Little help with constant velocity

[GPM870]

Good afternoon all,
Need a little help on constant velocity today. I have used it in the past for some projects but I need to understand more for the latest project.
Here is the project: A "sign" for identifying containers at my day job. A 6" square with some tabs that will get folded around some wire supports and back into the slots to secure. I can use exact stop and the outside looks great but the numbers are all chopped up. If I use the constant velocity setting then the numbers look OK but the outside corners are large radius.
Here is the Question (s), is there an inbetween setting? OR What settings should I be using?
I am using a Burntables table with sheetcam, Mach 3, and a Hypertherm PM85 set at 25 amp, 250 ipm,and 45amp shielded consumables.
There are a couple more settings on the right hand side of the general config screen in Mach but I don't know what they are used for so I don't want to mess up anything.
The pictures are of the same part out of the same material one with cv on and one with exact stop on.
I appreciate any help available.
Thank you,
Gerry

Both parts

[BTA Plasma]

Exact stop does not work well with a simple control system. In exact stop mode you will burn your corners and stop on every arc beginning and end. Dont use exact stop mode it isnt going to work with that sinmple of a controller. Go to config/motor tuning and reply with your accel and steps per numbers.

In constant velocity numbers the accel and deccel rates are very important. For example a number like 25 will only net you crisp results at 140ipm max if your less than that you wont be happy cutting thin materials. For this reason you cannot cut thin metals at book speeds or near book speeds for the maximum amperage of your consumables. Hypertherm has a 150ipm and below speed cut chart but its a comprimise. To achieve cut chart numbers listed by a mfg like Hypertherm you just wont do it with fine geometry and tiny stepper motors.

[SignTorch Vector Art]

Two things affect CV results, 1) acceleration must be sufficiently high to minimize corner rounding, 2) there is also a setting for CV read ahead buffer which often helps to be increased.

[GPM870]

Thanks for the info, I will have to check the numbers on the shop computer tomorrow night when I get home . I will admit i don't know the first thing on setting up the parameters or what I should be looking at.
The only thing I forgot to mention in the first post was the material thickness , 30 gauge. I didn't think it would make a difference.
Gerry

[GPM870]

Good evening,
I have some numbers from the motor tuning page.
X and Y are both the same... 382 steps per --- Acceleration is 18"per sec --- Velocity is 400.02 --- .0466 G's --- Step pulse 5 ---and Dir pulse 5. _ Z axis is: Velocity 40 --- Acceleration 9 ---- .0233 G's and 6400 steps per.
I attached a word document with a screen shot to help make sense of it.
I hope this all means something to someone that can help me out. I have been asking for a document from Burntables as to what the numbers should be and got ________________________________________________ nothing.
I'm thinking I may have to make a lot of changes depending on what I cut according to BTA Plasma's post as i cut a wide range of material. I have gone from 30 gauge (.0125") to .625" in the same day.
It's been a long uphill struggle to figure out how all the programs and controls work together with out any learning materials. Trial and error doesn't suit me very well.
If you need more information or if I have forgotten to include anything let me know.
Thanks again for all the help, I can only hope to gain enough knowledge to help others out in the future.
Gerry

[SignTorch Vector Art]

it's probably the steps per inch to CV Dist Tolerance ratio

a 180 unit (default) CV tolerance / 382 steps per inch in this case would be about .47" CV radius (the way I figure it)

try increasing the look ahead buffer (red)

and decreasing the CV tolerance (green)

if not increasing the acceleration (if possible)

those default CV values work for me but I have like 2122 steps per inch and 40 accel, (for a .08" CV radius)

here, CV radius is just a guesstimate for comparison, it doesn't mean the machine turns that radius, the actual CV turn radius depends on the speed and turn angle and CV settings, and is usually actually a lot smaller than the CV radius I have calculated, I think of it as the maximum possible radius at maximum possible velocity, but it's still just a guesstimated value for comparison

here is the CV documentation

[BTA Plasma]

Your using a machine with a .0026" minimum resolution in X according to that steps per. That is what I would call way too course for plasma. If you were to build with a single small motor you would want a small number for velocity or you will loose steps. With this small number you would have to limit your resolution or you would loose steps on decelleration. I am sorry that your machine has such a course resolution. I have found over the years that anything that has a resolution of great than .001 doesnt cut very well and anything with acelleration smaller than 25 will round edges starting at 80ipm. Those numbers are very poor. To make your machine give you a decent part you will need to use the smallest amperage possible and the slowest feedrate possible without getting into that jumpy course threshold that will come about at anything less than 10ipm with those numbers.

Honestly I would say your limited to 40ipm for decent geometry and nothing less than 10ipm. Those are good parameters for a machine with figures like that.

Calculation 1"/382=.0026" resolution
The coursest machine we even built had a 1019.1 steps per so it was 1"/1019.1=.000981" resolution
For example we hover from 2300-2500 steps per with an accel of 38-50 depending on the size of the machine

[GPM870]

Well I checked the look ahead buffer and it was at 20, I increased it to 40 so I can see if it makes a difference. I can't decrease the CV tolerance that you have at 180 because mine is set at 1 unit. I'm not sure how high I can go with the acceleration because I don't know what will happen if I go to high.
As far as the steps... what exactly does it refer to? How does it affect cut geometry? Can it be changed? As you can tell I don't have any experience or knowledge of the settings. I am trying to figure it out though.
I have a couple of drawings made up for some more parts so I will be able to try the settings tomorrow morning.
Gerry

[SignTorch Vector Art]

steps per inch is how many step pulses mach has to send to move one inch

that depends on the drive system

I'm trying to figure out (guess) how you get 382 steps per inch

a stepper motor has 200 steps per rev so with 4x microstepping for 800 steps per rev

and a 1.8 belt reduction (20 tooth to 36 tooth pulleys)

to 24 tooth 20 degree (rack and) pinion gear

that gives 381.96 (382) steps per inch

------------------------------------------------------------------------------

at 250 inches per minute, the motor is turning about 120 rpm

that's only using a small percentage of the motors useful speed range

and heavily loading the motor with a low gear ratio

which greatly limits acceleration potential

------------------------------------------------------------------------------

a 4x belt reduction (15 to 60 tooth pulleys)

would be a 120% improvement over 1.8x reduction to 849 steps per inch, 267 rpm for 250 ipm, less than half the load

even more reduction would be even better

-------------------------------------------------------------------------------

if your board supports 8x microstepping that might help (it would double the steps per inch)

-------------------------------------------------------------------------------

otherwise the only way to reduce the CV turn radius is to reduce the travel speed

and maximize acceleration

if you go too high with acceleration, it will simply start missing steps (mainly during acceleration)

way too high and it will miss steps and not accelerate smoothly

you probably won't damage anything by trying more acceleration

if you have big strong motors, they just might be able to accelerate much faster than 40

40 accel sounds like it should be enough to me, but I'm not used to cutting at 250 ipm, and never really experienced that much corner rounding, so all I can figure is the gear ratio must be too small, and 250 ipm must be too fast....

it looks like the machine is moving correctly, just at that speed and gear ratio it can't turn a very tight radius

[GPM870]

It was a nice day so I did get some cutting done today. Some successfully and some not so much. I cut out a part for a farmer to fix his disk, it was two pieces of 3/8" plate 7" by 16" with a couple of holes. That cut better than I have ever seen! I will pay dearly for that later!! I also cut some machine number signs for my day job out of 28 gauge galvanized steel. Some cut alright and some didn't cut all the way through. The THC couldn't keep up so some parts didn't cut all the way through. Not the end of the world, I can grind the edges to make them work.
The last part I cut was the same part shown in the first post. The only thing different was the look ahead units were changed from 20 to 40. I attached a picture of the one I cut today. It is slightly better than the first one.
The last picture is from the tag on the Y motor. (X and Y are the same.) If I have time tomorrow I will try to raise the acceleration a bit. What should I watch out for when I do? I would rather put up with crappy cuts compared to the machine self destructing.
Thanks again for all your help. I really appreciate all help I get on this site. Way more information then from the manufacturer.
Gerry