Mill Spindle Direct Drive Conversion

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
Just a little teaser, I got the rest of the parts in today for the vari-drive to direct drive conversion on my mill. I have a bearing failing in the existing drive and I'm getting tired of listening to the noise it makes even with new bearings in it. This conversion will also allow a max speed of 6000 RPM as opposed to the current 4200 RPM.

I'll get started on this in the next few days, I need to get some other work done first and I have a couple of small mill jobs that have to get out the door before I can have it down for a week.;)

So watch this space :grin:
 
Last edited:

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
I got around to unpacking stuff.

Here are most of the conversion parts. Not a heck of a lot to it. A 3 HP, GS3 sensorless vector VFD, two 5 inch, 12 groove Poly-V pulleys, 35mm Taper-Loc bushing, and a 28mm Taper-Loc, and one 30 inch Poly-V belt. I'll need to build a motor adapter plate so I can have belt adjustment, the existing system just has the motor bolted in place and no way to adjust.

upload_2016-7-29_14-56-15.png
 
Last edited:

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
Finally getting started

I need to get it out from under the mezzanine first. Running the final job with the original drive

upload_2016-8-3_7-37-28.png

I think I can rotate the head just a bit and remove the lamps from the light fixture and just slide it out.

upload_2016-8-3_7-39-57.png

And a picture of my spindle dust collector, works great for working with wood and plastic

upload_2016-8-3_7-41-59.png

More later............
.
.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
More fun and games.

Ready to drag it out
upload_2016-8-3_14-23-0.png

This just cleared the vice with almost an inch to spare :cautious:
upload_2016-8-3_14-24-26.png

Just ran the chain around the base and pulled. Just over an idle was all it took. Slow and careful. :eek:
upload_2016-8-3_14-26-56.png

And there it is. :)
upload_2016-8-3_14-28-48.png

I didn't get any pictures of pulling the motor, but I just chokered it with a strap, and lifted it out.

I made this little block a while back to put in the vari-drive pulley to keep it open so it's possible to loosen the belt. Fits right around the shaft. Just a chunk of 1 1/4 pipe.
upload_2016-8-3_14-32-47.png

One small problem with the way the cover is designed, there is no room to get your hand in there to place the block, so I stuck a welding to it for a handle. I'm pretty good at sticking welding rods anyway:D
upload_2016-8-3_14-36-33.png

Now for the problems. I have some parts here from Alloy's Kent mill, and the heads look the same from the outside, so I used those to measure for the Taper-Loc bushings. Well that didn't work out so well. It turns out that all mills are not created equal. The Kent driven shaft is 35mm over it's entire length. Mine is 38mm except for the top bearing area. The good news is that 38mm is a standard Taper-Loc bore. So no problem there. It will be on order this afternoon!

The other thing I failed to concider is the 7 inch cast steel lower pulley half, which is also the brake drum. I sized the new pulleys to take the motor rated speed of 6,000 RPM, but that 7 inch pulley half is only going to take 4250 RPM. Can you say cast steel grenade :eek::eek:. Now what to do????? I could just limit the speed to 4250, or take the cast pulley off and not have a brake, or maybe make a new brake drum out of some 4350 so it would take the speed. Operating without a friction brake would not be a terrible thing, it only gets used for tool changes, never for stopping the spindle. Dropping it in to back gear gives plenty of resistance to do tool changes. Decisions decisions.:oops:

This about where the new driven pulley will sit.

upload_2016-8-3_14-49-55.png

Here is about where the motor pulley will sit in the housing. The good news is the the pulleys will line up just fine with no modifications to anything. BUT..............I ran into another small problem. :confused: The Kent mill motor shaft seemed to be 28 mm (1.102) over the entire length, a little hard to measure because is was torn up so bad. So I ordered a 28mm bore Taper-Loc. Well it turns out that the design of mine is completely different. In the picture you can see the step in the motor shaft, 23mm (0.906) then steps up to 29mm (1.142) just above (actually below, it's upside down on the bench) the pulley, and of course stock Taper-Loc bore sizes are 28 or 30 mm, no 29 mm:mad: I guess I'll bore the bushing out to 29mm and make a special key or maybe turn the motor shaft to 28mm

upload_2016-8-3_15-24-9.png

Next I have to figure out how to tighten the belt. With the vari-drive, the motor is just bolted in place with no provision for adjustment. I think I can get away with just slotting the motor base and call it good. I won't know for sure until I'm able to install the pulleys and see how much travel I need in the motor.

More later......
.
.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
The other thing I failed to concider is the 7 inch cast steel lower pulley half, which is also the brake drum. I sized the new pulleys to take the motor rated speed of 6,000 RPM, but that 7 inch pulley half is only going to take 4250 RPM.
After running some calculations on this I have decided to go ahead and run it as is.

upload_2016-8-4_10-26-9.png
http://keisan.casio.com/exec/system/1271292951

The centrifugal force on the extreme outer rim will be about 10,000lb max at 6000 RPM. So assuming that the cast steel has a conservative yield strength of 30,000 PSI, I should have a pretty good safety factor. This is kind of a back of the envelope calculation, to actually do it correctly it should be an integral calculus problem taking into account the cross section area of the part but I'm to lazy to write a program to do the calculation.

Also concider that automotive flywheels and vibration dampers and other large rotating parts turn well over 6000 RPM on occasion and don't blow up (most of the time) :cautious:

Feel free to jump in here and point out any errors in my calculation;)
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
Well another problem rears it's ugly head. I noticed some looseness in what I thought was in the lower driven shaft support bearing, as it turns out the bearing spun on the inner brake hub and that's what was loose. The bearing seems fine, but I'm going to replace it. For about $10-$15 it's not worth taking a chance on.

Worse than that, the two piece bearing retainer you see in the picture used to be a one piece bearing retainer.:mad: It broke when I was trying to get the bearing off of the brake drum hub. The bearing was stuck hard just above the spun area. Normally these are just a very light press fit. The genius that designed it decided to make it out of cast. The new one will be steel. The good news is that I have Alloy's Shizouka here to make a new one, and I need to test it anyway.

upload_2016-8-4_14-16-27.png

Then the question is: Do I sleeve the hub on the brake drum, or do I just eliminate it altogether and build a spacer to take it's place. o_O There is not a lot of meat (about 2.5mm) there at the keyway so I'm reluctant to remove much material to sleeve it because it's cast also. I need to think about this for awhile. I'll do that while I'm drawing up a new bearing retainer. :blue:
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
After thinking about it a bit, I think I figured out why the bearing spun on the hub. Last time I rebuilt the head, I bought inexpensive bearings off of the Internet for the drive components (not the spindle bearings, went top of the line on those).

When I took this apart I noticed that there was grease oozing out of the bearing, black gooey, tar like grease and this is at 90° F. And a lot of it, probably a solid mass when it's cold in the shop. Even warming the spindle up before cranking it up to working speed would not thin this grease out. No way to see what was in the bearing when I got it because it's a shielded bearing. Never seen a gooey stuff like this in anything but very low speed bearings in heavy industrial equipment or bull gear grease for things like excavator rotators.:confused:

I never gave much thought to bearing grease before for utility applications in normal speed ranges. I will be getting a new bearing from a local vendor and a known quality brand, probably SKF. I learned my lesson, I'm done with cheap bearings.:mad:
 

sanddan

Active User
H-M Supporter - Gold Member ($25)
Joined
Jan 22, 2012
Messages
651
Looking good so far.

As to the brake drum, I'd model it on SW and do a quick FEA using your loading value. For a quick check the standard FEA that comes with the software would give you a ballpark number that would either make you feel good or raise concerns. If your cad doesn't have this capability I can model it for you and run it. I like to use it as a gut check but you have to be careful when using the absolute numbers as with any FEA analysis. Let me know if you are home tomorrow, I can come by with my laptop. I need to see your shop anyway.

Dan
 

mksj

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 12, 2014
Messages
2,316
Why do you need a drum brake with a VFD? With electronic braking it will probably be just as effective, if not work better. On my new mill with a digital head, they do away with the mechanical brake all together, and the fans that are normally incorporated in the mechanical varispeed head are eliminated. You could also program two braking rates, so the default is 2-3 seconds, E-Stop or brake lever would revert to single stage which could be something like 1/2-1 second. You would need a braking resistor. You would need to interlock the brake lever so it also deactivates the run command when activated and requires a restart. They also have a similar switch for the back gear.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
Why do you need a drum brake with a VFD?
I really don't, the only time the brake gets used is to lock the spindle for tool changes. Dropping it into low speed should give enough resistance to do a tool change. If not, I can always design a mechanical lock. I have the VFD set for a 2.5 second stop. We'll see how the new VFD reacts, I'll set it at the minimum that it will take without tripping out. I do have a braking resistor for it.
 

sanddan

Active User
H-M Supporter - Gold Member ($25)
Joined
Jan 22, 2012
Messages
651
No brake on my bench top mill was the main reason I went with an ER32 collet system. Made tool changes much easier than using the R8 collets.
 

mksj

Active User
H-M Supporter - Gold Member ($25)
Joined
Jun 12, 2014
Messages
2,316
I should have read your earlier post more carefully, I assumed you had a power drawbar and it wouldn't take much friction to use it for collet changes. Like Dan, I mostly use and ER system, but I am putting a power drawbar on my new mill so that may change. I have seen a number of spindle locks that use a spring loaded pin to lock the spindle, if you want to get fancy you could use an interlock switch to make sure it is fully disengaged when the run signal is activated. I would think that would be an easier and safer approach then using the cast iron brake system, and maybe less vibration. Looking good otherwise, look forward to it up and running and your final thoughts on the conversion.:cool:
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
I should have read your earlier post more carefully, I assumed you had a power drawbar and it wouldn't take much friction to use it for collet changes.
My spindle is an Erickson NMBT30 QuickChange, no drawbar at all. You loosen the nut with a special spanner wrench.

$_35.JPG

upload_2016-8-5_10-47-24.png
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
I would think that would be an easier and safer approach then using the cast iron brake system, and maybe less vibration.
I agree. I have decided that the brake drum is going away. I'm just going to build a spacer to take up the space where it used to be. I will build a lock pin system if it needs it.
 

derf

Brass
Registered
Joined
Oct 3, 2015
Messages
808
The Erickson quick change system is not all that quick, that's why installed a power drawbar.
My mill had a solid spindle, so I removed it and gun drilled it for a 1/2-13 drawbar. I never use the spindle brake to change tools. But on the other hand, I still have the lower pulley on my conversion and have run it up to 6200 rpm. If I were to do it over, I'd eliminate it and use a disc brake.
I had the same problem with that bearing retainer, and ended up making one from steel.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
Looking good so far.

As to the brake drum, I'd model it on SW and do a quick FEA using your loading value. For a quick check the standard FEA that comes with the software would give you a ballpark number that would either make you feel good or raise concerns. If your cad doesn't have this capability I can model it for you and run it. I like to use it as a gut check but you have to be careful when using the absolute numbers as with any FEA analysis. Let me know if you are home tomorrow, I can come by with my laptop. I need to see your shop anyway.

Dan

Dan did you get my PM?
 

alloy

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Jul 5, 2014
Messages
990
I like the poly v belt your using. The one on the Kent (single small 1/2" V belt) slips under even a medium load at low RPM. I would love to convert it over to the different style belt.

I don't like having a nice heavy 3HP mill that I can't use the full power on. But, it's still light years better than my Bridgeport.

Is the Shizouka ready to go?
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
I was going to go with a single V-belt, but I was talked out of it by some of the members here, so I went with the poly-V. Glad I did now. Converting yours over should be pretty easy, and about $180 in parts.

Is the Shizouka ready to go?
Almost, just need to finish the tool height comp. And then test run. See if you can get me some G-code for a simple part generated with SolidWorks and HSM or whatever CAM program you are using. Maybe a square with a pocket in it and a hole pattern.

I see your BP is for sale, at that price it's going to fly out of there.
 

alloy

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Jul 5, 2014
Messages
990
Yes just posted the BP. I went with $3k because there is a vari speed on for $4k. I only have about $2k in it and it made me about $12k in 2 years, so doesn't hurt me sell it for that price.
I got tired of waiting for my engineer friend at work to buy it. I need to room and the $$$ wouldn't hurt either. I haven't used it since the day the Kent went online.

I am on vacation this next week so I don't have access to any code files with pocketing right now. And I haven't been working on HSM, been concentrating my efforts on Solidworks. Believe me, learning Solidworks is enough to do at one time.

I think I sent an example of tool height call outs awhile back. I'll see if I can find that.
 

alloy

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Jul 5, 2014
Messages
990
Where did you get your poly belt and parts? I'm seriously considering converting my mill over. I already have motor adjustment, so hopefully if would be easy to just add the new pulleys and belt.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
http://www.ametric.com/default.aspx?

321495 12J50.2012 ANSI POLY-V PULLEY 2 $65.90 $131.80 OK

998912 2012.35 TAPER BUSHING 35MM BORE 1 $12.11 $12.11 Wrong Size

384213 2012.1-1/16 TAPER BUSHING 1-16" BORE 1 $14.81 $14.81 Wrong Size

781027 300J12 POLY-V BELT 1 $19.37 $19.37 OK, I think

This is what I originally bought, BUT...... the bushing sizes were wrong, but I took a chance based on the Kent sizes. Bushings are cheap, so it was worth it. I didn't know what sizes were actually needed until I tore it down. Took about 2 days to get the correct bushings.
 

alloy

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Jul 5, 2014
Messages
990
I looked up the pulleys and belt, but there is no spec on how wide the pulley is.

Can you measure it and let me know when you have a chance?

I would like to keep my brake, so need to go above the front pulley with the new setup and need to see if the new pulley will clear everything.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
1.283 thick. You pretty much have a mile of room in there. My shaft sizes are 38mm on the spindle end and 29mm on the motor. Your mileage may vary ;)
 

alloy

H-M Supporter - Gold Member
H-M Supporter - Gold Member ($25)
Joined
Jul 5, 2014
Messages
990
Yeah I have plenty of room for a pulley that thick. Just couldn't tell at all how thick it was from the pics of it or the belt on their web site.

Now I just have to get that heavy old motor off the top of the mill to measure everything. I've had it off twice since I got it here and dang near dropped it both times.
 

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
I got a bit more work done today.

I decided the brake was going away so that means some redesign.

First a new bearing retainer to replace the broken one (see above), also the new design uses a smaller bearing because the old bearing hub was built into the brake drum.

see here for the build on it:
http://www.hobby-machinist.com/threads/work-holding-ideas.34781/#post-415990

Still need to drill the drill & tap the bolt holes.
upload_2016-8-7_16-54-8.png

Then I needed a new spacer that sits below the bearing, it also acts as a belt guide for the low speed drive belt. The original was 50mm ID, the new one needs to be 40mm ID. I looked all over the shop for a piece of 2 1/4 inch or so round stock to make the spacer out of, but no luck. But I finally found a 1 1/2 inch grade 8 bolt that is 2 3/16 across the flats, so yeah, I can get a spacer out of that. Nice hard-ish steel, will make a nice spacer:)
upload_2016-8-7_17-1-19.png

I didn't get any pictures, but it didn't take long to make. Grade 8 bolts cut just fine with carbide, even interrupted cuts. I use only the finest AL-8 carbide lathe bits from Harbor Freight, about $2 each.

I built another stub arbor and pressed it on to face the other side.
upload_2016-8-7_17-6-32.png

Before pressing onto the shaft.
upload_2016-8-7_17-7-48.png

Then I built the spacer that goes between the bearing and the snap ring groove above and pressed the whole assembly together.

Here is the new shaft assembly parts. I still need to drill & tap the bolt holes in the bearing retainer.

upload_2016-8-7_17-13-41.png

And the rough location of the new pulley

upload_2016-8-7_17-14-50.png

The next steps are to build new shaft keys because the keyways are not standard sizes and don't fit the Taper-Loc bushings so I need step keys. Then bore the Taper-Loc bushing for the motor shaft. Also figure out how to tighten the belt. Oh, and drill & tap the bearing retainer bolt holes:rolleyes:

Stay Tuned..........:)
 
Last edited:

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
I thought I was going to get the mill back together today, but it was not to be.:(

First thing I did today was to bore the Taper-Loc bushing to fit the motor shaft.

First I just snugged up the set screws just a bit to seat the Taper-Loc bushing in the pulley bore.
upload_2016-8-11_19-9-50.png

Then I set it up in the 4-jaw. I used a bull nose center to get it sorta close to center, to minimize setup time. It was about 0.005 radial TIR. Note that the pulley is mounted so the small end of the taper is out, so the cutting forces will be against the set screws and not trying to push the bushing deeper into the pulley.
upload_2016-8-11_19-14-42.png

With the jaws snugged up a bit, first indicate the face so the pulley doesn't wobble. Shim as needed, you can see the white piece of paper behind the pulley.

upload_2016-8-11_19-15-43.png

Then indicate the radial runout. I picked the top of one of the ribs as my measuring point. This is a 1.125 bushing and it needs to be 29mm. (1.1403) I chose this bushing because I only needed to increase the bore by 0.015 to fit the motor shaft. It came out fine.

upload_2016-8-11_19-20-22.png

Then I had to make step keys because the keyways in the shafts and bushing are not the same. The shaft keyways are non-standard. So the first order of business was to mount the vice on Alloy's mill and cut a step in the vice soft jaws to hold the key stock. 0.250 wide by 0.100 deep. I closed the jaws up on a pieces of paper in each side just so I could make sure they were closed. I normally keep soft jaws in my vice, they are sacrificial and can be machined as needed for holding odd parts. I normally just leave them in until they are so chopped up so bad that they are no longer useful. These have been used for a number of different setups. These are cold rolled steel, but I also make them out of aluminum.
upload_2016-8-11_19-38-5.png

And the new step keys in place. The one sitting on the pulley doesn't look like it has a step, but trust me it's there. I have to say that Alloy's Shizouka cuts exactly where you tell it to. Nice machine to run. And the 4HP, 40 taper spindle didn't even know it was doing anything with that 1/4 inch end mill in there. BUT...............:mad:
upload_2016-8-11_19-39-51.png


Here is the problem: THE BELT IS TOO SHORT BY 6 INCHES:mad::confused:

I guess I need to go back to school and take a math class again. :rolleyes: When the pulleys are the same size, it's Center to Center distance x 2 plus the circumference of 1 pulley. Somehow I screwed that one up. :confused: There is a new belt on order.:oops:

Looking down into the drive housing that is sitting upside down on the bench. Oh, I got the bolt holes drilled and tapped in the new bearing retainer too!

upload_2016-8-11_19-47-8.png

Tomorrow I'll get the new VFD mounted and the new controls hooked up. I can't do any more on the mechanical right now anyway:(
.
.
 
Last edited by a moderator:

JimDawson

Global Moderator
Staff member
H-M Platinum Supporter ($50)
Joined
Feb 8, 2014
Messages
8,080
While waiting for the new belt I got the new VFD mounted and wired up. The only reasonable place to mount it was on the back of the electrical cabinet. The old VFD was mounted on the head, but the new one is a bit bigger so I wanted it out of the way. The breaking resistor is below the VFD. Sorry for the crappy picture, the sun was shining in the door.
upload_2016-8-13_23-34-47.png

This required building a remote control station to mount on the side of the head. It will be mounted where the old VFD was mounted.
upload_2016-8-13_23-36-19.png

Spent the day tuning the motor and getting rest of the wiring connected in the computer cabinet. I now have all of the wiring in place to have the VFD under full computer control. The VFD has a pulse output that is equivalent to the output frequency, so I connected that to one of the unused encoder inputs on the controller. So I now have the RPM readout on my screen. Getting that to all work correctly was a PITA because the frequency output seems to be unstable and it does not seem to be electrical noise. I had to write some averaging algorithms and take the arithmetical mean averaged over several samples. I may have to build a hardware averaging filter to clean it up, or just put an encoder on the motor. I can get away with taking the speed value from the VFD because the system is direct drive now.

A forklift makes a great temporary motor mount for setting up. :)

upload_2016-8-13_23-51-39.png
 
Top