Machsupport Forum
Tangent Corner => Tangent Corner => Topic started by: Overloaded on February 22, 2014, 11:18:09 AM
-
Just wondering .....
How stable is, and what affects the actual rpm (stability) of an AC motor ?
Very basically. Say, an AC, 110v 60 cycles motor, nothing on the shaft, plugged directly into a wall socket.
Will the rpm be rock solid and constant ? Dependent on the 60 cycles being steady from the utility ? Quality of the motor construction ?
Or what else would / could cause an unstable rpm ? Absolutely no load. I assume over time due to thermal effects it might change slightly, but I mean fluctuations ... + / -
while running.
I have no real way to test, other than to put a disc on a motor shaft and move my index sensor to it. If I did, should I expect to see 0 wavering of the reading on the Mach screen RPM report ?
As usual ... just curious.
Thanks,
Russ
:)
-
If you are talking an induction motor, one without windings on the armature, then the motor will always run less than synchronous speed. For 60 cycle AC a 2 pole motor will not run at 3600 rpm it will run a little less at no load. The 'slip' is needed so that the armature is moving through the magnetic fields to generate current and magnetize the armature. If it turned at 3600 rpm there would be no relative motion and no magnetization so no torque. This means an AC induction motor will run at a different speed for every load.
There are sychronous motors with wound rotors and slip rings that will run at 3600 rpm on 60 cycle power. However it only pays on really big motors.
Hope this makes sense
-
Hi Gary, makes perfect sense, thanks.
I was reading about synchronous motors and your explanation makes it even easier to understand.
I guess that's why an induction motor is either 1725 or 3450 instead of 1800 and 3600 respectively.
I understand that part.
What I am meaning to ask is not the actual rpm, but the fluctuations in the speed .... if any and what would affect that. (induction motors)
Thanks again,
Russ
-
The two things that affect the rotation speed are line frequency and load. I once installed a large diesel generator for a customer because he was monitoring line frequency and it varied from 59 Hz to 61 Hz all the time. This was affecting his process, and when he complained to the power company they told him they don't guarantee 60 Hz continuously, they provide an average of exactly 60 Hz over the course of a year! Load is the big variable, because the motor needs the slip to work.
If you need a really constant speed you will need feedback or a stepper motor. Microstepped motors will run at a very constant speed because the step frequency is typically crystal clock controlled in the drive. Another way to get a very constant speed is with a motor and an encoder. Then it doesn't matter to much what kind of motor it is, DC, Servo, 3 Phase with Inverter etc. As long as the drive has an encoder input you can get very steady speed. Lots of choices, and modern electronics has made this pretty cheap to do.
If you tell me what is you are trying to do I can make some suggestions.
-
Not looking for any specific rpm.
Just looking for a steady, stable non wavering NO LOAD rpm from a common AC induction motor.
.....and what might affect that, if anything.
" it varied from 59 Hz to 61 Hz all the time."
What was the frequency of these variations ? Per second ? per minute ? per hour ? week ? ........
Thanks,
Russ
-
So I did a test.
Put a slotted disc directly on the .5 hp motor, nothing else, no load. Mounted the sensor.
Mach reports 1797-1799. No Index debounce set, Not averaging.
Put 2 precision machined V pulleys on, new v belt plenty tight, remounted the sensor at the spindle disc.
Mach reports 195* - 199*. WAYYYY to goofy to cut good threads.
(pulley ratio is just what I had available for the test, I'd normally reduce about 2:1)
The belt is not perfect, but it's new and shouldn't wander the spindle rpm THAT much, but maybe a little.
Is the motor too small maybe ? It doesn't slow during threading, it's just not stable.
Does it need a timing type belt and pulleys maybe ?
I know, I know .... a servo would be the bees knees, I know. Or a vector type motor and drive with feedback, both a bit pricey.
I'm seeing more and more why commercial units cost so much. :-\
Thanks,
Russ
:)
-
You got my curiousity Russ.
Don't have any rpm checkers here that are accurate enough to check the motor rpm.
If you get a loan of one make sure you know what the % accuracy since don't think some of cheaper ones will give you +-1 rpm at 1750.
Way back, when we were checking the index on my lathe, we used a very accurate pulse counter and compared index input to Mach vrs displayed rpm and pulses sent to the axis and it was very close. I know that varying belt tension on say a stepper that it will change steps per unit, but the steps per unit are repeatable per rev.
I do have faith in the DRO readout as being accurate, again based on what Mach is being told.
The problem is one needs to isolate each step for comparison, as you simply did, but that must be compared to a known / calibrated standard. IE. indexer puts out x pulses compared to an electronic device that is more accurate then the index output.
RICH
-
I pretty much trust the Mach display, my test kinda shows that. (fluctuations)
I've almost come to the conclusion that one really needs to overkill the spindle motor HP (short of feedback).
I might slap a 2hp inverter duty motor on this puppy just to see.
If that's what it takes, so be it.
Thanks Rich,
Russ
-
Russ,
I think a V-belt drive is problematic too. As you apply a load the belt pulls down into the V and changes the effective diameter. This might cause a fairly large speed change. The real problem all the way around is that in thread cutting you load and unload the motor at the beginning and end of the cut. What you really need is for the X axis (carriage) to follow the A (spindle axis). Think of it as a linear move using the X and A together. I am adding a little lathe attachment to my mill and I am driving the spindle with a stepper so I can use it as a fourth axis and I am thinking threading too. Don't know how that will work. However I have a large driver on the A axis and I have 960 in/oz stepper and can get a 1600 oz/in stepper driven at 68 vdc so that should be too bad for a mini lathe.
-
Russ,
I think a V-belt drive is problematic too. As you apply a load the belt pulls down into the V and changes the effective diameter. This might cause a fairly large speed change.
I can certainly vouch for that ! Even at no load, the slight variations in the belt width throughout its periphery WILL introduce a variable rpm at the output shaft/spindle. Also a larger, thicker cross section belt will have an oval shape preset in it from packaging (maybe the mfg process too ?) that WiLL again introduce a sort of "surging" in the output as each end of the preformed oval comes into alignment with the pulleys. You can easily feel this effect while turning the spindle by hand. An undersized motor will definitely chase this variation up and down .... I recently proved that here with testing. Thin section, 3/8" v belts are better, but still not as good as a flat XL type. As a test, a V belt, running on its back, on XL type pulleys is quite stable, provided the belt does not have a strong memorable oval set into it, but torque transfer is not very good .... this is just a test.
What I'm finding ...... Overkill on motor size (open loop, induction), and flat, flexible belts.
Thanks Gary,
Russ
-
... also meant to mention that many of the "Farmer/Lawn and Garden" variety V belts have a slight overlap in the ends of the fabric sheath that encapsulates the belt. When the joint goes through the pulleys, it changes the effected PD of the pulley as well.
When threading, EVERY dinky detail must not be overlooked.
Russ
-
Russ,
The multi-V belts as used on a treadmill would likely give you a much more constant speed. You'll notice that the treadmill motors all have the motor pulley made as part of a very heavy flywheel which I would think really helps stabilize the speed. That type of pulley is easy to make too.
-
Have you guys tried the multiple link belt that are red. I can't remeber the manufacturerof them. It might be fenner? They work real well and don't require dissassembly of the machine for install. They shouldn't suffer from the ovality issue.
Mike(full of red wine)
-
Hey Mike, thanks for chiming in !
Yes, I have used those in quite a few apps., mainly with conveyor drives and such. Some with a fairly substantial load but most with light to moderate no-variable loads. Although the ovality issue would be eliminated, there is still the issue of varying widths of the V throughout it's periphery and this type would probably tend to dig deeper in to the pulleys under intermittent loading due to it's thinner cross section ...... as Gary mentioned earlier.
Thanks for the input, appreciated,
wine ? have 1 or 6 for me,
Russ
-
Russ,
The multi-V belts as used on a treadmill would likely give you a much more constant speed. You'll notice that the treadmill motors all have the motor pulley made as part of a very heavy flywheel which I would think really helps stabilize the speed. That type of pulley is easy to make too.
I like this idea Gary, have used that type of belt before.
Did some experimenting:
Mounted a slotted disc and sensor to various motor/drive combinations just to see the differences.
Disc on 1/2 hp 110v ac single ph motor .. rpm pretty stable +/- 1 rpm at around 1750
" 230v 3 ph Fugi vfd (rotary phase converter) +/- 5 rpm at around 1000.
same as above but with a new Automation Direct GS2 vfd (120v single phase in) rock solid, barely 1 rpm fluctuation.
One thing i didnt try, but will, is the 3 phase directly from the phase converter to the 3 ph motor.
The Fugi is an older unit, maybe newer technology is more steady at the output ?
Seems "jittery"..... is this typical with this type of set-up ? Rotary Phase Converters ?
And the flat belts are the best at transferring the rpm consistency. All V combinations introduce variations to some degree.
Regards,
Russ
:)
-
Note: all tests above were done with "Averaging" OFF.
Russ
-
Russ,
Right on in saying all the little things add up and need to pay attention to detail.
One of the things that would be interesting to do is see / test out all the details ( and there are a lot of them) to have a flavor of each influence.
Could probably do a spread sheet to mimic senerios.
Note that if a belt dosen't dosen't slip or is not pefect in grab it would be quite repeatable per each revolution.
You would find that belt adjustment / tension will affect steps per unit and would do likewise the same relative to output RPM.
Tough to test properly.
RICH
-
Hi Rich, I like the spread sheet idea, just not learned that yet.
Yes, can vouch for the tension too.
BTW, all tests were done with absolutely no machining load on the spindle.
When the heavy utility type v belt was on, I used 3 tensions. Light, med.(normal) and very tight, all of which produced different amounts of variations at the Mach reported rpm.
The looser settings were the best, but still unsatisfactory.
The tightest setting varied the most.
The relative output rpm isn't really an issue at the spindle as it is with an axis, just needs to be steady and consistent .... but, I see what you mean.
Thanks Rich,
Russ
-
Russ,
The multi-V belts as used on a treadmill would likely give you a much more constant speed. You'll notice that the treadmill motors all have the motor pulley made as part of a very heavy flywheel which I would think really helps stabilize the speed. That type of pulley is easy to make too.
Hi Gary, when first reading this, I casually overlooked the significance of the flywheel effect.
My machine is very light and just uses 4C collets so the rotating masses are minimal.
I might make up a flywheel to add some mass to the spindle. That is a good (another good) suggestion.
Even running a 4 or 5" chuck would seem substantial.
Thanks again,
Russ
:)
-
Russ,
The load on a treadmill isn't constant, it changes with every step you take. Every treadmill maker has a flywheel and I would guess it is because a person walking or running would really feel the speed changes when their foot hits the belt and it suddenly is loaded.