Project 1 from EmptyNester. Category: Custom Shop Tools - CNC Table

[EmptyNester]

I did start leveling the slats. Once I got enough to do support a test cut I mounted the floating head plate to my z-axis. I had to modify the mounting since I have a Precision Plasma carriage and a Shopdroids Z-axis. The mod I did put bolts in the way of the floating head. It reduced my z-axis travel from 6 inches to 2 inches.



John from Shopdroids was helpful with some ideas. I ended up making an adapter plate out of aluminum. Not pretty, but effective and it got the job done (I want to cut and I'm frustrated at falling behind where I want to be).



I got my energy cable today. I ordered (3) three meter sections. They were only $19/each on ebay. I looked at getting a wire mesh cable tray (used for hanging network cables in the ceiling). They were $100 for a ten foot section locally. Mail order they were only $20 each, but truck freight was $220. So, that was a non-starter.

I ended up getting a wire shelf at Home Depot for $20. I shortened it and made it narrow. I do need to give it a little more support, but I'm not sure the best way. I think that aluminium angle will work, but I don't know how much that would be.



I had thought of using conduit, but that probably has too much flex. Any ideas are welcome.

[EmptyNester]

I thought about a couple different ways to support the wire shelf for the energy cable. I ended up getting a U shaped wall bracket for holding shelf supports.



I didn't install the engery chain because I don't think the torch is long enough to make it. Hopefully, I'll have a longer torch in a couple weeks.



All I have left to do is:

• put in the X limit switches
• complete the wiring on the limit switches
• use silicone caulking in the problem area that may still have a leak
• finish leveling the slats

Before I try a test cut, I need to finish wiring my THC/interface. I don't plan on trying to get the THC working just yet, but it handles the relay for the "torch on".

[sportbike]

That thing looks pretty darned nice. I am looking forward to seeing how it works out. I have still been toying with the idea of building a small one for myself.

[acourtjester]

HI Empty

I ran my plasma cable using a rotating pole in the corner of my table. It ended up with a shorter run, no worry of interference with the motor cables. My motor cables are run like yours. One picture shows my table corner with the rotating pole base.
Like this one shown here other picture.

[EmptyNester]

Sportbike - thanks. I started thinking I'd build a small one too, then just kept coming up with reasons why it should be a little bigger.

Jester - that's a great idea. I had seem some like that but didn't realize the poll rotated. I ordered a 25' to use already, but that will let me leave one permanently mounted on the table, but still have one for portable/hand use.

[acourtjester]

HI
I used a different arrangement for my slats in that I used a 2” angle base and 1-1/2” slat. His way I save a little when replacing the slats. I can also remove the whole slat fixture for cleaning. Different strokes type thing.

[EmptyNester]

HI
I used a different arrangement for my slats in that I used a 2” angle base and 1-1/2” slat. His way I save a little when replacing the slats. I can also remove the whole slat fixture for cleaning. Different strokes type thing.

That's actually much closer to what I had hoped for. Unfortunately, welding the length of the tube frame caused the bottom of the water table to warp badly. Ideally, I would like to put steel down on the sides and middle lengthwise to provide a level surface for the 1 1/2 high slats (similar to what you have). I just haven't figured out the best way to cut the lengthwise supports to meet the curve of the table.

For now, I'm just using washers to level the 3" tall slats so I can do some test cuts when I have the torch interface (relay, etc) working.

[sportbike]

That's actually much closer to what I had hoped for. Unfortunately, welding the length of the tube frame caused the bottom of the water table to warp badly. Ideally, I would like to put steel down on the sides and middle lengthwise to provide a level surface for the 1 1/2 high slats (similar to what you have). I just haven't figured out the best way to cut the lengthwise supports to meet the curve of the table.

For now, I'm just using washers to level the 3" tall slats so I can do some test cuts when I have the torch interface (relay, etc) working.

How about building an intermediate slat frams lime acourtjester did, but use 1.25" angls and 1.5" slats. Then you can level/shim the 1.25" angle as needed floating it over top of your warped frame, then slip the slats in. Slats woudl be cheap since they would only be 1.5" wide and you could have a flat table. Maybe you could even just bolt the intermediate angles in to eliminate any distortion and allow whatever shimming you needed.

Maybe put multiple angles across rather than the single angle in the middle.

John

[EmptyNester]

I just completed doing all the wiring in the cable trays/energy chains and all the limit switches.

The stock torch wasn't long enough and I had wanted a longer torch since I got the unit. I ordered a 25' one on e-bay from "3Shop River" last Saturday and it got here yesterday. So I was able to finish my cabling.

The fit in the energy chain was tight, so I ended up cutting off the plastic sleeve on the torch for the portions that ran inside the chain. That made everything fit comfortably. Everything seems to work fine.

I need to do some more configuration on Mach. It sees the limit switches and stops, but you can't shift off the limit switch. I recall there was a setting somewhere for that, just haven't looked yet.

I had leveled some slats with washers, but then took them out to seal all the seams. I'll use the washers to level them until I get the table going and figure out a better solution.

All I have to do if finish wiring the torch on/arc good lines in my Arduino interface so that I can try cutting.



As much as I want to work on it tomorrow, I'll probably spend most of the day trying to get the hovercraft running again. I'd like to get it going well enough for at least one high speed run before it gets put away for the winter.

[EmptyNester]

I'm starting to configure Mach and test my system.

I've turned the motor speed up and am pleased with the performance. I do have an issue with my 'A' axis. When running at full speed through the center section of the table, there are two places where it makes a "ginding" noise. Its a very sharp noise that is short in duration (like 1 gear tooth?).

It happens twice on the same side within a couple of inches of each other.

When running at a lower speed, it doesn't make any noise.

Any ideas?

[sportbike]

I'm starting to configure Mach and test my system.

I've turned the motor speed up and am pleased with the performance. I do have an issue with my 'A' axis. When running at full speed through the center section of the table, there are two places where it makes a "ginding" noise. Its a very sharp noise that is short in duration (like 1 gear tooth?).

It happens twice on the same side within a couple of inches of each other.

When running at a lower speed, it doesn't make any noise.

Any ideas?

I suggest you give up on the project all together and sell it as-is.

I will be at your place in about 2hrs with $400, which is a very reasonable price for this.


Is it a single length of gear rack or is it pieced?

Can you push the drive across the area by hand to see if you feel anything?

[aaxiss]

Kinda sounds to me like a bind. You started in the center of the table think you may have some sad in the center maybe s and the extra pressure on the gears is causing the gears to click a bit?

Just a thought

Sent from my Nexus 7 using Tapatalk 2

[EmptyNester]

Sportbike - thanks for the offer, but I wouldn't be able to deal with the guilt of sticking you with a defective product

I can push the granty the entire lenght and the only time it hangs up is if I don't keep it square.

I spent last night working on the limit swiches and then this morning working on the grinding. I had thought that racking might be part of the problem. So, I have the limits and homing working, so I could rule that out.


Aaxiss - I thought that might be an issue with sag or bowing, but it seems to be based on the speed.

I did more testing. After everything was squared and homed, I could make it happen at high speed. At high speeds it seems to sound like the motor is running away but it's stuck in place. Then, if I'd rehome again, the X-axis was off inches (8" on one of the runs).

At this point, I'm wondering if I've driving my steppers too fast. On thinking about it, when I sit at the end of the table and move it toward myself "that's scary-fast" probably isn't the right speed.

I had read to set the motors as fast as you could and just slow them down until they work well. Mine work fine unless I use half the table to accelerate.

So, I started looking at my speed. I had it at 3800 in/min. From doing a quick search, it looks like I won't need to go over 100 in/min at the most. (I would assume even slower if I mounted a router.)

I haven't continued playing to determine max speed. I'd like to figure out how to use the touch and go so I can try a test cut.

[sportbike]

Any chancec the gear or something is slipping on the shaft? A buddy of mine has a mill that does that on occasion and loses its position.

[EmptyNester]

Any chancec the gear or something is slipping on the shaft? A buddy of mine has a mill that does that on occasion and loses its position.

I hadn't thought of that. It would probably explain the noise a little better.

I'll have to take the gear off and check for scoring from the set screw.

[EmptyNester]

I'm using the demo version of SheetCam TNG and the licensed version of Mach3.

I've used Corel Draw to draw a square and save it as a DFX. I import that into Sheetcam and use an "outside" tool to cut it. I post process with the MP-1000 THC post processor.

When I load it into Mach and regen the tool path, it shows the square I drew with a circle about 4 times it size.

I'm not sure why I'm getting the extra circle.

I know I still have settings issues. Sometimes it runs okay, sometimes it takes the Z axis too high on starting (due to rapid clearance setting?), other times it just hangs. But, I'd like to figure out what's going on the the square first.

[Rambozo]

It would help if you post your Gcode. I'm not familiar with Corel's output, so you might want to have a look at that DXF file in the AutoCAD viewer. I take it you don't see the circle in SheetCAM? (be sure to zoom out far enough to make sure). Make sure your clearance plane in within the work envelope of your machine. The toolpath you see in SheetCAM should match what you see in Mach. I can't think of anything right off where the post would add a circle, but it is possible. Have you tried the generic Mach3 post?

[EmptyNester]

It would help if you post your Gcode. I'm not familiar with Corel's output, so you might want to have a look at that DXF file in the AutoCAD viewer. I take it you don't see the circle in SheetCAM? (be sure to zoom out far enough to make sure). Make sure your clearance plane in within the work envelope of your machine. The toolpath you see in SheetCAM should match what you see in Mach. I can't think of anything right off where the post would add a circle, but it is possible. Have you tried the generic Mach3 post?

What do you mean by "generic Mach3 post?".

The square was right at the inside edge of the material part of the window.

I installed SolidWork's Edrawing viewer. It looked appropriate - I only saw the square. The G-code is:

N0000 (Filename: square.tap)
N0010 (Post processor: MP1000-THC.scpost)
N0020 (Date: 10/7/2012)
N0030 G20 (Units: Inches)
N0040 G53 G90 G40
N0050 F1
N0060 (Part: square)
N0070 (Process: Set variable switchoffset = 0.21 )
N0080 (Process: Outside Offset, Layer 1, T1: Plasma - outside profile)
N0090 M06 T1 F200 (Plasma - outside profile)
N0100 G00 Z2.5000
N0110 X2.1226 Y-0.1319
N0120 Z0.5000
N0130 G28.1 Z0.12
N0140 G92 Z0.0
N0150 G00 Z0.2100
N0160 G92 Z0.0
N0170 G00 Z0.5000
N0180 M03
N0190 G04 P0.3
N0200 G01 Z0.1000 F60.0
N0210 G03 X2.0202 Y-0.0295 I-0.1024 J0.0000 F200.0
N0220 G01 X1.9907
N0230 X0.0000
N0240 X-0.0443
N0250 G03 X-0.0295 Y-0.0443 I0.0000 J-0.0148
N0260 G01 Y0.0000
N0270 Y1.8220
N0280 Y1.8663
N0290 G03 X-0.0443 Y1.8515 I-0.0148 J0.0000
N0300 G01 X0.0000
N0310 X1.9907
N0320 X2.0350
N0330 G03 X2.0202 Y1.8663 I0.0000 J0.0148
N0340 G01 Y1.8220
N0350 Y0.0000
N0360 Y-0.0295
N0370 G03 X2.1226 Y-0.1319 I0.1024 J0.0000
N0380 M05
N0390 G04 P0.5
N0400 G00 Z2.5000
N0410 X0.0000 Y0.0000
N0420 M05 M30

[sportbike]

Is the circle some sort of lead in / lead out for the cut?

[Rambozo]

The code looks ok to me and here is a backplot of what it does.



The only arcs are the small circles at each corner to make sharp corners. The feedrate seems a little excessive for them at 200 ipm, so unless your machine can move at those speeds, you may want to go into SheetCAM and do some more machine setup. Make sure Mach is setup to rapid at a rate your motors can handle, too. The Z is set to go to 2.5" above zero, but there is a canned cycle for the THC G28.1 that I don't know anything about. It takes control over the Z axis and uses a coordinate setpoint that probably resets zero to the surface of the material at the start. Your pierce time seems to be .3 milliseconds so that points to another setup issue, unless that mystery G28.1 takes care of that, too.
In SheetCAM there is a Mach3 post that will generate code without the THC stuff.
The main feedrate is 60 ipm so that is probably ok, but parts are being done at 200 ipm which might be too fast and cause problems. At first you might want to really slow down the feedrate so you can watch things better, too. 10-20 ipm is a nice speed for small parts like this. You can also use Mach to step through one block at a time to better see what is happening during test runs. Does your display in Mach show the corner arcs properly? If not that might be what is generating your circle. Make sure your settings match for units, too.