Printable View
Hi all,
looking to build a small table for my plasma, before getting in too deep with plans I have one question...
My torch is a contact start, the head can be dragged or used at a small distance, is this unit suitable?
its an R-Tech 30A unit but as I only want to cut stuff up to about 5mm it's worked ok so far.
It's a bit of a deal breaker so is it any good or not?
I dont know much about Plasma heads but i do remember seeing one being used where the head would come down and touch off the metal and then start. So i think you could use it this way as well ?
.Me
Hmm, just checked the website again, it states it has "HF arc start", this is the unit... P30C Plasma Cutter, 240v, R-Tech Plasma Cutting Equipment So hopefully it will do the job, I was looking at manual torch height for simplicity/cost. Dave
Hmm, to save time and money I thought I'd ring R-Tech as they have been very helpful in the past. It seems my torch would not be good for CNC as it is a contact-start head and CNC really needs an HF Pilot arc. He said it "might" work but the arc may not always strike when told and that would waste the work-piece. Sounds like my project is not going far unless i sell my plasma and upgrade the unit which would blow most of my budget.
Thinking on some more, I could likely build the table first then upgrade my plasma later on to one with a pilot arc, that should work better.
next question is, can this be done without torch height control?
adding this seems to be very expensive and would blow all of my budget on that part alone, is there a manual height option that works?
Upon doing some more web searching, I found a couple of excellent pages... Home Page And Scratch Build scratch build cnc table (plasma and router) - bit-tech.net Forums Both show budget builds and the last one is so amazing that it practically says "Just bung it together and it will work". The first thread uses a drag-torch as well although he does mention it's limitations. So in light of these discoveries I think I will proceed to the fag-packet stage and start drawing up some ideas of frame and motion methods. I'm using 44mm structural aluminium for the chassis as i have tons of it, might run the carriage bearings direct onto the frame rails or possibly fit some plate to each side and use V-rollers top and bottom. My criteria so far are... Cheap, Small, Light, Table-top. I'm looking at T5 belt drive to both sides of the gantry via a main drive shaft with one stepper driving it via a reduction system. Again the X axis will be T5 belt drive as i have a box full of useful T5 pulleys Probably 3Nm Nema23 motors, drive gear from ebay, mach3, sheetcam etc Anyone wants to throw ideas in please feel free.
I seem to have lost control of paragraph formatting????
Its a problem with IE10 & IE11.
Use Chrome?
or something else...
edit: this looks ok in Chrome
Attachment 11280
I lashed together a frame today today to get a grip for the size I can build...
http://www.davekearley.co.uk/LinkPics/CNC1.JPG
Its about 4'6" x 2'6"
im thinking of mounting some 75x6mm steel along the long sides to use as guide rail for the gantry carriages, I'm driving it from both sides so the carriages can be quite small, will be using T5 belt and sprockets to drive the gantry with a cross-shaft along the short edge.
Rough calculations, with a 3:1 reduction on my gantry drive motor, 1 step of the motor will give me 0.16mm of travel, if using micro-stepping as recommended, then 8 micro-steps will give me 0.02mm which seems very fine for a plasma cutter??
If I can reach 500rpm on the shaft, I should get around 16m/min of speed but that's a guess as I have no idea of acceleration curve yet, I'm looking to keep the gantry light by mounting the drive motor on the main chassis and driving through the belt along the side-frames, I think that makes more sense than mounting the motor on the gantry and using to "rack" its way along a fixed belt.
I have two 3.1Nm motors, 3 x 5A drive boards, a breakout board and a 36v PSU coming.
Have also reloaded XP onto my target PC and have been playing around with the demo mach3 plasma install, downloaded a manual too. Just waiting for material to arrive now.
BTW, I have had to reduce the size of my intended machine as it wouldn't fit in my workshop :( so it's now got a cutting area of 2' x 2', not as good but easy to expand later on.
When you say torch height control are you talking about Initial Height Control for setting the start height or Arc Voltage Height control for keeping a stable height while cutting?
It was arc-voltage control I was mentioning. Not a mechanical height setting etc.
i can at least add it as an upgrade now :)
Seeing as your cutting area is small you might get away without AVC if you get your table levelled to your torch.
Excellent, that's a plus point then :)
should be be some stuff coming today.
Touch start really means angling in which is tricky. THC is for Americans who weld up massive plasma cutters with flood tables and then discover that they are so far off square there is no way they can possibly work without one. Then they read about pierce height and cutting height and they are so full of stepper motors that they can't see any other way to do it, usually ending up with the same motor they are using to move a 200lb gantry just to lift the torch.
.
Do you really need mega speed? It is nice if you have to go to the opposite corner on an 8'x4' table to start the next cut, but are you really going to be pressed for time? That long repositioning may just factor in nicely as the duty cycle on the cutter.
.
I think you do need 0.1 G acceleration minimum to cut corners. More is better. Working in G makes the sums nice and easy if you are running low friction linear bearings. If the gantry weighs X pounds then a force of X * 0.1 lbf will accelerate it at 0.1G
Brilliant, so doing away with THC, what do i need on the z axis?
does it just need a manual setting or is it best to have a stepper motor connected to mach3 so it can set piercing height then cutting height plus lift it clear to rapid.
my torch will strike at 0.5 to 1mm and cuts at 0 to 1.5mm or so, up to about 6mm thick, after that it must be dragged.
Speed is of little importance to me.
I am sure it is possible without HF arc starting but I have never stopped to figure it out.
If you are only 1mm above and fixed vertical, where do the sparks go during the pierce?
Is there some software that will add a sideways component?
Torch height control isnt just for large machines or under water cutting. It is there to keep the correct torch height so that the correct part of the plasma arc is cutting the material and giving the straightest possible cut.
Speed is important to plasma cutting. Travelling too fast will give apoor quality cut and you can easily lose the arc transfer or end up goughing the plate rather than cutting it. Travelling too slow will also give poor quality cut and also a wider kerf.
Thanks,
i am looking to add a z-axis after watching some videos on YouTube and playing with sheetcam and mach3.
It it will at least give me the option of adding THC later on with little hassle.
my torch does have HF start but not pilot arc.
You will need this Complete Compact THC MD system ( CNC Plasma Z axis controller ) | eBay, or at least half of it, it works with both type plasmas. Look at the videos in You tube or their web.
As to 0.02 precision, with a build like the one inspired you, you will be happy to stain in the 0.5mm ballpark. But having many alu profiles as you say, you could do far more precise build. Something like the cncrouterparts kits. Or round shaft inserted at the slots, could make cheap and precise rail
You can clock the belts on the long axis via shaft at one of the ends. In fact you should. My current machine is build like this.
Thanks, the proma THC is the unit I am thinking of using when I fit THC, the THC and driver does not seem to be available anymore but the THC on its own is.
0.5mm or better is plenty good enough I think.
Here's the reduced frame with side tracks fitted...
http://www.davekearley.co.uk/LinkPics/CNC2.JPG
And the two Y-axis trolleys...
Attachment 11306
Coming along nicely.
I'm looking into a bigger project but I also want to use my chinese plasma in it. I always keep my torch in contact during cutting. I plan to spring load the torch to follow undulations in the surface.
Yes, someone else has just suggested a sprung loaded torch as well, I am fitting a stepper drive z-axis so I can get the torch clear while rapid moving, a floating or gravity/sprung head sounds a good addition.
Errors accumulate in the build :-) , its not only one thing. I strive to make everything perfect in my build, even knowing that could not be.Quote:
Originally Posted by Davek0974
Also accuracy is ok when depends for whom and for what purpose.
FYI after reading your answer i went in the garage and started precisely measuring the 10mm thick steel elements that ordered lasercut for my build. I haven't done it before as everything fit nice together. Guess what - 0.01 -0.03mm at the most, different from the drawing.
A properly constructed , aligned and calibrated belt driven machine can offer real life precision in the ballpark of 0.05mm. Yes, even with roller bearings and belt. This for the unbelievers. My machine uses V bearings / roller bearings inside- 608/ on supported rail. When i bought it new it was in the 0.5mm ballpark, a couple of days thinking and bettering it and now is in the 0.05 -0.1 mm , i am talking about router not plasma.
So, take your time and strive, you could make a very precise machine if you invest more time and precision right now when building it.
Sound advice.
im trying to build as controlled and tight as possible, but this is a learning machine and never having seen one up close, I have no idea of speeds, acceleration, deceleration forces etc so maybe it will work wonderfully first time or maybe I will need to rebuild bits of it.
one thing I am surprised at is how difficult it is building small - every millimetre of axis travel counts and things keep getting in the way - gantry mounts, trolley brackets and so on.. Building a nice big table would have been much easier but I have nowhere to put it :(
As it would be only plasma machine, you could go light and simple on everything. Take a look at the ebay kits they sell and improvise. The Proma kit they sell at they web also. I pointed it to you as the second piece is the Z driver and will simplify things even further if you have not bought the drivers for the motors. Look at his simple Z axis.
it not my machine, cause the part i want to show you is hidden inside, so i can not make photo of it but that's what i meant for clocking the gantry belts:
Attachment 11344
that is good idea for simple Z DIY CNC Plasma cutter axis rail car for nema 23 stepper motors | eBay
If i were at this point where you are, i would buy a 1 second hand wide type square rail with bearing block or two for around 100-200eur shipped to you from ebay for the gantry, as it will be far better than the other solutions. This wider type of rail will hold a very light z perfectly, move it fast and precisely and simplify a lot the build.
something like this LAW21EL+710-2 Linear Bearing 1Rail 2Block NSK Used Wide type K1 lubricate seal | eBay
That picture is almost identical to what I have built :)
i have designed a carriage for the x axis on the gantry and will start machining it as soon as the aluminium arrives. It rides directly in the gantry beam, pretty much like the one in the eBay listing you added, but mine is made out of aluminium slab, same sort of bearing layout though.
hopefully have it made by Monday as I'm still finishing the gantry trucks and motor mount.
Made the X-axis trolley today, came out lovely...
Attachment 11366
Attachment 11367
The motion is smooth as silk and no twist or slack at all, very happy with that. The front plate is drilled for the motorised Z axis I have already waiting.
Steel bearings running on ali profile? Smooth now but that profile will wear; maybe not as fast on a plasma m/c as no cutting forces but eventually you'll find 'loose' sections...
Yeah I know, but I have a mountain of the stuff :) and it's not going to see production use so we'll see how it goes.
maybe the mk2 will have a steel gantry and welded frame, who knows.
It could be good idea to fashion some kind of cleaning brushes in front of the bearings, or a sponge.
That's possible, maybe some draught-excluder brush stuff.
I have seen a very expensive plasma table and it had small motors with brushes in front all the bearing blocks. Amazingly funny.
Wow, that's a step too far I think ;)
How does it handle piercing?Quote:
Originally Posted by Peter Fouché
my torch will pierce at 90deg but prefers being angled over a bit until pierced.
it may be over engineering it but maybe the torch could be mounted on a cradle that when driven into the work makes the torch tip over and then self-corrects when the z axis is lifted, so maybe it would pierce at -10mm and cut at +1.5mm??
just wondering.
i got the carriage mounted today and fitted the z-axis to it, pictures to follow, looking at connecting my drive belts to the gantry now.
htmlEncouraging discovery, after another few hours googling I found this excellent thread...
http://www.machsupport.com/forum/index.php/topic,20076.0
The guy modified a cheap HF start plasma by adding a ground wire to the tip thereby tricking the machine into thinking it is in drag contact permanently, this gave him full power plasma without hitting the metal and also cut the HF interference problem a lot.
its an inspired idea I think.
As it's starting with full plasma, could I then start my piercing cut at a more reasonable distance??
this needs experimenting I think....
Quote:
Originally Posted by Davek0974
I remember that one. A great idea.
The beast lives :)
wired it all all together today, just temporary lash-up for testing without the torch.
after a little debugging and tweaking motor steps and speeds, it is now running very nicely as a pen plotter :)
better than expected, very fast and accurate so far but will be testing for lost steps tomorrow if I get a chance.
are there any figures to aim for I.e.acceleration G's or mm/sec etc, never having run a CNC anything, it's hard to know what to aim for.
A short vid of the beast playing...
http://www.davekearley.co.uk/LinkPics/CNCVID1.mp4
fun to play with :)
i have the motors tuned to 15m/min , can't remember what acceleration was, this test file was programmed to 3m/min cut.