Experience with Plasmacam going from exhaust to water table

[Metriccar]

I have had a Plasmacam for the last 6 years. I used to have it in my dads shop and used an exhaust fan blowing the dust outside. I still had smoke coming up through holes and the dust blown out still just swirled around anyways as I had an open garage door. It was a 2000cfm fan and I'd cut a sheet at a time and just left remnants on the table.

I recently moved the table to my garage and made a water table for it. Ohmic sensing does not work on a water table with a Plasmacam despite what people say. The water continuously goes back into the nozzle and shorts it out causing a "torch hit material" error. The machine continues to blow air after it is done cutting and this causes water to spray up into the torch, shorting it out. What I found works is changing the cutter section to "cut from previous cut height." While this setting does not work with the downdraft because the metal always warps, it works great with a water tray because the metal does not warp. In fact this setting reduces the cutting time because it's not going down to find the material before each cut. I also changed the setting from "cut to reduce heat" to "cut to nearest cut" or whatever it's called so the machine spends less time moving around between each cut. I can do this because warping is no longer a concern.


However, there is a lot more dross on the back of the parts now, and consumables last considerably less time. I want to say 1/4 the time of not less.

Overall though, I am very happy with the new setup because I am getting a lot less smoke.

I keep the water about 1/4-1/2" below the plate. There is still some very fine dust particle in the air but it's nowhere near as bad as it was. I also wear a mask to prevent inhaling dust over concerns of long term effects of inhaling dust.

[abmetal]

I wouldn't go from a downdraft to a water table for anything! I tried a water tray for a short while and found it was messy and I still had smoke. I have no smoke UNLESS I have a small piece of steel on the table to cut and the rest (5x10) is uncovered. Then I just throw a couple pieces of sheet of some sort on it. I do realize there are a lot of folks out there that wouldn't have anything BUT a water table but that's just been my experience. Different strokes...

Allen

[acourtjester]

AS with many things if it works for you that is what you use, great thing here is nobody points a finger and states you must do as I say.

[Rodw]

I was really interested to read your comments about ohmic sensing on Sunday because I was working on a more reliable solution for ohmic sensing on water tables that same day. its a bit early to claim success but based on some testing by another user today, we should have it nailed. We just did not have an appropriate components to use it in anger. It seemed to be more of an issue for Hypertherm machines than my Thermal Dynamics. Unfortunately, the solution is currently specific to LinuxCNC but I'm wondering how to make it a more generic solution.

Instead of activating a relay to switch a signal to an input port, we measure the voltage while probing and when it exceeds an arbitrary threshold voltage, we enable a probe signal in software. In our case, we probe to the surface then move away from the surface very slowly until the probe signal turns off. That is the point we set the material surface height at.

The problem is there is relatively low impedance between the torch tip and the material clamp. This means that when a bit of water shorts the electrodes in the torch, near full ohmic voltage passes through the circuit. There is not a sufficient difference in voltage between the on and off states to set a reliable trigger point as seen by the slight step in this plot of the voltage on the right hand side of the graph. You can see its barely enough to overcome the noise in the signal.

waterlogged_probe-event.png

But if we increase the impedance while probing, we can create a 5 volt or so differential which is more than enough to reliably sense the on and off triggers of the ohmic probing event. I found my Thermal Dynamics has a cleaner signal and I could get reliable probing on a 0.2 volt differential so 5 volts is enormous by comparison. We just need to make sure the components increasing the components can handle the current in the circuit when the torch urns on. Thats the easy bit but its dependent on your sensing circuit and its components. here was a plot from today.

probe_with_22kOhm.png

Finally, let me say that I recently installed a water tray to my open bottomed plasma cutter and I found a significant improvement in cut quality and a lot less dross on the underside. But yes, it gets a bit splashy! If I forget to add a park position in the middle of the table, in sheetcam, I loose a lot of water...

Anyway, sorry about the meandering post but it is kinda relevant. Hope you don't mind!

[DXF]

I have been using a Plasma Cam machine for over 20 years with a 1500 CFM dust collector that works great. A water table just never made sense for me as the system I have made works and doesn't create rust or a pond to fish in. I am in the process of improving my system with a 2300 CFM dust collector and other changes. If a water table works for you that's great!

Dave Hanks