What CNC

[m_c]

Comment about routers was they have to work with considerable side loads cutting, so wondered if the same CNC would be better with plasma.

Thing with plasma, is you don't need that great strength or accuracy, whereas a decent CNC router table will be engineered for strength and accuracy.
The downside is strength and accuracy increase weight speed, which reduces speed and especially acceleration.

However a decent router table running servos, should be adequate for plasma, provided you're not trying to cut intricate shapes in thin sheet that really needs a sustained cutting speed for plasma.

 

[Hood]

You still need rigidity for a plasma, likely as much if not more than a router, well if running servos with high feedrates and high acceleration as I would imagine you will cut a lot faster on a plasma than a router.

 

[m_c]

You don't really need rigidity as such for a plasma, as there isn't any cutting forces.
You just need enough rigidity so the torch holder isn't flapping around every time anything accelerates hard.

It's a bit like the one arm 3D printers. Some of them will accelerate faster than a router and maintain good accuracy, but they're no where near as rigid as a router.

 

[Hood]

The rigidity is required because the gantry has a fair weight in it as it needs to be rigid enough to carry the X axis motor/gearbox/carriage and also the Z axis motor/gearbox/carriage and rails.
What span are these printers you cite? I suspect they will be relatively small in comparison to that of a plasma, hang a 1700mm arm on it and I suspect it will not be so capable.

 

[James1979]

The rigidity is required because the gantry has a fair weight in it as it needs to be rigid enough to carry the X axis motor/gearbox/carriage and also the Z axis motor/gearbox/carriage and rails.
What span are these printers you cite? I suspect they will be relatively small in comparison to that of a plasma, hang a 1700mm arm on it and I suspect it will not be so capable.

That ESAB crossbow thing has a cantilever X axis. I always wonder if it dips when its out full. Must either be really solid or have a good THC

 

[Hood]

I don't think the Crossbow is that fast, if my memory serves me right they are max 3m/min but could be wrong. I remember looking at them closely when I was about to embark on a CNC build as I thought they would have been a good solution for me with limited space but reports of their performance was often not great.

 

[Brad93]

The ESAB Crossbow is the biggest piece of crap.

it’s only place is in a field or quarry making one off parts. It has no business in a workshop

 

[m_c]

The rigidity is required because the gantry has a fair weight in it as it needs to be rigid enough to carry the X axis motor/gearbox/carriage and also the Z axis motor/gearbox/carriage and rails.
What span are these printers you cite? I suspect they will be relatively small in comparison to that of a plasma, hang a 1700mm arm on it and I suspect it will not be so capable.

3D printers are relatively small, but the point I was making was they do the job they're designed for.

The point I'm trying to make, is a plasma cutter only has to hold a torch in position, so can be relatively lightweight compared with a router that has a cutter trying to resonate the gantry to bits and push it around.
It's like trying to say a CNC router is as good as a CNC mill, as the router has to be more rigid to withstand the faster speeds.
Everything is relative.


Since somebody has mentioned the Crossbow, I have seen a commercial woodworking router using a cantilever design, and given it was from one of the major machinery companies (I can't remember what one!), and likely with an eyewatering price tag, I'd guess it was more than good enough for the job.
Although what gets me about the Crossbow, is it uses a £150 controller.

 

[henry Kadzielski]

Well good morning, I built my machine to do both and have done so. The gantry is now much bigger and heavier than it was in it's initial form. The drive motors (DC servos 850 in/oz) with 15:1 reduction have enough torque to snap off a 6mm tungsten carbide cutter at the spindles er32 collet, don't ask me how I know this . So there is sufficient drive to do routing. I have a 2.2kw spindle and the limit is the spindle power not the linear drive system. My machine is mainly for plasma and will do 19m/min in rapid (I set that speed limit). Since the gantry is now so heavy, speeding it up or slowing it down is a problem with the entire table getting the wobbles on the quick change of direction. Cutting the letter 'S' on thin sheet is a problem as Newton is very heavily involved with this process.

a plasma cutter only has to hold a torch in position, so can be relatively lightweight compared with a router

The best analogy for this is. A plasma is a Ferrari, wheras a router is a draught horse. A plasma needs good speed and acceleration / deceleration to cut the thinner parts well and the best accuracy to expect from a plasma is about .3mm. My total accuracy is now about .3mm over 6000mm. When I get time I will get that down to about .02mm in 6000mm, but for plasma it's ok at the moment. The router on the other hand has to push the tool (what ever it is ) through the job whilst maintaining the accuracy required for that job. When I checked my routed parts they were within .05mm of the drawings, so I thought that was enough. I have some wear now, so have to come up with a method of reducing the lash in the gear racks to get back to that .05mm or better accuracy. A plasma still needs quite good rigidity as it has to hold itself together accurately when zooming around.

One of my customers bought himself a router (2000mm by 4000mm bed size) to cut his own parts out. I have only recently found out that this commercial machine over $100kAu has stepper motors to drive all the axis, and if it misses some steps the parts come out wrong. The service callout for that router is about $400/hr Au and they are not just up the road. I just tear out my hair and scream obscenities till I work out what is wrong and get it fixed. Six of one, half a dozen of the other. There is no perfect machine