VFD conversion - drill press

[precisionworks]

A 3 hp motor & VFD had been purchased for installation on the mill. After buying the push-pull vise for the drill press, it seemed like that machine would really benefit from a larger motor & variable speed drive. The drill press is a Craftsman, purchased new around 1982, made by the Atlas Press Company - the same company that made the Craftsman metal lathe. I saw one in an Atlas catalog & it had a larger motor & a conventional (wide) belt drive. The first operation is fitting a four step pulley.

The eight step aluminum pulleys gives a speed range from 380-8550 rpm. And the "belt" has barely the ability to transmit 1/2 hp. The front pulley was pulled & the shaft measured:

The shaft is 25mm metric, and a metric bore pulley couldn't be found, so I ordered a 7/8" bore Baldor/Maska four step.

The bore indicated about .0001" total runout using the 3-jaw chuck, so it was ready for boring out. Since the bore needed to be about .109" larger, three passes were made removing .025" each time, followed by .020", followed by .010", then .005". These cuts were made using the dial on the cross slide, which is good enough for roughing work. A dial test indicator was used to get close to the target dimension.

The target dimension was between .9843" and .9845". Any larger than .9845" would give a loose fit, and any smaller than .9843" would make it hard to install the pulley on the shaft. The last pass showed a mic reading of .9842" which was just a bit small, but the pulley was removed from the lathe for trial fitting on the shaft - you can always go back for another cut if needed. Firm blows were able to get the pulley started but it was too tight, even after warming the pulley with a propane torch. Back to the lathe for one more cut, without disturbing the last setting. The 3/4" boring bar will deflect about .0002" when extended 3", so one more pass should remove that amount. The boring bar was run in with power feed, and run back out with power feed. The insert did not touch on the infeed, but just barely made contact as it backed out. The mic showed that the bore was .9844", pretty close to ideal. The pulley was again warmed with the torch & seated nicely. The dial test indicator showed .0002"-.0003" total runout on the bore 😀

The old motor & mount were removed & distance from top of pulley to top of mounting holes measured at 5.75". To make the new motor fit, a piece of 5" wide, 1/4" thick flat stock was welded to the motor base.

The motor plate is marked & ready to drill. More to follow tomorrow.

 

[wquiles]

Very nice. That new vs. old motor picture is very telling of how underpowered the original setup was!

Looking forward to more pictures 😉

Will

 

[precisionworks]

The motor is moved into position using the scissor lift table with a length of 4x4 and 6x6 giving the table added height.

Nuts and flat washers are threaded on both sides of the motor plate. They allow the motor to be adjusted exactly so that the rear pulley & the front pulley are in perfect alignment. Note that the bottom adjusters are threaded out further than the top ones.

The Starret 385-48 straight edge rests on the front & rear pulleys while the rear pulley is moved into alignment.

All the mechanical work is done, time now to start the electrical connections :twothumbs

 

[wquiles]

Getting closer and closer 😉

Will

 

[precisionworks]

All the mechanical work is done,

Well, not quite 😀

My plan was to wall mount the VFD & use a remote switch & speed pot, but it looked to be more simple to mount the VFD just to the side of the drill press. First step is cutting 2x4 lumber to strap across two studs:

This forms the support base for the VFD mount. The mount is made of 2" angle & 1" square tube, welded together:

The VFD bolts to the mount & is in ideal position to operate when drilling:

time now to start the electrical connections

The junction box at the motor is the first item to wire:

More to follow 😀

 

[precisionworks]

Next step is to bring the four wires from the motor (phase A, phase B, phase C, and ground or PE) to the drive. The first three wires on the left side are the motor phase connections, then the common ground with green wires, and the 240 volt single phase power are the black & white wires on the right.

The black, white & green input wires go to a disconnect (like a very large light switch, rated for 60 amps, that breaks both the hot legs). The drive wires are connected to the bottom disconnect terminals, and the supply line wires are connected to the top.

All hooked up and ready to go :thumbsup:

The disconnect was flipped to the on position which powers up the VFD. Then the VFD start button was pressed to check motor rotation ... and it rotated in reverse. Not a problem with a three phase motor, as swapping any two leads reverse direction. After that, it was just a matter of setting parameters & running the auto calibration.

The lowest pulley step gives a range from 71-1300 rpm.

The second step gives a range of 115-2100 rpm.

The third and fourth steps may never see any use, as the first two should do everything I need. Power is immense even at 71 rpm, which is just over one revolution per second.

 

[wquiles]

Next step is to bring the four wires from the motor (phase A, phase B, phase C, and ground or PE) to the drive. The first three wires on the left side are the motor phase connections, then the common ground with green wires, and the 240 volt single phase power are the black & white wires on the right.

(snip pic)


The black, white & green input wires go to a disconnect (like a very large light switch, rated for 60 amps, that breaks both the hot legs). The drive wires are connected to the bottom disconnect terminals, and the supply line wires are connected to the top.


(snip pic)


All hooked up and ready to go :thumbsup:

(snip pic)

The disconnect was flipped to the on position which powers up the VFD. Then the VFD start button was pressed to check motor rotation ... and it rotated in reverse. Not a problem with a three phase motor, as swapping any two leads reverse direction. After that, it was just a matter of setting parameters & running the auto calibration.

The lowest pulley step gives a range from 71-1300 rpm.

The second step gives a range of 115-2100 rpm.

The third and fourth steps may never see any use, as the first two should do everything I need. Power is immense even at 71 rpm, which is just over one revolution per second.

Very nice. Thanks for showing the wiring in detail - since I am using the same drive/controller, mine will be about the same 😉

Will

 

[precisionworks]

After running the drill for a few days, these are the parameters that work well:

Acceleration (or accel) set to 4 seconds. A shorter setting like 3 or 2 seconds would also work well, but a 4 second accel leaves enough time to hit the stop button if something isn't right.

Deceleration (or decel) ended up at 0.4 seconds. I don't yet have the braking resistor, and probably will not buy one, as the motor stops quickly without one. It also reverses very quickly at the lower rpm settings - 70 rpm (like you might use for power tapping) is instant reverse, while 300 rpm takes about one second to stop & reverse.

Setting the decel time is interesting as it's trial & error. I start with 5 seconds & drop down one second each try until the drive faults out. With this drive, 4 - 3 - 2 - 1 - .9 - .8 - .7 - .6 - .5 and .4 seconds all stopped the drive. At 0.3 seconds, the drive faulted - which is a built in protection program to keep regenerated voltage from killing the transistors. Switching back to 0.4 seconds gave the shortest decel setting.

Minimum frequency set to 5 Hz. In America, the standard utility frequency is 60 Hz, so 5 Hz makes the motor run at 5/60 of full speed ... about 8% of full speed. The 1725 rpm motor on the drill turns 144 rpm at this setting, and the pulleys reduce that to 72 rpm.

Maximum frequency set to 90 Hz. This makes the motor run at 90/60 of full speed (150%), so the 1725 rpm motor turns 2588 rpm. This is decreased or increased by the pulleys.

So far, the results have been better than expected 😀

 

[wquiles]

Awesome - I am surprised at how quickly you can get it to stop - pretty impressive!

I am finishing some flashlight work, and then I will have to do my own VFD conversion. I am sure I will have plenty of questions for you 😉

Will

 

[precisionworks]

I am surprised at how quickly you can get it to stop

Without using a braking resistor, there's a lot of variation from machine to machine. The machines that have a low rotating mass - like the drill or the wire wheel machine - will stop on a dime. The belt/disc sander, which turns a 12" cast iron disc plus two large drive drums, will fault out around 3 seconds - it needs a braking resistor.

I'd be willing to bet that your lathe, with its heavy chuck & gear train, will be like the sander. When the stop button is pushed on a VFD, the spinning motor becomes a generator. If you select the "coast to stop" menu option, the generator is unhooked from the VFD & spins down at its own pace. If you select "ramp to stop" the generator is feeding power back to the VFD, and the drive has the ability to absorb a limited amount of energy. To stop a generator plus a gear train plus a heavy chuck (plus any work held in the chuck) is too much for the drive without the braking resistor.

You might as well order that now ... EZXDB2222A1 ... always happy to help spend your money :nana:

 

[Atlascycle]

What HP was the Drill Press before the upgrade?

Jason

 

[precisionworks]

It was a (weak) 1/2 hp, 380 rpm minimum, with a belt that slipped 😱

 

[wquiles]

I'd be willing to bet that your lathe, with its heavy chuck & gear train, will be like the sander. When the stop button is pushed on a VFD, the spinning motor becomes a generator. If you select the "coast to stop" menu option, the generator is unhooked from the VFD & spins down at its own pace. If you select "ramp to stop" the generator is feeding power back to the VFD, and the drive has the ability to absorb a limited amount of energy. To stop a generator plus a gear train plus a heavy chuck (plus any work held in the chuck) is too much for the drive without the braking resistor.

You might as well order that now ... EZXDB2222A1 ... always happy to help spend your money :nana:

I am not yet sure if I will need it on the knee mill, but I am pretty sure I will have to order this for my lathe. If you watched the video I posted in the multi-use lathe tool, you can see how long it takes the spindle to stop on its own - a long time, in fact, way too long for my taste

 

[wquiles]

Barry,

Some email traffic on the yahoo groups talking about VFD's and the breaking resistor mentioned being able to make your own high power resistor. The AC Tech data sheet on the breaking resistors says that for 3-4HP:
R = 42 ohms
Wperm = 145
kWmax = 3.6

and for 5.5HP:
R = 28 ohms
Wperm = 210
kWmax = 5.3

Does Wperm mean the constant/nominal power (as in 145 watts continuous) while the kWmax refers to the peak power the resistor must handle over short durations? Does the R value mean the minimum value, or max. value for that power range?

There are lots of high power, 150-200 watts resistors in Ebay that could be made into a nice external breaking resistor, but I am not sure how to interpret those values above 😕

 

[precisionworks]

There are lots of high power, 150-200 watts resistors in ebay

Braking resistors are made of large coils of resistance wire wound over ceramic forms. Then the resistors are mounted on a heat sink and enclosed so no one gets burned, which also keeps from burning down your shop or house.

By the time you purchase just the components, it's as cheap to buy the unit designed for your drive.

http://news.thomasnet.com/fullstory/500968

 

[wquiles]

Barry,

I am not going to make my own, but I do want to understand how they work and how they are rated. Do you have any insight on those numbers and what they mean? Please?

Will

 

[precisionworks]

Do you have any insight on those numbers and what they mean?

For the smaller resistor, kWmax = 3.6 means that the resistor can absorb 3600 watts of energy based on a temperature rise of 375°C above ambient (40°C). As some of the web sites state "If the resistor elements glow red you may need a higher rated braking resistor".

3600 watts = 36 incan light bulbs at 100 watts each. For a few fractions of a second, that's how much energy the resistor has to absorb & dissipate without exceeding 375°C above ambient. It's a big job, even for large resistors mounted on a heat sink, to stay within the temperature parameters.

 

[wquiles]

For the smaller resistor, kWmax = 3.6 means that the resistor can absorb 3600 watts of energy based on a temperature rise of 375°C above ambient (40°C). As some of the web sites state "If the resistor elements glow red you may need a higher rated braking resistor".

3600 watts = 36 incan light bulbs at 100 watts each. For a few fractions of a second, that's how much energy the resistor has to absorb & dissipate without exceeding 375°C above ambient. It's a big job, even for large resistors mounted on a heat sink, to stay within the temperature parameters.

Glowing red - that is absolutely insane!. I had no idea it would have been this much above ambient 😱

So not only we have to select the correct breaking resistor for the VFD, we also have to worry about the duty cycle? So if the 3HP breaking resistor were to get too hot, the "solution" would be to go to the next larger one (5HP)? Would you have to adjust/program the VFD so that it knows that size/rating breaking resistor is using?

 

[precisionworks]

the "solution" would be to go to the next larger one (5HP)

There are different ways to configure the brake resistors to work. One would be to go up in size - on a 3 hp motor with a huge inertial load (like a pump jack) you might need a 10 hp or even a 20 hp brake resistor module. With the SMVector drive, you also have the option of "stacking" modules by connecting multiple modules in parallel.

Would you have to adjust/program the VFD so that it knows that size/rating breaking resistor is using?

The monitoring software built into the drive is primarily concerned with one thing, which is keeping the DC bus voltage at 114% of max or below. A module that is too small for the inertial load will not keep the DC bus voltage in that range & the over voltage will trip the drive into a high voltage fault. As more or bigger modules are added, the drive dumps all the regenerated energy into whatever is available, so no adjustment is needed.

This tech doc may answer some questions:

http://www.actech.com/documents/techlib/AN0039A.pdf

 

[wquiles]

Thanks much Barry for the info and links. It is also good to know that I can connect modules in parallel is the initial one is not "enough" 😉

As you recall besides the Baldor 3HP that I will use in the Knee Mill's VFD conversion, I also have a brand new Baldor 5HP motor and I would like to use for my 12x lathe. If I recall, I need the 10HP VFD, and use it on a "de-rated" mode? Would I then use the 5HP breaking resistor to match my motor, or would I then need to use the 10HP breaking resistor? 😕

Of course, if this gets too complicated I might just use the 3HP motor on the lathe as well, but I really like the idea of the larger motor, which is of course what "Toolman Taylor" would do :devil:

Will