[SPENG]

I own a cnc machine shop and I'm looking to build a cnc plasma table with some pretty wicked fast speeds and extremely high accuracy. That said...If I were to start with the Z axis, would any of you be interested in buying one?

[plain ol Bill]

What kind of speeds are you thinking of anyway? I find it hard to believe you can get extremely high accuracy from a process that use's an electric arc basically and compressed air or gases to blow molten metal out of a kerf. But then again it depends on what your definition of extremely high accuracy is also. My home built table can cut thin sheet metal at 240 inches per minute and 1/2" plate at 35 IPM and if I cut a 3/4" diameter hole with it you have to screw a 3/4 bolt into that hole.
Not trying to discourage you from doing it just wondering what would make your proposed machine different from anyone else's.

[SPENG]

most cnc plasma tables have sucked with holes. Now I fully understand that a "normal" hand type plasma even with a machine torch isnt going to produce something like hypertherms truehole. I have built a few of these plasma tables in the past and now I need one for my shop again (even though I own an omax waterjet).

Positioning is something that is pretty important with cutting arcs and circles, if you dont interpolate correctly in both axis, well those holes come out all jacked up. Much of the improper holes can simply be fixed with good motion control which I fully know how to do easily. IE removing backlash, being able to have encoders to track the position etc.

Ideally I expect a .005 repeat-ability/accuracy in motion control regardless of the torch accuracy. I just need the best advantage for clean round holes.

[SPENG]

Attachment 255164



Give you an idea of "some" of the "toys" in my shop...

[BTA PLASMA]

You don't have to re-invent the wheel there are solutions out there like our Star-Lab that are already working with accuracy far beyond those that were made years ago and still some made today.

You could build a machine with blinding speed that could take a 10hp servo on each axis, rapid at 10000ipm and still not be better than what is out there for less than those components cost.

You are absolutely correct in that you need to have positional and linear accuracy to start with before you even add the plasma. Too many times you will see folks running flat bar sold rolled steel accurate to "/- .020 over the length of the table. Showing part they have cut. Put those parts under an optical comparator and then next to a print and you will see where it appears the plasma was just surfing outside the toolpath. Hypertherms 'true hole' uses different gasses than you would find in air cooled plasma cutters. It also requires their logic to activate the selenoids, ramp up and down the air, cut current and speed. Could it be integrated on a Hypertherm plasma cutter without buying their complete system? Sure but you would spend more time and energy figuring out the logic portion and how that would integrate with their plasma units that you would spend on a system that integrated their control. They aren't #1 and maintain #1 status by not keeping innovation rolling.

The question is ..What kind of end accuracy are you looking for? Are you looking for a CNC laser at the price of an entry level plasma? Most folks are they want a CNC laser for $1 dollar in their shop.

[SPENG]

No thanks... I can build something in my shop for far less money. I already ordered enough components to make 10 z axis assemblies from good ol china. Once the parts are in I will take some pictures of the process.

I purchased the HPR260XD from hypertherm today.

I didnt post up to have people tell me not to do what i plan on doing, i posted to gauge interest in purchasing some z axis assemblies that im making extras of and possibly entire tables since im ordering everything in bulk.

but thanks for shamelessly plugging your system....I WILL NEVER go back to a HOBBY type cnc plasma table setup again...just for the motion control aspect alone...

now if you are looking for a real high end shop to start making/engineering components for your own tables..shoot me a pm and id be happy to make your parts for you. Hell shoot me a pm just so I can send you a list of equipment in my shop

[BTA PLASMA]

I think you took it the wrong way. You don't need to make parts for us the picture of those machines are what its like in one corner in our machining area. I commend you for trying to do something better or at least different though.


Shoot some pics along the way, folks here love progress. Also shoot some pics of the HPR260 when you get it in.

[SPENG]

perhaps I did take it the wrong way but the simple fact remains is this..most out of the box cnc plasma solutions arent much help and require TONS of modifications out of the box to get them to be "reliable"

[BTA PLASMA]

Yes, some definitely do

[dnelso]

What are you going to use for electronics? What price range are you shooting for for the z axis. Thanks DN

[SPENG]

I know servo is the way to rock and roll but I am more inclined to use planetary gear reduction boxes for all axis except the Z and use closed loop stepper motors. My omax waterjet has stepper motors with a magnetic encoder strip so a closed loop stepper system might be a decent option.

Price range for the z axis will be in the $500 shipped with a plasma torch mount. based on what I've seen out there, $450+70 for the mount plus shipping.

[dnelso]

I have built several tables with steppers and have no problem with backlash or repeat-ability. I use rack and pinion with belt reduction. Any of the gearboxes that i looked at were expensive and i think overkill for the job. I am thinking of building a new table with a HPR130XD .I am wondering what electronics you will be using to control the motion and run the cutter. I have used CANCNC controls on all the tables i have built and have had very good luck with them.

[SPENG]

belt reduction works decently. my omax waterjet uses belt reduction only on a much more massive scale and its all enclosed with boxes

the reduction gears can be had for under $300 each. thats not a lot of money in the grand scheme of things...

electronics wise i am leaning towards the kieling brand closed loop steppers with the candcnc thc. I cant justify the price of the full candcnc package, i feel its overpriced. They can keep that plastic box...

Ill post some pictures of my waterjet here in a few so you can see how a high end machine is built.

[dnelso]

I have tried the 300 400 dollar units and was not happy with the longevity or the accuracy. I have cnc machines and make all my own parts for the tables i have built. I don't try to build or design things that can be purchased for a fair price. I buy the z axis i currently use from velox because for the price its not worth my time. I cant build electronics and mach 3 screens for what candcnc charges. I think your biggest challenge will be to get the electronics to survive the high freq noise emitted by the hy def plasma units. Keep posting as you progress.
DN

[SPENG]

point me to what you have used in the past so i can stay away.

[Torchhead]

Closed loop stepper is an oxymoron. You cannot force steppers to increase torque to over come lost steps. Because steppers LOSE torque with RPM adding in more steps does lttle They are by definition a constant power device because the driver limit the current. Adding in steps just makes the situation worse. An encoder on a stepper will detect lost steps and you can make it stop motion if the loss is greater than a preset. You CAN make a stepper increase its torque by decreasing its speed but you have to do that on ALL the axis at the same time or you lose coordination. True closed loop needs to be a lot more granular than what a stepper can provide. By the time you add encoders, control circuitry and firmware to a stepper to try and make it into a servo you might as well use a real servo and driver circuit. The key to a good plasma Z axis is it needs to be tight (no lateral slop) , have a good rigid torch mount, have the option for a Floating torch mount (with swtich) and the proper leadscrew pitch to match the motor RPM. Bigger torches like for the Hypertherm 125 or for the HPR series need bigger mounts and heavier components but the requirements are the same.

TOMcaudle
www,CandCNC.com

[SPENG]

i disagree with you on that.


a quality servo system is not going to be the same price as a stepper system that has encoders.

[Torchhead]

Sounds like you know best. Can't wait to see what you come out with.

[SPENG]

I think ill commission some friends to whip up a standalone THC unit.

[Torchhead]

Ok .....I'll wait....