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

[Rambozo]

Unfortunately, its a little leaky as pools go. I went through one round of patching holes but still have some left.

I probably shouldn't have painted it until after I tested. Now I'm not sure how I'll get the paint out of the corner that needs "patched".

This is true. You also don't know if the paint is sealing other very small leaks. When I used to build oil pans, I found that testing for leaks with solvent or mineral spirits is better than water. Less surface tension and it seems to find everything.
I'm sure a little heat and a wire brush will burn the paint out of your corners enough to re-weld them. MIG and stick can be pretty forgiving about bits of paint, at least. Of course just coating with a sealer is the easy way out.

[SeanMurphy265]

I probably won't be doing anything heavy. (I just wouldnt' be able to lift it.) I'm planning on more decorative/artsy stuff.

If you look at post #12 you can see the cross pieces I put in. Would you put in support in addition to that for heaver work?

I think it depends on how heavy you go! You have your 4 vertical corner tubes now, but if you add one piece of tube in the middle (vertical) on all four sides that would greatly reduce the capacity of the table. My table is heavier 5'x9' and I used 3/16" and 1/8" tubing; I can slide the table fairly well. I'm going to add some castors in the rear to help move the table.

The main thing that would scare me is the the gantry (depending how you mount it). If you get too much weight on the table it could distort your gantry. I have seen where some people have made their gantry where it could stand alone instead of mounting it to the table.

I called precision plasma about their Z axis and they would not sell me one. Since you bought a kit will they sell you a Z axis?

[EmptyNester]

I think it depends on how heavy you go! You have your 4 vertical corner tubes now, but if you add one piece of tube in the middle (vertical) on all four sides that would greatly reduce the capacity of the table. My table is heavier 5'x9' and I used 3/16" and 1/8" tubing; I can slide the table fairly well. I'm going to add some castors in the rear to help move the table.

The main thing that would scare me is the the gantry (depending how you mount it). If you get too much weight on the table it could distort your gantry. I have seen where some people have made their gantry where it could stand alone instead of mounting it to the table.

I called precision plasma about their Z axis and they would not sell me one. Since you bought a kit will they sell you a Z axis?

I had thought about a single leg in the center of the table, but figured that would make wheels much harder. I did figure I could always add legs later and though adding one to the long side of the table would need it first.

I hadn't thought about the weight distorting the gantry. I guess that's something I'll need to keep an eye on.

I didn't pursue buying a z-axis from precision plasma. They seemed too expensive and were beyond my budget. I got one from No Sleep Studio. In hind sight, I may have gone with their gantry kit had I known about them at the time I purchased my kit.

It's funny that they wouldn't sell to you. I had a similar thing happen with C and CNC. I sent him questions about his low cost THC. I figured that was the best for the money. I asked if he had experience with Everlast machines. He told me a Longevity machine blew up one of his boards. He then said that since they are from China and Everlast is from China - I wouldn't get a warranty. After that, he didn't response to email.

As a result of that, I'm going to try using an Arduino board to make a THC. I have the board with a display (about $45) and am adding a prototype board ($12). I think I'll need about $10 to $20 worth of parts. I write embedded control software for a living, so I don't think that will be a challenge. The only hard part will be discovering all the "gotcha's" a THC has to deal with.

So, when I get it "mostly working" - I'll make it open source for anyone who wants to use it.

[EmptyNester]

Since I have a sqaure tube frame, I decided to use angle to make an adjustable mount for the gantry rail.

I cut a wood pattern so that I could cut slots in the angle. It would allow bolting the angle to the square tube legs. The slots are vertical so that the entire piece of angle can be moved up and down to level the gantry rails to each other.

At the top of the angle, I put another piece that had a slot in it. This is to allow movement of the gantry rail closer and further in relation to the other side.

[EmptyNester]

I've had the drive electronics set up for a while. I just hadn't gotten around to taking any pictures to post.

I used an old PC case to house everything. I had thought about using the power supply in it for 5V and 12V, but decided it was too big and too much of a pain since it needed the power control logic overridden (no PC to power it on and off).

This is a 4 axis kit from ebay. It came with two power supplies for the motors. They are mounted on the top and bottom of the case. I arranged them so that one of the x-axis motors was on each power supply. It seems like it would balance the load a little better.



I mounted the four motor controllers in the middle.

I left the original PC fan in. I got a small 12V power supply (upper left corner) to run the fan. I have additional fans and will add them if necessary.

Right under the fan is the break out board. I have it mounted so the parallel port connecter goes out an existing hole in the case.

I have a five volt power supply (again, from e-bay) mounted in one of the drive bays. It's used to run the logic voltage on the stepper motors and the motor side of the break out board.

I decided to use isolated power supplies for the BOB. I had an old 5V wall wart from an iPod, so I used that for the PC side of the BOB.

I couldn't find any reasonably priced connectors that I liked for the drive cables. So, I went the cheap route and got a couple terminal strips. Given that I'm not going to be taking everything apart often, I think that will be fine.



I've already had it all running the motors. That was the first thing I did when I got the kit. They had a 14 day return period, so I wanted to make sure that they worked.

[EmptyNester]

I cut the slats and put them in the table.



I had orginally put in slots every six inches. When I looked at it, I decided that was too far apart. So, I went to four inch slots. As a result, I didn't have enough steel to fill them all.

I just painted the gantry rail mounts, so I'll be able to mount them and the gantry rails in a couple days.


Question:
I had been thinking of painting the slats with heat resistant paint. Anyone have any experience with that? Is it a good or bad idea?

[EmptyNester]

Got the gantry rail adjusters and gantry rail installed. All the adjustments went well and it was easy to get everything square and level.

[EmptyNester]

Got the gear racks installed. Then did a third water test after using the leveling feet to set the table up. (Y-axis is just siting there, its not finished.)

Leveling feet worked great. Only one small leak left in a corner. I'm not sure if I should use silicone caulking or the spray on rubber sealant. (The leak was in a tube joint and water was going into the frame tube and coming out a pinhole in the bottom weld. I drilled a drain hole in the bottom of the tube so that water wouldn't accumulate.)



Because of concern that I might need additional support in the middle of the table, I took measurements before filling with water and after. There was absolutely no deflection. I didn't have the slats or any material to be cut loaded, but I did overfill it with water. So, I'll probably test it again with slats and steel to be cut, but I'm thinking it will be okay without additional feet.

[Rambozo]

Looks really nice. I'm really liking the low deck height, I can just imagine how much easier it will be to load. I wonder if anyone has built one at pallet jack height.
I'm sure any minor deflection will be taken care of by the torch height control. As long as your axis run true and do not contaminate each other a little Z error should be no big deal for plasma. If you also want to mount a router, you might have to be a little more precise. Since you have everything adjustable, I'm sure you can get it all dialed in.
Have you taken any measurements to see how true your rails are? I know square tube isn't as close as DOM, but I don't know how much different. Did you measure your rails to use the best sides for the most critical axis?

[EmptyNester]

Thanks.

I did the "stack the tubes and look for warpage". They looked fine.

I also measured the distnace between the rails in a couple places along the length when I put them on the table. They seemed very good. But I probably need to start derating my eyes for measurement accuracy If they are off, it's under 1/16th (probably my current eyeball accuracy).

I only did height at the ends, but should probably do that along the length.

I'm not sure I understand the "measure the rails to use the best sides for the critial axis"?

Long term, I am hoping to make a router mount. I've seen some pictures of steel inlaid in a table top. I really like it and figured that would be a good use of a router.

[Rambozo]

I meant with a straight edge and dial indicator. One of the nice things about a gantry design, is that you can use parts of it to check other parts. By only mounting your crossbar at one end you can put an indicator on the other and adjust your rails to have minimum runout all along them. You will also know how parallel they are and how much deflection there will be. Since things won't be perfect, you probably want to set your X rails so that the straightest sides guide Y. And setup Y so that there is minimal runout in X. This will put the bulk of your error into the Z axis which the THC can take care of. A lot depends on the final accuracy you are looking for. If you plan to do router engraving, the Z will need to as accurate as you can make it. V groove bits are not forgiving at shallow cuts. I would say plus or minus .005" or less would be ideal. Obviously for plasma you don't need to be anywhere near that close.

One nice thing about modern software is that it can compensate for a lot of error in the hardware. I know there are tools to map the periodic error in gears and gear rack, then you tell your control software and from then on it will automatically adjust so that it looks like you have a perfect gear system with zero error.

[SeanMurphy265]

I was looking at your carriages and they are almost exactly like the ones I purchased from No Sleep Studio. I think the guy from PP called out the guy at No Sleep Studio on CNC Zone for making his kit very similar to PP. Is this kit the same as their Iplasma? Have you kept up with how much you have spent?

[EmptyNester]

I was looking at your carriages and they are almost exactly like the ones I purchased from No Sleep Studio. I think the guy from PP called out the guy at No Sleep Studio on CNC Zone for making his kit very similar to PP. Is this kit the same as their Iplasma? Have you kept up with how much you have spent?

It's funny for calling someone out on a simple carriage design like that. I would bet that's a pretty common design approach. Sadly, they are different enough that my NoSleep Z-axis mounting holes don't match those on the PP carriage.

The guy at PP said they don't make any tables with that carriage any more because it is the least precise and the rails need to be kept clean. I did think the price was a little high ($325), but the hovercraft project taught me that its worth paying more for a kit so you have somebody you can call if you need to. (But in hindsight, I might have gone with the NoSleep gantry had I known about it at the time.)

I have kept track of the spend. I haven't totalled it lately but will post a total cost and parts list. It was a mixed bag of spending more or less than expected, but I think I'm still coming out significantly less expensive than buying. The biggest costs were $325 for the gantry, about $450 for steel (including water table) and $400 for motors/drivers/power supplies (I don't recall the cost of the cabling and gear and racks).

[EmptyNester]

When I welded the carriage, I didn't think about the cross tube between the two x-axis.

I ran a weld bead along both the top and bottom of each joint. The top of the top joint is where the y-axis bolts on. The weld bead would have prevented any adjustment in length of the y-axis (it's a slotted plate).



A little bit of grinding and the situation was fixed.

[EmptyNester]

So, I updated the spreadsheet I'm using to track cost with. I think (hope) this will be my final cost. I tried to format it in a table, but all the spacing keeps getting deleted.

The quick summary is:
- my table and electronics $1972 (versus $4305 - purchasing system)
- z-axis and floating head $337
- Mach3 and Sheetcam $325

Grand Total - $2634 (versus $4967 grand total for commerical system)


Here's a breakdown of what I paid for stuff........

Stuff I didn't buy
PC - Had an old one
Case for electronics - Used an older PC case
misc nuts/bolts

Stuff I bought (for a 4' x 6' table)


Precision Plasma Carriage Kit $325.00
Steel Frame $365.00
Steel Water Pan (not stainless) $120.00
McMaster (3) 6' Racks $162.00
McMaster (4) Gears $62.00
McMaster Flex Cable $50.00
McMaster VHB Tape $23.00
McMaster (4) Anti-vib leveling feet $30.00
Table Total $1,137.0

Motor Drive/Electronics Total
Ebay 4 Axis Control (428 oz motors) $407.00
Ebay 5 Volt supply $17.00
Ebay 12 volt power supply $7.00
Ebay Relay card $3.00
ebay Limit Switches (12) $10.00
ebay E-stop switch (4) $16.00
ebay Arduino for THC $24.00
Amazon Terminal strips $14.00
Motor Drive Total $498.00

My Total 4'x6' Table & Electronics $1,972.0

Comparision 4x4 iPlasma w/ stainless pan $2,995.0
C and CNC motors (300 oz) $1,310.0
total w/o Z-axis mechanicals $4,305.0



Additional Stuff you'd buy either way

Z-Axis
McMaster Torch attach collar $7.00
No Sleep Studio Z-axis, floating plate $330.00
Z-axis total $337.00

Software
www.Artsoft.com Mach3 $175.00
sheetcam.com Sheetcam TNG $150.00
Software Total $325.00

[EmptyNester]

I have the Precision Plasma carriage kit and the ShopDroids Z-Axis. The mounting holes didn't match.

I had to drill new holes on the carriage. ShopDroids sent me spacers so that the bearing bolt head wasn't an issue.



I also have the touch and go plate, but couldn't install it because I forgot to order the bushings for it.

[SeanMurphy265]

Do you think the Z axis is strong enough for a router? I've thought about making a separate plate with mounts that would hold a dremel for cutting designs from 4x8 sheets of foam.

[EmptyNester]

Do you think the Z axis is strong enough for a router? I've thought about making a separate plate with mounts that would hold a dremel for cutting designs from 4x8 sheets of foam.

On the No Sleep Studio web site, they have a video showing different weights being hung off the axis. I don't recall what the most weighed, but it seemed significantly more than a router. Just based on the weight of everything, I would think flexing of the z-axis would be the biggest issue.

I've seen aluminum mounting brackets on e-bay for routers. I figured I would just go that path and mount it directly to the aluminum frame of the z-axis. I would be buying a router motor for it too, but figured I'd wait until the Plasma was fully up and running.

One of the things I've seen that I think was really cool was a table that had plasma cut steel on the top and was covered in glass. With the router, it could be done as an inlay.

[EmptyNester]

While waiting for the Y-axis I made a shelf to hold my PC, drive electronics and PP50. I got a used shipping crate I got at work. I cut it in half and then just put it on a ripped 2x4's that went the width of the table.

Just got the gantry cross tube back. I haven't put the track on it yet, but put it in place so I could start hooking up the drive electronics to start playing with them.

The first thing I did was install the E-stop. Since the switches were cheap and I'll be using the table where I may have trouble moving around it fast - I decided to put an e-stop switch on each side of the table. I mounted them in caps that go on the 2" square tube.

After finishing that, I started running the first wire for the motors. I immediately saw that having an e-stop on the side with the cables won't work.



Since that's the side that will have the PC on it - I'll just ditch that e-stop.

I also worked on the menu system for the THC. I have the top level menu working. When I finish all of it, I'll post a video on the Plasma thread that talks about it.

[EmptyNester]

I started final assembly of everything.


The X axis is fine and seems pretty smooth (only have 1 motor moutned on it currently).

I'm having trouble with the Y-axis. It will bind as I run it along the tube. When it binds the bottom of the vertical piece of the carriage seems to kick out from the tube.



I did a quick check and it appears that the top and bottom plate are much closer than the x-axis carriages. Around 1/10" on the side you see in the picture. I'm surprised I didn't see it sooner.

I'm guessing that the dimension is the cause of my problem, so I'll have to figure out how to fix it.