CNC Plasma Table Build

I'm, supposed to ground my water pan?

 

Table is moving again, and I did a bit of testing of the initial height / touch sensor.

https://youtube.com/shorts/Dfa2V8RsB3U?si=NO6SdT7gtlCjQ-He

I need to change out a fitting on the air dryer and hook the plasma cutter back into the cabinet and I can do some more test cuts.

Now I need to get the water table made so I can get start getting this finished.

 

It has been awhile since I did any work on the table, had a close friend go through a serious life altering event. but after dealing with that situation and making sure she is ok, I have been able to focus a bit more on things around the shop an home.

One of those was getting the drag chain for the Y axis installed.



this simple little bracket took far longer than expected to make, it adapts the bolt pattern from the drag chain to the extrusion for the gantry.






The chain rides on a piece of 4" aluminum C channel that I notched to sit over the mounting tube,





I also attached some ziptie brackets the C channel the gantry chain rides on. these are the same ziptie mounts I designed for my roof rack years ago. I had a bunch cut on laser, and sold a few, but still have a bunch leftover and they ended up working nearly perfect





The cables have been routed through the cable chains, and the home limit switch for the X gantry finally got wired in at the same time.

 

@Kwikvette what consumables are you running in your UPM-105?

 

Funny you ask cause I just fired up my air compressor to cut.



I have gone through a lot of PrimeWeld consumables and they are my main go to for all my work.

Before refrigerated air, I was averaging 400 pierces per nozzle and getting a max of 800 pierces per electrode.

However, I am also in the process of experimenting with others.

I have Plasmadyn electrodes on hand, and currently in use. The electrode currently has 554 pierces, with over 400 of them being on non-refrigerated air and the remainder on refrigerated air.

Not fair enough to compare as I'm now on a refrigerated setup but I do have more left over to continue experimenting down the road.

I also have plenty of ATTC consumables which are American made and a direct competitor to Hypertherm. They did such a good job that Hypertherm even sued ATTC several times over supposed patent issues. Have yet to use them but will be interesting to do so once I start cause they are non-shielded.

Long winded I know, but to sum it up PrimeWeld consumables do the job.

 

@Pyrotech

PrimeWeld on the left, ATTC on the right











PrimeWeld





Apparently non-shielded are supposed to cut better, or better detail, compared to a shielded variant running the same kerf width.

I forget the whole reason why but there's old stuff about the debate, as this is based on an older consumable type (T-45).

The plan currently is to continue using the 550+ electrode (150+ nozzle) until I hit 1,000 pierces on the electrode (500 on the nozzle).

I'll pull the set, replace with a fresh set, and see about exceeding that pierce count to about 1,200 pierces (600 on the nozzle). It's hard keeping tracking of stuff as I don't want a bunch of old consumables laying about, especially since I can't do a direct comparison anymore as I once was before the air dryer.

If you want, I can send you my bottle opener file as well as sending you an actual bottle opener I've cut out. This way you can directly compare a physical product to one you cut for yourself using the same exact file. Give you an idea of where your machine is at even. Bottle openers on hand were cut using used PrimeWeld consumables and on non-refrigerated air.

 

Been trying to get my torch to fire on a reliable basis with the CNC4PC THC controller. I just have not been successful with it. So, today I did a fresh reinstall of mach4, and stuck with the M3/M5 commands to control the torch instead of the M62/M63 that was being used by the plugin for the THC.

So far, it seems to be working. I ran a quick test cut using the F-Bomb file @Kwikvette sent me for testing, including a control sample to compare my cuts to.

I did a cut in in 16g, I have a sheet I been using for initial test cuts. saving the thicker stuff for actual projects at the moment... While the cut quality is far from good, it did cut the entire file which was my goal.









this was not a test cut for quality, but the goal was to see if the file would cut. In previous test the table would cut part of the file, then fail to fire the torch and sections would not cut at all. all had to do with this..

I still have the previous mach4 Profile saved, So I can still work on getting that one to work. But for now, I want to remove a source of problems and get the machine dialed in, along with my CAM process. having one less thing to troubleshoot will keep me motivated to get this table done 100%

 

Awesome to see it fully cut out!

Think you posted here or in the other thread about a a file not being completely cut out.

Now you can confirm each axis is running smoothly as needed with square cut outs!

My go to is a 2.5" square, with two .75" circles, and a .25" radius at each corner.

Allows one to ensure all sides are equal length, each radius is the same size at the corners, and both circles are in fact circular.

 

Feels good, doesn't it when it finally does what it's supposed to.

Welcome to the club!

 

Coming back to this because aside from opening my file and sending you the DXF, I don't recall the exact details of this bottle opener (until now).

But now that I'm looking at it, I have to ask -



Does your controller specify what post processor to use? Or is it generic in that "many" will work?

I remember that test piece you made with a large radius at one corner; the whole piece looked like it was intentional and nicely done.

But for this piece, if I read correctly, you're using a different post processor right?

I'm glad that I sent that physical opener to you as well but now wonder if X and Y axis movements are interpreted differently with each post processor not just by the hardware itself, but how it relays to "steps" when communicating with the motors attached.

Edit: Just cut some openers, attached picture with said openers right off the table. Although I think I left smoothing disabled as it's an old NC file I haven't revised.

 

I was using sheetcam for the previous cut, then switched over to fusion360 and a generic mach4 post. my controller prefers sheetcam as they have a custom post written for it, but most of my cuts that have been anywhere near decent have been done in fusion with the generic post. so for I switched back for now.

sheetcam has a lot of setting to fine tune, I will fuss with them after I get known good and repeatable cuts, with fusion.

I was experimenting with the pierce delay and the lead in setting, cuts improved a bit, but I before doing much more testing I will need to get the gantry back to where it was precrash, an get rid of the sloppy movement. I am hoping that the rounded corners will sharpen up and the slight deviation in overall size will also go away

I cut out several of them today, and other then mistakes I made in the g-code, everything cut on a consistent basis. not sure what happen on the last one, but 4 of them cut with the same issue, so for now I guessing it was a mistake in a setting when generating the g-code. which will be easy to test just by using the previous version again and seeing if that issue is resolved.

@Kwikvette control sample on the far left...

 

Is there any "slop" or free play in your axis movement?

I ask because the first sample's opening at the bottom is larger than the other two

The top right opening in the second one looks narrower than the rest

And lastly, the lower left outer edge is visibly thinner

Seems as if one of your axis movements has play in it causing items to not be symmetrical

For your lead ins on the small cuts, you can incorporate something like a .06 lead in radius, using a .12 lead in distance, and disabling lead out

If you don't like the lead in position, you can manually select where you want your pierce to begin during your profile creation process

This is done in the last tab, where you adjust lead in/out settings

 

Here's an example of those lead ins and distances



You can see how small the opening is, only 1/2" tall



Wish I had pictures to show you, but here's a screenshot of the video - I make cut-ready files for someone in Honduras and he pays me weekly for help with Fusion



Just one of several people I either help with on Fusion, or do Zoom meetings with lol

Ignore the parts that didn't fall off just yet, he has moisture in his air supply

 

Spend the day trying to figure out a way to add rigidity to the gantry. After a bit of looking other machines. I came up with pushing one of the carriage blocks out about 3 " further. This will not impact the travel distance i get but should solve the issue i am having.

So spent the evening working on the cad model for the new gantry brackets..

 

Been keeping an eye out for some parts for a little addon project for the table. I came across a couple parts for a good price .

$343 Chicago Pneumatic CP9361 Industrial Air Scribe Engraving Pen and Precision Chisel Hammer with Hose

Chicago Pneumatic CP9361 - Air Scribe.



SMC Air Slide Cylinder Table MXS20-100 20mm Bore 100mm Stroke

also went ahead an order a couple air solenoids to control the scribe and the air slide

 

The new brackets are finished, just need to get them installed...

Since I have the advantage of hindsight, I went ahead and bought the aluminum plate in widths that where usable as is.



I cut them down with the bandsaw, an cleaned them up on the 2x72 to get them roughly the same length to each other.


Printed out a 1:1 drawing of the carriage plates


Test fit, was dead on for the right side. but the left side I needed to open the holes up a bit more.



repeated the 1:1 drawing for the upright plates.



After drilling an counterboring the holes, it was time to drill and tap the side holes for the upright plate. I scribe a center line and drilled an tapped the center hole. this time I set up a better jig to hold everything in position





After the center hole was threaded, I assembled the upright plate, squared everything up an used a transfer punch to mark the hole for the other 2 holes.

So far these seem to have turned out much better than the original brackets, I will know for sure once they are install. I am looking for two things. most important is the play in the gantry while moving along the Y axis is gone, Second the gantry is square to the table bed, last time I had to shim the front edge to bring it square.

 

Man, your work is fucking awesome.

 

The new gantry brackets are installed... Remind future me that when I take it apart to to paint the frame.. to open up the holes on the right side gantry plate. it will make reassembly in the future much easier.

The gantry wiggle I had along the Y axis is gone, and the gantry slides on the rails MUCH smoother.

 

Could that "wiggle" be what I suspected was the issue in those F bombs?

 

a large part of it..

this was after the new gantry plates.



Still a few issues on the corners though