My wheel spacer-like project is a bit held up with my circles being a bit too out of round to be of the intended use and maybe a re-calibration might get it there.
So. . . Are there any official special instructions for RE-calibration? or do i just go NIKE on it and it will take?
I found an older video from a youtuber named Amp Mechanic and mentioned all sorts of G-code commands necessary to re-calibrate. Are these kind of extreme steps really necessary? Was he possibly just showing off a bit?
He also mentioned standing the calibration triangle off the base as potentially helping. That, i believe, I did see on one of Andrew’s videos. did the firmware update resolve this particular issue or is it still recommended? will there be a new calibration triangle offered for sale? (no, i do not want it for free- i got in early on this knowing there could be quirks and yet still seeing the amazing potential)
@launch_pad_mcduck200 Recalibration is no different than the initial calibration. Spacing the triangle out 1” is helpful. It gets the arm movements a bit further out from the base and more into the sweet spot of mechanical motion of the arm. Of importance is that the triangle is anchored and the Droid base is level. You can put quite a bit of downward pressure upon the stylus (or torch tip). I look for the arm to flex up a little. This helps much with maintaining contact with the sled as it trundles along the guides.
Previous to my first post I did an initial calibration and it worked successfully, i just have not been getting round ENOUGH circle cuts. I have a youtube video of the machine performing beautifully if you account for my too short dwell time on the initial piecing which was immediately corrected for the following cut. Thusfar, my cuts have been 100% repeatable making stacks of identical parts- just not true to round and therefore stacks of identical waste. lol.
I guess as to RE-calibration so of the unknowns i was hopping to find out: if the machine would just over-write its old file, if the power re-cycle orders that Amp Mechanic mentioned were important and if Gcode had to be manually entered to get the machine "cleared’ of the old calibration.
in digging thru the manual- it says to use a NEW blank thumb drive for calibration (Sans the RE). So I am going to proceed as if the machine does NOT overwrite an old calibration. I guess in this manner if i make the next calibration worse i can then revert? maybe? These things are kind of the MEAT i wish got included in the manual at some point.
i will definitely also stand off the calibration triangle from the base of the machine and may make a permanent standoff for future use.
i will also level the machine to the calibration triangle to much higher tolerances than a carpenters level. . . . but at the same time i am curious if now that the machine has a built-in feature of torch leveling how that might affect things for those working on the fly. i do not think this will be an avenue i will explore because i have 70 lifetimes of projects to make parts for and plenty of other distractions. Maybe someone in the community will explore that issue?
I don’t see how the new torch leveling option can come into play during calibration. Your moving everything manually by hand.
When a calibration is performed the previous info is replaced. It’s not that you need a new USB stick, it has to be freshly formatted and blank.
When leveling the base to the calibration jig it is most accurate to check the height between the stylus tip and the three points of the jig. An arbitrary gauge block or shim works well. I have a machinist level that is extremely accurate but I found the gauge block really gave great results. It helps to actually see and feel what is going on and takes errors in the arm joints into consideration. When getting real close it is sometimes easier shimming the tramming jig at 3 points than messing with the 4 base adjustments.
I also find it best not to do the calibration on your cutting table. I actually did mine on a surface plate. A steel bench or countertop, flat table big enough for the Droid and jig is the way to go.
Those look pretty round to me! Your Plasma settings need some work though!
Please note: AMP mechanic’s video is only his own experimentation with a very early version of the product. It does not reflect current state of the system nor the official line on the function of the ArcDroid.
Notes on calibration.
Calibration is for XY only it does not affect Z
you do Not need the USB inserted to complete the Calibration (it is for saving data we may need for analysis later).
Calibration is only as good as your set up. So Flat, everything firmly clamped down and maintaining good pressure on the sled makes for best results.
The Sled should be adjusted so it moves smoothly but with just a little drag so to avoid any wobble
if your calibration is providing unsatisfactory results you can re-calibrate using same procedure. it doesn’t hurt to start fresh using the "Use defaults " option.
It is important to take the Calibration ALL THE WAY to the end some people are exiting the calibration before the final steps this will not save the completed settings.
Yes the parts are coming out ALMOST too good to tell by eye the out of roundness. However, they are not accurate enough even with 1/8 inch leeway over a 4.75 inch diameter main circle for the 5 smaller holes which are also given 1/8 dia increase to their diameters to fit the lug nut pattern.
(These are much like wheel spacers but NOT- they are actually frame jig components to hold my wheels, from the outside of the wheel, firmly where I want them while working from the corners of the car in for prototyping purposes)
Please clue me in on the plasma settings!!! This is all a learning process by jumping in with both feet. no classes around here for these things.
My machine is a Lincoln Procutt 55. All that seems a bit off to me is the air seems a bit excessive from the sound. However, when running it barely remains in the very low end of the green on the machine included regulator.
the material is 1/8 plate scrap from a local high-end, 18 wheeler style, race trailer manufacturer. so not clear exact lineage of this particular piece but it IS magnetic so going to lean to not being stainless. The consumables were new, never used, for this cut.
Now that said. . . the pieces on the LEFT of the picture were cut three times in place until i had enough dwell on the torch pierce to actually begin an arc before the arm started moving. youtube shows the first actual cut: Arc droid first project - YouTube i am not sure why this particular plasma torch has a LONG pause on its first cut between air solenoid opening and the arc initiation. All subsequent arc starts were almost instantaneous but still had the long dwell so it sat with arc going for the extended duration of the dwell.
the part on the right was done in a single run.
I’m not sure what exactly I am supposed to be looking for to judge a great cut- these are better than i ever managed by hand with this machine. i watched a youtube video of the Eastwood guys trying to give tips today- and well these cuts were far above their class in smoothness and looking more sheared from the side than cut. if anything the eastwood video examples of 1000 ways not to get a good cut.
both parts match up great but are egg shaped. the center hole is 3 inches one direction and 3.125 the other. not sure right now which access those match up too. but the holes should actually be 3 inches all the way around. i have not had a real chance to try another calibration yet.
For 1/8” steel I would try these for settings.
45 Amps
.06” cut height
.150” pierce height
.4 sec pierce delay
140”/min cut speed
Do a dry run and pause the program and verify the cut height. Adjust the number to get .06” for real.
Get a good measurement of what your actual kerf width is and use that as a setting. My tips give consistent .030” wide cuts. Using that as a cut setting my hole diameters are usually .005” undersized.
Good luck. Do another calibration! I’ve done very similar, Pipe flanges, in 1/8”, 1/4” and 1/2” thick steel and they were damn near perfect. About the same size your doing.
Set your part a bit further out from the Droids base than you showed the pics. May help some too.
I have had a similar problem and I had to replace all of my consumables. It turned out that an o ring in my torch had warped causing the air stream to come out at a funny angle.
Very true! I’ll see if I can put a note into the beginning of that video since that information is no longer relevant. The new calibration routine is much improved.
I apologize for the large delay in an update but sudden 2 of my project cars got paint appointments in the same week so it was a real scramble to get them stripped and bodies delivered to the painters. Also, placement of one of the chassis is right in front of the arcdroid cutting area so actually running any cuts is held up until the car sitting there is re-assembled.
After running a few more calibrations and doing the test at the end i am quite certain that it is my calibration that is off. The absolute best i could do was 1mm in the X and 2 mm in the Y. The calibration sled was sliding a bit stiff on rails #2 and #3 and i was having a bit of deflection on the calibration triangle away from the arcdroid arm causing the sled and stylus to separate and giving me several bad data points.
i have since buffed the two tight arms on the triangle so the sled slides much better and supported the underside of the triangle with a thicker piece of plate on top of my cutting table. i also stood the triangle off about 2 inches from the base of the machine as several suggested. The larger plate also helps me clamp things down with my clamps well away from the arm travel area. i was having problems staying out of the homing path with my clamps.
but alas. . . my sled micro switch has given up the ghost after all the extra use. . . .so looking herein for how to purchase a replacement. once that is remedied i am quite sure MY errors will be less and a better calibration will result.
