What's new

First plasma table build

I love Carling rocker switches but damn I hate trying to get them installed... I started buying the plastic housings to make it much easier to only cut the one hole.

I'm finally putting the Xmas presents for my wife's SXS, LED whips and rock lights.
This 50" rig doesn't have the real estate I'm used to in regards of adding switches, with winch enable and in/out there's only a few left in the prime area and 2 locations to the rear of the shifter. That location would be perfect for these two lights but I fear the dogs platform (between the seats) would constantly turn them on/off.

In front of the shifter there is a perfect area for switches but no pre cutouts in the plastic like all the other locations.

So I drew up shape that I thought would fit and cut it out of some .100 aluminum sheet.

Had to modify it a few times but finally got it dialed in. Pretty amazed that I could cut this part with my El cheapo rig. This type of part is absolutely what I had in mind when set out to build this table.

1000012399.jpg

1000012402.jpg

1000012405.jpg

1000012406.jpg
 
Spent a bunch more time figuring out OnShape sheet metal parts and finally tried to use the "spotting" function in Sheetcam/LinuxCNC. Took way to fuckin long to get all that working but I did get it in the end.
That feature is going to be really awesome as I much prefer drilled holes in the smaller diameters over the plasma cut holes. (With the understanding I have not dialed in much of my cut quality)

Figured out how to put the tabs on bend relief lines, not sure I should have used them on these parts, I'll weld them up when I add the gussets.
1000012797.jpg



1000012798.jpg

1000012800.jpg
 
I never really used the spot feature… but it was super nice to be able to do it.

Yeah… drilled holes win:
 
I never really used the spot feature… but it was super nice to be able to do it.

Yeah… drilled holes win:

It's probably due to my inability to spend the time to dial in cut settings...
But I don't mind drilling holes of a smaller diameter so this should really speed up general layout.
I can think of many times just having a bolt pattern layout would have been so awesome.
 
It's probably due to my inability to spend the time to dial in cut settings...
But I don't mind drilling holes of a smaller diameter so this should really speed up general layout.
I can think of many times just having a bolt pattern layout would have been so awesome.


I would say less then 2X material thickness and you are going to have a hard time regardless.... At that holes under .5" are not easy on a plasma. I wouldn't beat yourself up too much on it.
 
I would say less then 2X material thickness and you are going to have a hard time regardless.... At that holes under .5" are not easy on a plasma. I wouldn't beat yourself up too much on it.
That does make me feel better actually.
:beer:
 
Do you still have the listing where you bought the trucks and otion pieces. Thinking about building one and trying to budget cost but no pattern of terms I put in the search gives me anything like what you bought
 
Do you still have the listing where you bought the trucks and otion pieces. Thinking about building one and trying to budget cost but no pattern of terms I put in the search gives me anything like what you bought

This is his store.(and mine came from Turkey as well, fast shipping too)

While they work fine I will be replacing them with the SBR linear rails I have bought.
The SBR rails will be more work to get going but will result in a better table I think.

 
When I built the frame it didn't have any gussets and I didn't think it would be a problem. I figured I'd add them later but the way I did the Y axis rails I can't add anything to them or the trucks will hit them, so I blew it off.

I started noticing some weird shit happening when the table would make high acceleration moves around corners etc. and I wondered if that was due to the movement of the table either loosing steps or sliding the material. I figured I better add some gussets to restrain the table from gyrating.

Spoiler alert these gussets made a HUuGGGEEE difference In the table performance, I just cut some parts at 140 IPM where I could not do that previously without obvious cut quality problems.

On the non control cabinet side I was able to get a full X. I used 5/16" riv nuts in the legs.
I made to support rods out of 1/2" EMT conduit. Smash the ends flat, drill holes. Easy, cheap and flexible of future upgrades.

1000012857.jpg


I had some laser cut "squares" from the local short iron store. They made great gussets for the front and rear of the table. Used 5/16 riv nuts for them too.

1000012859.jpg

1000012861.jpg


And lastly on the control cabinet side I put the cabinet right directly in the way of any sort of X like the other side so I did one gusset tube as long as I could get it.

1000012860.jpg
 
Should've put some turnbuckles in there so you can both pull it into square and pretension it at the same time. :laughing:
 
Should've put some turnbuckles in there so you can both pull it into square and pretension it at the same time. :laughing:
Trust me I thought about just using cable to do this but I'm not sure that would be easier than the conduits.

I had ratchet straps on it for a while after I built it but they kept getting in the way I figured I was gonna fuck something (limit switches/wiring) up if I kept messing with them.
 
When I built the frame it didn't have any gussets and I didn't think it would be a problem. I figured I'd add them later but the way I did the Y axis rails I can't add anything to them or the trucks will hit them, so I blew it off.

I started noticing some weird shit happening when the table would make high acceleration moves around corners etc. and I wondered if that was due to the movement of the table either loosing steps or sliding the material. I figured I better add some gussets to restrain the table from gyrating.

Spoiler alert these gussets made a HUuGGGEEE difference In the table performance, I just cut some parts at 140 IPM where I could not do that previously without obvious cut quality problems.

On the non control cabinet side I was able to get a full X. I used 5/16" riv nuts in the legs.
I made to support rods out of 1/2" EMT conduit. Smash the ends flat, drill holes. Easy, cheap and flexible of future upgrades.

1000012857.jpg


I had some laser cut "squares" from the local short iron store. They made great gussets for the front and rear of the table. Used 5/16 riv nuts for them too.

1000012859.jpg

1000012861.jpg


And lastly on the control cabinet side I put the cabinet right directly in the way of any sort of X like the other side so I did one gusset tube as long as I could get it.

1000012860.jpg
If the woman don't find you handsome, at least they will find you handy :grinpimp:
 
On the hole pattern layout, when a plasma hole won't do and I can't cut under and use a reamer in the drill super quickly, there is a dot on Autocad that you put at the center of all drawn holes, cut program will come down and pierce, and now you have a center punch mark for the iron worker to poke out a perfect hole in roughly a second!. I'll stop bragging now. :flipoff2:
 
On the hole pattern layout, when a plasma hole won't do and I can't cut under and use a reamer in the drill super quickly, there is a dot on Autocad that you put at the center of all drawn holes, cut program will come down and pierce, and now you have a center punch mark for the iron worker to poke out a perfect hole in roughly a second!. I'll stop bragging now. :flipoff2:
I'm not sure if I'm doing it right but I can design all the parts with correct hole size but in Sheetcam do a "punch" operation on the holes instead of a hole cut.

You have to check a box when you import the file "drill cross hairs" or something.
 
I'm not sure if I'm doing it right but I can design all the parts with correct hole size but in Sheetcam do a "punch" operation on the holes instead of a hole cut.

You have to check a box when you import the file "drill cross hairs" or something.
I just program a cut path that will just give me a pierce when I need to do that
 
I tried Sunday to get the stepper drivers flipped over to closed loop. I started with the Z drive and was able to connect to it with a USB-RS232 adapter and the current motion software on the stepper-online site.

When I changed it to Closed loop I was getting error 128 position error, looking at the "scope" it seemed to me it was being commanded one direction but the encoder was reading the opposite direction. I flipped the direction switch to invert the direction then flipped one phase of the stepper wires to reverse the direction of the motor. Basically nothing changed in the actual direction. That actually worked and I was able to jog the Z as fast as I have it set in LinuxCNC without any loss of position. I lowered the position error from 4000 to 100 and it still ran great.

BUT when I tried this same method on the X I couldn't get it to work, same error 128.
And while trying to home the Z again after a power cycle I couldn't get the Z to move and even after connecting back to the PC I still could not get it working, I flashed the software back to the default and changed it to open loop where it started working again.

Summary, Closed-loop steppers kicking my ass :shaking:
 
I tried Sunday to get the stepper drivers flipped over to closed loop. I started with the Z drive and was able to connect to it with a USB-RS232 adapter and the current motion software on the stepper-online site.

When I changed it to Closed loop I was getting error 128 position error, looking at the "scope" it seemed to me it was being commanded one direction but the encoder was reading the opposite direction. I flipped the direction switch to invert the direction then flipped one phase of the stepper wires to reverse the direction of the motor. Basically nothing changed in the actual direction. That actually worked and I was able to jog the Z as fast as I have it set in LinuxCNC without any loss of position. I lowered the position error from 4000 to 100 and it still ran great.

BUT when I tried this same method on the X I couldn't get it to work, same error 128.
And while trying to home the Z again after a power cycle I couldn't get the Z to move and even after connecting back to the PC I still could not get it working, I flashed the software back to the default and changed it to open loop where it started working again.

Summary, Closed-loop steppers kicking my ass :shaking:
Haven't fucked with this anymore still running in open loop no issues.

I need to rebuild the Z, it still sux but it also still works.

No complaints, it could be better, a Hypertherm would make everything a lot easier to troubleshoot cut settings but honestly it's not that critical, I look at the chart for nozzle amps then compare that to my spreadsheet for metal thickness and go.

I need more resolution in the axis drives so I could get higher accelerations for thinner metals but I started lowering my nozzle size and amps and slowing down which works fine.

1000015457.jpg

1000014418.jpg
 
Bottom of my water table has never not been bone dry, you worried about osmosis? 🤣
I didn't give this a lot of thought honestly, was going to make a VESA mount to put it on the back of the monitor but that didn't seem to be a great place either.
So this location seemed much better than inside the control cabinet which will cause cable and heating problems.

Thinking about this now I need to put the table somewhere else and use that actual computer cabinet for the PC. :idea:

1730752425853.png
 
I did update LinuxCNC QTPlasmaC for the first time. It didn't seem to break anything but I didn't actually run the machine so who knows.
 
Top Back Refresh