Thread: VSD power tests

bob and stu. I have to say thats very impressive. You guys rock.
thanks heaps for taking the time to do all this. You know build the test rigs
and run the tests,
There is a lots of us guys on this forum that have been considering differant
options for powering our machines. And to now know is fantastic if not a little disappointing.
I only wish that the reality of it all wasn't so true and some sort of magic tric
allowed us maintain higher toque values over a bigger freq range. with out really
spending hardly any cash. or having the know how.

When i decide what i want for my other two wishes ill let you now.

there was a lot of other contributors to this thread. Including my self. That im not
real shore how helpful some of us were.
thanks again.
aaron

bobl gets ship done

Originally Posted by azzrock

. . . . .bobl gets ship done

Not according to SWMBO, she reckons I'm as slack as they come completing tasks on the "Honey Doo" list.

Anyway, thanks for the feedback.

Originally Posted by azzrock

bobl gets ship done

Thanks Aaron, but I didnt do much,,,,,,,just set the ball rolling and walked away.. Bob's done most the work.
One of my failings is I start a project, work on them until I'm happy it will work............ then stop and move onto something else. Trying to fix that(which is the reason the dyno got put on hold lol)

Originally Posted by BobL

Not according to SWMBO, she reckons I'm as slack as they come completing tasks on the "Honey Doo" list.

You to?

Stuart

I had a bit of spare time this arvo to do some more measurements.

Firstly based on some of the discussion I decided to recheck the HP output of the motor above 50Hz.
When I did this last time I measure the HPs at 50 Hz then 40 Hz, 30 etc, and then later I went back to the 55 and 60 Hz. It appears I must have gotten used to the very slight additional handbrake pressure that would stall the motor at low frequencies and was worried about doing this at high frequencies so I did not properly "push" the motor at these frequencies.

So I now have checked the HP output up to 100 Hz.
Unless the top of each set of graph lines flatten out and starts to kick over (as can be seen for the 40, and to a certain extent for the less than 35Hz lines) then the curves do not show the stall point of the motor and it is possible that the motor can produce a bit more HP than is shown on the lines for some frequencies. I do drive the motor briefly past its stall point every time and all of the top most HP points are within a a small handbrake squeeze from stalling. It is almost impossible to find and sit just below the stall point with the hand brake as just a touch too much pressure stalls the motor and the data for a stalled motor is somewhat meaningless

The points with the coloured lines represent the new data and you can see that the motor is capable of generating 1HP between 35 and 100Hz. I don't know why the higher HP values for the repeated 60Hz data are greater than last time.



Then I reconfigured the motor as a "Y" and repeated some of the above measurements.
The stall points are pretty clear for the 30, 40 and 80Hz curves.
It looks like a Y motor can output about about half the HP compared to the the motor configured as ∆ at any give frequency.

Bob,
If I read these graphs right, would the Hp curve roughly follow the tops of the lines like this?
HPVFdelta.jpg
If so, that means my use of above standard motor frequencies makes more sense that what others (like CBA) have tried to leed us believe...
So using the lower frequencies really only makes sense in connection with a reduction gear/belt ratio.
However, my experience is that it might be torque that is more relevant to machining than Hp. Why I say this is that both my lathe and mill work very nicely at very low rpm - as low as 12Hz for example. According to this graph, the power output is pretty negligable, but the torque must still be quite reasonable to overcome cutter loads.
Perhaps there is an explanation for that which evades me at present? My total reduction on the lathe is just shy of 2.7:1, my motor is an old 3kW 960rpm one and I get about 900rpm maximum measured spindle speed on that ratio at 120Hz and about 60rpm at 10Hz,
So my useful spindle speeds of 60:900 on the one belt drive ratio is nearly 30:1! That is WAY over the speed range predicted by most people.

Hi Joe,

As per Ray drawing torque is roughly* the same below 50hz,
So it should be able to take roughly the same cut at 700rpm as it can at 1400rpm........ it just takes twice as long**.
Now I would assume the same cut at half the speed would less an half the power, so you gain a little there.

15:1

Stuart

*I doubt its really a straight line all the way to 0 even with a vector drive, but maybe it is?

**now granted we normally spin things slower because the diameter is bigger. So if you kept the same DOC and feed you would need more power

Originally Posted by jhovel

Bob,
If I read these graphs right, would the Hp curve roughly follow the tops of the lines like this?
HPVFdelta.jpg

Yes, that's true if you always run the motor at the stall current.. I'd drop down a bit from the stall point and try to guess where it hits nameplate current.
But generally the graph looks pretty much as expected.

Originally Posted by jhovel

So using the lower frequencies really only makes sense in connection with a reduction gear/belt ratio.
However, my experience is that it might be torque that is more relevant to machining than Hp. Why I say this is that both my lathe and mill work very nicely at very low rpm - as low as 12Hz for example. According to this graph, the power output is pretty negligable, but the torque must still be quite reasonable to overcome cutter loads.

Perhaps there is an explanation for that which evades me at present? My total reduction on the lathe is just shy of 2.7:1, my motor is an old 3kW 960rpm one and I get about 900rpm maximum measured spindle speed on that ratio at 120Hz and about 60rpm at 10Hz,

Torque is pretty much constant below the base speed. see the graph I posted earlier for the explanation.

Originally Posted by jhovel

So my useful spindle speeds of 60:900 on the one belt drive ratio is nearly 30:1! That is WAY over the speed range predicted by most people.

Certainly way over the way I calculate it... try 15:1

Ray

Originally Posted by RayG

Yes, that's true if you always run the motor at the stall current.. I'd drop down a bit from the stall point and try to guess where it hits nameplate current.

I will calculate the name plate current for each frequency and mark these on the each of the plots and repost.

But generally the graph looks pretty much as expected.

That is comforting to know. I'm really only now feeling like I'm getting the hang of doing these measurements without cooking the brake band and the drum, making sure everything is aligned and turned on (like the DMM and luggage scales have an auto turn off feature that I cannot disable )

Must have been late.... fr basic maths to escape me....

Originally Posted by BobL

Not according to SWMBO, she reckons I'm as slack as they come completing tasks on the "Honey Doo" list.

Anyway, thanks for the feedback.

id call that list the auxiliary to do list.

Originally Posted by jhovel

Bob,
If I read these graphs right, would the Hp curve roughly follow the tops of the lines like this?
HPVFdelta.jpg
If so, that means my use of above standard motor frequencies makes more sense that what others (like CBA) have tried to leed us believe...
So using the lower frequencies really only makes sense in connection with a reduction gear/belt ratio.
However, my experience is that it might be torque that is more relevant to machining than Hp. Why I say this is that both my lathe and mill work very nicely at very low rpm - as low as 12Hz for example. According to this graph, the power output is pretty negligable, but the torque must still be quite reasonable to overcome cutter loads.
Perhaps there is an explanation for that which evades me at present? My total reduction on the lathe is just shy of 2.7:1, my motor is an old 3kW 960rpm one and I get about 900rpm maximum measured spindle speed on that ratio at 120Hz and about 60rpm at 10Hz,
So my useful spindle speeds of 60:900 on the one belt drive ratio is nearly 30:1! That is WAY over the speed range predicted by most people.

Joe, what it takes to remove a given volume of metal in a given time, is a given horsepower. Think about that.

Using the lower frequencies of a VFD drive really only makes sense:

A) for those tasks that can be done with reduced power (chamfering, threading, grooving, light cuts and the like). In practice, this means that much less belt or gear changes are necessary, because tasks with low power requirement usually do represent a large portion of machine time on a manual lathe.

or

B) If the complete drive has been sufficiently overrated at base speed, such as not to fall below the desired minimum power at lower frequencies. In practice, this means that the design power of the VFD drive has to be doubled, for each mechanical gear (or belt position) that is to be left off. Example: if a lathe had 4 speed settings and a 1HP motor, and you desire to completely do away with gear changes, you would need a VFD drive of 8HP. If you decided to keep one on the 4 gear settings (say the back gear) then a 4 HP VFD drive would suffice.

Of course you can also combine A) and B) to reach a compromise between cost and performance suitable to your needs. In practice, most hobby users will choose to keep the mechanical gearbox or belt drive and go for a VFD drive the same size the fixed speed motor was. It is the most cost effective solution.


Edit: the typical hobby user does not go variable speed to do away with mechanical gears. Doing so would be far too expensive. There are many other very good reasons to upgrade to variable speed. Like being able to avoid speeds that cause chatter on lightweight machine tools, being able to keep cutting speed nearly constant when facing or parting, being able to do critical tasks very slowly, being able to polish at very high speeds, being able to start the spindle slowly.......... and many more.

Originally Posted by BobL

I had a bit of spare time this arvo to do some more measurements.

Firstly based on some of the discussion I decided to recheck the HP output of the motor above 50Hz.
When I did this last time I measure the HPs at 50 Hz then 40 Hz, 30 etc, and then later I went back to the 55 and 60 Hz. It appears I must have gotten used to the very slight additional handbrake pressure that would stall the motor at low frequencies and was worried about doing this at high frequencies so I did not properly "push" the motor at these frequencies.

So I now have checked the HP output up to 100 Hz.
Unless the top of each set of graph lines flatten out and starts to kick over (as can be seen for the 40, and to a certain extent for the less than 35Hz lines) then the curves do not show the stall point of the motor and it is possible that the motor can produce a bit more HP than is shown on the lines for some frequencies. I do drive the motor briefly past its stall point every time and all of the top most HP points are within a a small handbrake squeeze from stalling. It is almost impossible to find and sit just below the stall point with the hand brake as just a touch too much pressure stalls the motor and the data for a stalled motor is somewhat meaningless

The points with the coloured lines represent the new data and you can see that the motor is capable of generating 1HP between 35 and 100Hz. I don't know why the higher HP values for the repeated 60Hz data are greater than last time.



Then I reconfigured the motor as a "Y" and repeated some of the above measurements.
The stall points are pretty clear for the 30, 40 and 80Hz curves.
It looks like a Y motor can output about about half the HP compared to the the motor configured as ∆ at any give frequency.

thanks for the star figures.
I think half is a good result and ill go and hook my deckle up today. on the lower rpm winding of the motor thats all ready
on the machine.
This is the basically what joe predicted to me recently. thanks joe..
now i wish i had of all ready found a mt4 boring head arbor..

Originally Posted by jhovel

Bob,
If I read these graphs right, would the Hp curve roughly follow the tops of the lines like this?
HPVFdelta.jpg
If so, that means my use of above standard motor frequencies makes more sense that what others (like CBA) have tried to leed us believe...
...................

I never said that Joe. On the contrary. I routinely use my VFD's at frequencies up to 130Hz. But you got to look at the limitations. Unfortunately the plot we are looking at stops at 100Hz. If it did continue, you would see that with a common motor, shaft power quickly drops down at frequencies above 100Hz, the motor will usually stall spontaneously at about 130-140Hz just from the load required to turn its own fan. There are special motors though with a base frequency of 400Hz, as used on aircraft for example. At work, I use turbomolecular vacuum pumps whose motors are driven by VFD's working at 500 and up to 1000Hz.

Hi Bob,

How do you feel about some more tests?

Tweaking the base freq in Star to 29hz. This will increase power below 50hz but wont effect power above 50hz.

Feel like playing around with the V/F curves?

Looking at the manual I wonder if I've stuffed up on some of my settings, mixing up Min Freq and Freq lower limit.

Stuart

Originally Posted by Stustoys

Hi Bob,

How do you feel about some more tests?

Sure, . . . . but it depends on what.

Tweaking the base freq in Star to 29hz. This will increase power below 50hz but wont effect power above 50hz.

And then measure HP? at what frequency interval and over what range?
Doing the measurements is relatively easy, its the up loading of the photos and transferring the data to the spread sheet that takes up the time.

Feel like playing around with the V/F curves?

This means measuring V? I haven't got the true rms V meter from work yet.