New plasma build - planning stage

[spumco]

All,

Initial stages of a 5'x6' plasma table; mechanicals are designed & 80% complete, starting to refine the control system.

Dual-Y motors, steppers on all 4 axis, proxy switches at each axis end for limits, and a mag mount with breakaway switch. Nothing fancy. Planning to use the Neuron Simplicity THC and a 300ETH/UCCNC for motion control.

Question - just saw the new UCCB BOB and wanted to get some input from anyone more experienced with plasmas & Neron whether that BOB + the Simplicity will be sufficient I/O?

I gather UCCNC sends signal to 300ETH, then to BOB, then to Neuron for THC and arc on/off & other torch/Z controls. About the only thing I can think of that might require a relay (besides E-stop/contactor 220VAC stuff) would be the pre/post air blast, unless that's handled via the Neuron through it's connection to the Hypertherm 45.

UB-1 was my first choice for this, but that really seems like overkill. Anyone have suggestions or thoughts?

Thanks,
Ralph

[Robertspark]

Hypertherm 45 + Prox sensors do not work on my setup at present (too much interference). Worked fine on the pmx350.... Not 45

Pre and post airflow is all internal to the pmx45

The new board looks fine. Note you will need a 24v and 5v power suppply

[Robertspark]

With the neuron simplicity you will have more than enough io's given it only requires 2 outputs (M3, THC on/off) and 1 input (arcOK) to the Bob, and your Bob does not drive the z-axis which is done automatically by the neuron.

https://www.cnczone.com/forums/attachme ... 99&thumb=1

All the io's for the touchoff (ohmic and or floating head) are all internal to the simplicity too

[spumco]

Hypertherm 45 + Prox sensors do not work on my setup at present (too much interference). Worked fine on the pmx350.... Not 45

Thanks a ton. What voltage proxies were you using that didn't work with the Hypertherm? I was thinking some 24vdc sensors would be more resistant than the typical 5vdc, but I can substitute mechanical switches. The switches are just limits/homing and aren't intended to be anything really precise.

So I'll need a 300ETH, a motherboard, and the UCBB. Already have a 24/12/5VDC multi power supply, so good to go there.

Any other tips or pitfalls to avoid? We'll be using ohimc, no floating head, and a mag mount with breakaway switch.

Thanks,
Ralph

[beefy]

Hi Ralph,

personally, I put a lot of importance on my homing switches being as accurate/repeatable as possible. My reason is for when you lose position in the middle of a cut and need to start again.

Imagine you're in the middle of a cut and something happens that causes the system to lose position. Power failure, reset button, you feel like a beer and decide to shut the table down, whatever. How do you get the system reprogrammed so it knows exactly where to start cutting again.

When I first turn on the machine, I home the table, so the system "knows" where it is. Then I jog to my "Work Zero" point, which for me is normally the bottom left of the sheet I'm cutting. I zero my work coordinates at that point.

Next I check my absolute coordintates values for X and Y at the Work Zero position, and I write them down. You can save them in UCCNC but I also like to write them down. Now, so long as the plate does not move on the table, I can shut the table down, push the gantry and Z axis around, etc, and still get back to Work Zero position when I turn the table back on. The accuracy of doing this is completely dependent on how accurate my home switches are.

Limit switches are a completely different matter and don't need accuracy, but "HOME" switches are so called called because they are how the machine knows EXACTLY where the home position is, and thus where everything else is referenced to.

Food for thought.

Regarding voltage of the switches I've only ever used 5v and never had an issue. But there again I've done all the grounding stuff, and use shielded twisted pair cabling. Nothing wrong with going 24v though if you have a breakout board that accepts that.

Keith.

[thegreatwaldo]

Hi Kieth.
What kind of switches would you recommend for home switches?
Cheers.
Andrew

[beefy]

Hi Andrew,

just stock standard little plastic microswitches, without any of those springy arms, etc. Directly pressing the tiny plastic nib on the microswitch gives best accuracy.

I once put one of these microswitches on my milling machine to test the repeatability of operation using the digital read out. I was very surprised to see it was about 1/100 th of a millimeter. Good enough for plasma LOL.

[spumco]

Keith,

Thanks for the switch advice. I'm used to my mill - the proxy switches are pretty darn good, but if I lose position for some reason I just probe a feature or the fixture - not so easy on a plasma. Your comment about using the home switches as a repeatable reference to the corner of the piece is well-received.

I suppose when I mentioned that I didn't need super-accurate switches I was thinking in milling terms - i.e. 0.001" or less repeatability. I don't think it's reasonable to expect that sort of precision on a home-made gantry machine, but I will need to get it fairly close as I'll be using the switches to home and auto-square the gantry.

-R

[thegreatwaldo]

Hi Keith

Thanks for the info.
Cheers

Andrew

[Robertspark]

What voltage proxies were you using that didn't work with the Hypertherm? I was thinking some 24vdc sensors would be more resistant than the typical 5vdc, but I can substitute mechanical switches. The switches are just limits/homing and aren't intended to be anything really precise.

So I'll need a 300ETH, a motherboard, and the UCBB. Already have a 24/12/5VDC multi power supply, so good to go there.

Any other tips or pitfalls to avoid? We'll be using ohimc, no floating head, and a mag mount with breakaway switch.

Thanks,
Ralph

Proxies, 24v, the bit everyone forgets is proximity sensors have a leakage current, hence you should have a parallel resistor arrangement of you common up a few sensors to one port or the voltage on the Bob input pin can appear high enough to be a digital high state

Switches are better than electronic sensing in that there is no leakage current potential and the action is mechanical and positive. Allow for axis deceleration though at limit switches from rapid motion to stop.

Ohmic is ok, not so good with thin plate or anything coated (or a water bed obviously), also if you get crud inside the cap or on the cap it can give a false reading, where as floating head is a positive mechanical action again

Acceleration is key in design of the table, including z axis (lightweight), good step size to minimise jolt / jerk (mm or in /s/s/s

Do you have any target numbers (rapid velocity, acceleration, axis stepper unit, what units)?

What do you intend to cut (material, thickness range)