[Fastest1]

Actually nothing is critical on the tramming tool as long as it isn't loose. Due to the rotating of the tool to tram, you just divide the difference in the dials. Check and repeat til perfect. Usually takes 2 rotations for me, double checking. Also look up "machinechiks" tramming aid. Wonderful additions that assist the adjustment and help keep it in place.

[mduckett]

Actually nothing is critical on the tramming tool as long as it isn't loose. Due to the rotating of the tool to tram, you just divide the difference in the dials. Check and repeat til perfect. Usually takes 2 rotations for me, double checking. Also look up "machinechiks" tramming aid. Wonderful additions that assist the adjustment and help keep it in place.

I saw Hoss demonstrate a way to use a small step block on the table to calibrate the dials. Each indicator is stepped onto the block at the same place on the table and adjusted to zero. In theory they are then calibrated to the same height, so lower until they both engage the table and adjust until they read the same. It's not bad with a single digital indicator, zero on one side, rotate to the other, split the difference and adjust, repeat. Works good except on my machine I'm in back making the adjustments and it's hard to move the indicator without bumping it out. I'm looking forward to using this tool next time.

Yep, that adjustment screw is a simple, great idea. I saw where some guy had put some L brackets at the base of his X2 column too, with an adjustment bolts in each going down to the base to do something similar for left/right tilt adjustment. A lot of great ideas captured on this forum. Thanks for the comments. --md

[Fastest1]

As long as you zero the indicators and rotate them 180deg it will tell you how much to correct. 1/2 the difference.

[djbird3]

Hi MD,
I see that you used a few of my designs on your build, cool. I've been busy with other stuff
so haven't been frequenting the CNCzone but I just wanted to compliment you on your work
on the SX2, very nice job. Once you get the bugs straightened out you should do my power
drawbar design, it's about the most compact design I've seen for the SX2 and it's easy to
build. But anyway keep up the good work and if you need any help with anything let me know.

~Don~

[mduckett]

Hey Don,

Thanks for the nice words and encouragement. Your SX2 build thread was a top motivator for me on this build. I am absolutely planning to use your version of the air cylinder/power drawbar design with some kind of tool changer arrangment. I downloaded your plans and took the time to put the design into my CAD package, so now it's just a matter of working it into the schedule with the other mods and projects that are pending. --md

[mduckett]

Got some time on Saturday to get the tramming tool project underway. I had already cut a 3/4" x 2" x 7" blank from some scrap aluminum bar, so I got started by mounting it in the vise and running the hole drilling G-code setup. There are 4 holes that will be tapped 1/4"-20 to hold the dial indicators, and I added one more for the shaft hole, thinking it would be a good idea to remove some of the material with the drill while I'm at it. I ran the G-code without incident, and the extra hole for the shaft gave me an extra benefit later on as it turned out. After drilling the #7 holes, I removed the part and made a jig from another length of scrap 3/4" x 2" stock to hold the part while milling its outline. Marked/drilled/tapped 4 holes for #10-32 screws into the second piece and attach the part. The jig then went into the vise on parallels. I used 4 bolts out of caution, might have been ok with 2. You can see the jig in the video with the bolts holding down the part, a pretty common technique but my first attempt and it worked well. The benefit of the extra drilled hole came at this point, as I was getting ready to re-install the part and re-register the left corner. The last operation on the machine had been drilling the last hole, so the spindle and chuck were still at that position. I put the #7 drill back in, loosely positioned the part, and basically moved the part to center on the drill. It only took a few seconds and is certainly good enough alignment for this part. Put in the end mill and started milling. At .040" deep cuts, this operation was loud and amplified by my open cabinet, but it was cutting well and increasing the feed from 10 IPM to 12 helped smooth it out. Flood was working great, I increased the flow from the last job and it was really moving the chips. Anyway, at the end of the .480" depth cut, the motors just stopped moving, while Mach3 continued to show the tool moving along. It took a few seconds to discover that my flood system was flooding more than just the work area! Unknown to me, both of my drain seals (100% GE silicon shower caulking) had failed, apparently loosened by the Syn-Kool. In fact, all of the silicon, RTV, and Plumber GOOP on my machine in contact with the SynKool has become soft and loose. Anyway, the dripping into my cabinet reached the electronics and appears to have shorted something in the power supply, as it is only outputting 15V instead of 36V. I had planned to put an extra layer of protection in the cabinet for this very reason but got overconfident by the last run which was completely dry. Never thought the silicon would give way like that. Anyway, I got some marine epoxy and will be making some improvements on the drains. Also found a couple more leaks in the corners but easy fixes. Ordered a new PS and should have that in a couple days. Big lesson learned here about not following through preparing for the worst case scenario. One layer of plastic would have prevented this. Probably should mount the electronics outside, but I'm really strapped for room, so we'll see--it will be made safe one way or another. As best as I cant tell the rest of the electronics are ok, but won't know for sure until the new PS arrives. BOB and stepper drives show normal leds, laptop and adapter are fine too. I think the part can be salvaged; its still in the vise waiting. I'll post the completed tramming tool when I get back up running again.
Here's the link to the video:
Tramming Tool Cutout - YouTube

[mduckett]

As hoped, the new power supply got the mill running again. I bought some marine epoxy and JB Weld, and used both products to fix the coolant drains, figuring to use the one that works best on some other problem areas. Turns out that both are resistant to the Syn-Kool, and both worked well sealing the drains. Still having some problems in the corners, despite adding some plastic angle and some rubber stripping on the corners behind the doors. There was a path behind the doors that I thought was the source of the leaking, but now it looks like the rubber O-rings on the bottoms of the door pivots are not sufficient to keep coolant out of the pivot holes. I will need to make some aluminum inserts for the pivots to sit in, and epoxy those in to solve the problem. The good news is that the cabinet is dry again, although I'm running with the electronics outside the cabinet for a while…...
So, with the mill operational I got back to the tramming tool project. The tramming tool part was still in the vise where it was when the power supply crapped out. Since I had a hole drilled in the shaft location, I just used a drill bit in a chuck to locate the hole and re-register the part. As expected, I didn't get it perfect, so the part had a slight offset on the final outline cuts, but I figured I would need to file these edges smooth anyway so no big deal. The outline milling completed without incident, including cutting the shaft pocket. I found that increasing the feed to 16 IPM (feeding .040 depth/pass, 2500 RPM, 3/8” TiN endmill) made for much quieter and smoother running.
With the outline completed, the part was removed and flipped over on the jig to mark the holes again, this time to drill clearance holes in the jig for 1/4-20 bolts to hold the top face of the part down on the jig so the bottom could be milled and parted from the remaining stock. The holes were tapped and the part bolted to the jig and put back into the mill. I used the NewFangled wizard in Mach3 to make the facing g-code. I used a 1 1/2”", 3 flute carbide insert end mill for the facing. I set up for .020"” cuts, running 8 IPM. This cut smoothly, at least until it got close to cutting through the thinned areas, then it got loud from the vibration. I stopped the cutting and just broke off the pieces with some pliers, then manually controlled the finish cuts. As expected, the shaft pocket was a little out of round, due to the Y-Axis issues described in previous posts, but good enough for this. Next was to make the shaft, so cut a 5” piece of 1/2"” CRS and turned the end for a press fit in the shaft pocket. Had to rotate the backs on the dial indicators to horizontal, and attached them to the beam with 1/4"-20 x 1”" socket screws.
I did a lot of filing and some light emery sanding to get the finish on the part to look decent. Hope as I get better with milling operations that I won't need to do so much manual work on parts. The only really acceptable feature on this part was the finish on the faced surface, which was very good. The attached pictures show the final product. Tried it out tramming the drill press table with excellent results, and will use it on the mill when it’s down for maintenance soon. Attached some pictures of the part and final assembly, and also the flood coolant drains and the mill right after the facing operation, which really threw out a spray of swarf ringlets everywhere, but it was contained INSIDE the enclosure .

[NewHobby]

Hi there, just curious, every 0704 or larger mill turns into a full rebuild project. Looks like you got right to cnc. You Did not need to scrape or lap the ways and/or gibs?

Well done as well. Congrats on progress!

[mduckett]

Hello, not sure I fully understand the initial comment--it felt like a full rebuild to me! , but to answer the question re: scraping/lapping--The base and saddle were purchased replacement parts that did require some targeted work to make them slide smoothly. The replacement spindle carriage required a small amount of work too. I replaced all of the gibs with new ones puchased from LMS, prepped with the usual fine emery and oil on a flat surface. Scraping is a skill I would like to learn, but I didn't want to invest the time to do that at this point. Appreciate your taking a look, and thanks for the nice comments. --md

[NewHobby]

Ahh yes it was indeed a full build and a darn nice one! Was not disparaging it at all it is a model build for this machine. I was not clear, I mean I fairly often see guys with pages and of scrapping and lapping going on. Did you go through that and just not post it or did you require minimal work in that area with lapping? Was nto sure if you did more and did not post of if that type of work was minimal in your build is all.

I ask as I just bought the mill and a ton of tooling and I am dreading scraping as I can see how one can get solid seating on the mill ways but also have them nonparallel to each other and at different heights from each other. Almost seems like scraping can cause more problems than it solved.

Really great build.

[mduckett]

I did spend a fair amount of time on the new saddle/base, probably around 6-8 hours because it was binding/tight when slid to the front position. I didn't go into a lot of detail on this part of the build because it seems like many of the threads that get into working out dovetail/slide issues devolve into endless discussion and controversy re: lapping/scraping. For me, I understood the problem was to find the tight spots, and applying very targeted removal of material from the tight spots. My method involved using the mating surfaces to help do the work. My results might have been better had I used scraping methodology, but I currently don't have the skills or tools to do it. There is an excellent series in the magazine Home Shop Machinist from somewhere in the 2011/2012 timeframe that provides a terrific overview of scraping, and I would highly recommend studying how scraping is done and how one goes about identifying the high and low spots on the surfaces as it helps to understand the underlying problem. Once you identify the problem, then it becomes a task of removing material from the problem areas without damaging the surrounding. Although frowned upon by some, I used fine emery paper strips attached to the mating surfaces of the dovetails to target the areas (used plenty of synthetic motor oil) I found to be binding. Initially I used a Sharpie to coat the surfaces and slid them back and forth (small travels) to locate the problem areas. Later I got some of the blue dye (Dykem HiSpot) and tried it with some success. Anyway, I went real slow and spent a lot of time studying and inspecting the parts, and finally got the parts sliding smoothly for the entire Y travel with a given gib adjustment. Before, the gib could not be adjusted to satisfy the entire travel without binding at the forward position. All this was done out of necessity; I would not have done anything if the parts had a decent fit to start. --md

[mduckett]

After finishing up the tramming tool, I needed to fix some coolant leaks coming from the door pivots. The pivots go into holes in the base. The pivots were resting on O-rings, but this didn't seal very well. Turned a couple of 1/2" diameter x .7" cylinders from 3/4" aluminum bar, left a flanget at the top about 5/8” diameter with a thickness of about .05”". The cylinders are about .7”" long, and bored .375”" diameter to about .5”" deep. The corners of the table were drilled out with a 1/2" 2-flute end mill to about .75”" deep. Mixed up some of the marine epoxy and coated everything and pushed the inserts into the holes. Used the remaining epoxy to seal all around the insert. The first two pictures show the results for the right side door. Another thing I did was to install some of the clear plastic angle in each corner between the Plexiglas panels and the corner angle uprights. This prevents coolant from getting behind the Plexiglas panels in the corners, probably not necessary now that I sealed all of the corners with the marine epoxy, but it was easy to do. Should give it a test soon when finishing the X switch block.

[mduckett]

I’'ve been thinking about Mach3 spindle speed control and after looking at what’'s commercially available and what others have done here in the forum, decided to give a try at making my own. In djbird’'s build thread someone pointed out a website for inexpensive PCB fabrication called ITEADSTUDIO. They will fabricate quantity 10, 50mm x 50mm boards for about $14 delivered via registered mail. After finishing the schematics and layout, I ordered the boards before Christmas and they showed up about 15 days later, not bad given the busy season for mail. Anyway, the picture shows the boards front/back sides. They appear to be pretty good quality, and they claim to test them 100%. Plan to buy parts and assemble a unit in the next week or two. I’ll give more details on the capabilities and features in future posts, but as designed the board accepts 5V power, PWM speed control, Direction, and Enable inputs, and an optional spindle index pulse from the controller side. It implements RF isolators, similar to opto-isolators, to keep the control side isolated from the mill’s electronics, and uses the mill’'s 5V power for the mill side electronics. Later, if this all proves out, I plan to make a version of the board that is specific to the SX2 connectors so that it can plug into the existing cabling instead of using wire terminals, with one additional connector for the control signals and associated power.

The Snap-On stool was given to me for Christmas, it's a welcome addition to the setup!

Edit: Sorry for the blurry pics, used a different camera and didn't notice that they were so blurry.

[gcofieldd]

MD, nice work on the pcb boards! From all the design you have been doing I would have never guessed you were an electrical guy Looking forward to seeing how they turn out. Great build log by the way, I looked through it again and forgot how much you had covered.

[mduckett]

MD, nice work on the pcb boards! From all the design you have been doing I would have never guessed you were an electrical guy Looking forward to seeing how they turn out. Great build log by the way, I looked through it again and forgot how much you had covered.

Thanks for the comments, but only hobby level electronics for me; although working with EE's daily does tend to rub off a little...Without free schematic capture and layout, and a cheap source for PCB manufacture I probably wouldn't have tried it. This is my first attempt at surface mount parts so a co-worker is going to help me with the assembly/soldering techniques, otherwise the odds would be seriously against me. Parts should be here before the long weekend, hopefully I can get one together for testing. --md

[mduckett]

I've been working on several fronts, and therefore have little to show for accomplished tasks. But here's an update on some progress. For the spindle controller project, my friend at work helped me to get the circuit board assembly started by putting on the most challenging part--the micro-controller—which has the smallest pads. I made several mistakes on the circuit board layout, mostly omissions in connections to some of the parts, and that has made the assembly slower, but so far workarounds haven't been too difficult. I bought a visor magnifier and a desktop lamp/magnifier to help see these tiny parts, but am getting the hang of the surface mount soldering. I've been putting together the major functional sections of the circuit and testing the sub-functions along the way. Also finished writing the bulk of the software, so the next big push is to incorporate the control software and start testing the board functionality. I plan to complete only the minimum assembly on the prototype board required to prove out the design, as I will be making a new board revision to address all of the fixes. I've also made a new layout for the SX2 plug-in compatible board and will probably make that board next, as soon as the prototype design has been wrung out. More to come on this project…

It's been really cold and therefore hard to get motivated to work on the machine in the garage. I have a kerosene heater and a small electric heater so it's not too bad after about half hour of running them. Anyway, decided to tackle a couple items from the ToDo: list today: 1) Finish and install the re-designed X-axis limit switch replacement, and 2) investigate and fix the Y-axis backlash problem that has been observed. Both of these involve the same area on the machine, so it's 2 for 1 payback on the disassembly. For the X-axis limit switch housing, I took the part made with CNC a while back (the one with the broken drill bit) and finished it using manual CNC and the Mach3 wizards to finish drilling the holes and facing it to final thickness. Tapped the 10 (!) 2-56 holes for the cover and switches, and drilled/tapped for the 1/4-20 holes in the sides for the plungers. Since the new switch housing is more compact with thinner walls than the original, the plunger stems are too long. Rather than make new ones, I'll just add some spacers and re-use these. The pictures show the nearly completed assembly attached to the machine. Note also the slotted holes for attaching it to the machine—nicely drilled CNC holes that were somehow mis-located(!), so I had to drill adjacent holes and turn them into slots with a file. So, just need to drill the final cable hole in the housing, re-solder the connections, and make the Plexiglas cover and this will be done. Replacing the switch housing also requires replacing the Y-axis way cover as it previously went around the old switch housing. As it turns out, the scrap piece of polycarbonate that I used to make the first way cover got eaten up by the flood coolant (see the pictures), so I needed to replace it anyway. The scrap polycarbonate that I used had some flaws in it to start with, like something had spilled on it. The coolant ate it right along those areas. The new cover will be a similar shape, but made from Plexiglas (acrylic), and will be attached to an aluminum angle as before using the two screws into the saddle. When this is done, I'll start looking at the Y-axis bearing block and bearing to see where the excess backlash is coming from. More to come…
One other update for the enclosure leak fixes—the front door pivot inserts fixed the leaks in the front corners, however, I still have a leak one back corner, so another go around on that.

[stumbleN2cnc]

I am slowly reading through your thread, but I had to stop to thank you for the time you are taking for this very in depth write up! You really have a sweet setup going on!

I am hoping to order a LMS 3960 Mill very soon to complement my HF mini lathe. Until I came to this forum recently, I had no idea a CNC conversion was financially feasible for me.

Then I came across Don's plans, and thought for a minute that maybe with his kindly shared work I could machine the conversion parts to result in a better setup than the CNCFusion kit. However, after reading most of your thread, I am beginning to think that I would be in way over my head taking this on, LOL! Maybe after a long while learning with the mill left manual controlled that may change, but I sure have a lot to learn...

Thanks again for the time you put into this thread!

[mduckett]

I am slowly reading through your thread, but I had to stop to thank you for the time you are taking for this very in depth write up! You really have a sweet setup going on!

I am hoping to order a LMS 3960 Mill very soon to complement my HF mini lathe. Until I came to this forum recently, I had no idea a CNC conversion was financially feasible for me.

Then I came across Don's plans, and thought for a minute that maybe with his kindly shared work I could machine the conversion parts to result in a better setup than the CNCFusion kit. However, after reading most of your thread, I am beginning to think that I would be in way over my head taking this on, LOL! Maybe after a long while learning with the mill left manual controlled that may change, but I sure have a lot to learn...

Thanks again for the time you put into this thread!

You're welcome. Sorry if my wordiness is discouraging you; I type fast and it's just a lot of detail. And I understand about limited budget too. As for skills, I had zero machining skills with metal, but have been a woodworker for a lot of years which helps a little. It's a good idea to tune up the machine, learn how it all works, and then make some parts to get the feel of how it all works. Once basic skills are in place you'll probably feel less overwhelmed. Buying the machine and the minimal working set of tooling does cost a bit, but careful planning helps get the essentials. I'm jealous of the solid column mill, would have bought it if they had it when I got mine. Hope you'll post back or start a build thread when you get your machine set up. Thanks for reading, feel free to post back if you have questions. Good luck! --md

[stumbleN2cnc]

Au contraire! Its an awesome read, it isn't the wordiness that is discouraging me at all. I just underestimated the scope of the project, and your write up has opened my eyes to whats in front of me. I still think it will be much more rewarding to make the conversion parts myself rather than just buying a kit, and your success is evidence of that. That is a real nice setup you have there any way you look at it, and I'm willing to bet I will borrow at least a few of your ideas

[mduckett]

Over the long holiday weekend I got some time to work on the machine. Finished up the new X-Axis switch assembly, which was resoldering and testing the switches, turning/parting some spacers for the plungers, and making a new cover. After testing I used some silicone to seal the cover and fastened with the #2-56 screws. I didn't seal the previous unit (forgot!) and there was just a hint of some moisture on the inside of the lid, so this will be better. After finishing the switch I tore into the Y-Axis bearing assembly to see what is causing the bad backlash issues. Well, it appears that there isn't a single source, but several contributors. 1) The bearing block was not cinched up tight. I remembered that I had been playing around with some binding issues and apparently forgot to finish up that investigation, so there was some play from that, maybe .005 max". 2) Upon dis-assembling the bearing block lid, it was evident that there was too much clearance between the lid and the bearing. I thought I had measured these before installing but....anyway, installed one .005" shim and that helped a lot, but there was still some evidence of more, and the lid was down tight, so I put in another shim. Now the lid has a gap of about .002, so that means there was about .008" of backlash in the bearing movement. So that is about .013", which is in the ballpark for what I've seen. 3) The third source is one that will require a change of the ballscrew. Even though I installed oversized balls, there are places on the screw that seem to have .002" of backlash. Other places have close to zero. This particular screw had issues with straightness, and I think that it was just a bad piece from the start. I'll probably replace it at some point, depending upon how bad the final measurements are. On the positive, the place where the screw is sloppy is at the far extent of movement, so at least it performs well where most of the milling takes place over the middle of the table. The Y-axis was re-assembled, and took a fair amount of time playing around with the bearing block to keep it from binding. Ended up using .007" shim on the right side to allow cinching down tight without binding. I guess the front face of the table is not square with the saddle, or at least not with my ballnut assembly installed in the saddle. So the next step is to do some testing to determine the final backlash results. Sorry no pictures, it was cold and my hands were too dirty to use the camera. I'll post some pics when I get the test results.
While I had the vise off the table, I took some time to check the tram using my homemade tramming gauge. It turns out that the Y-direction (nod) is out only about .001", and the X-direction is out about .0025". Not bad, but I'll tune up the X direction before using the machine again. Tramming tool works great, highly recommend everyone making one.