[B-RAD]

I'm to the point where I need to adjust my gibs better, and it being CNC I really am not sure how to go about that. I have tore off the motors and did the adjusting by feel the first time, but I don't think I made them tight enough.

So I have some questions. (preferably the newer Taig design)

1. What size steppers do you have and what is your acceleration set to? (assuming your gibs are tight)

2. How do you adjust your gibs correctly on a CNC machine?

[bjones]

Seriously, they banned B-RAD? He was a breath of fresh air in here. I saw on another forum they banned cameraguy again, so we won't be seeing him for another year I guess. Apparently from what he said, it was for calling out a mod on lying (which the mod clearly was, but thats still a stupid thing to do, especially on here). Cameraguy may not care much about his account, but I do, which is why I am not using my name. I'm not stupid.
The newbs getting banned is pissing me off though. This and those other guys in benchtop last month at least. Come to think of it, that last festival of nuking I saw was on a thread where people pushed linuxCNC over Mach, so maybe that mod employed by Artsoft thing has something to it. Where there is smoke and all. Keep your trap shut about anything not an advertiser here I guess.

why did you get banned b-rad?

[BAMCNC.COM]

#1 Set bearing pre-loads and Lead Screw half-nuts so there is no backlash
#2 set acceleration to 20, call that a constant for small work
#3 set velocity to 80, testing only
#4 tighten gibs so there is no play when you wiggle the bed perpendicular to the axis your setting, try to move the bed so you have the most leverage when doing this, I usually have the axises at their extremes for this.
#5 Set up a dial indicator and set it to zero then zero your DRO and place it against something solid on the bed.
#6 now jog the axis back and forth 5 times the full length of the bed, then 10 times quickly back and forth, leaving you somewhere in the middle of the table.
#7 send your axis to zero..... Does your dial indicator still say zero?
#8 lower velocity 10 and acceleration 1 and repeat, that is if you lost steps...

[Fastest1]

I dont have a Taig but the techinique should be similar. I lube all the axis, then loosen the gib adjustment screws. I use the hotkeys to jog the particular axis back and forth. The gib slides in further only going 1 direction. I always tighten the gib after traveling that direction. Then I try full travel. If it moves smoothly, I tighten a little more. If it stalls I loosen it. I keep doing this until all axis are snug and stall free.

[BAMCNC.COM]

That's a good point, which leads me to a question...or comment..

You say "stalls", stalls are dramatic and a loss of many steps, what if you lose 1 out of 8000, how would you know? Even 50 out of 8000, this is just an example.

My way takes some of those steps into consideration, by seeing if your axis is "dead on" after playing around for awhile.
If just to the point of stalling, and back off, I'm afraid your slooooowly missing steps..

[Fastest1]

I am glad you made me rethink my technique. It works well enough for most set ups I had done. Sure enough there were missed steps.

I started by using the indicator (.0005) to detect deflection while tightening the gibs and applying pressure to the table from the side. This led to a discovery of .001 of play somewhere in the carriage/ball nut mount or thrust bearing preload of the Y axis. X had no such loss in that regards but was missing steps.

I am to the point of getting repeatable results on X at 120ipm with an accel of 20. Anything more resulted in a loss. I tuned the X down to a max of 80/20 and couldnt make it lose a step. I am happy with that. The Y is in the same realm of speed minus that pesky .001 of movement (which will be resolved later today as I am pretty sure it is the bearing preload as I can push the table straight backward and pull it forward and see the movement on the indicator).

I havent gotten to the Z yet but I expect no difference in the process.

I appreciate the correction and explanation of another and more accurate way. I had read it but chose to be lazy. I do believe I am running the gibs a bit looser.

That's a good point, which leads me to a question...or comment..

You say "stalls", stalls are dramatic and a loss of many steps, what if you lose 1 out of 8000, how would you know? Even 50 out of 8000, this is just an example.

My way takes some of those steps into consideration, by seeing if your axis is "dead on" after playing around for awhile.
If just to the point of stalling, and back off, I'm afraid your slooooowly missing steps..

[Fastest1]

After a little tinkering today and adjusting the Y, I was able to attain a 331 ipm with an accel of 40. I repeated 10 cycles of most of my Y's travel with a warm up wizard. I didnt miss a step repeatedly. I settled in the 80 ipm for the moment for X & Y accel of 20. That is plenty fast especially if I am only cutting at 8-14 ipm.

[BAMCNC.COM]

Excellent info! I'll have to try your speeds.

[Fastest1]

Here is the video!
www.youtube.com/embed/8CPqGaUcyMc