.org]

You'll need to know the driver type and the current rating of the stepper motors.

I should probably 'fess up that this is entirely theoretical for me, I'm still working on my first build. Hopefully someone with more experience will recognize the symptoms you describe.

One thing to note: I have seen jumpers without the metal inset and wrong size jumpers, so just the presence of the jumper doesn't mean that they are doing their function. I will recommend you to use a multimeter and check that the pins are actually connected by the jumper. You can swap jumpers from the other working axis to test them too.

I don't know if it is possible, but if for any reason the pin layout was modified in the firmware, the controller may discard or misunderstand the micro stepping. You can try a fresh Marlin firmware to see if that fix the problem, but that it's last resource.

If you come to the conclusion that it is a board issue (cold solder or broken trace) you should contact Folgers tech with some photos of the broken trace and or cold solder spot and request a replacement, I don't see no reasons why they will not honor some manufacturers defect replacement.

The z axis rise is not making the noise now with the 2031 setting but it did before with the 4000 setting.

Ok so I pulled the driver board and measured all 3 jumpers and they are providing continuity and they appear to fit correctly as far as sizing. I separated the ramps board from the arduino and the solder joints look good to me. See detailed pic attached.

Folger offered to send out new ramps for now. I am going to reassemble for now and check the VREF on the z axis again

I might go with that new Marlin Firmware if all the above fails.

Thanks!

Quote
ggherbaz
One thing to note: I have seen jumpers without the metal inset and wrong size jumpers, so just the presence of the jumper doesn't mean that they are doing their function. I will recommend you to use a multimeter and check that the pins are actually connected by the jumper. You can swap jumpers from the other working axis to test them too.

I don't know if it is possible, but if for any reason the pin layout was modified in the firmware, the controller may discard or misunderstand the micro stepping. You can try a fresh Marlin firmware to see if that fix the problem, but that it's last resource.

If you come to the conclusion that it is a board issue (cold solder or broken trace) you should contact Folgers tech with some photos of the broken trace and or cold solder spot and request a replacement, I don't see no reasons why they will not honor some manufacturers defect replacement.

Quote

The z axis rise is not making the noise now with the 2031 setting but it did before with the 4000 setting.

With the mismatched steps/mm setting the Z axis would have been trying to work with double the configured max velocity and acceleration settings, so maybe that was too much for the motors. Glad to hear that they are sending you a replacement - sounds like things are headed in a generally positive direction

So I double checked vref volatges and they are right on .35 on the z board. How weird....any other way I can check if im getting the 16x stepping vs the half stepping?

You can check the voltage at the pins on the stepper driver. Are you using A4988s? If so, MS1, 2 and 3 should all be high for 16x microstepping (according to [www.pololu.com]). If MS3 is low that will explain the factor of 2 you are seeing. If all three are high then maybe it's a faulty driver - you could try swapping them around and see if the problem migrates to a different axis.

If you are using 8825s then it should be low low high for 16x, high high low would give you 8x again (https://www.pololu.com/product/2133).

I'm not sure what the recommended vref for Z is, but if it's moving smoothly now I guess you're good.

COOL its a4988's so on the z axis driver MS1=4.97v MS2= 4.97v and MS3= 4.97v all other drivers are same values for MS1 2 and 3.

I guess its micro stepping at 16x with the 2031 setting.....weird

Quote
JamesK
You can check the voltage at the pins on the stepper driver. Are you using A4988s? If so, MS1, 2 and 3 should all be high for 16x microstepping (according to [www.pololu.com]). If MS3 is low that will explain the factor of 2 you are seeing. If all three are high then maybe it's a faulty driver - you could try swapping them around and see if the problem migrates to a different axis.

If you are using 8825s then it should be low low high for 16x, high high low would give you 8x again (https://www.pololu.com/product/2133).

I'm not sure what the recommended vref for Z is, but if it's moving smoothly now I guess you're good.

Quote

I guess its micro stepping at 16x with the 2031 setting.....weird

No, it means that the problem might be in the driver itself or in the signal sent from the board.

Now it will be a good time to swap drivers with another axis and test. If the problem is in the driver, wherever you move it too will show the issue if its stays in Z then a reflash of the firmware might be recommended (be sure to save all the values so you can transfer them to the new firmware).

Quote
ggherbaz

Quote

I guess its micro stepping at 16x with the 2031 setting.....weird

No, it means that the problem might be in the driver itself or in the signal sent from the board.

Now it will be a good time to swap drivers with another axis and test. If the problem is in the driver, wherever you move it too will show the issue if its stays in Z then a reflash of the firmware might be recommended (be sure to save all the values so you can transfer them to the new firmware).

Ok so I swap the driver with the x axis which uses the same v ref of .35 and check if the x axis starts priting too wide? I do this without modifing the firmware.

If it doesnt mess up we can rule out its the driver which leaves the ramps board, audrino board or the Marlin firmware from folger tech to be the root cause of the issue right?

You are correct.

Good luck and let us know your progress.

Quote
ggherbaz
You are correct.

Good luck and let us know your progress.

Dam the x axis ran itself to the end opposite the end stopper sensor and the belt started skipping printing a simple 1 x 1x1 inch cube....

Folger included a spare driver so let me try it out after calibrating it.

PROBLEM SOLVED BAD DRIVER. They included a spare in the kit and I calibrated it and installed it and reverted the Marlin h setting for the z axis back to 4000 and i'm getting a good print.

WHOW A MONTH TO GET IT TO PRINT RIGHT!!!

Big THANK YOU to all of you that patiently helped me!

Edited 3 time(s). Last edit at 05/05/2015 10:21PM by Ralphy.

Glad you solved, now enjoy printing.

sounds great!