PDA

View Full Version : Need Help! Trying to sort out an angularity problem.



CoolHand
02-02-2010, 01:44 AM
OK, I've got a Dynatorch 5x10 table, with a DASH, Plate Marker, Tube Cutter, and a HyperTherm Powermax 1250 G3 Plasma Generator.

My air compressor is ~60 feet of airline away from my IR Refrigerated Air Dryer, which is about two feet away from my plasma generator. I have an IR 1 micron coalescing filter upstream of the air dryer, and the plasmas has another water trap on the back of it (though I very much doubt it's doing anything). The air goes through a bulk tank and 60' of steel line before it gets to the filter and air dryer, so it is good and cooled off when it arrives. IIRC, the compressor is a 12 CFM model, but since it doesn't run continuously when the machine is cutting, I'm assuming it's large enough to properly supply the generator's needs.

The problem I am having is with edge angularity.

I know you will always get taper, I understand that, but my problem is that I'm getting excessive taper. I am seeing a pattern, but I cannot decide what it means as far as troubleshooting. It happens in a way that seems to rule out mechanical issues, but I could be totally wrong on that score.

I'm still sorting out my methodology as far as setup and proving the process goes. I am finding this part of plasma cutting to be MUCH harder than the same task with a CNC milling machine, and it is driving me up a wall.

You cannot isolate a single variable like you can with a VMC, 'cause when you change one variable on the plasma, the rest of them change in response, and seemingly never in a good way. I can't attack the problem systematically like I normally would, so I find myself feeling rather disoriented.

Here's my methodology: I cut 2" x 2" squares using the cut data I want to test, and then check for size, cut quality, and edge taper. Here's a pic of the whole batch that I will be talking about:

http://www.logicpaintball.com/PlasmaCouponGroupFoto.jpg

I've got my 16ga down to where I can live with it, no more than 5-7 degs of taper, and it appears to be uniform regardless of cut direction. The data is as follows:

Material: 1010 Steel CR
Thickness: 16 ga
Nozzle: FineCut
Amperage: 40A
Arc Voltage: 77V
Travel Speed: 100 IPM
Kerf Width: 0.040"
Air Pressure: 75PSI

Results:

http://www.logicpaintball.com/16ga40AFineCut.jpg

I can live with that. Dross is minimal and can be chipped off with a scraper.

The 11 ga isn't too bad. The first cut was a bust because the damned DASH set the pierce height too low, so the pierce happened with the nozzle basically pressing down on the plate, instantly smoking the nozzle orifice. It cut fine, but the angularity was shot all to hell on two of the four sides:

http://www.logicpaintball.com/F'dUp11ga40AFineCut-Side1.jpg

So, I changed the nozzle, re-calibrated the DASH, and cut another one using the following data:

Material: 1010 Steel CR
Thickness: 11 ga
Nozzle: FineCut
Amperage: 40A
Arc Voltage: 90V
Travel Speed: 40 IPM
Kerf Width: 0.045"
Air Pressure: 75PSI

That gave me this result:

http://www.logicpaintball.com/Good11ga40AFineCut-Side1.jpg

http://www.logicpaintball.com/Good11ga40AFineCut-Side2.jpg

http://www.logicpaintball.com/Good11ga40AFineCut-Side3.jpg

http://www.logicpaintball.com/Good11ga40AFineCut-Side4.jpg

As you can see, sides 1, 3, and 4 are good and square, no more than 1-2 degs of taper, but side 2 is clearly tapered much more, I'd estimate at least 10 degs.

What's going on here? The torch is square with the world, the nozzle was still good, and what little dross there was just popped right off the back side. This would be a perfect cut if that one side wasn't so tapered.

Any ideas?

The next cut was in 1/4" HR that had some rust on it. I had to disable the DASH entirely to even make this cut, but I didn't get that brainstorm until the DASH has smoked two more nozzles by jamming them into the plate surface. Every day I use the machine, I become less a fan of the DASH. It's a neat deal, but it could use a more robust operating window, if you know what I mean. But I digress.

The first one was done at the following data:

Material: 1010 Steel HR - Rusted
Thickness: 1/4"
Nozzle: FineCut
Amperage: 50A
Arc Voltage: 100V
Travel Speed: 40 IPM
Kerf Width: 0.055"
Air Pressure: 75PSI

That gave the following result:

http://www.logicpaintball.com/QtrInch50AFineCut-Side1.jpg

http://www.logicpaintball.com/QtrInch50AFineCut-Side2.jpg

http://www.logicpaintball.com/QtrInch50AFineCut-Side3.jpg

http://www.logicpaintball.com/QtrInch50AFineCut-Side4.jpg

As you can see, we have the starts of a pattern here, in that sides 1, 2, and 3 are good, but side 4 has excessive taper. Three good sides, one bad one, same as above. Maybe it's just coincidence, but in my experience, there is no such thing as a coincidence when dealing with machine tools.

Again, any ideas?

Lastly, I switched out consumables for a standard shielded setup and made another test cut on that same piece of 1/4" HR.

I used this data:

Material: 1010 Steel HR - Rusted
Thickness: 1/4"
Nozzle: Std 40A Shielded
Amperage: 40A
Arc Voltage: 149V
Travel Speed: 40 IPM
Kerf Width: 0.055"
Air Pressure: 75PSI

And got this result:

http://www.logicpaintball.com/QtrInch40AShielded-Side1.jpg

http://www.logicpaintball.com/QtrInch40AShielded-Side2.jpg

http://www.logicpaintball.com/QtrInch40AShielded-Side3.jpg

http://www.logicpaintball.com/QtrInch40AShielded-Side4.jpg

I'm getting closer to calling this a pattern. With so few instances, it could still be just coincidence, but I'm rather starting to doubt it. Again, sides 1, 3, and 4 are within the normal range of taper, but side 2 is way out in left field.

I'm drawing a blank here, so any help would be greatly appreciated.

Help save my sanity, I beg you! :D

WSS
02-02-2010, 02:09 AM
Coolhand,
It could be in the torch body. Is the taper relative to the same side of the table? If something is not set right with the swirl ring, that could push the taper in one direction. Try the cut again with the same parameters but turn the part 45 degrees, this way both x and y motors are counting/turning. If the problem moves or is cut in half it could be motion. If it covers the same area as the original (half of one side and half of the other) then it could be torch. Are you the original owner of the 1250? I had looked at used machines and found that water damage in the torch cable was common.

I have cut some 12ga over the last few days and found that the taper was considerably more than on 1/2" or 3/4". Over half inch you can consistently get under 2 degree taper. I was surprised to find how hard it was to get good cuts on thin stuff. All that said, the taper was even all the way around the parts.

I offer that as food for thought. I am relatively new to CNC. So I could just be rambling.

Good Luck!
WSS

CoolHand
02-02-2010, 02:39 AM
Coolhand,
It could be in the torch body. Is the taper relative to the same side of the table? If something is not set right with the swirl ring, that could push the taper in one direction. Try the cut again with the same parameters but turn the part 45 degrees, this way both x and y motors are counting/turning. If the problem moves or is cut in half it could be motion. If it covers the same area as the original (half of one side and half of the other) then it could be torch. Are you the original owner of the 1250? I had looked at used machines and found that water damage in the torch cable was common.

I have cut some 12ga over the last few days and found that the taper was considerably more than on 1/2" or 3/4". Over half inch you can consistently get under 2 degree taper. I was surprised to find how hard it was to get good cuts on thin stuff. All that said, the taper was even all the way around the parts.

I offer that as food for thought. I am relatively new to CNC. So I could just be rambling.

Good Luck!
WSS

Everything is brand new, bought all at once in early December of 2009.

The swirl ring is the one Hypertherm shows in their parts stack in the manual.

The consumables were all new and installed as per the manual.

Torch is a T80M machine torch, and it is square to both axes.

Excellent suggestion, I will rotate the square 45 degs and cut another. I will report back with details tomorrow after I get the new ones cut.

Just to be clear, I could live very well with the taper shown on the three good sides, it's just that forth one that I can't deal with.

Hopefully Jim Colt will stop by and give his two cents worth.

BTW, I read your build thread. That is one hell of a table you built. Love the giant press and all the heavy plate work you guys do.

If you don't mind my asking, what are those big abrasive edged plates used for? Are they teeth of some sort, like on a saw, or are they a wear edge for a crusher or digging machine?

That looks like a very cool process.

magma-joe
02-02-2010, 08:08 AM
Coolhand,

If there is something mechanical, perhaps it would be easier to define by cutting a larger part? Maybe a 12" x 12"? I would be curious to see if the results were still consistent with the 4" x 4" part?

Here is some info that might help. The section on bevel is on page 9 of the PDF file. http://www.centricut.com/Articles/Centricut_Training_Articles.pdf

Setting the parameters for the best plasma cuts can be a challenge. As you mentioned changing one affects the others. Since the plasma flame is flexible and spinning in a circular motion it takes some getting used to when you have been accustomed to cutting with a rigid tool such as an end mill.

Understanding the characteristics of the plasma arc in relation to cutting heights, speeds, short paths, circles, different thicknesses etc. along with hands on experience will be the fastest route to the best cuts.

Also take a look at parts cut by other shops to compare to yours. You might be surprised at what you see.

Magma-joe

CoolHand
02-02-2010, 03:38 PM
Coolhand,

If there is something mechanical, perhaps it would be easier to define by cutting a larger part? Maybe a 12" x 12"? I would be curious to see if the results were still consistent with the 4" x 4" part?

Here is some info that might help. The section on bevel is on page 9 of the PDF file. http://www.centricut.com/Articles/Centricut_Training_Articles.pdf

Setting the parameters for the best plasma cuts can be a challenge. As you mentioned changing one affects the others. Since the plasma flame is flexible and spinning in a circular motion it takes some getting used to when you have been accustomed to cutting with a rigid tool such as an end mill.

Understanding the characteristics of the plasma arc in relation to cutting heights, speeds, short paths, circles, different thicknesses etc. along with hands on experience will be the fastest route to the best cuts.

Also take a look at parts cut by other shops to compare to yours. You might be surprised at what you see.

Magma-joe

Thank you, I will read that PDF and see what I can come up with.

I don't doubt that a lot of places would just let this little big of taper go, because the cut quality, dross, and size are right on.

I don't do a lot of work for outside places, I am cutting parts to use in products I manufacture here. For some things, that taper wouldn't make a bit of difference, but for this one project, I cannot abide that taper on the 1/4".

The taper is pushing the holes around and forcing the parts that are inserted in them off center, and I can't have that.

Or rather, I should say, I couldn't abide the taper I had been getting before I undertook this last round of testing, which was a lot worse than what is pictured. Now that I think about it though, that was likely more due to a combination of excessive speed, too much heat, and too little air pressure. This test is a lot better, there is still excessive taper on that one side, but maybe it will not cause as much trouble as before. I may just cut a set of parts tomorrow and see how it turns out.

I didn't get in to the shop today like I planned, all cold sweats and stomach troubles, dunno what it is, but it sucks.

Hopefully tomorrow I will be back to full steam. I'll use today to read that PDF and see what else I can dig up.

Thanks for your help thus far.

jimcolt
02-03-2010, 07:54 AM
I just found this thread.....I'll be happy to help you troubleshoot this later today....I'm tied up through early this afternoon. Looking at the pics and your results I suspect nozzle/shield damage due to pierce height......one pierce to close will produce a microsopic nick in either the nozzle or shield orifice that causes non uniform angularity around the perimeter of the part. There of course could be other issues as well....but I'd be happy to try to help you improve that quality of your cuts.

Jim

jimcolt
02-03-2010, 09:20 AM
I had another chance to read through your posts...here are some observations/comments.

-I notice some of the pics show cutting 1/4" with FineCut consumables. The FineCuts are recomended (by Hypertherm) for cutting up to maximum thickness of 10 ga. In reality they will cut thicker....but the drawback will be with piercing.....an edge start on 1/4" may work well....but the nozzle (I can pretty much guarantee) will be damaged on the first pierce on any material thicker than 10 gauge.....nozzle damage results in a misshapen arc, which will cause variable angularity and slower cut speeds.

- I see that you are monitoring the arc voltage very closely....which is a good thing......but more important than maintaining the arc voltage listed in the Hypertherm manual...is with maintaining the physical standoff during the cut. For FineCut parts the physical standoff (on carbon steel) needs to be .08" ......higher than .08" will produce a positive bevel, closer will produce dross and possibly a negative bevel, as well as a higher chance of plate collision. Voltage.....if you are cutting at a fixed speed and Amperage...will vary with height of the torch. Set the voltage to ensure that the physical height is correct......don't worry if the voltage is way different from the Hypertherm specifications. Arc voltage is often different with different height controls....it is dependant on circuit board calibration, consumable wear, cut speed, air pressure and amperage.

-Pierce height is absolutely critical.....rule of thumb is that it should be 1.5 to 2 times the cut height.....and a pierce delay time needs to be long enough so that the arc fully penetrates the plate before any x, y or z movement ocurrs. One bad pierce pretty much destroys the nozzle (need to use an eye loupe or microscope to see the damage).....I walk into countless shops that think plasma cuts with varying angles...and they live with it. Often after helping them solve their pierce height issues they have a whole new perspective on plasma cut quality. On materials thinner than 10 gauge....you can pierce and cut at the same height with a Hypertherm system.

- Here is a test that I would try. Load some 10 gauge...or 1/8" steel on your table. Use new FineCut consumable in your torch. Ensure the torch is perpendicular to the area of the plate you are cutting (use a square....not a level to do this!)....then shut of the automatic THC and the initial height sensor.....adjust the torch to work distance at .08"......and program a 4" square part. Dry run the torch around the part shape and carefully check to ensure the torch is staying at .08" (+- about .010 is adequate).....then cut the part using the following specs (out of the Hypertherm manual)
40 Amps, 50 IPM, .7 seconds pierce delay, plasma pressure (dynamic (flowing)) should be 65 to 70 psi. let me know how the cut looks.....this eliminates the THC and will help determine where the problem is coming from.

Best regards, Jim Colt

CoolHand
02-03-2010, 07:24 PM
I believe your assessment is correct.

I had to cut some 1/4" today, and given your advice to not pierce with the FineCut nozzles, I switched back to a regular shielded 40A nozzle, and backed my speed down to 32 IPM.

Data today:

Material: 1010 Steel HR - Clean
Thickness: 1/4"
Nozzle: Std 40A Shielded
Amperage: 40A
Arc Voltage: 145V
Travel Speed: 32 IPM
Kerf Width: 0.055"
Air Pressure: 75PSI

The cut was good, I got a lot of slow speed dross, but it chipped off fairly easily.

The good part was that the edges were all within the 2-4 deg range, and all of them were the same.

I don't have any pictures of the edges, no time today, but I think this might be a mystery solved.

I used that same FineCut nozzle to cut a sheet of 16 gage today, and it did fine, so it's not a total loss, just can't be used for thicker stuff.

Thanks for your help Jim.

WSS
02-04-2010, 03:00 AM
If you don't mind my asking, what are those big abrasive edged plates used for? Are they teeth of some sort, like on a saw, or are they a wear edge for a crusher or digging machine?

CoolHand,

Thanks, 99% of the stuff we do is for ground engaging tools, anything that wears out in or on the ground. A lot of the parts are the cutting edges of buckets, blades, dozers and scrapers for Cats.

Looks like you are on the right track again. Guys like magma-joe and Jim colt know what's going on. I learn something every time these guys post. Who would have thought to add into the equation that the arc is flexible? Thanks magma-joe for adding new depth to my confusion (LOL)!

Cheers and have fun with your new DT.

WSS

slammedxonair
02-04-2010, 11:50 PM
You might want to speed it up just a little bit I run at 40imp with a 40 amp setup and get pretty clean cuts with little to no dross.

CoolHand
02-05-2010, 01:39 AM
I ran some parts at 40 IPM, and for most stuff that will work fine, but with these particular parts the reduction in edge bevel was worth dealing with the dross.

For stuff that the bevel won't matter on, I think I'll step up to 60 amp consumables so I can kick the feeds up to around 80 or 100 IPM.

jimcolt
02-05-2010, 08:07 AM
You are on the right track.....increase speed will likely produce less dross, reduce speed will reduce angularity. Low speed dross usually comes off very easily....and its a good trade off when better edge quality is needed.

Jim

CoolHand
02-05-2010, 07:45 PM
You are on the right track.....increase speed will likely produce less dross, reduce speed will reduce angularity. Low speed dross usually comes off very easily....and its a good trade off when better edge quality is needed.

Jim

That's what I thought as well.

As it happened, 5 mins with a sanding disk and I was ready to move on to secondary operations on the milling machine.

Thanks for your help guys, I'm getting a better handle on things.