My (kinda) budget build 4x4 plasma table

[Dmaxpwr]

kl3jEOlnT5ycwi2FPmtCxw.jpg

I've been snooping around on here for a while now (there's tons of info to absorb) and just like most fab guys, I've been making my own stuff for years. And because it takes so much time to fab any kind of bracket I've also wanted a CNC plasma table for quite a few years now, but the costs just couldn't justify it in the past. I finally decided that I was going to buy a 4' x 4' table and have done a ton of research into what's available. There's defiantly some decent 4' x 4' tables out there in the $5,000 range, but nothing that I was really happy with so I decided to build one myself.

The plan is to try and build a heavy duty table for $3,000 to $3,500. It will have a 50" x 50" work surface, sheet storage underneath, linear guides, z-axis w/THC, rack and pinion drive (not happening at this time), dual x-axis drive steppers, a scribe or engraver and hopefully be strong enough to support a router in the future. I am still on the fence about a water table or downdraft.

I started ordering steel and parts about a week or so ago. it's Memorial Day weekend and with all this COVID-19 crap going on there aren't any cookouts or outings, so I figured it was a great time to start the project.

I got the base frame made out of 0.188" x 2" x 4" tube and the legs were made from 0.188" x 2" x 2" square tube. I had a small cart that that's rated for 1,000 lbs. so I robbed the wheels off of it and made plates for the bottom of the tube to mount them, I will probably add leveling feet later on to stabilize the table. I drilled and tapped the x-axis tubes for the linear rails. This was the when I had my first hiccup, I broke my m4 - 0.7 Snap-On tap after just 4 holes WTF, the funny part is that I was able to tap the other 96 holes (including y-axis) with a hardware store Vermont American tap, go figure.

After installing the linear rails I made 1/4" x 8" x 12" generic side plates and drilled them for the linear carriages. I made gantry uprights out of the same 2 x 2 tube and capped them so I can bolt them to the gantry beam. I made that out of the 2 x 4 tube. When I drilled all the gantry parts I oversized the holes by 1/16" to allow for adjustment. I bolted the gantry together and squared it up. The base frame is out of square by about 1/16" so I wanna sure how this was going to go but the gantry slides nice and smooth.

The last thing I was able to get done was install the Y-axis linear rails, and fabricate a Z-axis base plate, I did this with dual rails for extra strength for potentially running a router in the future.

[Dmaxpwr]

I forgot to add, another must is a break-away torch mount

[Blackslacks]

Nice. I like your design. I built my own 4x4 with water table too. One thing though. Try and keep your gantry as light as possible. It looks fairly heavy already. You will need big motors to get the speed and acceleration required for quality cuts. I used aluminum extrusion for my gantry and aluminum everywhere I could on the y. Total weight including motors, guides, carriages, z and torch is about 60 lbs. Im only using 300 oz motors with direct drive rack and pinion and it flies. It will have no problem handling a spindle either. Just a tip for you. Keep us posted.

[acourtjester]

Nice start keep going, I have built 9 table by myself lots of fun. I would add cross bracing to the legs, adding a router later is not hard and very handy for wall plaques. Easy to add options after you have the table completed.

[Dmaxpwr]

Nice. I like your design. I built my own 4x4 with water table too. One thing though. Try and keep your gantry as light as possible. It looks fairly heavy already. You will need big motors to get the speed and acceleration required for quality cuts. I used aluminum extrusion for my gantry and aluminum everywhere I could on the y. Total weight including motors, guides, carriages, z and torch is about 60 lbs. Im only using 300 oz motors with direct drive rack and pinion and it flies. It will have no problem handling a spindle either. Just a tip for you. Keep us posted.

The original plan was for an aluminum gantry but because I've gone a little overboard with a few things I had to cut back a little and use what I had on hand. It will be driven by two 570oz Nema 23 steppers so I think power wont be a problem, but they will be belt driven so we'll see if that holds up

[Dmaxpwr]

Nice start keep going, I have built 9 table by myself lots of fun. I would add cross bracing to the legs, adding a router later is not hard and very handy for wall plaques. Easy to add options after you have the table completed.

I have pretty much zero storage space in my garage so I'll be tying the legs together with angle and 1" x 1" sq. tube that will be used for sheet storage. And once I have a CNC table available I'm sure modifications will be made.

[djreiswig]

Just remember inertia. If you make it too heavy it's going to have a hard time changing directions quickly.

[acourtjester]

Dmaxpwr PM sent reply to yours

[Dmaxpwr]

I decided to order an Openbuilds C-Beam XLarge Linear Actuator for the z-axis. It seems like a pretty decent product for the money, but time will tell. I forgot to order t-slot nuts so I had to modify some 1/4"-20 bolts to be able to mount the assembly. I made the breakaway mount out of 1/4" x 4" flat steel, (4) 1/2" steel bearings and (5) 60lb.magnets. I mounted this to the z-axis plate. I then made a bump up using 1/4" x 1-1/2" flat steel and two 12mm x 100mm linear slides offset from each other about 5/8" to give some additional stability. I welded on some split flanges to use as the torch holder. I bolted on a piece of angle for the hard lower stop, the linear carriages create an upper stop due to the offset. I added two proximity switches and mounts, one for the breakaway emergency stop and the other for the bump up signal. I also added an ohmic sensor bracket.

wWLONi31SgiHqjY33swwtw.jpg

l5QGb4MOQ82CB4P3VCn34A.jpg

tZ5uFHb9SaaY+3b%98Qh5A.jpg

%2+8qJi%T4yG7a4pPdlS9A.jpg

0i306PuWS7iKEz9fZ3uK0w.jpg

vbteh6KTSGmMSrID03EJMw.jpg

OtBy7kBvSxiNlNJGBdsMFg.jpg

[Dmaxpwr]

Heres a few more pics & video of the z-axis, everything seems to be working pretty smooth, although it takes some effort to release the breakaway so I may have to remove one of the magnets but I'll hold off on that for now.

[Dmaxpwr]

I made a decision last weekend that I hope I don't regret. I was still waiting for a fair amount of the electronics to come in, but wanted to keep going with the build so I decided to move forward with mounting everything and skip the bench testing and just test in place.

I was able to finish all of the stepper mounts and belt drives as well as get all the proximity switches mounted. After that I started working on wiring everything. I use Deutsch connectors on most of my wiring projects so I didn't see a reason to not use them on this one. They provide an extremely high quality connection, easy to use, recycleable, 100% waterproof and most importantly, I get most of them for free by cutting them off of old parts at work. I also recycled a couple of cut off sections of multi-conductor from work and will use that to connect the table to the remote mounted electronics enclosure. I made sure to label everything and draw up a pin out list for all the connectors. I have no doubt that I'll be diagnosing something later on so I want to make sure that I know exactly what wire goes where.

On another note, I found out pretty quickly that I grossly under sized the cat track/drag chain. I originally bought some cheap 18mm x 50mm, it seems like it might be ok for the z-axis, but definitely not for the gantry. I ordered some larger track for that, I stepped up to 45mm x 100mm hoping that its just big enough that I can run the torch lead in it to. I'll see how well it works for this and maybe upgrade the other one later.

Next I'll start on the 6U server cabinet layout and wiring

[acourtjester]

Nice update, I am not a fan of running the Plasma torch cable with the other cables. I see you did use zip ties to attach it so you can remove it if needed, I like the boom method better IMHO. I also use the book for the power and vacuum for the router easy to switch to either when needed.
Also I would add cross bracing to the table legs for better wobble protection IMHO again.

frame build A (2).JPG

[Dmaxpwr]

I think the torch will end up half in the cat track and half out. The new track from the frame to the gantry should be here tomorrow so I'll see how it looks. If I can easily slide the torch end through the track then I'll put it inside it. If not then who knows which way I'll go, The plan has changed multiple times already so who knows where it will end up.

The legs will defiantly be getting bracing, the plan is to incorporate under table sheet storage which will tie all the legs together.

[djreiswig]

Some people have issues with interference when the torch lead is next to their control wires. I have had mine run through the cable chain since I got my table and I've never experienced this. I think it's a cleaner setup with everything running together.

[Dmaxpwr]

Some people have issues with interference when the torch lead is next to their control wires. I have had mine run through the cable chain since I got my table and I've never experienced this. I think it's a cleaner setup with everything running together.

I'll defiantly be running the torch in the lower half in the cat track. The new cat track came in today and its huge, so space won't ben problem. I guess I'll pay a little more attention to my metric conversions more in the future, plus the 45mm x 100mm is the I.d., defiantly quite a bit larger than I expected.

[Dmaxpwr]

Its been awhile since Ive posted and update so I figured it was time. I got most of the components mounted and wired in the server cabinet and connected to the table. I used Deutsch connectors in the bottom of the cabinet to make it removable/serviceable. I forgot to order the plasma cutter CPC cable when I started all this so its not installed yet. It came in this week so hopefully I can get it installed this weekend. And the Price THC got lost in the USPS system somewhere so I don't have that yet. For now it only has proximity switches set up for homing, I'll get the other ones set up later. I mounted the stepper drivers on the left side access panel for ease of service, the power supplies on the back wall and the ether-mach board right in front. I've got some relay modules and a time delay module (I don't know what I was thinking) mounted but they will be getting swapped out.

At this point I was dying to see this thing work and since I didn't have the parts to connect the plasma I started tuning the stepper motors. And thats where I ran into my first big problem. I knew this part was going to be a struggle since I've never worked with electronics. But I figured, what the hell, I'm a great learner, how tough could it be, WTF was I thinking.

There wasn't any issues with the X & Z axis, but the Y axis (dual motors) is a whole different story. They are set them up at 8 microsteps because that's what the torque chart from the mfg was set at. But no matter what I do, I can't get any speed and the motors keep locking up (I assume loosing steps). I (think) followed the procedure for motor tuning but it was quite a bit off, I used the stepper wizard and was able to get the axises to within 1/64" over a 14" span, but this is moving at a crawl.

So I swapped out the 12 tooth pinion gears (Saturday) on the Y axis for 14 tooth pinions and re-tuned the motors, I had it running at 325 inches per minute and I thought I was all set. So I spent most my time Sunday working on finishing up the wiring and building the slat supports. With most of the critical wiring connected I fired up the system and its not working again WTF! After spending a few more hours trying to get this working I just had to drop the velocity down to 150 to get it to only miss steps occasionally.

Homing is an issues now too, it was working perfectly. Now, when I try to home the unit it goes past the switch and crashes into the hard stops. The motors will continue to try and drive for about a 1/2 second or so. I also noticed that i'm no longer seeing a solid home indicator on the board or the diagnostics tab on Mach4

[Dmaxpwr]

I think I figured out the homing issue, it looks like it was because I had two home switches on the Y axis (dual steppers) and no matter how I adjusted them I just can't get it precise enough to compensate for gantry flex. I was able to figure this out by setting the jog feature to incremental and lowering the distance to 0.01". Once I did that I could adjust the switches so they actuate together constantly when I move a few inches off of home. But once I travel full distance across the frame and come back to home, the two axises were tripping the switches at slightly different times (apron. 0.03"). This was causing the motors to kinda fight each other when homing. I messed with this for quite a while before I realized that I only need one switch to home the unit. I turned off one of the switches and haven't had another issue with homing since.

[djreiswig]

I'm not sure what controller you have (too lazy to read the whole thread) but my table has dual motors and dual switches. One is X and one is A. I have it set up so it will auto square. Each motor runs to its switch and then backs off. This way the gantry is always square. If you only have a switch on one side it's possible for one side of your gantry to lead or lag the other side. If you have another method of squaring, then disregard.

[Dmaxpwr]

I'm not sure what controller you have (too lazy to read the whole thread) but my table has dual motors and dual switches. One is X and one is A. I have it set up so it will auto square. Each motor runs to its switch and then backs off. This way the gantry is always square. If you only have a switch on one side it's possible for one side of your gantry to lead or lag the other side. If you have another method of squaring, then disregard.

I'm using Mach4 demo with an Ether-Mach CS motion controller.

That's why I originally added the switch for each axis. It was working fine and then the motors started fighting each other. I'll see if I can figure out how to adjust those settings so the motor backs off after tripping the switch.

[Dmaxpwr]

So I think I've got my motor tuning and homing somewhat figured out. Since both motors and drivers were the same so I figured that I could use the same SPU on the master and slave motors but it wasn't working. So I decided to disassemble the gantry and use the wizard on each motor independently. By doing this I found that there is a 64 step difference between the two. Since i adjusted this setting I can now get 250 in. per min. travel speed. I still have an issue with the A axis (slave) moving before the X axis. The homing situation is fair now, works most of the time. I know both of these need to be much better but I have much bigger issues to deal with first.


Still can't get the table working, I've got a bunch of issues going on that are driving me nuts. When I try to run a test g-code without the plasma hooked up, the unit I get all kinds of errors, home switches tripped (I have to jog off of the limit switches), runtime errors, macro errors, z axis height, in-op z axis probe switch etc. If I skip through and start later in the g-code I can get the thing to run and trace out the test pattern. I think part of the problem is that I haven't figured out how to trigger the torch without using a THC module.

Another new problem is that I'm not getting a solid signal for my homing switches now, I have a consistent 5v at the input but the indicators for all four switches on the diagnostic display flicker, they do not turn on solid like they did before. The 5v terminals also have inconstant voltage. I think I must have fried the C25 terminal board, so I ordered a replacement.

[acourtjester]

What is the origin of the 5 volt supply, you may be connecting to many loads to the 5 Volt circuits. I had a similar problem using the +12 volts from the computer's internal supply, in the controller. I have since gone to a separate 12 volt power supply. I changed the circuits, I use the 5 volts from the computer to trigger a SSR that switches the input power to other power supplied ( 2 driver power supplies, separate 12 volt supply,and the monitor for the controller). The SSR take very little power to switch it and they come in many amp sizes for AC circuits. I use small switches to control different sections of the controller like the THC, Laser light, and power to a router and vacuum setup I use optionally on the table.

conroller power 1.JPG

controller power 2.JPG

DSCN4194.JPG

[robertspark]

By doing this I found that there is a 64 step difference between the two. Since i adjusted this setting I can now get 250 in. per min. travel speed. I still have an issue with the A axis (slave) moving before the X axis.

Something is wrong there, are the dipswitches on the stepper motor drives exactly the same for the X and slaved X' (A) axis?

Are the steps per unit and acceleration in Mach4 exactly the same and without setting any backlash?

Do you have a rough schematic of your wiring (trust me it will save you time in 3 years / 5 years time when you want to change or add something!)

[Dmaxpwr]

Sorry, I left out the details of how I wired the home switches. The proximity switches are 24v, but they trigger relays which close to send a 5v signal from the power supply (unregulated linear 45v/13a toroidal power supply with 5v) to the ether-mach BOB. The 4 home inputs are the only thing connected to this 5v power supply so the draw is extremely low.

The Ether-Mach CS motion controller is fed with 24v from another power supply. I originally had the limit inputs fed from the BOB but the 5v output was intermittent, this is part of the reason I ordered a replacement C25 BOB.

45v stepper drivers

24v Ether-mach board
24v relay modules
Proximity switches

After reviewing my wiring again I realized that I have a few items running off of this 5v source:
5v C25 BOB home inputs
Bump up/ Port 2 pin 13
X, Y, Z & A Stepper driver Pul+ & Dir+

[Dmaxpwr]

By doing this I found that there is a 64 step difference between the two. Since i adjusted this setting I can now get 250 in. per min. travel speed. I still have an issue with the A axis (slave) moving before the X axis.

Something is wrong there, are the dipswitches on the stepper motor drives exactly the same for the X and slaved X' (A) axis?

Are the steps per unit and acceleration in Mach4 exactly the same and without setting any backlash?

Do you have a rough schematic of your wiring (trust me it will save you time in 3 years / 5 years time when you want to change or add something!)

Yes, the dip switches and SPU /velocity / acceleration settings for both motors were identical. The backlash is set to "0" on all motors.
I don't have an actual schematic drawn out you but it is defiantly in the plans

[Dmaxpwr]

I don't have a wiring schematic yet but here's what I have for worksheets and connections.