Printable View
Having just got my first CNC table up and running, has anyone got any practical advice on axis tuning in mach3?
i have no idea of what speed or Acceleration G's to Aim for in motor tuning.
its only a small table, I have a work envelope of 650x680mm to play with.
any tips or tricks?
Unfortunatly Dave motor tuning is quite specific to each machine with many factors playing apart to what can be achieved.Quote:
Originally Posted by Davek0974
Best advice is start low and work up and only raise one thing at a time, ie Velocity and test the affect.
The way I do it to get a ball park quickly is to double then half.? So say start at 2000mm/min on Velocity and 300s/s Accel.
Test and if ok double Velocity to 4000, test again. if too much then drop to 3000, test again. If ok raise to 3500 and so on.
Do the same with Accel but just be aware you can't have high velocity and high Accel so start with Velocity and find the limit where motors stall then back off 25%. Now do the Accel in same way and you should be in a nice balance. Just tweak from here to get the best performance.
The figures I've quoted are just guide lines and the volts you are running the machine motors/drives and the pitch of the ballscrews etc will determine the starting speed. For instance with only 24-36V and cheap drives on 5mm pitch screw then you may well be near the limits at those starting speeds I quoted.?
Give some info on your setup and will be able to give a better idea of a good starting point.
Sounds good,
my machine is all aluminium build, carriage runs on steel rails with bearings.
im using nema 23 motors, 3.1Nm, 4.1A, and a 36v 11A psu.
The drives are all T5 belt, 3:1 reductions on both axes and I'm using 1/8 micro-stepping.
Well you should have high rapid speeds due to high pitch but this obviously depends on pulley sizes thou with 3:1 ratio I can guess at around 25-35mm pitch.
With only 36V then you'll have lower speed and there fore torque from the motors than they are capable of so for this reason I suggest you keep the motor tuning down so the motors are in the lower Rpm range where more torque is available.
Like I mentioned before your problem could come from lack of resolution and motors running too slow causing rough movement.?
There is no problem JAZZCNC, I'm just looking for facts and figures at the moment.
I now have it set for 15,000mm/min and 1300 mm/sec/sec and it's been running a pen over various files repeatedly and has not missed a step yet.
Seems pretty good so far, I think that speed is more than enough for a small machine???
To be honest I forgot this was for a plasma machine so very little forces involved and at those speeds your running about half the motors speed which is about right where I'd have predicted would be ok without missing steps etc. You may find you'll need to tune slower if ever cutting forces are involved.?Quote:
Originally Posted by Davek0974
If plasma is limited to say 4500-5000 mm/min, is there any point at all tuning a small machine to 15000-20000mm/min???
Now there's a question that gets people butting heads. . .Lol. . . . Why do you ask.?Quote:
Originally Posted by Davek0974
My answer is NO absolutley not. High rapid speeds are only any use for A-2-B positional moves like drilling lots of holes or with very large machines. Small machines then IMO it's pointless and can easily get folks into trouble with crashing machine into hard stops etc.
With a small machine your much better off sacrificing Velocity in favor of accelleration and often it brings cycle times down anyway.
Brilliant, that's exactly what I was thinking too.
the reason I asked is that many threads go on about tuning for high speed, must go this speed etc. but then I realised that plasma itself is limited (at least in home size machines) to speeds of around 750 to 4500mm/min so what the heck is the point of mega speed.
as you said, acceleration seems to be important and on a machine with an envelope of 600x600mm, 20,000mm/min is pretty scary and damaging.
i think I'll limit mine to 5000 max.
Bragging rights and showing off that's all there is to why they chase Rapid speeds.!!Quote:
Originally Posted by Davek0974
If your working with small machine for cutting woods then I'd aim for between 8000 to 10,000mm/min Max Rapids. If cutting mostly Aluminium or hard materials then you won't need much above 3000mm/min and 5000mm/min is a good Rapid speed.
Also Resolution is important so think about that when building a machine.
Yes, resolution is fine on my build, I've had it drawing intricate shapes over and over for a good few days and it hasn't lost a step, these shapes are far more intricate than can cut with plasma so it's a pretty tough test.
I've got it in bits for final wiring etc, my new plasma cutter should be here this week, can start testing properly then.
I was more talking about cutting machine rather than plasma, Cutting loads will change things slightly. Not in resolution terms but how well it handles the torque required for cutting and whether it misses steps or not.?Quote:
Originally Posted by Davek0974
Throwing a spindle on it will soon show you thou and cutting finish will show how the lower resolution is affecting things, esp with 3d surfaces and tiny stepovers.
Last belt machine I built was heavily geared, approx 5 or 6:1 I think to get the resolution down for the mostly 3D work it was doing. Thats was 3-4yrs ago and it hasn't missed a beat since and done some lovely work from the few pics that have been sent to me.
Ahh, I see, I very much doubt this build will see a spindle, it might however see a 3d print head being fitted;)
would need to be a fair bit tougher for a spindle I think.
where did your max of 4500 to 5000mm/min come from?
Purely from reading many hundreds of posts and asking endless questions. As I said, I have never used a CNC machine before but have cut a fair bit by hand and that seems to back up my theory on speeds.
dont forget that I am only concerned with lower power cutting, as I'm in a home-shop and very limited on power etc. A 50A cutter is about the limit, I know that would cut ridiculously fast on say 1mm sheet but it's variable and more control can be had by lowering it to 20A and running at a more controlled pace.
Just checking. It seemed 4500-5000 was being stated as some general maximum - which as you know is not the case.
50A - at home - see the street lights dim when you start that puppy :smile:
Lol, it's only 30A draw on a 220v supply, should be ok.Quote:
Originally Posted by stirling
Its no more than an electric shower unit draws, the old unit was rated 15A but worked fine on a 13A socket.
Just don't turn on the radio at same time hey.!!.. . . LolQuote:
Originally Posted by Davek0974
Next to me is a Sub station and you should hear that sucker buzz when me and my lad are on full tilt. . with 2 welders going at same and all other kit in workshop.!!. . . . . . Just hope the Elecy board don't spot the Cable running thru back wall of Station. .:hysterical:
He he, that's the way to do it ;)Quote:
Originally Posted by JAZZCNC
nah, worst it can do is pop the breaker to the shop, if it does and likely will, I'll have to rewire and uprate the breaker, I'm pushing my luck well into the realms of stupidity as I have the shop running on a 32A breaker and intending to use a 2hp compressor, 6 fluorescent lights and a 32A plasma cutter :)
luckily the the cutter can be run at lower power on thinner stuff, but I am expecting to rewire soon.
New torch and psu arrived today, this thing is a beast!
not had a chance to power it up yet as I need to make a torch holder to fit the massive machine torch it has.
The tip is totally different to my smaller unit, must be half a pound of copper in the tip :o
Hopefully get the table cutting this week.
Made some test cuts today,
I cut repeated test lines at various speeds and settled on the one that gave me seemingly acceptable results, this turned out to be 20A at 1275mm/min. I am trying to keep speed down as my machine is happier at lower speeds (this will be looked at later) hence the lowest power possible, 1.5mm steel again, pierce height 4mm, cut height lowered to 1.5mm.
This setting gave me a 1mm kerf width, picture of my final test for the day below..
Attachment 11815
One point I noticed was that I had plenty of dross on the good (job) side and almost none on the scrap (cutoff) side, does this indicate my cut direction is wrong???
System is behaving itself though which is good.
Great start. I bet you feel very pleased. Well done. G.
I don't quite see how, but wouldn't a better electrical connection be a more likely explanation?Quote:
Originally Posted by Davek0974
From what I've read so far, it's a bit like milling, the jet prefers to cut one way rather than the other, I'm trying to find the post again because I can't remember if it was clockwise around the outside or anti clockwise, easy to test though.
its connected with the way the plasma swirls as it exits the nozzle.
once the first part of the cut passes a bed slat, it's pretty much tack welded down so the contact should be good.