What's New


Workshop Stuff


Moving the mill

Spindle Noises

ShumaTech Digital Readout

ShumaTech DRO Continued

DRO-350 Repairs

South Bend 9" Lathe

South Bend 405 Lathe Bench

Grizzly Mill Revisited

Surface Grinder Rebuild

Surface Grinder Continued

Grinder April 6, 2008

Grinder April 20, 2008

Grinder August, 2008

Grinder September, 2008

Grinder November, 2008

Grizzly G3103 Mill

Grizzly G3103 Mill
Rebuild - Part 2

Moving the Shop

Moving the Shop 2

Bringing Home a Sheldon 12" Shaper

Sheldon 12" Shaper 2

Sheldon 12" Shaper 3

Sheldon 12" Shaper 4

Sheldon 12" Shaper 5

Sheldon 12" Shaper 6

Sheldon 12" Shaper 7

Sheldon 12" Shaper 8

Sheldon 12" Shaper 9

Way Alignment Tool



Email Jim

Moving the mill to its new home
September 27, 2005
 About 4 PM on September 16, the truck finally arrived with the new mill.  I had paid an extra $75 to get it  delivered by a vehicle with a lift gate and pallet truck so that the 950 pound crate could be lowered to the ground and pushed up the driveway into the garage.  The extra $75 wasn't a problem, but trying to get it setup with the trucking company was a bit of a pain and ended up taking quite a few telephone calls.  You'd have thought that Grizzly had gone through this enough times to make that part of the purchase less of a hassle - or maybe I was the only person to have trouble, but I finally got it worked out.

The mill was wheeled on to the lift gate and swayed back and forth a few times as it was lowered to the ground.  I just stood back and watched as there was no way I would be able to do anything  if it started to fall.  The lift gate hit the street with a bang and the trucker seemed as glad as I was that the crate was safely down.
                3103 Milling Machine
Between the trucker, his helper and me to pull and push, it wasn't too much of a chore to get the crate into the garage.  It even cleared the garage door by about 2 inches.

I grabbed my big yellow screwdriver (~3 feet long) and pried open the front panel of the crate so I could open it and check the machine for damage.  I had been warned by one of the guys on the 6x26 Milling Machines Yahoo group to pay special attention to the two crank-wheels that would be on the floor of the crate.  Sure enough, when I opened up the paper wrapped wheels, both had some rust on them.  I noted this on the trucking bill and grabbed an end mill holder and slid it into position in the spindle, then tightened up the draw bar.  This was another check that was recommended to me as there were a few mills that had been shipped with spindles that were not properly machined.  The spindle on mine was fine.

After signing off on the package, it was mine.  I was pretty excited.  This mill had been a long time coming and I had researched this purchased for so long that I had wondered if it would ever see it.

Now, phase two had begun.  I would need to strip the mill down far enough that I could move it from the garage, around the house, over the lawn, through the back yard and into the sliding glass doors on the walk-out basement.  I was hoping that I would just need to pull the head and table.

Remove head
Off to the local Harbor Freight store to get a fold-up two ton engine hoist that was on sale for $150 and a big, pneumatic-tire equipped hand truck.  I broke down and also bought some other stuff while there.  There is more stuff at Harbor Freight that I don't need, but "might just need" than I can take.  A lot of the stuff is of questionable quality, but the hoist was well worth the price.

The first step in stripping down the mill was to open the electrical box and disconnect the wires for the motor.  I drew up a quick sketch of the wiring so I wouldn't have to try and remember where everything went.  I used a nylon tow strap and wrapped a few turns around the turret, head and motor,each time passing through the hook on the end of the engine hoist boom.  A few wraps of some nylon cord to add a little tension to keep the head level, then unscrewed the three acorn nuts that held the turret to the column.  With the engine hoist's front legs up on the pallet, I lifted the unit up until I had a couple of inches of clearance.  I was planning to just roll the hoist back until it reached the edge of the pallet, then drop it a inch or so to a couple of 2X4 blocks, then
on to the floor of the garage.
                is off
Unfortunately,the thin wood on the floor of the pallet couldn't hold the weight of the engine hoist and head assembly and one of the casters broke through the plywood.  Oh crap, I'm in trouble now!

A yell to my wife got her into the garage and within a couple minutes we had also enlisted the help of a neighbor.  Between the three of us, we were able to lift the front legs of the hoist out of its hole and on to the garage floor.  With sweat pouring off of me, I thanked all concerned and looked at the head assembly in amazement.  Dam, this thing is heavier than I thought!  If I had to do it over again, I think that a couple of 2X4s shoved under the pallet with a piece of plywood or two to take up the space, would have prevented the engine hoist caster from breaking through.  Live and learn.

Head close-up
One of the modifications that I may be performing on this mill is to install a "riser block" between the top of the column and the turret.  I have seen them done in both 4"and 5".  I'm pretty sure that I will be making a 5" so I will be able to raise the working height enough to get over my rotary table and work on larger pieces, but I will do some set-ups and see exactly how much more room I will need.   I took a break from the moving to sketch a riser block and transfer the measurements of the turret and the column top to my drawing.  It looks like a pretty straight forward piece to machine using the rotary table.

The next piece to come off was the table.  This comes off by removing the hand wheels on the Y axis, the power feed on the right side of the table, a circlip on the same side and a few spacers and a needle bearing race.  Finally, remove the gib adjusters and gib. Once the table ends are off, it is a matter of sliding the table to
the right until the dovetail of the table clears the saddle, then lifting straight up.
Packing the Z bearings
I don't have to tell you that the cast iron table is heavy, but I will anyway.  It's heavy!  Unless you're built like a power-lifter, keep it close to your body and have a place cleared to set it.  Boy, I'm going to be sore tomorrow!
After removing the saddle, I foolishly tried to slide the hand truck under the base and move it.  No chance!  So we move on to removing the knee.  This is easiest if you remove the three bolts that hold the vertical lead screw to the base and the connection at the knee.  Once I had the screws out, I was very dismayed to find that the pair of bearings at the top of the screw felt like crap. Crunchy!  I was about to mark these on my list to talk to Grizzly about replacing when I decided that it wouldn't hurt to clean them first.  I used some brake spray as solvent.  Within a minute,the bearings were as smooth as glass.  The protective grease had just hardened and was being ground up
between the balls of the bearings.
Stripped to parts
Once I had both the bearings cleaned - without removing them from the screw- I used a bearing "cheater" (a tube that fits on a grease gun that has been flattened to a thin semi-circle) to pump grease between the races and balls.  This could be done by hand,but is much quicker and less messy with the tool.

Once I was down to just the column and the base, I again tried to move it.  No go.  Oh well, only 4 more bolts to go and it is completely stripped down.  So there it is.  Only 24 hours since it was delivered and it was stripped to its parts on the garage floor.  Between using the wheelbarrow for the head assembly and the hand truck to move the base and column, it all ended up in the basement and I began to reassemble it. 

Since the mill had been painted after being assembled, there were a few paint chips where I had to disassemble parts, but I am not too concerned by this.  If I get real
energetic, I may find some paint that matches and do some touch-up. It turns out that beige appliance paint is a very nice match. Frigidaire brand was what I had.
                3103 Milling Machine Base
On the way back up, I took my time and ran taps through all of the threaded holes to make sure that the threads were clean.  I also used some anti-seize on the threads to keep them from seizing if moisture should get in.   Using the engine hoist, I placed the column back on the base and torqued down the bolts.  I made up a little sanding block and cleaned the surfaces of the mating parts and applied a thin coat of grease between the pieces to help protect them from rusting.

For the time being, I have leveled the base with some wooden shims.  I haven't decided if I will fasten the mill to the cement floor or whether I want to build a movable base with casters that will pivot out of the way when not being moved.  I have seen a very nice
design using 3"X2" boxed steel that has casters and leveling pads that  would make it very easy to move to a new location. 
                3103 Milling Machine Knee
Since I doubt that I will be in this house forever, I am leaning toward making it somewhat portable.  With the amount of equipment I am acquiring in the basement, I don't even want to think about a possible move, but I guess that it is better to plan for the future. - I ended up going with Enco adjustable machinery pads which made it very easy to level the mill.

 Next, the knee gets lowered into place.  The knee is not too heavy to lift, but to try and lift it AND align the dovetails to get it to fit is a little too much for one guy.  The money spent on the engine hoist made what could have been a hard job, easy.  The casters on the hoist moved smoothly on the basement floor, which made the alignment of items like the knee, "a piece of cake."

With the knee in place, it was time to turn my attention to the knee's gib.  It is a hand scraped piece made of cast iron.  The finish was pretty far from being flat.  I had read that it would be good to have at least a 70% contact between the gib and the ways so that there was some room for lubrication.  The center of the knee gib was low by a few thousandths and as you can see in the picture, the center of the gib looked the worst after a few minutes of lapping.  I continued lapping with progressively finer grades of wet/dry paper until I had a pretty nice finish.  I cleaned the ways and dovetail, then lubed them up and installed the gib and two adjusting screws. Eventually I will get in and re scrape the ways and gibs, but that's not part of my immediate plans.

This procedure was repeated on the other two axes.  Hand lapping takes some time, especially if you are starting with something like a 300grit paper, but trying to speed up the process on my power sander was discarded as it is too easy to take off too much material.  So it was back to figure "8" patterns on a whole sheet of paper on top of my cast iron table saw table.  That old Craftsman saw table may be ugly, but the table is pretty flat and solid as a rock.  To quote an old saying,"they don't make them like that any more."
Grizzly 3103 Knee adjustment
The attention to detail is an area where the inexpensive Chinese equipment is not as good as their Japanese, Taiwanese, European, and American counterparts, but if you are willing to take the time, the basics are all there and it is just a matter of some TLC to tighten up all of the specs.

Before I started lapping the gib, there was about 0.025" lateral movement in the knee and there were
loose and tight spots over the entire range of movement.  When I was done lapping the gib, I had reduced the play to about 0.007 with a pretty constant smooth feel while cranking the knee up and down. No doubt that will grow a bit as things wear in, but nevertheless, it is quite an improvement.  The difference in smoothness after cleaning and repacking the bearings on the top of the Z (vertical) lead screw was - as one would suspect - nothing short of amazing.
                3103 Saddle
With the knee back together and feeling good, it was on to the X and Y axes.  One thing that I was not real impressed with was the backlash adjustment on both the X and Y axes.  It is accomplished by a slot being machined into the brass lead screw nut with a couple of cap screws to pinch the nut on the lead screw.  (See the image below.)  Before disassembling the mill, I checked the actual measurements against the sheet of readings provided by the factory. Backlash was not one of the provided measurements;however, I checked and noted backlash for all three axes, so I would have a guide to assemble to.  The X axis was the worst of the three, with  0.011".

                the lube tubes
After lapping the gib and cleaning up the dovetails, I mounted the saddle and checked the X axis backlash.  It was easy to bring it to a very smooth 0.009",but as soon as I tightened up the two cap screws to try to remove more of the backlash, the nut seemed to bind on the lead screw.  I went through this a couple times with no improvement until I surmised what was happening.  The slot in the nut was allowing the nut to pinch the threads, but because there was a slot instead of a two piece nut, the pressure on the lead screw was uneven with more pressure being placed on one side of the nut than the other.  I'm going to need to fix this as soon as the mill is together.

 The mill is equipped with a "one-shot" lubricating system.  When I first broke open the crate, there was oil all over the base of the mill.  At the end of the first day's work, I had noticed that two of the clear plastic lines that kept the X axis oiled were just stuck into their holes.  The remainder of the lines had screw-in fittings.  I thought that the lines that were not screwed in were probably the ones that were leaking, so I used a little
high temp,General Motors orange silicone sealant to seal the tubes to the holes. 
adjusting the x axis
The next day I filled the lube tank about 1/4 full before I broke for lunch.  When I returned, the tank was empty and there was oil all over the place.  It appears that the oiler is siphoning the oil from the tank and it leaks out from all of the oiling points.  None got past the two lines that used sealer, but it did come out below the holes where the oil exits to keep the ways lubed.  I will have to look around to find a check valve that will only pass oil when I press the pump handle and pressurize the lines.  For now, I will keep the tank empty and lube the ways with an oil can.

 While I am on the topic of future plans, another project, no, make that purchase, is some new adapters for my dial indicator.  I purchased the dial indicator decades ago,
primarily to measure run-out on brake rotors and for measuring the spring tension on Mazda rotary engines. It came in a heavy duty plastic case, which is probably why after twenty-some years, it is in great  condition.  I can see that it will now have a "second life" in the workshop. 
Mounting the Table
I have already used it more in the past week than I have in the past ten years! I need to find a couple of adapters that will make it easy to use it for aligning the machining vise, tramming the head, and general set-up work.  In the searching I have done so far, I see lots of magnetic bases, but no shafts and right angle adapters.  I'm sure they're out there, it's just a matter of finding them.  It would seem that in getting a mill, there are many, many accessories that I will need.  It looks like I may need a second job to support this new "habit."

Anyway, more measuring and more assembling.  With the table on and the end-play adjusted on the 
table lead screw, I had my first chance to try the Y axis power feed. It works as it should and will be a nice feature.  The power feed main gear rides on a needle bearing and is smooth as silk. Nice.  Finally I was getting close to having the mill up and running.
Mounting the head, turret, and motor
Once more I grabbed the engine hoist an lifted the head, turret, and motor assembly out of the wheelbarrow and set it atop the column.  After some more measuring, I had the head assembly aligned.as close to perfect as I could get it.  I rewired the electrical box and finally plugged in the two power cords and hit the main switch. Success!

After making a couple cuts to make sure that the alignment was true, I rechecked the backlash on the X axis.  It was horrible.

I fired up a computer and started looking at different styles of anti-backlash nuts on the web.  I found that the majority were made from two pieces that either screwed together with a shim or spring between them.  This would allow the pressure of the nut on the lead screw to be constant for the full rotation of the thread as opposed to the current pressure on one side and no pressure on the other.
The new
                Y axis anti-backlash nut
I reasoned that if I cut the slotted section off, drilled and threaded a couple more holes for an additional two cap screws and made a shim of the proper thickness (I.E., to put the right amount of "pinch" on the threads),it would work better.  To do this, I would either need another nut, or remove the existing nut.  I could remove the existing nut and still use the mill if I locked the X axis and worked only with the Y axis.  I would extend the slot until I cut the nut into two pieces.

This was the first "real" cut I made with the mill and even though I only had the y and z axes to work with, it was sweet!  I'm going to love working with this tool!  The end mill cut the material like butter and soon I had a two piece nut with a copper shim to adjust the pressure on the lead screw.  I decided that instead of clamping the threads together, I would shim it so that the force would be toward the outside - an expanding, rather than pinching force.  I finished the new nut, reinstalled it  on the lead screw and rechecked the backlash. 0.002" and no binding!  After a couple of days, this has grown to about 0.004" and seems to have stopped there for the time
Mounting the saddle
Of course, the backlash will grow as the threads on the nut wear, but the adjustment is pretty easy and can be performed by removing the two cap screws that hold the hand wheel and lead screw assembly onto the saddle and adding a new shim.  For adjusting in small amounts, a piece of aluminum foil makes a nice shim.

I reassembled the table to the saddle and measured the backlash on the Y axis.  With the clean-up, lapping of the gib and touching up a couple of burrs on the long table lead screw, it came in at just under 0.005".  I decided to hold off on modifying the Y axis nut until I needed to adjust for wear.

After some searching on the web, I have found a couple designs for adjustable lead screw nuts that will allow me to adjust without using shims.  I think that this will be a project for the future unless I decide to convert to CNC. 
The mill is up and running
I have seen some shots of a Grizzly G3103 that has been converted to CNC with ball screws and it looks nice.  Whether I go that route or not remains to be seen, but it is fun to ponder.  Another possibility - and a much cheaper project - is to mount up a digital readout (DRO).  There is a pretty neat do-it-yourself setup using Chinese scales and a controller box by ShumaTech. There is also a Yahoo group devoted to assembling the project.  I have read about half the posts in this group and am getting pretty close to ordering the circuit board and scales.  However, before I start a new project, I really need to clean up the workshop.  I picked up some shelving, which will give me a place to put all of the heavy stuff for the mill and I still need to get some pegboard for the items that can be hung up.  In the near future, I'll not only have a working mill, but a clean shop as well. 

What a treat that will be!