On Wednesday, December 28, 2011 10:10:54 PM Steve Stallings did opine: > Sorry, but a different opinion here.... > > Gecko publishes the 67% number based on > real world experience with a large number > of motors and power supplies tested. > > The reality of calculating the current > needed is daunting. First you start by > looking at power, which is what is really > being delivered. The Gecko drivers are > modern chopping drivers which behave > somewhat like a switching power supply. > They convert a high voltage moderate > current into a lower voltage at a > higher, but regulated, current to drive > the motor. For the same driver and motor > the current drawn from the power supply > will be different (lower current) for a > higher voltage power supply than for a > lower voltage one. Indeed, I have seen > drivers overheat simply because the > power supply voltage was too low. This > is a natural result of the driver delivering > a mostly constant amount of power, but > having to draw more current to do so. > Heating in the driver is dominated by > I squared R when the MOSFETs are fully > turned on, and twice the current results > in four times the heat losses.
Another effect of a miss-match between power supply and driver is often overlooked, and that is the I*r loss in the mosfet as it is turning off this higher current during the 10-100 nanoseconds it takes to turn it off. Because more current was flowing, there are more charge carriers that must be cleared away and this loads the mosfets (hexfets) driver stage and slows in down. What this translates to in the real world is that a 48 volt driver, running on 24 volts will probably need to get rid of about 4.5 to 5 times as much heat as the *4 doesn't take this switching time difference into account. OTOH, I think we are getting closer all the time, thanks to the push for ever more efficient computer psu's, which has led to the development of ever faster switching transition times AND ever lower on resistances. I believe I benefited much from that effect when I blew the hexfet in my micromills motor controller. I looked up and printed out tha data sheet on that device, picked up a dead psu I'd had laying around for a couple years, checked the hexfets in it, and found that except for the gate capacitance going up by about 150%, every other specification was anywhere from 2 to 10x better. So I put it in, and watched it with an IR thermometer while the motor was being run at about 75% of full rpms for around an hour. I could not find a single part that was more than 15F above the background room temp and the hexfet itself was only warmed maybe 3F. So I put it back together and its still fairly well sealed inside a shack project box, about a 3.5x5.8x7 black plastic thing, sharing room with your PMDX-106. It does warm some but more than 15F of ambient. Running nicely for at least 2 years now. > > Things get complicated because the > power required is the sum of: > > 1) mechanical power delivered by the motor > 2) mechanical losses within the motor > (bearing and air drag) > 3) resistive losses ( I squared R ) > 4) hystersis losses in the iron (this > is often the main cause of motor heating) > 5) efficiency of the driver itself > > Modeling all of the above is a lot of > work, even if you do have accurate data > to start from. > > Gecko has stated that the 67% is the most > you are going to need assuming you have > selected a reasonable power supply voltage > and are driving the motor to deliver its > maximum mechanical power output. > > The fact that most machines only move two > axes at a time for most operations will > mean that the mechanical part of the power > is less, but the other factors are not > much effected by the mechanical power > output. > > I generally tell my customers that 50% > is good enough for typical machines, 67% > will provide for anything you can ever > hope to achieve, and anything more is > purely for bragging rights. Such a statement covers it quite well Steve. > Regards, > Steve Stallings > PMDX Cheers, Gene -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) My web page: <http://coyoteden.dyndns-free.com:85/gene> People who take cat naps don't usually sleep in a cat's cradle. ------------------------------------------------------------------------------ Ridiculously easy VDI. With Citrix VDI-in-a-Box, you don't need a complex infrastructure or vast IT resources to deliver seamless, secure access to virtual desktops. With this all-in-one solution, easily deploy virtual desktops for less than the cost of PCs and save 60% on VDI infrastructure costs. Try it free! http://p.sf.net/sfu/Citrix-VDIinabox _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
