Machsupport Forum
Mach Discussion => Mach4 General Discussion => Topic started by: eschx on December 09, 2018, 04:01:17 PM
-
Hi all, in My spindle speed controller, 0-10V does not map to a linear increase in spindle speed.
I have fiddled with the scaling parameter in Mach4, and doing that, I can get it "close" to the requested speed, on either high or low end, but not both.
(sacrificing on the low end at the moment, but just got bit with too low speed, because I forgot about it)
Is there a way to apply eg: a quadratic or other function to this output in march4 to match what my controller seems to want?
Any ideas?
(Sorry if this has been answered before, search doesn't seem to be working right now)
-
I use the PoKEYS57CNC and the 0-10v output is linear. Also, my LSIS C100 VFD uses this and generates a linear RPM. MY spindle can do 8000-24000rpm and if I specify 12000rpm, I get 5v on the output and the VFD shows 12000rpm. My guess is your controller output is linear, but your VFD probably doesn't respond with linear speed. The cheap Huanyang VFDs are probably suspect in this area.
-
Hi,
how accurate do you require the speed?
For years I've used manual mills, lathes and drills and used the speed range closest to the desired optimum, and usually not
much worse than 20% off but seldom within 10% either....and all those jobs turned out fine.
Are you saying that because we are CNC machining the speed has to be spot on?
Craig
-
20% would be fine, but in my case it looks more like this. (guess it could be the VFD in the speed control as suggested)
Call for -> Get %diff
300 -> 163 -46%
400 -> 281 -30%
500 -> 396 -21%
1000-> 989 -1%
1500-> 1563 +4%
2000-> 2155 +8%
Normally I mill around 2500 which is the top end of my spindle speed (and that is all totally fine), but in this case, I needed to run a face mill at around 350 rpm. It came in under-speed, with a stall and a crash
Of course, I can work around this by jacking up the requested speed to ~800 in CAM, or by adjusting SRO slider on Mach4 (which is what I did). However, both of those methods introduce other potentials for human error on a future date, so if there was a way to fix it, I would like to save future self the headache =)
I am using a Vital systems HiCON Controller (and set the Analog Spindle Scale to 48.8 in the mach plugin to get it 'close'). That output feeds into a Precision Matthews PM-25 stock speed control. I did measure the output of the HiCon card to be generating expected voltages. From there wired into the location where the analog Speed Potentiometer on the PM-25 would have attached. (The rest of the board is potted with black epoxy, so I have no idea what is in there, or able to test further)
Thanks for the info!
-
Hi,
Mach can read spindle speed very accurately if it is supplied an index pulse from the spindle motor. Plenty of Mach users have made
one by gluing a magnet onto the rotor and mounting a small Hall sensor adjacent to it. Mach measures the time between
pluses and calculates the spindle speed very accurately indeed.
I'm not sure about the Hicon but I would guess its similar to my ESS and the ESS has a PID feature, fairly new feature at that, that would ensure that the
spindle is under closed loop control. That would solve your problem.
Craig
-
If your HiCon can't be configured you might want to take a look at this hack:
http://forum.machsupport.com/forum/index.php/topic,36020.msg246708.html#msg246708
HTH
RT
-
Thanks! Will give these a try!