Ask Grbl CNC questions here.

Dear Peter,
First of all thanks for the reply.
I forgot to mention that drivers are DRV8825 (I'm from Italy so i think that buying from pololu site or out of UE, i will pay duty and shipment more than drivers itself, so I bought them from local Amazon.it, I don't think they are genuine).
From when I wrote the first post and now, I've switched from 12V supply to 24V supply, and I've noticed so much heating and stepper drivers resets every 3-4 seconds.
Other thing, not really needed, it's an MPCNC but designed from scratch by me, not available online. I've checked all leadscrews and couplings, no frictions or lockings on movements.
Now i'm printing a new case with 2 fans, trying to solve overheating.

Thanks for the attention
50Special

 

A couple of thoughts.

First, GRBL in absolute mode will "accumulate" any lost precision your machine will have. You can test this by doing a series of short moves in one direction, say 20 moves of 5.005 mm, and compare to a single 100.1 mm move. They should be the same length.

If you haven't already bought the drivers, look at trinamic step sticks (assuming you are using that format). They run a lot cooler and are significantly quieter than the Allegro based drivers. Not that much more expensive, either.

Try running at 1/8 step for better resolution.

For a CNC machine, you might want to consider better drivers not in the stepstick format. The stepstick format maxes out around 1Amp and were really designed for 3D printing. CNC has higher power needs.

 

Those certainly will work and are an excellent choice, if your budget allows. There are other driver that are quite a bit less expensive - TB6600 based, for example. - that will work a lot better than a stepstick. If you think you are going to be continuing on with a more powerful CNC machine, the 542s make sense as you can use them on fairly powerful motors. Lots of choices for different budgets.

 

Hi Everyone,

Currently at the beginning of a new project and I have been playing with GRBL 1.1.

I am planning to drive the table with 4x stepper motors using 4x DM542 controllers. I was plannning to wire the controllers in parallel back to the 3 pins on the UNO (Enable, Direction and Pulse) however I got to thinking about whether the Uno could handle the signal current. From the data sheets the typical current draw for the signal inputs on the DM542 is about 10mA so if I wired 4 in parallel it would be equal to 40mA which is double the current that the Uno can support (So I'm expecting to see the magic smoke if I did it that way).

I then got to thinking, it is typical (I believe) to wire the 3 controllers (in a 3 axis machine) to the Enable pin which would draw 30mA on the pin which again is over the 20mA upper limit the UNO can support !! I haven't seen this issue raised before so assume its not a problem but I'd be grateful for any views about this.

I'd also be grateful for any suggestions about changing my design; perhaps putting 2 stepper motors in parallel on to one controller (and then using 2 controllers in parallel to get 4 steppers in total). I realise this doubles the load on the controller so 2x2A motors would put a 4A load on the controller.

I've also toyed with the idea of putting in a simple transistor circuit after the Uno to act as a buffer between the Uno and DM542.

Anyway, any comments gratefully received

Ian

 

Thanks David!

That is exactly what I am doing ... except the part about 'sketch' .... sadly C is still on my "round-to-it list" .... While grlb would have made development easier, the complication of controlling grlb from a second microcontroller would have negated any time savings ....

So, leaving grbl out of the loop, I have everything working on an ATTINY2313A (have plenty of them on hand). As expected, it took some time & fiddling to get acceleration / deceleration worked out ... (there is still room for improvement) ... but the time to benefit ratio is getting steep. Mostly what I tap is Aluminum, generally 5/16-18 or smaller ... and the ATTINY2313A code is adequate for most of those tasks ... not as fast as the same mechanical & driver set-up using a "real" controller with a tweaked script run on a PC, but in most cases faster than hand tapping. With the ATTINY2313A all that is requisite is a 12V wall-wort, the Stepper Power Supply and the stepper driver ... I replaced the "speed knob" & "travel knobs" with three pairs of momentary switches (the 2313 doesn't have an ADC...) and they are easy enough to use.

Perhaps after some real-world use (and//or I find a "round-to-it" for learning C), I will revisit the controller and a more robust target.

Thanks for your help!

Fish

 

I was able to connect the black box to windows 10 about 12 hours ago. Now I cannot get it connected any more. Please what shall I do?. USB not recognized

 

@Peter thanks for your reply. I have disconnect and reconnect severally. I can no longer see the ftdi driver on the usb port menu. Please what can I do next?

 

Hello, as new as new can be. I will apologize ahead of time if I am not supposed to ask control s-ware questions here. I couldn't find another place to ask. I just purchased a "toy" CNC router to teach an old dog new tricks. It is just a cheap little router, I figured if I destroy something, I better start out small and inexpensive. I have programs that I downloaded from the site I purchased the little router from. It has 3 axes. I have downloaded gbrl control s-ware and my machine works. I have loaded g code program and my router performs fine. My question is: is there a way I can open the project and change some of the code, to see how this works and learn? I didn't buy this to just download other people' programs and not learn anything. Thank you ahead of time.

 

Hello!
Pardon if this isn't in the write place, but I was looking for guidance as to if Grbl would be the right solution for a particular project I am undertaking (Senior Mechanical Engineer in College). I am essentially making an 3-axis (XYZ) machine that needs to drive screws into pre-made holes to a particular depth. Due to very tight depth tolerances (0.005"), I may have to us a laser to tell the machine when to stop driving the screw and move onto the next. Would Grbl support such a sensor integration?
I am a bit out of my usual area of expertise, so I greatly appreciate any sort of guidance possible.

 

I'm new. I assume, to ask a GRBL-related, I'm supposed to just reply to this thread.

My question: I have a laser-engraver type machine, Chinese, ~55 cm X axis, aluminium extrusions, 2mm-pitch T belt, Nema 17H3430 motors, one mechanical homing switch for X, Arduino-based.The X-axis accuracy is not up to par. When I make a mark on the X axis and jog to that mark (after homing), depending on the day, the position varies within ~4mm or so from G00 X-418 to G00 X-422. I checked the belt tension, it is seems OK, not too tight, not too lose. Is there something else I should check? Another idea I have: put a homing switch on the other end of the X-axis, then home both ends of the X axis, measuring the difference in machine position (assuming I can measure it), then modify the steps/mm value for each session to force to command that puts me on the mark to be constant (e.g. G00 X-420). Any other ideas? Thanks.

 

If the results are inconsistent then it will not be corrected by adjusting the grbl setting for steps/mm. Can you tell us what stepper motors (I'm guessing nema 17 - but give us any numbers printed on them) and what drivers you have (again guessing - if they have any numbers printed on them it says A4988 or something similar). You could also check that various machine movements are smooth - DON'T TURN THE STEPPER MOTORS BY HAND - disconnect the drive pulleys from the motor shafts before checking that each axis moves smoothly. Also check that the drive pulleys are firmly attached to the motors - a fixing screw in the pulley should grip the flat on the motor shaft.
Let us know if you find anything.
Alex.
Sorry - just read your post again - you told us what motors you have - what drivers are you using?
Alex.
PS I suspect this will turn out not to be a grbl question - if so I will move it to another thread.

 

It's possible something is loose or binding. Carriage, motor coupler, motor mount, frame. When checking axis movement binding, do the whole length. If the the carriage always moves less than it should, you could be losing steps - turning up your motor current could be the solution if there is no binding in the X movement.

 

Thanks for the reply. Yes, the motor model is mentioned above (Nema 17H3430), the (2) drivers are TMC2130. Y axis is mirrored (2 motors, one driver, twice the current of X axis). Microstepping is 16. Bought for their quiet operation. Indeed, the motors are 100% silent and move smoothly (at least, to my eyes). Someone mentioned that the issue could be missed steps. Hm, now I vaguely recall reading somewhere that TMC2130 by default are in some "StealthChop" mode, which makes the motors weaker and MAY lead to missed steps. I'll look into this. Hard to tell (yet) what problem this is, it could be mechanical (belt), electronic (steppers) or software (GRBL).

 

I agree with alex, it's very unlikely to be GRBL unless the differences are highly repeatable.

I like those drivers - amazingly quiet. I've used them in a couple of projects (not cnc, though). Try switching to 8:1 microstepping to see if the problem goes away. I tried 16:1 in a star tracker I built and didn't like the result, switch to 8:1 and it was excellent.

 

Sounds more like a mechanical problem than an electronic one.

Make a little g-code program (macro?) that is just G0 X+200, G0 X-200. Center the axis, make sure you are set to relative co-ordinates, and run the program several times in a row. It should end up back exactly where it was. If it does not, there is definitely a repeatability issue.

Check all belt anchors, belt should be quite tight, but don't bend the motor shafts. Make sure there is no play in the axis. Grab the carriage and jiggle it back and forth along the direction of the axis. If it moves AT ALL, or feels loose, there is a loose component somewhere. Make sure motor mounts are tight, pulley screws are tight and on the shaft flat, belt holder on the carriage is tight, etc.

If you eliminate all the mechanical possibilities, it is likely too low a motor current or microstepping is set too high.

MG

 

Hi I just got the leed 1010 and we have made a file with solid works but are having trouble importing it into the open build controller. We have saved it multiple way and we still can’t get it to work. It keeps giving us errors or unsupported code.


I know we need to use a post processer for solid works to make it into a grbl file but have not been able to make it work.


we also can’t seem to import 3D files into the open builds cam program. It only wants to accept 2D files or pictures.


any suggestions?

 

So We down loaded Fusion 360 and used it to make our build and then ran our file through the post processor you suggested and it still wont fully work? The simulation works fine but when I hit run job it just does small circles on the board... the code seems right so I'm not sure why we are having so many issues.

If I attach a file would it be easier to see any flaws?

 

Attach a file.

 

Hi @Chris Preece, in Fusion you can export a Fusion archive file. Zip it, together with your nc file from Fusion and post it here.
Alex.

 

HI All the GBRL masters.

Im looking for the best software that i can control my gbrl with for very long toolpaths for my rotary setup.

As this are now then i run it through UGS it reaches a freezing point in the software and just stops, then i need to check the line and edit the gcode which is a hassel.

Does anyone have advice PLEASE.

 

Better to get your Post processing in order to avoid generating the offending op codes.

With grbl panel you can hit continue to get past an unsupported GCode but that's not the preferred option.

 

hiiii sir my project is to control 3axis cnc milling machine using arduino with matlab ,build program grbl that control the machine and sending file g code to ardunino uno can someone help me in that i need the code in matlab ..thanks alot

 

Hi,

I'm trying to modify my simple CNC to drive it over GRBL
(LPT pins connected to GRBL)

The CNC is working by directly driving LPT pins under Linux CNC

Is there any online calculator how to calculate settings from "Stepconf" to GRBL ?

 

See here:

GRBL Settings and Calibration – 3D Tech Works

Scroll down to the Movement Calibration setting. This explains how to calculate the proper values.

MG