New Build-12x6ft Plasma Table

Hey,

I'm looking into building a 12ft x 6ft Plasma cutting machine.

I'm selecting a 12x6 as for I reckon it might be cheaper to buy 10x5 sheets and just in-case I might need the exceptional larger sheet for a job.

The type of machine I'm going to use is one of the machines in the :link:, http://www.engineering-machinery.ie/manual-plasma ,with the straight gun nozzle instead of the trigger gun, http://www.engineering-machinery.ie/manual-plasma/81-hypertherm-duramax-mechanized-upgrade-torches-for-powermax1000-1250-and-1650

I'd appreciate anybodies advice, help or thoughts on this type of build.

What is the best material to use?

What do I use for the rails?

What do I make the gantry from?

I'm gonna be building a 12x6 table anyway as for I'm gonna buy a plasma so why not CNC it altogether.

As soon as I have some rough plans done up I will put up some images. Any extra help with this would be greatly appreciated. :eagerness: :cool: :beer:

Hi

I would go straight to CandCNC for the electronics, they do some great kits and the stuff is first class, I have there MP3000-DTHCII torch height controller and it works great, great support too.

I am at the start of a second build now, this time i'm going with HiWin linear rails and ballscrews but my one is only a 1250x1250mm size, any bigger and i would have gone with rack and pinion drive.

Hi Dave,

Thanks, yes I've seen CandCNC pop up alot when I've been looking around.

Have you any pictures of your own machines?

Gav,

This is my first build - it was an educational exercise that got pressed into mini-production:redface: It worked really well on my Powermax 30XP but that was a hand system and difficult to use on CNC, I then upgraded to a PM45 and now the table cannot go fast enough !

I am now building a new, bigger table to match the PM45.

http://www.mycncuk.com/attachment.ph...id=15631&stc=1

Hi,

I am almost ready with my mixed universal machine which will do plasma also, usually at the back side of the table. That for the moment untill i have time to build me a dedicated plasma machine. So i investigated this and that.

here are my thoughts on building a dedicated plasma machine, especially one your intended size. I can further elaborate on the "why" if you have any questions

Preliminary:
- study commercial designs that size
- decide on a budget and don't cut any corners cause that machine size is not for hobby

Machine:

-make machine and table SEPARATE

-build left side and right side as beams only , supported by legs, leveled by screws that screw on integrated in floor plates

-for left right structures, use sth like 120x120x4mm at least, better would be sth like 140x140x4 at least, best would be beam like 200x200x4. The bigger the better, the longer could be the element, the simpler.

-RP obviously, use proven design with community support , don't go for belt to belt drives following a hype, unless you have made previous working machine, you want top make money, not test ideas . Avoid expensive RP, special shapes, special gear boxes, etc. In other words maybe go for the CNCrouterparts design and i would even buy from them the pieces for a piece of mind

-spend time on the design and make it truly modular, cause one day even this one could be short for you

-rails - on such a machine no doubt i would use at least Hiwin20 rail, but best would be the low wide rail / MGW15 size/ which is stainless by the way , so are the carriages and the balls. Other benefit of using the slightly more expensive MGW is that you could make the gantry rail only 1, same with the Z. One rail only.

-motors- no need to tell you to use servo motors, i hope you have that clear already. you will need the speed for cutting thin stuff without bending. If you have doubts forget them and use servo motors, at least for the closed loop, cause i dont want to think what would happen to a geared , even light, RP gantry flying at tremendous speed cutting thin stuff that suddenly racks due to lost steps. i will tell you- it will be bad.

- ATHC, one thing that is clear to me is it must be "real" , no semi THC with simple arc controls there. meaning it must control the motor on the Z somehow, a motion control. I bought the Neuron THC. Its a new one, more or less is same or even better compared to the CANDCNC , who by the way is not my favorite guy to deal with. If you dig deeper you will see some ridiculous statements made by him, here and there, even at his web, talking about chinese machines and warranty and so...
Now i still can not tell you how good the Neuron works, but i can tell you is made in EU, Andrei is very helpful and knowledgeable guy. I received mine without any hassle and Andrey said at the moment the machine is running we will make Skype contact and he will adjust whats necessary remotely for a first run. What more do you want as a support?

-Plasma is not 100% necessary to be Hyper therm . Now here many will cry, but they are the best. Yeah, but ...if i was buying from USA in $$, maybe. Not with that EU price. First of all consider Thermall Dynamics/Victor or re branded as ESAB and other, here in EU. The magic word is CUTMASTER/ . 20-30% cheaper and original consumables are much readily available everywhere and much cheaply.
Secondly consider German-Chinese machine from eBay with quality control better than pure Chinese and with 5 year warranty. I have such and confirm that the MIG, ARC and Plasma work like a charm. Now change the Chinese plasma crap torch with universal Thermal Dynamics torch like SL60 or SL100 and you have a winner.Still not tested though but soon will report. Now even cheaper if you find Radnor which is the sam
Andrey who makes the Neuron told me in some conversation that many people in Russia change the torches with Binzel torches and all works like a charm.


Hope that helps. I am talking a bout a superiour build, that does not necessary has to be very expensive.

http://img.directindustry.com/images...40-6010713.jpg

wow guys thanks for the replies there great so far.

silyavski I was thinking of something like the image at the end of your quote build my table first use it if I need to then progress to cnc it asap. I'm new to this whole scene but I willing to try anything as long as I'm learning from it and it has something to do with my industry.

I'm thinking of a bench approx 12'6 x 6'6 so I can sit a 12x6 sheet onto it with finger or hook room. Might even place it in a room 20x8 ish with a little overhead gantry to keep it all tidy away.

The table will probably be 75x75x5 square tube and 50x50x5 square tube with a water tray similar to this. http://www.mycncuk.com/attachment.ph...id=15633&stc=1

And the left/right sides will probably be the 140x140x6mm at least. What will I make the gantry out of?

Quote:

---

Originally Posted by Davek0974

This is my first build - it was an educational exercise that got pressed into mini-production:redface: It worked really well on my Powermax 30XP but that was a hand system and difficult to use on CNC, I then upgraded to a PM45 and now the table cannot go fast enough !

I am now building a new, bigger table to match the PM45.

http://www.mycncuk.com/attachment.ph...id=15631&stc=1

---

Is that the Machine on the you tube video??

https://youtu.be/vETkf1sqo3M

No, thats not my machine on youtube.

My one is belt drive not chain as well ;)

Quote:

---

Originally Posted by F1transportationserv

The table will probably be 75x75x5 square tube and 50x50x5 square tube with a water tray similar to this.

---

Hang on a mo', stop drawing and start thinking...
You have to support the slats that support the work piece and you have to support them through the water bed. I don't see slat support. A water table 5' x 8' is 40 square feet so every inch depth weighs slightly over 200 lbs. You have to support that to, add some triangles perhaps.
When you weld this thing up it is going to bend. Suggest you do not weld the gantry supports because you will only end up trying to correct the misalignment with a THC. Best if you correct misalignment with screw adjustment or, let the gantry assembly rest on the table so it conforms.
More if you want it... :very_drunk:

Quote:

---

Originally Posted by Robin Hewitt

Hang on a mo', stop drawing and start thinking...
You have to support the slats that support the work piece and you have to support them through the water bed. I don't see slat support. A water table 5' x 8' is 40 square feet so every inch depth weighs slightly over 200 lbs. You have to support that to, add some triangles perhaps.
When you weld this thing up it is going to bend. Suggest you do not weld the gantry supports because you will only end up trying to correct the misalignment with a THC. Best if you correct misalignment with screw adjustment or, let the gantry assembly rest on the table so it conforms.
More if you want it... :very_drunk:

---

That was an idea i came across for the table, I have this one also http://www.mycncuk.com/attachment.ph...id=15635&stc=1 as an idea separate table to the gantry with maybe a 4 hole base plate for leveling up (or is this too much)??

Wow, some heavy-duty ideas there:)

I was also advised never to weld the thing 100% and i am not doing so now.

The base frame will hopefully be welded and gusseted/triangulated as one piece, right up to the support pads for the motion beams.

The beams for the Y-axis will then be bolted on so they can be adjusted, paralleled and shimmed level to each other and the bed area. Once they are set, the X-axis/gantry will be easy to fit.

That is my plan but my machine is far smaller at only 1250x1250mm cut area. I am thinking of using 50x50x5mm for the whole build including the single gantry beam.

Here is another idea. You will need a torch height control. Once you accept that as a design feature it becomes obvious that the torch needs to be raised and lowered by a DC motor driven by a power op-amp. This hugely simplifies the THC and greatly reduces the weight of the torch assembly which, in turn, greatly reduces the weight of the gantry needed to accelerate it. It also reduces the wiring and without electronics at the torch head it makes you immune to interference from the arc.
If you think of torch height as an obvious job for a DC motor it is not hard to get cut and pierce heights using micro-switches rather than using a computer and vast hardware to do something absurdly simple. There are more ways to kill a cat than building an Arnold Schwartznegger, terminator look alike, cat stomping machine.

Yes I was thinking of an adjustable 'Z axis', the diagram of the table was only a rough idea.

I'm thinking a basic frame 75x75x5 with the braces out of 50x50x5. I'm looking to cut 6 and 10mm mostly maybe some 12mm sheets so I need to build a table braced to support a 12ftx6ft x15mm sheet if need be I will increase the sizes or just make it all outta 75x75x5???

The image of the table I attached had a .dxf file of the slat support designs I must redesign these to suit the size of table I go for and I will price them for cutting in a local engineering shop.

My biggest problem of the whole build is the motors and the rails setup. The fabrication of the table and the separate supports for the gantry is the reasonably easy part?

Do I have to buy the rails as a one piece or can they be joined?? Also how many do I need??

https://youtu.be/vETkf1sqo3M
This lad used angle irons and rollers like a roller gate set up for the liner rails will this system work?? is it a cheaper method??

Thanks again for the replies its great seeing that people are willing to talk and share there views.

You can say 75x75x5 until the cows come home but it still won't mean anything. It isn't what you use it is how you use it. At the moment you are trying to build a bridge without appreciating how a bridge works. More James May, less Jeremy Clarkson please.

It might be worth mentioning that all your problems are down to the idea that it will look really neat if you lift it off the floor. If you sit it down on the floor all your problems evaporate, well, so long as the floor isn't flammable :encouragement:

Quote:

---

Originally Posted by F1transportationserv

Yes I was thinking of an adjustable 'Z axis', the diagram of the table was only a rough idea.

---

If by adjustable, you mean a torch height control, then I would say for a build of your size it would be essential, not just a thought ;)

Quote:

---

Originally Posted by F1transportationserv

My biggest problem of the whole build is the motors and the rails setup. The fabrication of the table and the separate supports for the gantry is the reasonably easy part?

---

One thing I have learnt so far is to take it all into consideration, but build it in parts.

Quote:

---

Originally Posted by F1transportationserv

Do I have to buy the rails as a one piece or can they be joined?? Also how many do I need??

---

Liner rail like HiWin can be butt-joined if carefully aligned, longest bit in one length I think is 3m. As to how many, if you take a look at the specs, they can take a massive load, the 15mm section am looking at can roll a 1500kg load! It must be fitted correctly though with side support etc.

Quote:

---

Originally Posted by F1transportationserv

This lad used angle irons and rollers like a roller gate set up for the liner rails will this system work?? is it a cheaper method??

---

Cheaper, yes, better, no.

I would look at proper linear rails, one per side probably 25mm size, but thats just a guess. Get the carriages fitted with wipers and scrapers.

What i said in the long post before is still valid, so maybe read it a couple of times more, cause you will get there if you think more and more deeply :hysterical:

for example:

-Hiwin 20 size rail/ or MGW15/

If it was ok Hiwin 15 size rail, i would have told you that. The savings from HGR15 to HGR20 are very small if none, but with the small blocks on the 15 size rail you will have much more problems designing the thing.

I was talking that if you design it properly you could even use 4 rails and only 4 blocks in the whole machine. Using the long HA blocks, where each block is 90mm long. And for sure 4 rails and 8 blocks if you dont feel like minimalist design is for you. After all if you buy from China 4 bearing blocks more, it will be only 120 euro more in the whole build.

As you were told, there is no problem connecting Hiwin Rails, especially in plasma build.


-While on a small table speed is no problem, on such a size you need things to be really fast. I would go as far as to say that you need 800ipm real life speeds with serious acceleration there . Cause bad things happen on plasma when deccelerates and you have crappy THC . You will need to cut thin stuff something like more than 10000mm/min

-Again - THC is a must


So , make the machine separate from table. Forget the stupid table for now. Invest money and effort in the actual machine. Later having THC you can support and cut the shit on a basic frame, which once you start earning money can upgrade as per your desires.

How hard could be to sit down and design a modular sides design from that 75x75x5 or whatever. Like at the photo bellow. As i said personally will use at least 100x100x3 . You are talking about 3.6meters long+ 4 legs 1m high each side. But size really it does not matter if you use THC and simply is strong enough not rack the gantry.


http://www.mycncuk.com/attachment.ph...id=15639&stc=1


-simple 1 beam gantry with one rail. Similar to the video before. Forget about the monstrous machines from pictures. they are made to route and drill also, driven by powerful servos, etc...

yes brilliant thanks the previous massive machines (first images i found of the type of design) were only generating ideas ie separate table and rails of that type design.

so whats my next step start drawing? or what do I need to do?

something similar to the image above is what I'm looking to put together.

Also sorry if I seem eager I'm just unsure what way to go with things :)

Quote:

---

Originally Posted by F1transportationserv

so whats my next step start drawing? or what do I need to do?

---

I think that first you have to understand the problem :sneakiness:

Your problem is the change of direction on thin sheet and is all to do with accelerations.

For a first machine you will probably build your gantry out of 75x75x5 and fit enormously powerful motors to overcome the inertia. Strangely you might use the same motor for the Z as you do for the other axes. You might even use 2 motors for the long axis and try to connect them electrically :hysterical:

1 Newton force will accelerate 1 kg mass at 1 m/s/s

Quote:

---

Originally Posted by Robin Hewitt

Strangely you might use the same motor for the Z as you do for the other axes. You might even use 2 motors for the long axis and try to connect them electrically :hysterical:

---

I can see that a plasma machine won't need heavy lifting capability on the Z axis but why does slaving 2 motors for the long axis merit a ROFL, unless you mean connecting two motors with a choc block :whistle:

I find lots of things funny, especially since the blood supply to my brain failed last year. In fact I find it difficult to take most anything seriously :ambivalence:

We must remember that the decision to use 2 motors usually follows a failed Google search for extra long timing belts. You have to be oblivious to the problems it can cause or you would not give up that easily.

How many people check the driver initial step pattern on power up before they shop for this? Do they fit a stall detect system to ensure that both sides fail at the same time? Do they fit one location sensor and pray or two sensors and then decide prayer was the easy option? :hysterical:

The reason I have chosen to use two motors is because I have a belt drive machine with one motor and it suffers belt issues and does not go fast enough, along with the fact that there are many many commercial build tables with slaved motors and AFIK no hardware or software fitted to check for a failed move on one motor i.e. racking the gantry.

With steppers, its a moot point to a degree as they will not do any (or much) harm if stalled, a servo on the other hand will happily destroy itself and/or its driver when jammed. Gantry damage is possible with both systems i guess but to different degrees.

My motors have dual shafts so I guess I could install slotted opto-discs and detectors on each motor and a small logic board that trips the E-Stop circuit if one side does not move - could be an interesting project one day.

Power-up IS however one issue I am looking at, I know my steppers jump into lock when powered up, but what are the chances of BOTH jumping into the same lock position ?? Zero, 50/50 ????

That is the main point I think

Why would a servo system damages itself if it stalls?
I have been working with servo systems for over 20 years and even 20 years ago, you could stall a servo motor for 24 hours without damaging it.
On a gantry system where you have a motor on each side of the machine and a side stalls will not cause any damage to the motor or driver, but could cause mechanical damage to the machine due to one side trying to continue to move what the other is not.
Closing the position loop on these axis will have the added advantage of stopping the motion if this were to happen. (Stepper or Servo)

Quote:

---

Originally Posted by Davek0974

The reason I have chosen to use two motors is because I have a belt drive machine with one motor and it suffers belt issues and does not go fast enough, along with the fact that there are many many commercial build tables with slaved motors and AFIK no hardware or software fitted to check for a failed move on one motor i.e. racking the gantry.

With steppers, its a moot point to a degree as they will not do any (or much) harm if stalled, a servo on the other hand will happily destroy itself and/or its driver when jammed. Gantry damage is possible with both systems i guess but to different degrees.

My motors have dual shafts so I guess I could install slotted opto-discs and detectors on each motor and a small logic board that trips the E-Stop circuit if one side does not move - could be an interesting project one day.

Power-up IS however one issue I am looking at, I know my steppers jump into lock when powered up, but what are the chances of BOTH jumping into the same lock position ?? Zero, 50/50 ????

That is the main point I think

---

When you start up your system or do a new job, surely the first thing you do is home the machine (at shut down you home the operator :stupid:) The initial gantry alignment error will either not change if the steppers lock in the same direction or each or one of the steppers can be up to 1.8 degrees out. At worst this will give 3.6 degrees or 1/100 of a revolution. Divide the screw pitch by 100 and you get the gantry skew. For a 10mm pitch this is 0.1mm . Now you decide can you live with the chance of this until you Home the machine and correct the skew.

I have one machine programmed to home on start up. I lift the tool clear then home X and Y. By zeroing the co-ordinates at the start point and remembering where they got to before I reset X,Y to zero I can then fast return to start point which I do in 2 moves. Full X,Y with half Z, then Z completed.
.
My larger mill has no home switches, I set the work piece to the tool then the tool to the work piece. If I expect trouble I program a location mark and drill a centre pip. If I want accuracy I bolt my tool finder to the bed and home on that, it uses an Omron optical gizmo accurate to 1um which is very silly because my minimum step is 5um :smile:
.
An auto-home can be a right royal pain in the backside because you don't know if you have 1mm or 500mm to go and detectors with overshoot require forethought and planning. Meaning you probably can't whack it in to top gear, full steam ahead and depend on it to stop in time. Also if you have set the thing up and a power glitch sends your tool creeping away to some remote corner and no way to stop it, you can regret this feature.