Coordinates of calibration circles  

Re: Coordinates of calibration circles

When on the first calibration point, I find that my probe is off by about 1.2 mm in the X direction. On the last point, it is off by atleast 4 mm in X!!
1.2 mm at first point is acceptable and should work. 4mm at last point (generally any different value than at first point) means the geometry of Y-axis is wrong.

Re: Coordinates of calibration circles

1.2 mm at first point is acceptable and should work. 4mm at last point (generally any different value than at first point) means the geometry of Y-axis is wrong.

Thanks for your response, David. In my case, I am able to perfectly trace all the lines on the bed. E.g. if I run G1 X201.2 Y0 followed by G1 201.2 Y200 then I see that my print head perfectly runs over that line at X=201.2 ! I have verified that I can do the same for X=101.2.

I have also verified similar movement for X axis. Any other ideas ? Feel free to let me know if there's something wrong with my experiment too...

Re: Coordinates of calibration circles

Thanks for your response, David. In my case, I am able to perfectly trace all the lines on the bed. E.g. if I run G1 X201.2 Y0 followed by G1 201.2 Y200 then I see that my print head perfectly runs over that line at X=201.2 ! I have verified that I can do the same for X=101.2
Well, then your printer probably loses steps on X axis. Have you tried in high power mode?

Re: Coordinates of calibration circles

Is it just the X axis that is wrong?

Is the belt tight?

Are the grub screws on the belt pulley in the correct position and tight?

Is the X axis pulley properly aligned?

Is the Y frame square and level (all 4 feet on worktop)?

Peter

Re: Coordinates of calibration circles

Well, then your printer probably loses steps on X axis. Have you tried in high power mode?

Yes, I did try the high power mode. That did not get me any different results.

I used a thick piece of double sided tape to push the X right by the desired amount. Then the calibration passed. I have been able to print after this. However, I am not totally convinced by this result.

I am worried that this may still mean than something is off, and affects the results in subtle ways.

Is it just the X axis that is wrong?

Is the belt tight?

Are the grub screws on the belt pulley in the correct position and tight?

Is the X axis pulley properly aligned?

Is the Y frame square and level (all 4 feet on worktop)?

Peter

I will check these, Peter. I have to double check the X belt. The Y frame is square & level. I am pretty sure of that. My first build (done in 3 days) had worked pretty well. I was getting decent prints right from the word "Go". Then I moved the printer, and all hell broke loose. I am able to get some prints (as indicated above). However, I am still puzzled about what is going on.

Re: Coordinates of calibration circles

I tried tightening the X axis belt. I ended up making it loose.

Is there any special technique to tighten the X axis belt ? I seem to require herculean amount of effort, but still dont get it right...

Re: Coordinates of calibration circles

I think the X axis pulley had become loose. Tightened that, and BAM, calibration works ! Removed my additional piece of double sided tape, as that was now causing issues.

Big thanks, Peter ! Now I understand this a little better..

I see that my pulley screws are coming loose often. The one on the extruder has become loose a few times, and now this one. Weird ?

Re: Coordinates of calibration circles

I see that my pulley screws are coming loose often. The one on the extruder has become loose a few times, and now this one. Weird ?
I use threadlocker on all screws except those in plastic parts. Highly recommended especially when there is vibrating environment.
http://www.loctiteproducts.com/p/t_lkr_blue/overview/Loctite-Threadlocker-Blue-242.htm

Re: Coordinates of calibration circles

Tightening belts - I simply remove all except one screw from the motor; the remaining screw is loosened slightly. Swing the motor inwards, fix the belt (quite loose) and swing the motors back. You will be surprised at the tension possible, and may have to loosen the belt one or two notches.

When setting up, I moved my printer; the Y frame twisted so that only 2 feet were touching the table. Unfortunately I use anti-vibration feet, so I didn't notice the twisting. It took me a whole day before I realised what had happened.

Seems the MK2 does not like being moved...

Peter

Re: Coordinates of calibration circles

I use threadlocker on all screws except those in plastic parts. Highly recommended especially when there is vibrating environment.
http://www.loctiteproducts.com/p/t_lkr_blue/overview/Loctite-Threadlocker-Blue-242.htm

Thanks for the suggestion. I will need to find something equivalent to it in India.

Tightening belts - I simply remove all except one screw from the motor; the remaining screw is loosened slightly. Swing the motor inwards, fix the belt (quite loose) and swing the motors back. You will be surprised at the tension possible, and may have to loosen the belt one or two notches.

That's a fantastic tip ! Moment I read it, I was pretty sure of the amount of tension achievable. Today morning, I was wondering if some sort of belt tightening method would be possible to implement using a screw, some sliders, etc. This method makes belt tightening a joke. I had it working in a few minutes - near perfect calibration I would imagine. This method should be in the manuals. No, you must be writing the official manual 🙂 You have certainly saved me many hours of effort, and a lot of sweat. Big thanks again !

I have another issue related to calibration, but I think I will post this in another thread. This thread has already deviated a lot from "Coordinates of calibration circles".

When setting up, I moved my printer; the Y frame twisted so that only 2 feet were touching the table. Unfortunately I use anti-vibration feet, so I didn't notice the twisting. It took me a whole day before I realised what had happened.

Seems the MK2 does not like being moved...

Will calibrating after the move & checking everything again help ? Any standard calibration procedures you would recommend.

Re: Coordinates of calibration circles

Shreekumar

Thank you for your kind comments.

After moving the printer, I think it is always good to do 2 things:

1. Check the Y frame - ensure all feet are firmly on the table
2. Align the Z axis by moving up to the stops at the top.

Regarding calibration; I do things differently, so it is probably better that you follow the manual and not what I do... My method won't be good when using Slic3r.

I have ignored the "Live Adjust" and have adjusted the probe such that the nozzle gap (hot) is about 0.18mm. This means that the first layer usually sticks well when printing with a first layer height of 0.2 mm (I also over-extrude the first layer slightly). Basically it's very similar to the Mk1 calibration. I also amend the firmware to my own requirements and mainly use KISSlicer for generating GCode.

Peter

EDIT: I believe that "ThreadLocker" is very similar to Superglue, but personally, I would not use that on the printer. I just ensure the grub screws are tight, and for that I use my own allen key which seems to be a better fit than the one supplied by PR.

Re: Coordinates of calibration circles

I have to add some praise, the tip to tighten the belt is just brilliant ! It should be the official way to do it in the build manual !

Re: Coordinates of calibration circles

I believe that "ThreadLocker" is very similar to Superglue, but personally, I would not use that on the printer. I just ensure the grub screws are tight, and for that I use my own allen key which seems to be a better fit than the one supplied by PR.

I had a similar issue with the allen key supplied in my kit. It would slip, and not allow me to fully tighten the screw. I had assumed that this was by design 😳

After the belt tightening, I did a few calibration prints of a 1 cm cube. I got different results each time - with different layer shifting in X, and even a leaning tower of Pisa 🙂 . I found that the X belt had come loose again. Closer inspection revealed that the reason was the grub screw coming loose.

Long story short - your tip came in handy again. My cube now seems very nice - and close to being perfect.

I also noticed that I lost one of the grub screws of the X axis pulley. Right now printing with just the one screw securely fastened (on the cutout area on the shaft). Will this cause any problems ?

Re: Coordinates of calibration circles

Hi Shree

There should not be any issues with the screw as you describe, but you may wish to get a few spares (eBay is probably the best bet). I was very fortunate in that my R/C cars have a screw kit which includes both a variety of 3mm screws and also good quality grub screws.

Peter

EDIT: Screw kit similar to this one: http://www.ebay.co.uk/itm/D-Screw-Clip-Set-w-Case-f-RC-Car-Truck-HSP-Sakura-D4-Traxxas-Slash-E-Revo-Summit-/331729945821?hash=item4d3ca544dd:g:YjQAAOSwf-VWaixp They are called "Set Screws" in this kit (it's the 3mm one you require I think).