[Normsthename]

Hi
I have had my homebuilt Plasma up and running for a few days now, and I am trying to get to grips with all the different cutting parameters.
I have a Hypertherm 1000 with fresh Finecut consumables and it is using a Campbells THC300 Torch Height Controller.
I definitely have clean air, because by luck I did manage to produce a good cut once......but in my excitement I did'nt note the settings....DOH!

Two problems I am seeing a lot are 'Wavy' cuts and not great cut quality (See Pictures)
The material is 16swg 304 Stainless Steel.
I ran a load of test parts to find the best cutting parameters for the material as follows.

Internal Holes
40amp 47.25" (1200mm) per min at a cut height of 0.25mm

Outside Profile
40amp 138" (3500mm) per min at a cut height of 0.25mm

I am using a homebuilt Water Table, and have the water about 1/2" below the surface of the material.

Would cutting with Nitrogen improve the cut quality, and if so by how much?

Any ideas would be much appreciated.

Thanks

Andy

[Torchhead]

Those notches are not from using shop air. If the cut gap is staying constant with the THC the nodes are probably from a setting in MACH. Make sure you have the CV enabled and CV Feedrate turned OFF. Watch the torch. If you are seeing ragged movement and hesitation in the cuts then it's a software/setup issue. If tuned right and it should move smoothly through the cuts.

Nothing like starting out on material that is a challenge to cut. Learning your machine by cutting mild steel is a lot cheaper and less frustrating way to optimize the cuts.

Cutting at .25mm? The charts call for .063 inches for cut gap which is closer to 1.5mm.

That's some pretty ugly cuts for a 40A tip.

[Normsthename]

Thanks Torchhead

Cutting at .25mm? The charts call for .063 inches for cut gap which is closer to 1.5mm.

I am using Finecut consumables and the book lists 0.010" (0.254mm) for Stainless

Nothing like starting out on material that is a challenge to cut. Learning your machine by cutting mild steel is a lot cheaper and less frustrating way to optimize the cuts.

I did'nt think there was much use cutting mild Steel when I will be cutting Stainless Steel 99% of the time.
I thought that if I got it cutting good on mild steel, I would then have to start all over again on the Stainless Steel???

Make sure you have the CV enabled and CV Feedrate turned OFF

I am using Mach3 Version 2.60
CV mode is enabled, but CV feedrate was turned on.
I also read last night that people who use the THC300 with thin material turn off the anti-dive for the THC.
I will try all your suggestions.

Thanks

Andy

[jimcolt]

It looks like you have either a height issue or an issue with rough motion. Remember that the plasma torch is burning at a fixed power level....whenever the height control makes a correction....expect to see some sort of effect...usually a noticeable notch in the edge of the plate. For the same reason...if there is any roughness in motion...could be mechanical or part program or electronic.....then sudden changes in x and y velocity will create the same effect. If the velocity slows down...the plasma will just burn a wider kerf.

First step in trouble shooting is to dry run through the part program...put your hand on the torch body....do you feel any roughness. If you can feel any roughness during motion...it will show in the cut. If it feels relatively smooth....then do this test again while cutting a part (plasma on, THC on)....if it now feels rougher than before...either quick z axis corrections or velocity changes/mechanical stiction....then you may have to adjust THC sensitivity or you may have issues with electrical noise interference that only ocurr with the plasma on.

Jim Colt

[Normsthename]

Thanks Jim
I will try your suggestions.
I tried doing a dry run yesterday and touching the torch. It seemed smooth enough to me.
I will try again while it is cutting.

Thanks for the input.

Andy

[Normsthename]

Well I tried the suggestions, and the cut quality has improved noticeably.
The main culprit seems to be that the CV Acceleration was turned on (Thanks Torchhead )
I still think that there is room for improvement.
I think that the THC may need tuning because it seems to be going up and down many times a second to maintain a constant height (Approx. 5 adjustments a second)
I will post a query on the THC300 forum to see what it should be doing

Thanks

Andy

[Torchhead]

Mach and the THC300 have a couple of parameters that will help tune the response. The THC300 is an analog control and the loop response time is partially determined by fixed components in the unit.

You can tune the response of the actual THC by raising or lowering the THC Response number in MACH. That determines the reduced speed (percentage of max velocity of Z) that the Z moves will under THC control. A good number is 30%. It depends a lot on your Z speed. You want it set as high as you can (with a leadscrew) then tune the THC rate to make it cut evenly.

The important thing it to watch the tip gap. If it's varying a lot then you need to do the tuning. Just because the Up and DOWN leds flash a lot does not mean the torch will move that often. Another thing to look at is noise. If plasma noise is getting into the tip volts signal back to the THC it will dance around a lot. If your table and plasma unit are not locally grounded (close rod in the earth) then disconnect the earth ground wire on the THC Sensor card. It may be injecting noise into the low side of the tip volts rather than draining it off.

[Normsthename]

Thanks Torchhead
My machine has stepper motors, and the Z axis uses a 3mm pitch screw.
Its maximum speed is set to 960mm/min

Andy

[BIGCNC]

Just a small note to rule out any problems with motion make a small fixture with a pen that will slide up and down use a rubber band to apply pressure
attach the fixture to the torch, Put a flat steel plate on the table with paper
taped to the steel. And run a program to see what kind of motion you get
I use this to test problems on my machines. Perfect lines= Perfect Parts.

[jimcolt]

Pen tracing is a good way to ensure that the machine has relatively fluid motion and is staying on path....but perfect pen tracings does not necessarily mean perfect cut parts. Here are a few things that will affect a plasma cut part on a cnc machine:

1. Variations in speed/velocity. If the machine is continuously accelerating and de-accelerating...expect the cut edge to appear rough. Plasma requires a rather narrow (too slow/too fast) speed band for each material type and thickness and power level combination......and if the machine does not doo a good job of maintaining the speed even on intricate details....then the cut quality will defiinitely be affected. This is why more pricey cnc plasma machine manufacturers go to great pains to have very high acceleration/de-acceleration capabilities. When the plasma is operated at slower cut speeds...the kerf naturally gets wider...changing part dimensions.

2. Torch height control. If the torch height control system does not accurately maintain torch to work distance within about .005" of the correct height....the the cut edge will be affected in terms of angularity and kerf width. Also...if the THC is overcorrecting...best described as a "sewing machine motion", expect very pronounced lag lines on the cut edge.

3. Occasionally we see issues where with a pen tracing we see very smooth contours...but when the plasma torch is active....the cut edge is very rough. Often the rough edge is dismissed as "that's the way plasma cuts". In reality....roughness during the cut could be an effect of electrical noise interference that is generated by the plasma power supply (many are inverters operating in the 20 khz range) capacitively or inductively coupling to the feedback loop for the servo drives (or stepper drives)....which can cause anomolies with the motion...usually shows up as a rough cut edge. This can usually be felt by resting your hand on the torch body while cutting...if there is any vibration during motion...expect to see it in the cut face.

Jim Colt

[Normsthename]

Since my last post, I have been trying a few things.
Each one in turn has improved the cut quality

I had a problem with the THC300 causing the Z motor to oscillate.
I placed a small piece ot sticky tape on the Z motor shaft, and I could see how the Z motor was reacting to the THC
Mine was oscillating about 1/4 of a revolution (45°) about 4 times a second.
My Z axis has a 3mm pitch screw, so that worked out that the torch was going up and down approx. .4mm (0.015")
I had to turn the THC rate down to 10% to stop the oscillations.
I calculated that at this Z Speed, the sheet can warp by 50mm high over the length of my bed and the THC will still cope

After a lot of experimentation I managed to understand the workings of the THC
I found the best method was to do a manual cut, with THC and anti-dive turned off.
I write a program to cut a line approx. 500mm long at the cut height specified in the Hypertherm manual.
I then watch the Tip voltage and note the voltage.
I then do the same cut with the THC turned on, and anti-dive turned on.
I then turn the THC course knob to match the previous voltage.
This method produces a perfect cut

After all this I was still getting problems on more intricate cutting.
I was told to turn off all the CV settings, but to leave CV Mode turned on.

All these mods and it is now cutting O.K
I have noticed that sometimes even when cutting identical pats, you get differences of cut between the parts. So it always best to run several parts and take an 'average' to judge the cut quality.

Thanks for all your input everyone

Andy

[benergy]

i have cut stainless recently and after much cutting of normal steel am frustrated by the quality of stainless cut. is it customary to have black jagged finish?