Ground Control calibration problem

Okay, I started a long write on my phone but it turned into some rather incorehent ramble in the end. So I’ll try o summarize my findings during calibration this morning.

@bar - if you can use any of these suggestions it’s yours :slight_smile: (thank you for a great machine!)
okay, here goes:

  • First thing - during assembly it might not be a good idea at all to mount the chains. during calibration they will go on and off a few times and the nails that keep the idle end of the chain to the frame has a head that’s too big to allow easy removal. I ended up taking the nail out and feeding it through the chain and then hammering it back in. a good number of times. I’ve not changed to a smaller nail :wink:

  • First time you fire up Ground Control (GC) it doesn’t know the serial port. It can be entered two places, either by the settings menu (manually by typing in COMx) or by using the drop down box in the actions menu. However, if you select it through the actions menu (the easy way) it doesn’t automatically connect. I missed that and started calibrating. When asked to turn the sprockets to the o’12 position of course I couldn’t. pressing quit at this point doesn’t work. the ‘back, skip and quit’ buttons just filled up the screen. I ended up starting over.

  • The main problem - the sprockets going the wrong way… I couldn’t make it do that (of course!) on the other hand, reading this thread yesterday made me realise that I had been doing it wrong. It says in the text how to do it, but I managed to screw it up - I had taken the left-hand side of the chain and fed it over to the right sprocket. I don’t know if a ‘please read this carefully’-text would help on this step.

  • I’m using a tablet for controlling the machine so I guess I have it easy. When the chain is on the right sprocket and you’re tightening the chain, it came undone for me. The next time I was holding a finger onto the rightmost link and that worked. it feels like a two-person job, that step.
    Suggestion: Make a field for manually entering this number if it continues to cause trouble. after all it’s in the manual config page.

I was wrong earlier on this having anything to do with top or bottom chain feed - I’m sorry. the chain is not attached to anything at this point in the calibration so that can’t be it.
Since I managed to get this error during the initial calibration I suspected that maybe the direction flag for GRBL is not set initially so I redid the calibration where I started by wiping the EEPROM. Unfortunately this didn’t cause any problems and the sprocket was still behaving.

Lastly: Z-axis calibration

  • My setup requires a negative z-pitch. I forgot that initially so if it’s possible adding a ‘Flip Z direction’ on the calibration page would be most helpful! all it should do would be to negate whatever number is in the z-pitch. A real time-saver would be the option to enter the pitch directly (since i’m using a M8 rod and not a leadscrew).

TL;DR version
I couldn’t make it fail even though I tried a few times. I guessed that it could be a GRBL issue but I didn’t manage to provoke the error. However I have some suggestions for improvement/changes (see above)

3 Likes

I am having the same problem, hooking the left side of the chain to the left motor, when i hit “pull chain tight and measure” it rotates CCW and the chain falls off. I’m reading this thread but I dont see a fix to the problem.

@rubiks, is this your first attempt at calibration?

yes it is.

This seems to happen more often than it should. One of my suspicions is that it happens because there’s some bad settings somewhere and the controller is confused about what condition its in. @bar had recommended to @sig43 to do an EEPROM wipe. Have you tried that? It might solve the problem. Let us know one way or other.

Edit: @bar suggested resetting to default, not wipe eeprom… sorry

if you skip that screen go to the next and tell it bottom feed then go back and set chain is what finally worked for me

@madgrizzle ive done the reset and designated the port but had the same issue.

@Sig43 i’ll give that a try and report back. Thanks

That’s interesting… one of my other suspicions, is that the controller thinks its in bottom feed configuration while doing that step because that’s the only way I can see it turning the wrong way. However, when you enter that step, the code is supposed to set the controller to top feed temporarily and then set it back to whatever it was when you exit the step.

@madgrizzle @Sig43 ive done the reset again and designated the ports, zero’d my sprockets to 12 o’clock, skipped the next step (chain measurement), selected bottom feed, went back to chain measurement, click “pull chain tight and measure” , still rotating CCW chain falls off.

You need to extend the chain first until you get enough to hook it over the right sprocket as shown in the pictures.

Step 1. Hook chain on left sprocket as shown in bottom left picture.

Step 2. Press Extend Chain (it will turn CW to give slack) until you have enough to hook the end over the right sprocket as shown. - Here, I suggest two things. 1. As soon as you can get a hold of the chain, grab it with your hand and pull it down and to the right to apply tension on it. If you don’t, its very possible (or maybe even likely) it will get wrapped around the sprocket. 2. After a few extends of 100 mm, change the value to something like 1000 mm so you don’t constantly have to click it. Keep track in your head how much chain you have feed out. You should shoot for around 3020 or so. If you try to feed out too much, you will run out of chain and bad things will happen if the other end of the chain is fixed (like mine is)

Step 3. Press “Pull chain tight and measure”

1 Like

i think i get it now. im using the same side of the chain in step one and hooking it to the right motor in step 2.
i was attaching the other end of the chain for step 2.

thank you.

1 Like

Ah, I see what you mean. Nope… don’t do it that way (the way you were doing it). This helps so we can update the instructions to make it clearer.\

I think we need to add an animated gif on this screen showing the proper technique. so many people get confused.

4 Likes

An animated gif sounds really good!
I was doing it the same way as @rubiks did. of course it’s in the text but one seem to skimp on reading when there’s so many pretty pictures!

1 Like

Animated gif is a great idea! I’m not sure how to make one but I’ll look into it :grin:

1 Like

Just an FYI that I figured out on attempt number 5. Turn the first 1/2 link of chain that is not on the sproket up 90 degrees. As the motor turns it will fall with gavity off of the sproket and not wrap the chain.

2 Likes

I do that as well, but it’s no guarantee… I’ve had it still wrap when I wasn’t pulling on the chain (my chain has been abused a bit and may be more susceptible to wrapping than others).

Alright, I’m so very confused here.

So I’m doing all of the things according to the pictures.
I’ve done all of the things suggested here, including Resetting Ground Control via advanced settings. Re-zeroing the spockets, skipping the step and making sure it says bottom feed.
I’ve reset the the chain onto the sprockets as shown in the image. I start with the chain on the left sprocket, exactly matching the photo. Same with the chain on the right sprocket.
Am I suppose to be using the same chain on both sprockets (this way is how the instructions read)? or both chains while keeping it attached to the sled?
I’m using the bolt together frame. This is the first time running calibration.

Now I’m retrying, following Madgrizzle’s latest post here with a clear Step 1. and Step 2.
What I don’t understand is how Extend chain will pull it tight in the imaged set up, it just keeps rotating CCW and falling off. If either sprocket rotates CCW the chain will fall off.

With this setup, won’t rotating CW remove slack? Not give more? I’m so confused.
There is plenty of slack to get the chain to reach to the other sprocket.

Other observations:
Any action on the calibration step only moves the left motor.

OMG!!! I FIGURED IT OUT!!!
So, the machine is doing everything right. It is not entirely clear that the end of the chain in step one (left image) will “extend” all the way to the right side to become the SAME end of chain attached in step two (right image) I mean, now it makes sense and is totally clear, but it was not before for whatever reason.

Thank you all!

3 Likes

This has been a problematic step, hard to convey and seems to be easily misunderstood. Very glad you discovered the ‘trick’!

1 Like

Lol… been there, done that.

1 Like

While you are in that valuable moment of seeing the answer but also how it was confusing, what can we word differently in that step to make it more clear?

1 Like