plasma table idea

[robertspark]

In the interests of shooting the breeze

I've had a plasma for a while (hypertherm powermax 350, don't think it's regarded as a chepo), but think that I've always known what was missing was a good cnc xy table.

Plasma I have always found to be great and easy to use, but unless you've got a steady hand and can remember what you did the last time you used it with regards to cutting speed, I never seem to be able to get a good edge on whatever I'm cutting (guide or no guide), hence have thought about making a simple CNC table.

I saw this on ebay and thought it was a good idea http://www.ebay.co.uk/itm/161150918726 , considering space for me is a premium, hence something that could fold away is always a great idea (Triton workcenter is another great item, but not really LR related (unless you want one in wood)).

Thoughts?

Anyone tried anything similar as a potential fold away cost effective (read: cheap + does what it needs to with fair precision) idea?

Note: I don't think I'll be buying one, as I thought I could make one up with a weekends of effort, or something similar, and it offers the potential to change out the rails in the future if I ever needed to cut full sheets.

Instead of rack and pinion drive, anyone used anything else with fair precision for a plasma (ballscrews / acme seem overkill, + potential for grinding paste in the screw never appealed to me considering the amount of steel blown about by a plasma, may use ACME on the z though for THC).

Anyone tried chain drive? Not quite the same as here http://www.youtube.com/watch?v=vETkf1sqo3M but, welded track of chain with a drive sprocket on the steppers (chain type R&P)?

Thanks in advance

Rob

[simonr]

This is something I've been thinking about for a while! I was considering it as something I could produce as a kit.

The conclusion I came to, like you, is that you don't need particularly high precision - somewhere between 0.1 & 0.5mm is likely good enough.

The conclusion I came to was to use a laser cut frame with the teeth for the racks cut in to the frame, driven by a small 20 tooth pinion, directly couples to each stepper motor.

With 200 step per rev, this gives 94.2mm per rev based on a Mod 1.5 pitch - so about 0.5mm or 0.125mm with 4x micro stepping which is plenty.

Cutting the racks into the frame removes some of the most expensive components.

The rack would be cut in to the upper side of a slot and the carriages would run on roller bearings on the lower edge. With a couple of folds in the rail plates, it will be rigid enough.

Use one stepper motor for the X axis, connected to the gear on the opposite rail by a long shaft - reducing the stepper / drive count by one. Make the frame bolt together with wing nuts so it can be easily dismantled.

I saw a neat idea for controlling the plasma. You have a torch height control. This used a stop on the torch height arranged such that when the torch is lowered sufficiently, the stop presses the button on the torch, firing up the plasma. Thus it can be used with any old plasma without interfering with the electronics.

I'd figured on making something with a 1m x 0.7m cut area - but make it so you can clamp the frame to the sheet itself so you can cut any size sheet you like in small chunks. The way to do this accurately is to cut a couple of reference holes in the sheet which the frame bolts to as a reference when you move the frame to the next area. Thus, with a small table, you could cut a whole chassis!

The best idea I've seen for the surface under the sheet being cut is just to use an old pallet. Sure it gets burned a bit, but they are cheap. The frame dimensions were designed so the whole thing fits within the confines of a Euro Pallet 1x1.2m. An addition to that was someone who painted the pallet with cement which largely stopped the burning.

Somewhere I have a set of drawings for the major bits - I'll see what I can dig up. One day I'll actually get around to building it!

Si

[robertspark]

Si, I'm surprised that someone a not come up with a simple average accuracy (good enough for plasma) kit either.

To be fair to the kit of rollers etc on the ebay link above the bloke came across a simple idea, but it's still a little high for my pricing considering you need the drive mechanism (rack, screws, threaded bar (although not recommended)), steppers, drives, powersupply, PC interface given most of us have an old pc somewhere you can forget that bit, + MACH3 or one of the linux ones).

I've seen this interesting drive arrangement too, which is basically a chain + sprocket drive in reverse (self explanatory in the photos), although I think buying it from nexus is expensive.... again looking for a reasonable accuracy drive idea.

http://www.ecnmag.com/articles/2011/07/advanced-roller-pinion-technology-provides-increased-accuracy-motion-control-applications

I was kind of looking at a 1.5m x 1.5m frame, which once you loose a bit with the carriages will end up at about 1250 x 1250 (i.e. 1/2 sheet size).

I guess cutting the rack into the frame gives you the issue of wear, and what do you do when the rack is worn (+ any backlash adjustment??) .... but if the rails were cheap (i.e. easy to make .... thinking chain drilling type here, what's the loss for a DIY'r just changing the rails once in a while).

Like welding a standard chain to a 50x50 HS.....

Rob

[CwazyWabbit]

How about a coarse threaded rod instead of a rack? Use the threads as a rack, when it wears turn it a little and when it's all knackered buy some more rod.

[simonr]

That's a clever idea Mr Wabbit. You would really need helical gears to drive against it - but you could use a thin plate gear instead.

Si

[robertspark]

How about a coarse threaded rod instead of a rack? Use the threads as a rack, when it wears turn it a little and when it's all knackered buy some more rod.

Yup, it's been done, the problem tends to be the complaint of backlash .... next time you've got a length of studding (available in up to 3m lengths), spin a nut on it and wobble the nut about.... fair bit of backlash which MACH3 (cnc software) can compensate for, you can try to preload the nut, by using a second nut and a spring between the two (not explained very well), but a picture below may help.... problem is I think the threads are going to wear fairly quickly, given the metal particulate from the plasma which seems to go everywhere (given it's compressed air propelled is not surprising).

Thought about making or using a long nut (such as a studding joint nut), again, didn't think it would work well enough.... hence sounding out ideas

http://muc1.files.wordpress.com/2010/09/abl_6_sol.jpg

[robertspark]

That's a clever idea Mr Wabbit. You would really need helical gears to drive against it - but you could use a thin plate gear instead.

Si

Err, I think Mr Wabbit was on about driving the studding (threaded rod) with the stepper, and welding a nut to the carriage...

http://www.youtube.com/watch?v=b1W4EbMM3ig ... sort of..

[CwazyWabbit]

Nope, Simon is right. I meant the studding was held still and you used a small section of the circumference to run a gear on (gear driven by stepper motor), as that section of the rod wears away from the gear running along it you would manually rotate the studding to an unworn section, when all the circumference of the rod is worn just replace the rod.

My thinking is threaded rod is relatively cheap so would just be looked upon as a consumable.

[simonr]

I bought some M16 stainless studding - and the quality is outstanding. More expensive obviously, but very much better than the usual stuff.

Si

[Cynic-al]

A friend has one that he bought for around £4k. It uses a machine head which is straight so easy to clamp and can be triggered by a relay. It used Mach 3 to control it and had a none contact proximity sensor for the height so it doesn't drag on the surface. The material was supported on steel plates on end which obviously become damaged over time and water under it which I guess is just to catch the sparks?

For the motion you could use timing belts with stepper motors, they should be far enough away from the heat.

[robertspark]

Nope, Simon is right. I meant the studding was held still and you used a small section of the circumference to run a gear on (gear driven by stepper motor), as that section of the rod wears away from the gear running along it you would manually rotate the studding to an unworn section, when all the circumference of the rod is worn just replace the rod.

My thinking is threaded rod is relatively cheap so would just be looked upon as a consumable.

Interesting, I stand corrected, although it sounds like a load of work making a meshing gear for a large pitch length of studding (M16 = 2mm pitch)

For the motion you could use timing belts with stepper motors, they should be far enough away from the heat.

That sounds like a go'er .... HTD pulley, coulple of idlers, and a belt cut to length acting as the rack, I'm thinking M5-25, with a 20 tooth gear, 0.5mm resolution at full step (200/rev)

better images on here http://www.cnczone.com/forums/plasma-edm-other-similar-machine-project-log/16634-plasma-table-build-log-including-free-plans-36.html

[CwazyWabbit]

Ressurecting this thread again what size stepper motors would be sensible for this if you were to use the timing belt direct drive method? I suppose that is a how long is a piece of string question given you need to know things like how fast you'd want it to accelerate and gantry weight etc .....

[robertspark]

Well, my CNC plasma cutter has moved on a little, I'll add a full list of specs of what I'm running tonight.

I've gone with a toothed belt drive and direct driven timing pulleys.

Speed on steppers is a little bit of guess work, as steppers can be over-driven and go out of sync / miss steps quite easily. The main thing besides the steepers is the drives and more importantly the power supply.

(Before I get shot down by a passing expert, I am no expert just a leaning backyard mechanic, who reads a lot and can fabricate what I need, so each to themselves please, I can only say the way I looked at it and did it).

Being that space is at a premium I looked at various solutions and wanted the most adaptability to maybe use other heads (not just a plasma), and make it easy to assemble / disassemble and store away when not in use. From the cheap [err cost effective!] .... drawer slides and MDF .... to rack and pinion + fully welded rigid table ....

And ended up with a set of carriages that run on 2"sq RHS (box), so I can (space permitting in the future [next house]) extend the table as big as I want, and an old PC case for the drives + power supply etc + the bits were cheap [... err cost effective for my needs + tolerances and specifications ... not to make a business out of it [yet] ]

Like everything it's work in progress.

[Cynic-al]

a friend wanted a saw for cutting sheets of timber but every tool he makes has to fit in his garage with his car so he got a circular saw and some sliding door runners from a car boot and made it against the wall at a slight angle so you can lean the sheets on it and pull the saw across.

Would this work with a cnc plasma to save space? I think the problem would be this sheets not staying flat but you could hinge it to increase the angle?

[CwazyWabbit]

Well, my CNC plasma cutter has moved on a little, I'll add a full list of specs of what I'm running tonight.

.....

Look forward to it, sounds like my kind of build.

I'm thinking of making one out of box that's bolted together so I can extend it easily if needs be and also knock it down for storage.

[robertspark]

steppers, I've used 4x NEMA 23, 2.3nm steppers http://www.cnc4you.co.uk/index.php?route=product/product&path=83_84&product_id=362

yes they are probably overkill, why? ... because I may want to use them on my milling machine / lathe (which is what I started off using them for).

They have way more torque than is required for a plasma, but being adaptable, I may want to change the head for a rough milling machine to give it a go.

Also they are 6 wire steppers so offer a whole load of flexibility when wiring them (for speed or torque) and they are double shaft, so I could put a handwheel on the rear (not done yet).

power supply, 600W / 42V and another at 24v for the low voltage stuff (breakout to steppers & torch height control)

stepper drivers , TB6600, 5Amp http://www.ebay.co.uk/itm/TB6600-5A-CNC-Single-Axis-Stepper-Motor-Driver-Board-Controller-/231265858708?pt=UK_Computing_Other_Computing_Networking&hash=item35d884f494

breakout board, loads of them about very cheaply, : http://www.ebay.co.uk/itm/5-Axis-CNC-Breakout-Board-For-Stepper-Driver-Controller-Mach3-Cable-SYUK-/360895523551?pt=UK_BOI_Industrial_Automation_Control_ET&hash=item54070caadf

Torch height control: Proma Torch height controller.... think of it more as a torch voltage feedback circuit that basically tries to keep the torch height constant, based upon the voltage that is sensed at the head within a dead band .... it's not as scientific as a hypertherm THC system at several thousand, but it depends what you want to do. The common problem that is often highlighted is that a THC like the proma can put the head into a dive when you need to cut a tight corner (as the metal disapears away the voltage increases), the expensive THC's read the g-code and compensate for this, if you're using Mach3, it can have a setting enabled that ignores the THC feedback (either torch up or down signal) when it sees a g-code for an arc and keeps the z axis constant. It could be classified as a hack workaround, but it works with some success.

The carriages I got from the link in my first post, simply because I did not have the time to prototype things (and I was wasting a load of time drawing things up).

I thought about an inclined table, but the problem with metal is distortion when you cut it with heat... more so when you're doing complex shapes with corners, so it the table is inclined there is the possibility of the metal rising up and catching the torch head (personal opinion... not tested).

the carriages that I bought arrived well packed, quickly and were very well made and rigid when assembled. Yes I could have saved a bit of cash prototyping it but what was the point as my time is valuable (trying to build my 90 + have a young family).

I went with the timing belt kit, and thanks to Simons post above, selected a 5mm pitch pulley / 20 teeth, so that with with a steeper with 200 steps / revolution, on full step, gives me a travel 100mm per revolution, so on full step, the resolution is 0.5mm, which on a plasma is good enough I'm likely to ever do. On micro stepping the resolution can be much improved, but then the speed suffers because your controller may be sending out signals (steps) to each of the drives simultaneously.

(I also have a smoothstepper, but that is another story).

I bought a roll of 5M (5mm tooth pitch) timing belt (5meter roll) with wire internal + rubber external covering, so that I can get around a 1.5 x 1.5m table (2 X tracks and a Y track) so cut about 1.66m each, hence no wastage.

Because of the 2" RHS tubes its very easy to partially disassemble and store + extend / cut if you need it smaller, as my local steel stockholder has a good offcuts section.

I search for a lot of stuff on ebay.co.uk + .com or .au etc as some of the Chinese sellers just list their stuff on one of those websites and not .co.uk. Stuff takes a while to come in but i'm in no hurry and it always arrives (not lost one yet, touch wood). Yes you pay for what you get, but it's a hobby for me, and it will probably outlast me for what duty / workload I'm likely to give it) [each to their own please]

Stepper sizing, there is a huge amount of info on the web + calculation etc, I just settled on some that would suit my milling machine, which had a running current requirement of 4.2A (parallel), 2.1A (series) or 3.0A (unipolar).

In short:

Parallel stepper wiring arrangement, sort of a 1/2 or both speed and torque (best of both worlds)

Series stepper wiring arrangement, sort of high torque, half speed

unipolar stepper wiring arrangement, sort of half torque, high speed.

bit to read here: http://www.probotix.com/stepper_motors/unipolar_bipolar/

I figured that a higher powered stepper, 6 wire would give me all the flexibility I needed as on a low torque application (plasma) I could have a high travel speed operation, and if I chose to use the steppers on something else I could change the wiring config later.

With regards to the voltage for steppers, it's basically the voltage that gives the steppers the high speed as well as it's high frequency pulses that drive the steppers, hence think of it as alternating current, hence the inductance plays a large part on resisting the current going through the windings, hence if you increase the voltage available but use the stepper drivers to limit the current flowing into the steppers they are not over powered. Hence the power supply is 42V at 600W (so about 14 A), and the steppers only require about 3A (unipolar) [high speed, 1/2 torque, near flat profile], and it will drive all four of them no problems without too much ripple etc. [that was my viewpoint anyway]

[CwazyWabbit]

Cheers for the very comprehensive reply

I shall update here when my build is underway, hopefully.

Are you pleased with your results so far?

[robertspark]

Yes, happy with it so far.

If you have time on your side, I would suggest putting a few automatic searches for things that come up from time to time on ebay as you may get something sold under auction and not buy it now (I got my THC that way), you may get steppers and drives too (note the costs seem to have come down a huge lot as the market has become flooded and more people are building their own (or attempting... hence you may get an abandoned project you could build on).

utube is the best place for ideas and options on what others have done on a well trodden path.

[zim]

What sort of price are you guys looking at to makes small table ?

G

[simonr]

I like the torch height control - that's one of the most difficult / normally expensive bits, but you've found a good solution!

I've used a smooth stepper on my mill - and though it has occasional issues, on the whole it's very good! The main issue is if I switch on the coolant while a program is running, the smooth stepper stops taking data and the mill stops. The only way to make it run again is re-start Mach3 - so you loose your datums. It's not an electrical noise issue (though I had a lot of them too), it seems to be a feature of the driver software. Having said that, I'd use another one!

Looking forward to photos!

Si

[CwazyWabbit]

What sort of price are you guys looking at to makes small table ?

G

That's a really difficult question tbh.

The size of the table doesn't affect the price that much, obviously you need the extra steel but that's the cheap bit in the build.

You need the following to build a table (I'm still in the collecting bits phase so I'm bound to miss something) ....

PC with parallel port and 32bit windows ---- 30 pound ish ebay

Stepper motors/drivers/power supplies ---- 200 - 300 pound kits from china

Belts about 5m of 5mm pitch ---- 20mm wide is from 5 - 16 a meter on ebay

Pulley for belt x3 ---- 7.50 each

carriage kit from ebay ---- 230 quid or build them

THC ---- 170 quid or build one

Limit switches ----

Cabling ----

Case for drivers etc ----

earth spike if you have HF plasma ----

Mach3 licence ---- about 110

Sheetcam Licence ---- 110

Metal to build ---- No idea as I have managed to get hold of some scrap box section to build mine from.

I'm bound to have missed some bits (obviously you need a plasma cutter as well)

EDIT: Still looks cheap in comparison to a plasmacam at 10,600 + vat which also doesn't include plasma cutter or pc

[robertspark]

CrazzyWabbit has got it right.

I would have said allow about £750 to £1K. The problem is time..... I know Zim has a bit with your on rig / off rig lifestyle, but some stuff is probably just better bought.

There are schematics for stepper controllers, but for a £10 a piece, I just could be bothered.

note china focuses heavily in my build list (stuff takes 3-4 weeks and touchwood always arrived, 6 at the most).

the THC is another one... if you search for arduino and THC on google, you'll get hits, but again by the time you're prototype'd it, put it in a case ad wired it..... I'd just buy one (I was lucky one came up s/h on ebay....)

You could save money one the carriages, but I wasted a load of time drawing something up and getting it to appear workable, then just thought my times more valuable than re-inventing the wheel, hence bought them.

Make sure you get a few mushroom emergency stop buttons about too for when things don't look like they are going right (i.e. know where they are in arms reach).

I'll get some pics sorted out .... got to find somewhere to lay it out .... rolling chassis in garage finally with transfer case being built on rear.

I'll need to be using it in the next month so it may be a little while when I start of the gearbox to transfer marriage.

Software side, there are free programs about (and free stuff via Linux operating systems), some of the demos are just limited to a number of lines (such as Mach3).

[CwazyWabbit]

Good point about the software, I'd forgotten about the linux offerings.

[zim]

So which one of you will I be sending a cheque to ? ? Cos that's a fair bit cheaper than what you can buy off the shelf.

[robertspark]

the £750 to £1K excludes the plasma and is just the cnc bits

I'd quite happily sell a pack of bits I just lack the time to put the things together (too many parallel projects)

The main advantage of the 2" RHS carriages is the rails take up a load of space in shipping, and lets face it, you can beg borrow of steal 2" box section.

there are just so many companies competing for cnc stuff now the market is swamped (not being the business type, can't see a market], and you just need a clear build list of what you actually require [steppers, drives, pulley's & timing belts / pinions & rack, breakout board, old PC, power supplies ... software of your choosing].

There was one here on ebay a little while ago http://www.ebay.co.uk/itm/Mobile-cnc-plasma-cutter-kit-80-finished-1m2-cut-easy-to-make-bigger-/201172390618?pt=UK_BOI_Metalworking_Milling_Welding_Metalworking_Supplies_ET&hash=item2ed6cf12da&nma=true&si=N%252FDLdYDUOD7P0h6JLHrnQ50ANTM%253D&orig_cvip=true&rt=nc&_trksid=p2047675.l2557

and utube clip: https://www.youtube.com/watch?v=9gXOeYN6pYE&list=UUpgmzq6hO93weh8GFuvsYoQ