Mr Beavers HDZ and CM, hmmmm


(Griff Carpenter) #1

I seem to recall reading that we can’t/shouldn’t use CM with the HDZ? Anyway, I’ve only used 3rd party gcode senders because of that perception.

Until today.

I’ve been chasing a weird issue these past few days. Without thinking I decided to try CM 412. Using my home built HDZ, no changes to settings, I successfully probed and set zeros. Then I remembered, hmmm, that shouldn’t work so I did an air cut of the work that had been troubling me in CNCjs. Worked fine, no problems. OK, now we are making some chips, will report results.


(Dustin S Tilton) #2

Are you using a thicker piece of stock that is.not requiring the full travel? Or maybe they removed the hard coded Z travel.


(Griff Carpenter) #3

You’re probably right. I don’t remember the particulars about the restriction. Guess I’ll find out when I make the final profile cutout. Stock is 1.75 thick soft maple.


(Neil Ferreri) #4

What was the problem?


(Griff Carpenter) #5

@neilferreri, honestly, probably not an issue with CNCjs. Most likely some sort of hardware glitch.

I’ve been commissioned to produce some oval trays from soft maple. Stock is 32” by 16” by 1 3/4” thick.

Using VCarve pro I’ve designed a couple of trays, pocketed at 1.35”, sidewalls at 1.75, .5 thick.

0.25 2 flute carbide, 0.5” doc, 0.1 step over, 150 ipm - numbers I’ve used with red oak and ash. Never cut soft maple ‘til now.

What happens, initial cut completes, 2nd cut proceeds to 50% or so then machine “stalls” on the X axis and “walks” at a right angle on Y, either + or - Hit the e-stop, re-zero, adjust cut file to start .5” deeper, start again. Same thing happens again, near end of second cut. Fortunately, nothing ruined yet since this is just a giant pocket :smile:.

Re-set VCarve file and ran remainder with Carbide Motion 412, finished fine, no glitches.

THEN, set up next cut file, asked CM to home…and… Z and Y did their thing while X vacillated between east and west, not getting anywhere.

I eventually set hard limits to 1, got the machine to home. Turned hard limits off, set up a new tray job in CM, cut fine.

What?

Edit: link, video when homing,

https://share.icloud.com/photos/0ozIKFvUibFFhYMoTbN8-SiHw#Home


(Luke) #6

I think I need to run some tests…


(Luke) #7

@WillAdams - any chance you can tell us if the hard limit has been removed?


(Neil Ferreri) #8

@Griff Any chance you have a loose connection on the X stepper? If you rapid back and forth along the extents of X, do you see any issues? Without seeing it, it sounds a lot like an intermittent connection on a stepper. Could be a broken wire. I’ve seen behavior like that and tracked to wiring. That said, messing with hard limits doesn’t fix wiring.
Keep us posted.


(William Adams) #9

I can’t imagine that it would have been — it’s not a part of the code which would have been touched recently and none of the reasons for it (keeping folks from running their carriage off the bottom of the Z rails, simplicity / reliability) have changed.


(Griff Carpenter) #10

I’m guessing that’s probably it. Another symptom I forgot to mention is changes in pitch in the sound coming from the X stepper when doing long traverses,

Was planning a redo of my drag chain after Christmas. Sigh :pensive: I guess I’ll have to move that task forward.

Thanks all for the responses


(Griff Carpenter) #11

Did some air cutting today. Haven’t seen a failure yet but I did find the molex connector for the X extension to be extremely hot, clearly some resistance there. I will cut the molex out and solder the connection. See what happens.

I hate intermittent electrical issues!

Yep, definitely some heating going on.


(Griff Carpenter) #12

I’ve run 3.5 hours of air cuts since I cut the molex out, no problems. Granted, low load but given that I noted the molex heat following aircuts I’m suspecting the problem is fixed.

I also observed that, pre-fix, power off manual movement in X was very herky jerky. Now, movement is quite smooth, power off. Something to do with the steppers acting as generators when moved in power off state? And the impaired molex connection?

Now, back to the original topic, determining the limits of using CM with the HDZ.