It has been a minute since we’ve updated our Kearney & Trecker 2HL Mill overhaul. Life has intervened with a job change and relocation, and with the mill basically finished it now sits in storage awaiting its new home. Not to worry, though. Knowing this transition was in the cards, I saved a (to me) cool little mini-project to keep the machine overhaul gods appeased during the transition.
Continue reading
We are nearly done with our Kearney & Trecker 2HL overhaul. Having scraped the top of the knee flat and applied Turcite to the bottom of the saddle, the mating surfaces between the mill table and the top of the saddle now have to be scraped flat and the dovetail ways brought into alignment. As recounted below, this ended up being a chore.
Things started out easily enough. Using a camel-back straight edge and alternating with the glass surface plate, I was able to fairly quickly get the top of the knee flat and the flat ways co-planar to one another.
Continue reading
While we have the Kearney and Trecker feed gearbox out and the saddle off the mill, it is time to address the sorry state of the lubrication system. While we are at it, we’ll rectify two nasty surprises we’ve found inside and under the saddle.
Reapplying Turcite to Saddle Bottom
Recall, the y-axis woes of this machine can be traced back to Turcite (or some generic substitute) that had lost adhesion and allowed swarf to accumulate under the saddle ways. Upon removal of the old material (most of which I could do by hand) it looked as if there was not much work done to ensure that the bottom saddle ways were flat and co-planar before applying the original Turcite.
Using the bottom of the saddle as a reference surface, I was able to verify that there was indeed .004″ or so deviation across the ways. Rather than spend an eternity hand scraping them flat — and keeping in mind that this will be covered by new slideway material — I selectively deployed an extended pneumatic cutoff tool as a grinder, taking light horizontal cuts.
Continue reading
It has been a minute since I’ve posted on the K&T 2HL overhaul, but rest assured, things have been very busy in the background. Having fixed the stripped feed gearing, work has focused on correcting the poor Y-axis travel behavior we noticed on the initial tear down. Recall, we were seeing the table rise around .005″ over just 8″ travel.
Fixing this ended up involving quite a bit of somewhat tedious hand scraping — in particular, scraping the top of the knee flat and perpendicular to the Z-axis quill travel and making sure the dovetail ways were parallel and aligned perpendicularly to the X-axis of travel . There are TONS of much better scraping hands out there and, for that matter, there are tons of better hand scraping tutorials. An expert I ain’t, but I’ve usually been able to get decent enough results in the past (and did so here) so in case it may be useful to someone, I’ll walk through my basic steps.
Continue reading
From the recent teardown of the feed distribution gearbox on our K&T 2HL, we were able to trace our high speed feed problems to a stripped “tertiary high speed cluster gear” that has clearly suffered past catastrophic damage. Because this is a two-part cluster gear that rides along a splined shaft, the chances of finding a direct replacement that replicates the exact tooth count on both gears is about zero.
Fortunately, only one of the two gears that make up the cluster was damaged, and even more fortunately it only measured 1.25″ in diameter — less than the width of many off-the-shelf splined shaft sleeves that can be bought from places like McMaster. So here’s the repair strategy: we’ll take a stock splined shaft sleeve, turn one end to 1.25″, and cut a relief channel on a lathe. We’ll then cut the teeth into the sleeve, cut the old stripped gear from its undamaged partner, and then TIG weld the new gear to the old using the splined shaft to keep everything aligned. Because the splined sleeve is unhardened mild steel, we’ll have to case harden it to complete the repair.
Continue reading
From our initial diagnosis we know that our Kearney & Trecker (Milwaukee) 2HL Vertical mill has issues with power feed at any speed over 7 or so inches per minute. To figure out what is going on, we’ll need to remove the front of the knee and disassemble the feed distribution gearbox. I won’t go into too much detail since there is an excellent video already out there, but there are a couple of key points I can reinforce here.
Most of the preparatory work to remove the box requires work on the left side of the mill, where there is a small right angle feed change worm gearbox (#9 in the parts diagram below the fold). It is held in place with three slotted screws. Before removing it, best practice is to go ahead and set the feed to the lowest setting to simplify the feed timing when the feed distribution box is reinstalled (see discussion below).
Next, we’ll need to remove the trip dogs (#52) that ride on the vertical and cross feed rods (#55 & #53). These are each held on by a cap screw and slide off to the rear. In my case, some previous jackassery had resulted in one of the cap screws being positioned such that it rubbed on the underside of the saddle. I’m not sure if this was what was throwing off the Y-axis precision as the table was cranked out, but I’ll be keeping my eye on this possibility.
Continue reading
Over the last year or two, I replaced the abrasive wheels on my 8″ no-name bench grinder with a brass wire wheel and a nylon fiber deburring wheel, both of which I found I used much more on a day-to-day basis. Still, I do need to grind down welds, shape the occasional HSS lathe tool, and sharpen larger drill bits from time to time, and was getting sick of swapping out bench grinder wheels when the occasional grinding job presented itself.
The original plan was to buy a second, quality made bench grinder … maybe even a 10″ model … and devote it to actual grinding. However, the good bench grinders seem to be caught up in the post-COVID inflation spiral (or maybe they were always this expensive, I dunno), but for whatever reason the cost of a decent new grinder didn’t make a lot of sense for my situation.
As a “Plan B”, I began monitoring local equipment auctions, as one does, and as luck would have it, found a ratty looking Queen City 12″ grinder at a nearby auction in Dallas, Georgia. I ended up getting it for the princely sum of $120 and got it home without incident … where it sat for most of the next year. With my CNC project finally done, and before launching into the Kearney & Trecker 2HL mill rebuild, I thought I’d clean up the Queen City grinder and get it operational.
Continue reading
Having replaced the spindle bearings and fixed the issues with the Reeves drive intermediate shaft of the Clausing 20″ drill press, it is now time to reassemble the machine and put it into service. While they were apart and everything was disassembled, I degreased all of the castings and gave them a coat of KBS Coatings “top coat” in battleship grey (which really turned out to be more of a grey-blue). Even though the color wasn’t what I expected, it has grown on me.
Cleaning up the 4″ center column was a bit more of a chore. Thirty+ years of grime and surface rust left it severely gummed up. After struggling with a brush and various cleaners for a while, I decided to just chuck the column up in my LeBlond, supporting the far end using a steady rest. I then set it to a few hundred RPMs and attacked the grime with green scotch brite pads and mineral oil. It worked a treat. In the future, I wouldn’t tackle this job any other way.
Continue reading
Having safely transported and disassembled our second-hand Clausing 20″ drill press, we’ll now replace the spindle bearings and tackle the borked up Reeves drive intermediate shaft. The spindle bearing layout consists of an upper ball bearing (I believe a MRC 205SFF) and a lower double row angular contact bearing (MRC 5205SBF) that are pressed into the quill. The spindle shaft is pressed into the inside of the two bearings. It is splined and can move vertically in an upper spindle pulley assembly as the quill is rotated up or down. In my case, the upper bearings in the pulley assembly spun fine and don’t seem to have much impact on runout so I left them alone.
I was however getting several thousands of an inch worth of runout at the spindle nose and heard some suspicious grinding, so decided to go ahead and replace the two lower spindle bearings held in the quill.
Continue reading
I am very pleased with my Kearney & Trecker 2HL vertical mill. It is a marvel of mechanical engineering –robustly built and full featured. (In other words, more mill than I deserve).
As great as it is, one area where the Kearney & Trecker falls down compared to Bridgeport-style knee mills is the absence of a quill that can easily be extended from the head for drilling purposes. Instead of the familiar lever, the K&T sports a front-facing hand wheel that engages the rack and pinion assembly inside the head to lower the spindle. This setup is exceedingly rigid and works fine if you take the time to set it up (save for the occasional broken clamp collar). However, when casually knocking out a few holes, it is really easy to break bits or to drill too far because you can’t feel what is going on with the bit pressure.
To recover this casual drilling capability while letting the K&T do what it does best, I’ve decided to seriously upgrade my drill press game. I have a small CNC mill that can be used for precision drilling purposes, and can always use the K&T for larger work when I need to take my time. What I lack, though, is a more capable drill press with a large table that can stand in for the mill for everyday drilling where power, feel and ease of setup are the operative goals.
Continue reading
We have more or less completed our Benchmill CNC upgrade project, installing Sieg 2.7 castings in place of the original mill, reinforcing the head, upgrading the power distribution, and installing a 2 HP spindle motor. We have the auto-oiling and pneumatic power drawbar installed, and everything is up and running. As icing on the cake, we are going to take advantage of the Intelitek automatic tool changer functionality built into its CNC software to support a scratch-built sled-style automatic tool changer system.
In designing an ATC system, there are a few options to consider when getting the tool to the spindle nose. The most common approach is to use a rotary carousel where an indexed platter holds the tools. This platter rotates to predefined positions to call up the requested tool, which is then clamped into the spindle using various design-dependent mechanisms. Digging through the config files for the Intelitek software, I saw that it does in fact include settings for a rotary tool changer, but that functionality is associated with a different model (Promill 8000). Rather than run the risk of building something only to find missing wiring connections into the controller that the software assumes are present, I elected to go with a more simple sled-style design.
Continue reading
We at the finish line for our CNC upgrade project (no, really). We have installed the upgraded Sieg 2.7 mill castings in the Benchmill cabinet and have gotten our Benchmill controller talking to the PC running the control software. Now we’ll begin connecting all of the odds and ends to give us a working CNC mill with all the bells and whistles (which, if you’ll recall, was the point of starting with a trainer mill as a base for the project).
There is a lot going on here, none of it particularly challenging but taken together representing a lot work to make the final project polished and usable. Instead of an extended how-to, this post will be more of a tour showing all of the connections as made, since the connections themself are fairly simple.
Continue reading
Having enclosed the upgraded Sieg 2.7 mill using the old Benchmill Enclosure, we’ll now establish power and get the controller and control interface running. We know that everything works independently, so it *should* just be a matter of connecting components.
One initial challenge lies in the fact that we’ve seriously upgraded the stepper motors on all three axes from the little 1.35 Nm motors that came with the original Sieg iXK1 to 4Nm on the Y and Y axes, and 9 on the Z. These draw considerably more power (I have the drivers set to 5-7 amps each). In addition, the new spindle motor draws up to 8 amps at 220v. The original Benchmill transformer tops out (if I recall correctly) at only around 10 amps, so we have some power delivery work to do.
Continue reading
We are nearing the end of our Benchmill 6000 CNC project. The Sieg 2.7 replacement mill has been completed. We’ve reinforced the head, shoehorned in a 2hp servo motor, hand scraped the ways, and installed stepper motors and ball screws on all three axes. Now we’ll begin the process of enclosing the CNC mill using the old Benchmill enclosure.
With the reinforced head, epoxy resin fill, and reinforced column and back plate, I’d estimate we’ve added about 80 pounds to the mill (a good thing). Starting from a 245 pound base weight, that puts us around 325 pounds for the machine alone (excluding the mounting plate in the base and the enclosure). Whereas I was able to manhandle the little iKX1 in the original Benchmill, this took an engine hoist to get it in place.
Continue reading
We’ll be taking a detour today from our CNC benchtop mill project for a quick “one afternoon” build for the Kearney & Trecker 2HL vertical mill. The Milwaukee was missing its y-axis hand wheel when I purchased it, but I figured at the time that I could pick one up cheap off of Ebay. After a couple of months of searching, I discovered that this was not the case and decided to machine one myself from a blank.
The design of the K&T hand wheel and crank is relatively simple. A boss extends (on my machine) 3/4 of an inch from the back of the wheel and passes through the throw levers for the table feed. It engages what I suppose is the y-axis screw by way of four teeth cut in an “X” configuration.
The shaft of the screw passes through the wheel, which allows it to slide on and off to engage the teeth when desired. The boss is just the right size to keep the feed lever from engaging while the hand wheel is engaged and vise-versa: while the lever is engaged, the hand wheel teeth can’t engage the screw.
Continue reading
With the new Sieg 2.7 parts for our Benchmill 6000 CNC upgrade assembled, it is time to build something for the assembly to sit on. We could, well, just put the Benchmill enclosure on a bench, but we’ve increased the weight of the machine to the point that may become problematic. We’ve also increased the Sieg’s power over stock, and it makes sense to match that power with something a bit more rigid than a bench top. Besides, I want something that I can roll around the shop.
That said, I don’t want to spend an arm and a leg on what is essentially a fancy stand. To keep cost down, we’ll use scrap already laying around to build a 2″ square tube box with melamine panels for the walls and cabinet doors. The bottom half of the stand will be a basic cabinet, with one side dedicated to a coolant tank and the other to store a future 4th axis. Up top, we’ll have a slide out tray for the keyboard and two drawers for tooling. The whole thing will sit on adjustable, heavy duty (500lb each) casters.
None of this is particularly rocket science, so feel free to give this a skip if you’ve got cabinet making covered. That said, I have a few of these under my belt and feel that I have the process down to the point that I can construct something relatively sturdy, professional looking, fast, and cheap. If you are looking for design inspiration, read on.
Continue reading
Having reinforced the head and upgraded the spindle motor of the new Sieg 2.7 guts going into our Benchmill 6000 CNC training mill, we’ll now complete the head assembly with an air cylinder-driven power drawbar. This will support the eventual automatic tool changer feature we are planning and, in the meantime, simplify life in the tight Benchmill enclosure.
The power drawbar design we’ll be using sounds complicated but is actually simple in operation: tool changes are made by activating an air cylinder that will press the mill’s drawbar down against a series of Belleville spring washers. When uncompressed, the washers put tension on the drawbar to pull a collet in the spindle nose tight against the inserted tool. When the washers are compressed, tension on the collet is relaxed and the tool can be removed. Here, we’ll be using the Tormach TTS system for the collet and tool holders but the principle would work with any spring-type collet.
Continue reading
After upgrading the spindle and reinforcing the head of the Sieg 2.7 castings we are installing into our Benchmill 6000 CNC project, we can at long last turn our attention to the spindle motor. The original Benchmill sports a 1kW motor, but with servo prices dropping we’ll use this as an opportunity to upgrade our cutting power with a more capable motor to match our upgraded, more rigid platform.
The motor we’ll be using is a Hangzhou Bergerda Nema 42 model (number 110F-0625TDL). It is a 1.5kW AC servo with a rated speed of 2,500 rpm and should be capable of delivering constant torque across that speed range. The Benchmill software is set up for a 5,000 max speed, so we’ll gear the motor at a 2:1 ratio using 40 and 20 tooth HTD 8M timing pulleys that take a beefy 25mm belt.
Continue reading
Having successfully installed the CNC stepper motor mounts and ball screws for the Z-axis and Y-axis, it is time to finish with the X-axis. We’ll follow a similar design to the one we chose for the Y-axis, with an angular contact bearing held in a mounting plate attached to the left end of the mill table and a bracket tying the ball nut to the saddle.
The stepper motor will be mounted in reverse configuration using a steel mounting plate, tucked under the front of the table, and will turn the ball screw using a belt to achieve the 4:5 gearing ratio we need to use the Benchmill software with the new 1605 ball screws (the original machine uses 1204 screws). Unlike the Y-axis, where we used a bronze bushing to hold the floating end, we’ll use a regular sealed ball bearing held in a mounting plate on the right side of the table.
The main (left side) mounting plate will be milled from aluminum, the motor mount and the ball screw mount will be milled from 3/8″ mild steel, and the floating end (right) bearing mount will be 3D printed using carbon fiber filament, which will save us time and will be plenty strong for the floating end of the ball screw.
Continue reading
Having reinforced the Sieg 2.7 bench top mill column, reinforced the head, and dry fitted the Z-axis and Y-axis CNC components, it is time to finally pull base, column, and head together. Before doing this, it is time for a final modification to further improve rigidity and cut down on vibration.
By moving the Z-axis ball screw to the outside of the column, we’ve made it possible to fill the entire column with epoxy granite to add weight and stability. Recall, we inserted a piece of 1/2″ threaded rod where the Z-axis lead screw used to go, and secured it so it will extend into the base to provide an additional point to secure the column to the base. We’ve also used extra long bolts when fixing components to the column, to provide more real estate for the epoxy to grip.
Continue reading
Previously in our Benchmill CNC upgrade project we reinforced the Sieg 2.7 head casting and hand scraped the base. Now we’ll take our first steps toward adding CNC capability to the mill, starting with the Y-axis.
For ball screws we’ll be using a 360mm “C5” 1605 double nut model. I put the C5 accuracy designation in scare quotes because the ball screw I ordered was rolled, not ground, and most real C5 accuracy screws seem to be ground. Color me a bit skeptical. Still, the screw and nut that arrived were clearly of higher quality than the $30 Ebay specials one tends to see, and the nut/screw combination was significantly cheaper than a true ground C5 screw (at about $120).
Continue reading
In earlier posts on our CNC mini mill project we reinforced the mill head and column. Now that we have the rigidity we need, we’ll focus on an upgraded spindle. We are already off to a good start: as a cost saving measure, when I first purchased the Sieg 2.7 castings from Little Machine Shop I purchased a R8 spindle cartridge, which includes items 2, 3 and 30-37 in the parts diagram above. I knew this might only be a partial solution, however, since the description indicated that the cartridge was rated for only up to 2,000 RPM.
For CNC work, I want to be able to achieve at least 5,000 RPM. Initial research showed I could probably get there with this cartridge: replacement bearings for this spindle (used on the Sieg SX2.7, as well as the X3 and SX3) can be had that are capable of 6,300 RPM and above on grease. I also saw a number of comments from owners of X3 and SX3 machines who were getting 4,000-4,500 RPM out of this spindle, stock.
Continue reading
In the last installment of our Sieg Chinese mini mill CNC conversion we installed linear rails and constructed a sliding saddle mechanism tied to the head assembly to increase rigidity. Now we will complete the Z-axis assembly by installing a heavier duty 2005 ball screw and Nema 34 Stepper Motor.
I intentionally decided to mount the ball screw nut on the top side of the linear rail saddle to keep the attachment point as close to the head as possible. To do this, we need to raise the bearing housings through use of two riser blocks.
I designed these in Fusion 360 so that the stock bearing blocks would bolt directly to the risers. This was made more complicated than it needed to be by the spacing of the bolts holding the back plate to the column and by the bolts mounting the rails to the plate (dictated by the wide spots in the castings and pre-drilled rail mounting holes). While designing the risers, I added through holes so that the blocks could be tied together with a threaded rod and spacer assembly.
Continue reading
We’ve identified a lack of rigidity in the head and column of our import Sieg mill castings as the primary areas of concern for our CNC mill project. In a previous post we added a half-inch plate along the back of the column and made sure it was flat and parallel to the front ways. Now we’ll return and add a mechanism to stiffen the head along the column and increase overall rigidity.
The design is simple: we’ll install a pair of linear rails on the new half-inch steel column back plate. To that we’ll attach a mounting plate that rides on the rail trucks and to that, in turn, we’ll mount the ball nut holder for the eventual rear ball screw. To the sides of the mounting plate, we’ll attach two thinner (3/8″) plates that pass along the sides of the column and tie the assembly to the head casting. This will allow the ball screw to elevate the head, while adding to the overall rigidity of the setup.
Continue readingHaving temporarily repaired the quill clamp collar and gotten the Milwaukee milling machine running on a VFD, it is time to fire it up and see what we have on our hands. But first, as with any old machine, a good cleaning and change of fluids is in order.
A reprint of the original K&T lubrication guide is available at the excellent vintagemachinery.org site and walks through all the lubrication points. The spindle in these “swivel-head” mills is greased, which we handled when we had it apart to fix the clamp collar. Ditto the bearings in the motor, which we similarly cleaned and repacked when we had the motor out.
That leaves the oil in the oil reservoirs in the column, knee, and saddle (there are three).
Continue reading
Having initially scraped the saddle flat, the next step in converting our Sieg 2.7 mill castings into something actually accurate for the Benchmill 6000 CNC upgrade is to get the base casting flat and parallel to the bottom, and to make sure the base dovetail ways are flat and parallel to each other.
Throwing the casting on a surface plate showed that one corner was 7 thou (!) higher than the others. This is too much to attack comfortably using a hand scraper so I started by gently flattening the bottom of the casting with a die grinder turned at a 15 degree or so angle. Between rounds, I inked the bottom using Prussian Blue, just as one does with hand scraping, and checked on the surface plate between grinding rounds to make sure I headed in the right direction. When it was within 2 thou or so of being parallel to the top, I finished with a hand scraper.
Continue reading
Now that we’ve got the Milwaukee milling machine running on shop 220V, we need to make it operational so we can use it to complete our CNC project.
One of the first things I noticed when I got the mill home was the jammed quill. Attempting to turn the hand wheel resulted in no movement. A bit of research led me to this great thread at Practical Machinist. Turns out that overzealous tightening of the quill clamp lever (#49 in the exploded parts diagram below) can result in fracturing of the cast iron clamp collar (#27). I immediately suspected this was my problem.
Continue reading
So the first task after getting the Milwaukee 2HL Vertical mill home is to get it running on 220 volts, single phase. The motor on the 2HL model is a 3HP “U.S. Motor” brand 3-phase model, rated for 1800 RPM. My understanding is that the 2H models came with a 5HP motor, but three should be fine for my purposes, assuming I can run it on the 220 volts available to me.
Continue reading
Time for a new knee mill. My old Wells Index 745 has provided great service, but I’m looking for a new challenge while I finish the CNC project. I’ve been fascinated for a while with the various flavors of American “Heavy Iron” such as Cincinnati, VN, and Kearney & Trecker. Now … drum roll … I have one to play with.
Continue reading
The biggest drawback with using Chinese mill castings for this project is that they come from the factory with … less than perfect … accuracy. Iron is iron, though, or is so within my budget range, and as as long as the casting is reasonably solid, inaccuracies can be corrected by hand scraping.
I’m not making nuclear submarines here, so I don’t need perfect precision, but if I can get the base, saddle, table, and column ways flat, parallel/perpendicular, and co-planar to within 5 tenths over a foot, I’ll be happy. And just as importantly, I feel that the manufacturers of the original machine would have been ecstatic with those tolerances.