MKS sbase v1.3 setup

[Francois Colmant]

Hello to all,

I'm in the process of building an OpenPnP machine in France. It's based on Anthony's design. I take this opportunity to thank Anthony, Jason, and every single contributor to the project - really a great job !

The most significant mechanical hurdle I had was to get the Y-rails perfectly parallel. I ended up with 0,1 mm delta and had to pull the two rails together with a cable.

The machine is based on a MKS sbase v1.3 controller. It appears to function quite well, at least  with Pronterface. I could do most of the config.txt setup without any major issue. However 2 problems remain :

1- I cannot control the solenoids,

2- the Z-axis is acting weirdly.

Would someone with experience with the MKS base v1.3 controller be so nice to share the setup files for their machine ? Specifically config.txt and machine.xml ? I could start from there to solve the issues of my system. I could also do with any advice...

Many thanks for your advices.

Cheers,

François.

[Francois Colmant]

OK, I solved the two issues and my PnP is now officially working.

[Jason von Nieda]

That's great to hear Francois! Any findings that would be helpful to share? Perhaps you could post your configuration files for others to benefit from?

Thanks,

Jason

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/41d73ba5-cc7c-46ef-9d43-f2db831c33e6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Zeljko Zed]

Hi Francois,

I also have plan to use this controller, so please share your config file...

Thanks,

Zeljko 

[Francois Colmant]

OK, here are some files from the project :

- config.txt

- machine.xml

- openbuibldsdriver.java

You will find her my first test placing 0402 components. I will certainly have to add an upward looking camera.

https://youtu.be/uzT4Dy1ar_Q

Note that the machine is working with one nozzle only. I will add the second one by the beginning of 2017.

Some difficulties I had :

- One of the stepper (x-axis) performs very well. The y-axis has a small non-recurring error - we are talking 0,3 mm/200 mm, and I have no idea where it comes from.

- I tried to work with optical endstops, and they behave very erratically. I'm switching to mechanical endstops asap - they should be in the mail.

- I think (note the conditional) that the MKS card accepts xhp connectors from JST. I ended up soldering the cable to the connectors.

Do not hesitate with questions.

Cheers,

Francois.

[Daniel Dumitru]

Francois

Show us some more details of your hardware ! Head looks very nice to me, it's a kind of PEte's head ?

Br

Daniel

​

[Francois Colmant]

Not really. It's Robotdigg's

[evilwulfie]

The issue with the optical endstops is that they are manufactured incorrectly.
I had to modify them to function with the mks base.
Mine after the modification work perfectly.
see thread on this list a few months ago.

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/375111c6-7a03-4bfa-8ffb-59787f1e0280%40googlegroups.com.

[Jason von Nieda]

Thank you Francois, this is very helpful! 

One thing I will note: The GcodeDriver should work just fine with this board now and you should be able to use it without modification, unlike the OpenBuildsDriver which required some small modifications.

You probably don't want to go through the effort of changing drivers now, but I wanted to let you know in case you want to try in the future.

To evilwulfie's point about the optos, the thread he is referring to is here: https://groups.google.com/d/msgid/openpnp/9589bf30-19ba-d757-71cb-984ba64ea746%40gmail.com

Jason

[Zeljko Zed]

Thanks Francois,

I hope that I will soon be able to share my experience with this excellent project. First of all thank to Jason. Of course to all those who are assisting their participation in this project. 

Finally (after one month) I get all the necessary materials and, finally, tomorrow I'm starting to make me the most difficult part - FRAME. A while ago I bought aluminum profiles and I made a basic design, but the accuracy was poor due to the poor quality aluminum profile. Again, I bought this time profile of the "ITEM" company (not just cheap) ... I hope  this time the BASIS for this interesting machine will be ok.

Salute,

      Zeljko 

[Dave Hunt]

First, thanks for uploading your config this is helpful for others.

I would mention that your video the C axis turns around like 3 times.  You may want to look at the steps per mm on the extruder.  Try 8.8888 instead of 90

Dave

[Francois Colmant]

Hello Jason,

You are certainly right about the driver. However, the openbuilds driver was the only one working out of the box, with the exception of the solenoid management (which I discovered very late).

I should add that the page about using Eclipse is so clear that, even though I never used it or java before, it went like a charm...

I'll keep the fix for the optical endstops on hand, for when the mechanical stops fail...

Again Jason, thank you very much for all the leadership you put in that project !

Cheers from France, and have a good weekend.

[Francois Colmant]

Hi Dave,

You are right ! I will fix it asap.

Thank you

[Александр Зендриков]

Good afternoon. I use the same charge MKS Sbase 1.3 . Use GcodeDriv. Here are my settings, can someone need:

воскресенье, 4 декабря 2016 г., 22:52:14 UTC+3 пользователь Francois Colmant написал:

[Jason von Nieda]

Thank you Alex, this is very helpful! I have posted it to https://github.com/openpnp/openpnp/wiki/GcodeDriver:-Example-Configurations#alexs-mks-sbase-13-configuration so others can use it too.

Thanks,

Jason

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/f487734a-4af2-4d6a-b8d2-ef4fbe267e49%40googlegroups.com.

[Александр Зендриков]

Well. Although there may be errors)))).

[Jason von Nieda]

That's okay! It's a good place to start for anyone who needs it. 

Jason

On Tue, Jan 24, 2017 at 10:24 AM Александр Зендриков <zendr...@gmail.com> wrote:

Well. Although there may be errors)))).

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/300d6ed2-6eee-4a5b-86cd-53e05044b63d%40googlegroups.com.

[Александр Зендриков]

That's right.)

[Александр Зендриков]

Here found error. The capture of the second nozzle is activated the solenoid from the first nozzle( the signal goes to the same port). Where it is possible to change?

[Jason von Nieda]

Alex,

Do you want to know where to change it in your config, or in the config that I posted on the Wiki?

To change the Wiki, you can just Edit the wiki page at https://github.com/openpnp/openpnp/wiki/GcodeDriver:-Example-Configurations#alexs-mks-sbase-13-configuration. You can post a new file using Gist https://gist.github.com/ and then post the new link in the Wiki.

Jason

On Tue, Jan 24, 2017 at 3:34 PM Александр Зендриков <zendr...@gmail.com> wrote:

Here found error. The capture of the second nozzle is activated the solenoid from the first nozzle( the signal goes to the same port). Where it is possible to change?

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/dbdb42db-3d5d-4f93-8908-662c08ec23a0%40googlegroups.com.

[Александр Зендриков]

Help me to find a place where I need to fix the code.

воскресенье, 4 декабря 2016 г., 22:52:14 UTC+3 пользователь Francois Colmant написал:

Hello to all,

[Jason von Nieda]

Oh, I understand. 

You need to set a different PICK_COMMAND and PLACE_COMMAND for each nozzle. You can do that here:

Just select each Nozzle from Head Mountable, and select PICK_COMMAND or PLACE_COMMAND from Setting and then set the right code for each.

Jason

--
You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/5c6cca4b-efb7-4075-ad85-103f536ea799%40googlegroups.com.

[Александр Зендриков]

Thank you. Now I will try.

[Александр Зендриков]

The team looks like this?

[Jason von Nieda]

That looks good. Make sure to set the right code for each of the two nozzles.

Jason

On Tue, Jan 24, 2017 at 3:58 PM Александр Зендриков <zendr...@gmail.com> wrote:

The team looks like this?

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/108366f5-2f51-495c-a1a9-4a2218dc957a%40googlegroups.com.

[Александр Зендриков]

Fixed , everything works. Here is the file made for the two solenoids. Thank you

[Jason von Nieda]

Glad to hear it!

Jason

On Tue, Jan 24, 2017 at 4:22 PM Александр Зендриков <zendr...@gmail.com> wrote:

Fixed , everything works. Here is the file made for the two solenoids. Thank you

--

You received this message because you are subscribed to the Google Groups "OpenPnP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to openpnp+u...@googlegroups.com.
To post to this group, send email to ope...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/openpnp/e9c590a6-e411-47f0-b767-c6343b3180bb%40googlegroups.com.

[Robo]

Hello,

Could you please uploaded photo of your machine ?

i am trying to setup machine with MKS sbase v1.3 controller, but facing lots of issue.

thanks

[Arie Lashansky]

Hi Robo

My setup is working using a config very
Similar to the one here.I had issues with the drv8825 missing steps on the C motors.The diode mod improved it. But I still have high pitch noise from the motor.I will try using a external dive
as the motor is heating up at 0.6amps current limit

I am waiting for the drag chain and still have no End stops and still no base board or metal plate

http//www.engineerination.com/2015/02/drv8825-missing-steps.html?m=1

[Bernd Walter]

On Sunday, June 11, 2017 at 7:32:06 AM UTC+2, Arie Lashansky wrote:

Hi Robo

My setup is working using a config very
Similar to the one here.I had issues with the drv8825 missing steps on the C motors.The diode mod improved it. But I still have high pitch noise from the motor.I will try using a external dive
as the motor is heating up at 0.6amps current limit

Not that long ago I had been suggesting using the DRV8825 modules because they can handle 24V and high currents quite efficiently.
But in the meantime I've faced the same issue in another setup, which completely changed my mind on those modules.
This is not exactly loosing steps, but a problem with low motor current in specific microstepping positions, which could lead to lost steps of course.
What adds to this problem are 3 factors.
Higher voltage, low maximum current and low inductance in the motor.
In the given case I've had been using a NEMA17 with 1,7A, which means it is low inductance, since it gets it's magnetic field with current and not lots of windings.
But I was using a lower configured current of just about 1A.
Once I've switched from 12V to 24V the problem was there.
Higher current was good, but in this case unwanted as I wanted to run the drivers without a fan.

My personal current decision for my own PnP machine will be to use LV8729 for X, Y and Z.
I havn't tested them fully yet, but so far they seem to be ok.
I've tested them with 24V at configured 1A, but not with the high current NEMA17 motors yet.
The downside of the LV8729 is that they can't handle more than 1,5A.
I have the yellow boards from MKS with the chip on the bottom side.
There are also some other LV8729 boards on Aliexpress.
Not sure if I will keep the 128x microstepping, since I'm using an AVR based board.
Your ARM based board should be fine with the higher step rate required for fast movements.
The rotation axes will be using TMC2100 modules.
They just use 16x microstepping to the outside, but interpolate to 256x for smooth movement.
The preconfigured setup, even for the chines boards can't do much torque, but run smooth.
I don't think that much torque is needed to rotate.
The TMC2100 are more expensive than the LV8729, but I already have them.
I don't know how the LV8729 behaves at low currents to avoid motor heat, but the TMC2100 should do well.
That said: I'm still not having a productive machine myself.

[Robo]

So,

here is the my machine photo and almost everything is ready except pneumatic part. after completing mechanical part i am struggling with software configuration.

my machine is based on Anthony Webb design. so almost setup is the same.

i am facing issue for steps per mm - Calibrated machine as per instruction. when i click on 100mm jog my X and Y axis travel as per given jog like 10mm- 100mm etc but when i jog using camera window and pointing location my camera not targeting at same location this creating issue for feeder setup.

also not sure about axis moment direction while click on jog button direction. whether my all axis moves as per jog button direction or not.

Thanks

[Trampas Stern]

From my experience what I have found is to ignore the nozzles when first setting up, the follow the following steps:

1. Calibrate the X and Y steps per mm in smoothie config.txt.  

To do this use OpenPnP's Camera with the rectile (cross hairs) turned on. 

Move machine over 1cm (10mm) mark on ruler, jog 100mm to 110mm mark. 

Adjust steps X and Y to be correct. 

2.  Verify squareness of machine:

Using a framing square move machine near bottom left corner of machine and position framing square on machine with corner at camera corner. 

Move to the short end of framing square xxxmm, for example 300-400mm, rotate square such that camera cross hairs is at edge of square

Test that corner of square is still correct, adjust until square is parrallel to axis. 

Returning to corner of square move 300-400mm and verify machine is square, if not adjust have the error shown in camera  and repeat above steps. 

3. Rotate camera

Using framing square and cross hairs on corner of framing square set the rotation of the camera until cross hairs line up with framing square. 

4. Calibrate camera pixels per mm. 

Using ruler or item known size calibrate the camera pixels per mm. 

5. Backlash setting. 

Using ruler move head to 10mm mark, then jog to 20mm, then jog back to 10mm. The error in the camera from 10mm to where cross hair is now is you axis' backlash. 

Do this for each axis and adjust backlash compensation in OpenPnP.

6. Repeatability

Move camera to 10mm mark on ruler, other axis on edge of ruler. 

Jog 200mm X, 200mm Y and then back again, verify machine returns to same location. 

Do this with different jog settings. 

7. Home switch testing

Home machine then move to 100mm x and 100mm Y, place clean paper under machine

Place a dot on paper and center under cross hairs

Home machine and then jog to 100mm,100mmm. Verify that you over over the same dot

Repeat 10-20 times.

Reboot electronics and retest

8. Heads offset 

Using the silly putty trick calibrate the head/ nozzle offsets. 

9. Verify perpendicular of nozzles and camera

Take a 3-4mm piece of plywood, MDF or stacked PCBs place silly putty on that. 

Drop nozzle to touch silly putty, then move camera over where nozzle hit silly putty 

If the head offset from #5 is different then your camera and/or heads are not perpendicular to plate. This might cause some errors on parts with large height 

9.5 Build Plate Height

Check nozzle to build plate height at several point on machine, adjust build plate as needed to get parallel to X/Y motion plane. 

10. Nozzle and head play

Grab the brass on the nozzles and wiggle noting the nozzle tip movement and force required, to move tip, if it is easy to move tip you will have problems. 

11. Repeatability

Place a dot on your build plate, much like the paper dot test. For the next few days power off electronics home and repeat testing going to dot. 

Repeat head offset silly putty testing at least once a day. 

12. Strips setup

Only set up one part, one feeder. Test picking part and placing part. Reboot and home often. 

Verify that the strip alignment does not change and everything works. 

Print out test alignment and test placing parts at different rotations. 

I actually have a test where it places part and then puts back in strip. If it can not put back in strip you will have problems. 

If you have problems placing rotated parts check nozzle run out. 

13. Nozzle run out with up vision

Move nozzle over up looking camera. Center nozzle over cross hairs in up looking camera. 

Rotate nozzle 10 degress at a time, if the nozzle does not stay centered on cross hairs, you have run out. The error in the run out will be same error in placements with rotations. 

This can also be tested with silly putty testing but is easier with up looking camera. 

The important thing is to get the Smoothie steps per mm correct first, then squareness then backlash, then  camera calibration and then repeatability. Some may swap camera calibration and repetability. However you need to be sure you machine moves to where you tell it to before you worry about heads nozzles, and such.  If your machine can not move correctly nothing else will work.... I have spent days chasing my tail because I did not follow these steps. 

Also do NOT setup feeders until you are confident in machine and mechanics, I have changed my heads and tweaked my mechanics so many times, each time you have to setup feeders again. Hence do not setup feeders for a job until you know you have your mechanics down. 

Make sure all your feeders and PCB are at the same Z height if possible. This helps camera focus and head/nozzles/camera not being perpendicular to build surface. 

Trampas

www.misfittech.net

[Trampas Stern]

I forgot to mention on step #1 that due to backlash when adjusting steps per mm you should move camera to zero on the ruler, then jog to 10mm, then jog 100mm to 110mm. Never jog backwards except to return to zero then go to 10.  This will be sure to remove backlash from your math... 

[Trampas Stern]

A few other items... 

I mentioned the camera rotation and setting it correctly. Unless things have changed in OpenPnP any error in the calibration of the camera and the rotation of the head camera will also affect placement when doing vision on strip feeders. That is you need to make sure your camera rotation is square with the X/Y movement else it will cause pick location errors with the strip vision system.  

Many of these assumptions are not documented, so it is good to make sure things are setup correct else the code might follow the happy path, with unhappy results.... 

Also with you dual Y motor design you will want to verify squareness after reboot several times, if the dual motor starts up 4 steps off it will cause squareness problems, so try and create a problem and verify you can not get it out of square without effort.  

[Christophe Buchelot]

Hello, i don t know if somebody could help me ... Is it possible to add button to control the second rotation motor in Openpnp ?

Thanks,

Christophe 

[Robo]

I have MKS SBASE V1.3 and its working fine.

I ll upload machine and xml file soon.

[Tung Nguyen]

Hi 

i have download machie.xml file then put to PNP folder but have one error. then software can't start

can you help me solve it

[Tung Nguyen]

hi all

i'm just start make one machine use OpenPNP

i use file machine.xml but have one error. can you support me solve it?

  

[Adam Tamimi]

Dear All

I designed a small-scale OpenPNP Machine. I am planning to use MKS motion controller. 

Could I use the same configuration file for MKS SBASE V1.3 for my MKS Robin Nano V1.2 ?

I look forward to hearing from you.

Many Thanks,

Tamimi

[Adam Tamimi]

Dear All 

I managed to finish designing a small scale OpenPNP machine, I have got a quick question regarding the configuration file.

The motion controller is based on MKS Robin Nano v1.2. could I still use the existing MKS SBASE V1.3 configuration file ?

I look forward to hearing from you 

Regards,