SOLVED: how to make gg flex f#@%&! connect to ddcut and home finally

ill start this by saying that i am in no way affiliated with the devs of this project, just someone trying to make sense of it all and wanting this machine to work.This is how i got the gg flex to "home", after burning up numerious cnc shield v3.0 and arduino boards i finally got the damn thing to work! Attached is a pdf with install setup instructions along with images on how i have my setup working, i hope this helps guys. also the settings that need to be changed if they dont match the settings in your current config to match what i have. these instructions assume you have already flashed your "arduino uno" with "gg firmware" and  are using "a4988 stepper motor drivers" and "bi-polar stepper motors" with a "12v power supply" (your typical hardware setup). this is how i have mine working and i am not responsible for any damage you may do to your equipment if you follow my instructions and blah blah blah,user beware. if anyone sees an issue in the way i have set it up, please let me know as this too is my first time with grbl or for cnc machines in general for that matter like the majority of people in this forum, tho i am very familiar with 3d printing and printers in general. you can enter the settings that im using in the "grbl controller" or "universal gcode sender" or any other software that interfaces with you machine via serial port, in ugs (universal gcode sender) type: "$$" in the command window and it will give you your current config setup, based on your setup (ball screw legnth, ball screw size etc adjust as needed). The most important 2 setting i have noticed which makes it fail to connect to ddcut/home is if your config value $10=115 it will fail to connect to ddcut and needs to be changed to "$10=127" if its not already set to 127 and the second is what makes it fail to home the config $27=0.7, which makes it fail as the value is too small, so change it to a slightly higher value of "2.25" or any value above the one currently set that works for you, i was satisfied with 2.25. so if you followed the way i have my stepper motors wired and used the config i am using then after that the shit should home without issue. so im leaving this here for you folks, best of luck and hope this at the very least give you some insight on how wire this up,peace.

link to download the pdf as i couldnt upload it via the images button:
http://89.187.173.184/wiring_up_gg_flex_to_home.pdf

link to video of gg flex homing with above setup in ddcut:
http://89.187.173.184/homing_in_ddcut.mp4

config settings needed to be edited/checked to make sure they match youre config settings using either grbl controller or ugs (universal gcode sender)
==========================================================
$0 = 10    (step pulse, usec)
$1 = 25    (step idle delay, msec)
$2 = 0    (step port invert mask:00000000)
$3 = 0    (dir port invert mask:00000000)
$4 = 0    (step enable invert, bool)
$5 = 0    (limit pins invert, bool)
$6 = 0    (probe pin invert, bool)
$10 = 127    (status report mask:01110011)
$20 = 1    (soft limits, bool)
$21 = 1    (hard limits, bool)
$22 = 1    (homing cycle, bool)
$23 = 7    (homing dir invert mask:00000111)
$26 = 5    (homing debounce, msec)
$27 = 2.250    (homing pull-off, mm)

if anyone is able to get their flex homing with the above setup or has questions in regards to setup , please let me know and ill do what i can to assist you, peace

You are using 4 stepper drivers and jumped both of the 'X' pin sets on the CNC Shield board. If you use a splitter to join the 2 X steppers and only use 3 stepper drivers, I'm assuming that the 'X' pins will NOT need to be jumped, correct.

TIA

patkelley wrote:

You are using 4 stepper drivers and jumped both of the 'X' pin sets on the CNC Shield board.  If you use a splitter to join the 2 X steppers and only use 3 stepper drivers, I'm assuming that the 'X' pins will NOT need to be jumped, correct.

TIA

that should be correct, if ur using the same stepper motor wire thats split like the dual z axis wires are with 3d printing then u should not have to clone/ jump the pins as you stated.Let me know if your able to get your flex homing to confirm if above method works with the 3 stepper motor drivers. Lets get you "home" lol.

Cool write up. Good job in figuring it out.

s0urman wrote:

ill start this by saying that i am in no way affiliated with the devs of this project, just someone trying to make sense of it all and wanting this machine to work...

...you can enter the settings that im using in the "grbl controller" or "universal gcode sender" or any other software that interfaces with you machine via serial port, in ugs (universal gcode sender) type: "$$" in the command window and it will give you your current config setup, based on your setup (ball screw legnth, ball screw size etc adjust as needed). The most important 2 setting i have noticed which makes it fail to connect to ddcut/home is if your config value $10=115 it will fail to connect to ddcut and needs to be changed to "$10=127" if its not already set to 127 and the second is what makes it fail to home the config $27=0.7, which makes it fail as the value is too small, so change it to a slightly higher value of "2.25" or any value above the one currently set that works for you, i was satisfied with 2.25...

...
$0 = 10    (step pulse, usec)
$1 = 25    (step idle delay, msec)
$2 = 0    (step port invert mask:00000000)
$3 = 0    (dir port invert mask:00000000)
$4 = 0    (step enable invert, bool)
$5 = 0    (limit pins invert, bool)
$6 = 0    (probe pin invert, bool)
$10 = 127    (status report mask:01110011)
$20 = 1    (soft limits, bool)
$21 = 1    (hard limits, bool)
$22 = 1    (homing cycle, bool)
$23 = 7    (homing dir invert mask:00000111)
$26 = 5    (homing debounce, msec)
$27 = 2.250    (homing pull-off, mm)

Looking awesome, thanks for the info and documentation!

$27=0.7 is specific to the Mk3 type endstops used with the GG2 and that I've been providing on all the most recent iterations of FLEX kits, and is the default controller setting in both cases.
Other types of mechanical endstops will have longer throws or trigger arms or users might decide to use inductive or optical sensors as their endstops... in either case, this debounce value needs to get changed to match the particular endstops.
I'd usually recommend keeping the value as tight as possible, since it also performs an inherent backlash check when homing, but it can be increased if the stops are causing problems.

$10=127 should be the default for all the DD2v7 firmware packages I've seen/installed. A 115 value would lack RX and planning values which DDCut needs to connect.
Also FYI, resetting/updating firmware through DDCut will also reset all factory values in case you've modified extensively or are suspicious of the firmware installed on a particular controller.

OK, EEM-LLC had me install my limit switches wrong. I will change to what you have in your PDF.

I had: on SpnEnX+Y+ pins: Red off, Black -, Green +

You have: on SpnEnX-Y- pins: RED -, Black +, Green off.

EEM sent me 3D printed limit switch mounts. So I hope mine are in the same places as yours?

(I also have probe wires on Z+ pair)

-----------------------------------------------------------

You stated on the PDF something about SWITCHING the TWO inside wires on the stepper plug??? The motors EEM sent with my machine don't have plugs at that end as the wires go right into the motors...

Also EEM had me change the Z stepper plug at the shield to be upside down while the others are right side up. Stating that Z axis needed to run the opposite direction for DDcut as the others?

Also my stepper wires are; Blue, Red, Green, Black - but yours are Black Green Blue Red. So I can not copy your plug orientation?

Your photos show black wire at the TOP of the CNC Shield so I will match at least that wire.

-----------------------------------------------------------

Lastly besides the fact I can not match your exact stepper wires... You listed new code, is that to be edited into the DDcuts file;  .\Cutting Code\AR15 Code\AR15_pocket+selector+pins.dd file somewhere?

You were rather vague on that part... What or where do I edit this code?

-----------------------------------------------------------

So basically I changed my limit switch wires to your way and ran DDcut, pressed start and again the Z axis moves reward and hits end stop and DDcut reads:

"Machine Locked Due to Error"

I've never gotten farther than this...

Maybe this has been my issue from the beginning? Because my Z could not hit top of switch lever so I 3D printed a tab that clicks the switch arm which does not back off far enough to unclick when homing... How or where do I edit this?
$27 = 2.250 (homing pull-off, mm)

thanks @bob_da_guy, shit was buggin the hell outta me lol, @EEM-LLC ok cool good to know on the $10=127 value, then i guess maybe the stock grbl firmware has the 115 value then (must have seen it when i was confirming my theory on the $27 value with arduino/shield combos i knew that were not working before the change in the $27 value), @Pegasus  i would say as long as your limit switches are in the same general location you should be ok, and if your stepper motors have the wires coming out directly from the motor then dont worry about switching them like i posted, but i do have a 3d printer with the same wire colors so ill check to see the way they are wired up to the board and let you know, and where to enter the code ,my bad i think i erased that part when i was editing some shit i wrote,oops, yes the elusive code entering into the matrix lol, so you can enter it in either grbl control that is usually packaged with the gg files(i sourced all my materials , so i not sure it comes with the kit or not) or what i honestly prefer (maybe because it was the first one i tried and ultimately used to figure this out lol) is universal gcode sender (ugs) which will do the same thing as grbl controlleror, ill post some links to both. you will enter the commands as you would if you were issuing them in the terminal in octoprint, just look for the command window after connecting your arduino/shield to the respective program which is the little plug icon in ugs or the open/connect button in grbl control command window is lower right for ugs and lower middle center for grbl controller,or should be going from memory on thaat but you should see it right there, and simply type $27=2.25 and hit enter and the value will store for you. you can confirm it was stored by typing $$ in the same window in ugs but that didnt work for me in grbl control so there is a tab that will bringup your settings at the top but not terribly familar with grbl control. or an alternative to that is once you change the $27 value you can issue the homing command in the command window by entering $h just be ready to stop machine should it not home properly by aborting, soft reset, pressing the disconnect button, un plugging the usb cord from the shield etc. shouldnt damage anything just make clicking sounds from the motor as it might go in the wrong direction but nothing "bad" should happen, just be ready to do one of the above, and if your motors do go in the wrong direction  and you have stepper motor wires all facing in the same direction on shield, unplug the usb and disconnect the power to the board and simply reverse the orientation of  the plug by turning it 180 degrees and plug back in and try the above mentioned again after re connecting to the either program.hope that helps, im not the greatest at explaining shit, but i do what i can do when i can do it. lmk how it goes bro , ill be around, good luck!
=======================================
link for universal gcode sender:
https://winder.github.io/ugs_website/download/

link for grbl controller:
https://github.com/Denvi/grblControl/releases/download/v0.8/grblControl_0.8.zip

Pegasus wrote:

Maybe this has been my issue from the beginning?  Because my Z could not hit top of switch lever so I 3D printed a tab that clicks the switch arm which does not back off far enough to unclick when homing... How or where do I edit this?
$27 = 2.250    (homing pull-off, mm)

im willing to bet that it is, and if it is, problem solved, and now on to the next episode, z probing and then cutting some shit!

Ok, I have GRBL or should I call it Garbage. I have opening it, connected to Flex and ran $ code to unlock the GG firmware. But could not get it to home correctly.

Do I just cut and paste your entire command block into it all at once? Or 1 line at a time?

So I assume this changes the Flex firmware written to it from DDcut firmware restore process?

@Pegasus i checked my stepper motors on 3d printer but they are yellow,blue,red,black, but in any event, the most improtant part is that they all face the same orientation and they all move in any direction, doesnt matter as long ass they move and dont just make that clicking noise shit we all love to hear, and if they go the wrong way rotate them 180 like i mentioned above and you should be gravy bro

do 1 at a time and just the values, dont add the comments next to each one ($27=2.25) like that

i tried to add them all too, but was getting some cache error or some shit so i entered them 1 by one

Pegasus wrote:

So I assume this changes the Flex firmware written to it from DDcut firmware restore process?

yea i believe thats what its doing is overwriting the current config settings currently with these

Ok, do I enter just the part before the ()

Like: $0 = 10

Or also: (step pulse, usec)

For each of the below:

$0 = 10 (step pulse, usec)
$1 = 25 (step idle delay, msec)
$2 = 0 (step port invert mask:00000000)
$3 = 0 (dir port invert mask:00000000)
$4 = 0 (step enable invert, bool)
$5 = 0 (limit pins invert, bool)
$6 = 0 (probe pin invert, bool)
$10 = 127 (status report mask:01110011)
$20 = 1 (soft limits, bool)
$21 = 1 (hard limits, bool)
$22 = 1 (homing cycle, bool)
$23 = 7 (homing dir invert mask:00000111)
$26 = 5 (homing debounce, msec)
$27 = 2.250 (homing pull-off, mm)

Pegasus wrote:

Ok, do I enter just the part before the ()

Like: $0 = 10

Or also:     (step pulse, usec)

For each of the below:

$0 = 10    (step pulse, usec)
$1 = 25    (step idle delay, msec)
$2 = 0    (step port invert mask:00000000)
$3 = 0    (dir port invert mask:00000000)
$4 = 0    (step enable invert, bool)
$5 = 0    (limit pins invert, bool)
$6 = 0    (probe pin invert, bool)
$10 = 127    (status report mask:01110011)
$20 = 1    (soft limits, bool)
$21 = 1    (hard limits, bool)
$22 = 1    (homing cycle, bool)
$23 = 7    (homing dir invert mask:00000111)
$26 = 5    (homing debounce, msec)
$27 = 2.250    (homing pull-off, mm)

you read my mind i just edited my comment, but no just the $key=$value, not the comments

any luck @Pegasus?

Well, it seems your code changes have made my problem WORSE...

Before, as I said in my other thread. Pressing start in DDcut on any cutting code would HOME Z axis rearward click the limit switch and stop, with error: "Machine locked due to error". At least I had movement.

I went ahead and unlocked the shield in GRBL using $$ then $X, most of my code was the same as yours other then 2 or 3 so I changed them as you have them. Then read it again with $$ and new code showed as you can see in my attached image.

I then unplugged USB, closed GRBL, Opened DDcut, repluged in USB and DDcut connected fine. Opened cutting code and clicked start, NOTHING moved, NO sounds, then just the message: "Machine locked due to error"

This Flex is clearly not going to function!

At this point I have ruled out the following;
-the controlling computer with 2 laptops and 1 desktop.
-the USB cable with 3.
-the power source confirming 12v.
-the arduino with 2.
-the shield with 3.
-the drivers with 4 sets. confirming 0.6v.
-the stepper motors by switching connectors.
-the limit switches by removing them and using a wire to jump the pins while homing.
-Now changing the firmware in GRBL to your settings

At this point there is NOTHING else I can imagine to swap or try, other than the STEPPER motors???

Pegasus wrote:

You stated on the PDF something about SWITCHING the TWO inside wires on the stepper plug??? The motors EEM sent with my machine don't have plugs at that end as the wires go right into the motors...
...
Also my stepper wires are; Blue, Red, Green, Black - but yours are Black Green Blue Red. So I can not copy your plug orientation?

s0urman's kit was not built up with electronics provided by me; they were sourced elsewhere and so his configuration has several variations from the standard electronics setup as a result.

In case it wasn't evident from inspecting your 3DP, different manufacturers use different color codes for their wiring. On top of that, there are different coil pairings/pin positions on different connectors and motor manufacturers. s0urman sourced motors that had different pin configurations from the CNC Shield's output and had to swap the two middle pins to turn A+/B+/A-/B- into A+/A-/B+/B- to match the shield.

The electronics in my 'Complete Kits' are plug and play; no swaps are required.

Pegasus wrote:

OK, EEM-LLC had me install my limit switches wrong. I will change to what you have in your PDF.

I had: on SpnEnX+Y+ pins: Red off, Black -, Green +

You have: on SpnEnX-Y- pins: RED -, Black +, Green off.

Like the motor pairs, different endstop manufacturers use different wiring colors.
Black/Green was the correct pairing for the Makerbot type switches I supplied.

If you have doubts for either coil pair configurations or which wires form the switched pairs on endstops, the link I shot you about 40 emails back for the Uno/CNC shield assembly (on Zyltech's site) has info on using a potentiometer to test both (which is also a fairly standard test when putting together electronics of mixed origins). Since I specifically sourced parts to match the CNC Shield setup, this shouldn't be the case on any of my electronics kits... but it isn't completely impossible either.

Pegasus wrote:

At this point there is NOTHING else I can imagine to swap or try, other than the STEPPER motors???

My bet would still be the Uno having several isolated damaged pins. I thought that if both Uno's I sent you had the exact same issue then this was extremely unlikely... but we've ruled out almost everything else, and the only other similar issue I've had reported and the only issues documented with other similar electronics setups both point that way.

s0urman wrote:

Pegasus wrote:

So I assume this changes the Flex firmware written to it from DDcut firmware restore process?

yea i believe thats what its doing is overwriting the current config settings currently with these

Correct, updating firmware from within DDCut (or flashing a fresh copy through xloader) will (re)set working 'factory' settings to all the variables in the GRBL config.
Tested and confirmed to provide the $10=127 value, though it will also reset the $27 pull-off value so you'll have to go back in and change that value after any firmware updates if you're using another type of endstop that needs the larger value.

EEM-LLC, Why don't I mail back to you the 2 uno/shield/driver stacks for you to test. Get working and ship 1 back to me confirmed working?

@s0urman

Your files are no longer working. Can you put back up, even if just for another week?

@s0urman

Where exactly have you placed your limit switches. I got the 3dp mounts provided by EEM-LLC but I have no idea if I placed them correctly.
Thanks

Does Doe anyone have a copy of the PDF files that @s0uman posted? the link is not working and I would like to look them over.
thx