Power plot of a CNC machine cycle.

Status
Not open for further replies.

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150625-1923 EDT

Attached is a power plot of one cycle of a HAAS VF3 CNC machine drilling and tapping a 1/4-20 hole.

The measurement was made with a TED 1000 System. The primary supply is an open delta wild leg source with one phase being 240-0-240. This is the phase the TED 1000 was connected to. The power axis is scaled for the single phase measured. Assuming the three phases are approximately balanced, then total power is 3 times the single axis value. Note: that by powering the TED from 120 of the source that the voltage is properly phased with respect to the load currents. If the source was a wye, then a 240 to 120 transformer would be required to power the TED MTU.

The residual standby power required is about 3 * 120 W with the servos on. This power is seen as the left most and right most straight lines.

The power measurments are the average power over a 1 second period. One second = 0.000,278 hours. The x-axis is calibrated in hours. From 10.685 to 10.690 is 18 seconds. This is a connected line graph. The one second average power values mostly occur where you see a change in direction of thre curve.

The energy to accelerate the spindle to 6000 RPM is 12 kW for 1/3600 hour or 0.0033 kWh and cost about $ 0.000,5. The cost to drill the tap hole was about (0.6*3)*(4/3600)*0.16 = $ 0.000,32 .

If we assume the machine value is $100 / hr, and machine time is 41 seconds (41 is a better value than the 36 in the plot comment), then the price for the hole is a $ 1.14, and that does not include load and unload time. If many of the same size tapped holes were processed without tool changes, then the cost would be much lower per hole. No matter what, the electric cost is relatively insignificant compared to the machine cost. But this does not mean that reducing electric cost is not worthwhile.

If this was a heavy cutting operation that was removing a lot of metal, then electric costs would be much greater.

P-453-150520B.jpg
.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150626-0920 EDT

The plot I originally included was a fairly high resolution .PNG file that imported with greatly reduced line quality. Thus, not very readable. Here I am trying again. This time from a .JPG file.

This is somewhat better, but not good.

The plot would be more useful with a shorter power averaging time. But TED 1000 only quantizes to 1 second. 100 or 10 milliseconds would be better, but I would have had to haul too much equipment to the shop to do that. TED 5000 and other models do not even resolve 1 second. They are 2 seconds or longer.

But TED is a convenient tool to use.

150520B.jpg


.
 
Last edited:
Status
Not open for further replies.
Top