TaK's Sieg X3 CNC Upgrades

[TaK]

I was getting ready to put the bearing in the top of the quill when I noticed something. The top of the quill is stepped, and somehow I don't remember it being like that.
https://lh3.googleusercontent.com/-3...6/IMG_1366.JPG

The lower step fits the original thrust bearing, then a washer and the upper step is for the deep groove ball bearing. Here's a picture showing the thrust bearing in place.
https://lh4.googleusercontent.com/-p...6/IMG_1367.JPG

My plan was to replace the top thrust bearing / deep groove ball bearing combination with a single tapered roller bearing. This is mostly for a future power draw bar because that tapered roller bearing can take some force. Even if you don't do a power draw bar, it seems like a better way to go since you are going to be pre-loading the spindle in the quill and hitting the spindle with a mallet to get the tool out all the time anyway.

I was expecting the roller bearing to be shorter then the combination of the two original bearings and I would need to put some shims (washers) to get the level up to the threads for the locking nut pair. Luckily I think the plan will still work; it's just that the new tapered roller bearing sits in the top step and is raised above the level I expected. Here is a picture of me placing the old, original tapered roller bearing into that top step to estimate the height. It sits about 2mm above level with the top of the quill. The new bearing won't sit that high because the outer race of this one has welds all over it from the removal process, but its close enough for an estimate.
https://lh6.googleusercontent.com/-a...6/IMG_1368.JPG
https://lh3.googleusercontent.com/-n...0/IMG_1369.JPG

I did a little more measuring and put the quill in the oven. I finished polishing the spindle journals on the lathe. I ended up using mostly 80grit emery cloth; it still took a long time. Be sure to let the journals cool before test fitting.
https://lh6.googleusercontent.com/-X...0/IMG_1370.JPG

I got the upper taper roller bearing race in the quill. It took two tries again. This time I know the problem was that it didn't start going in strait. Here it is in the top of the quill. It looks like when the rest of the bearing is together it sits approximately 1mm above the top of the quill. I think it's going to work out nicely.
https://lh6.googleusercontent.com/-i...6/IMG_1371.JPG

Here's all the spindle parts waiting to be put together. It's time to grease the roller bearings and final assembly. I need to get this thing spinning again and see if it was worth it.
https://lh5.googleusercontent.com/-T...6/IMG_1373.JPG

[TaK]

I reassembled the quill and spindle. It went pretty smoothly. I packed the tapered roller bearings with the Kluber grease trying to fill 50% of the available volume.
https://lh5.googleusercontent.com/-8...6/IMG_1375.JPG
https://lh5.googleusercontent.com/-6...6/IMG_1376.JPG
https://lh4.googleusercontent.com/-b...6/IMG_1377.JPG
https://lh4.googleusercontent.com/-M...6/IMG_1378.JPG
https://lh3.googleusercontent.com/-E...6/IMG_1380.JPG

The new bearing in the top of the spindle sits about 1mm past the top of the quill. I'm using the washer that was just beneath the c-nuts and it seems to fit about perfect.
https://lh6.googleusercontent.com/-I...0/IMG_1381.JPG

Here is the assembly back in the head.
https://lh4.googleusercontent.com/-J...0/IMG_1383.JPG

Don't forget to put the rubber ring back on at the bottom of the quill like I did or you'll have to take the whole damn thing apart again and redo...

Here is a photo of what I'm calling the 'magic screw'. It took me a long time to find this little jerk to remove the spindle. The preparation guide didn't illustrate this thing well enough.
https://lh5.googleusercontent.com/-c...6/IMG_1386.JPG
https://lh5.googleusercontent.com/-Z...6/IMG_1385.JPG

The only real tough part was getting the spring back on and tensioned correctly. Here's a tip: use a large set of circlip pliers to grip the spring cage by those 2 holes. Then you can wind the spring back to your liking with little issues. First I tried using big pliers like the preparation guide shows, but that wasn't working well.
https://lh5.googleusercontent.com/-D...6/IMG_1387.JPG

Something interesting:
The spindle is much harder to turn by hand now then it was. I may have pre-tensioned it more then the factory did. Also, the grease I'm using is much thicker than what was in there before. It might settle and spread a good bit too. It is nice and smooth though, so I'm going to leave it for now and see how the motor handles it. I can back off the pre-tensioning if I need to. I think I'd rather have the motor work harder and have less vibration in general though.

When I first turned it on, there was a nasty noise. I'm fairly sure it's in one of the gears. I think its either the spline gear or the 2 metal main drive gears on the top of the head. I'll pull the spindle assembly tonight and just run the gears to see if that's where the sound is coming from. The main drive gear that powers the spline might have taken a bang during disassembly. Hopefully I'll be able to find a little dent and file it out.

I'll probably post a list of the parts and specialty tools I used to do this bearing change next time. In case anyone else is thinking about it.

TK

[PriddyShiddy]

any word on what caused the noise? gotten to run it yet?

[TaK]

It's been running since I got it reassembled. I have only cut one piece since then; about a 1.5 hour job. I measured the run-out with a dial indicator and it looks to be about 0.00025". My finest dial indicator is only 0.0005" per tick, and it moves about half a tick. I would need a better indicator to get a better measurement, but I think it's probably an acceptable amount of run-out and I don't think I messed up anything badly with this whole endeavor. I wish I had measured the original run-out though... can anyone out there fill me in?

The noise and vibration has gone down at higher speeds. I still believe it's the two big toothed gears on the top of the head. I guess whatever little dent I put in one of them has worn down. Either way, I'm not very concerned about it since I am going to be replacing those with a timing belt soon and I don't think the vibration is enough to be affecting my cuts right now.

An annoying detail has been that I have not gotten the force on the quill spring where I want it again. There is a tab bent into the quill spring where it engages on the quill bar shaft. On mine, it seems to have unbent so it is tough to get it to engage and get the tension correct.

A more substantial problem is that the head is getting very hot. After about an hour of pretty light cutting, I could hold my palm to the machine head for only about 15 seconds before I had to remove it. My guess is that I have too much pre-load on the spindle. I have noticed that it is quite hard to turn by hand even without the motor engaged, but it's very smooth. One of these nights I will remove the quill and back off the pre-load nuts. At this point, I should be about to do that in about 15 min.

Other possible causes are that there is a problem with the bearings. I don't think this is likely because the motion is very smooth feeling so they probably are not misaligned, and the run-out isn't bad. I was careful not to get debris in them when installing and packing. I doubt I over packed them with grease; I was careful to pack 30-50% of the volume. The 6007 bearing came packed from the factory and I will assume they put in the appropriate amount. The two tapered roller bearings are open on one side so if I did manage to over-pack them, it should have squished out a bit.

I did a tear-down and tune up on the machine in general. I am no longer satisfied with my ~3mil of backlash. I should probably spend some time tracking that down. I am using the CNCFusion premium ballscrew kit. My first guess is the helical couplers are the largest contributor. Any recommendations on what variety I should be using?

In other news, I am making progress in selecting a motor/controller upgrade. Right now, I am trying to figure out how best to get more power into my garage. My garage's normal 120V, 15A breaker is probably going to be stressing when I make this upgrade.

TK

[PriddyShiddy]

no recommendation on the couplers since the helical ones from cncfusion are the only ones I have ever used. I have gotten my backlash down under .002 on x and y just by scraping my gibs. I also found that only about 10% of my gibs can possibly make contact (which makes scraping pretty fast) because the angle used on the dovetails is too shallow for the placement of the 4 adjuster screws. all of the pressure is pushing into the bottom edge of the way NOT the face. every picture I have seen shows the exact same pencil thin wear mark on the x and y gib. sad.

when adjusting the gibs I get then where I think they should be, check the backlash and loosen or tighten till is .002 which is a good tight (for the x3) fit with little backlash. if I back then off a hair I can get 0 backlash, but twisting the table makes the indicator dance. .002 on a $1000 machine is pretty damn good and hasn't effected any of my parts. I'll have some more videos up on my youtube soon and will show it slot cutting 1018 with a .25" 4flt at.2" doc.

column flex/ head flex is more of an issue for side cutting, but a finish pass takes care of it.

the reason I reread your post today was BECAUSE of the heat... so interesting you mention that. I ran the spindle close to 10 hours this weekend and noticed it was getting really hot (4000rpm, mostly) last night 37 minutes into cutting it stopped. the fuse was in several pieces. replaced the fuse. it stalled. tried to turn the spindle....SEIZED. pulled the gear train off to make sure it wasn't the motor. it wasn't.

I just laughed since I had just priced the spindle and quill and bearings yesterday

after your exploits, and several others I'm gonna spend the $130 for both and start fresh- I can't afford having the mill down while I play with it and the likelihood of damage is less

as I started tearing everything apart I removed my florescent light that had an aluminum bracket holding it to the quill. what do you know
?! spindle was free again. seems the heat was making the spindle, bearings and quill expand, but the aluminum bracket washed its heat and stayed the same which was creating additional heat, resistance and eventually stalled the bearings.

left it off and ran another 3 hours last night perfectly and definitely cooler. I borrowed an infrared thermometer today to test the heat.

my spindle inner taper shows .0004ish TIR. I'm using a .0005 too to measure it too. I wasn't to measure it again when its hot because its noticeably smoother the warner it gets.

how fast have you run yours? just the standard 2000? if its getting that hot there... I'm scared. very curious to see what happens after you loosen the preload.

all in all I'm impressed with the stock bearings. I run up to 5600 and they get hot, to the point that I can only hold the nose for 5-10 seconds, but the head had never gotten quite as hot as yours did.

I'm thinking about making a heat sink for the quill. im thinking an aluminum round sink with fins pointing out and cut into 3 pieces and held to the quill with a heavy rubber band or two so it could expand with the quill and should draw at least a significant amount of heat out. heck I wouldn't even mind wiring up a couple tiny fans to the x3 power switch and mounting them under the head behind the heat sink.

I look forward to hearing from you on the heat. I will make a heat sink when I can and do some benchmark testing the day befor install so I can see if it really helps. if I can get 20% cooler I can run 20% faster I'm thinking 4000rpm measured at the nose, quill, fixed point on the head, spline and top of gear box (while running) every 15 minutes until the great levels out.


great work on the posting btw. really enjoyed reading this one again

[TaK]

OK. I did some work tonight.

I removed the spindle from the machine.
https://lh4.googleusercontent.com/-D...6/IMG_1411.JPG

First, I verified that the weird vibration is the two main gears on top of the head. I will be removing them both soon so no big deal. I probably damaged a tooth when I was removing the 6007 bearing from the spline drive gear. I looked for any obvious damage, but don't see any. I can't really feel the vibration but I can definitely hear it.

Next, I did a test to find out approximately how tight my spindle is in the quill. I don't have anything resembling proper equipment for this sort of test... I took a piece of 30 gauge wire and wrapped it around the spindle and tied the other end to a plastic bag. I added coins to the bag until the spindle unwound. Then I used a kitchen scale to weigh the bag.
https://lh6.googleusercontent.com/-U...0/IMG_1412.JPG

The first test bag measured 2lbs and 14oz. The spindle is 58mm in diameter so that should be about 40oz-in. Did I mention it was tight? It's actually easy to move by hand at that level, but is noticeably tight. I readjusted the pre-load nuts and that bag weighed 6oz or about 5oz-in. Both of these measurements are just the spindle/quill assembly and dont include the spline drive bearing or gears or motor. It's a real pain in the ass adjusting these c-nuts by the way.

While I had the spindle apart again, I worked getting the quill spring back in shape. I finally got the end of the spring bent back into a reasonable shape. For any of you that have not had the pleasure of working on the head of your mill, this is also a pain in the ass. I don't know why it moved from the original position, but it now seems happy and springy again.
https://lh3.googleusercontent.com/-l...6/IMG_1414.JPG

So hopefully tomorrow night I can run the spindle at 2k for a couple hours and see if the head gets noticeably hot with the lighter pre-load. I don't know if there is enough, or still too much pre-load. I'm not sure if there is some way I can measure the play in the spindle in the Z direction under loads. I'll have to think about that. In the mean time I'll be on the lookout for vibration.

Priddy-
If you are going to do work on your bearings, can you do a similar test so we can see how much torque it takes to move the spindle with a factory pre-load?

TK

[PriddyShiddy]

my hat is off to you for thinking of the bag trick to measure the tension! very smart idea.

tonight I measured the temp at the quill, and head. they both started around 56°. I ran the spindle at 1000rpm for 20 minutes in low gear and it climbed to 60 at the quill and 69 in the center of the head. then I ran it at 2750 rpm in low gear for 30 minutes and it topped out at 20 minutes at 66 on the quill and 80 on the head. I was a little surprised that the head was hotter. I guess the air exchange cools the spindle. interesting that it seemed to top out fairly fast. more tests are on the way next year

had to stop there and finish up my christmas ornaments I have been making for the family.

at this point I'm not sure if I will pull the bearings or not. I'm just killing time until we move to buy a tormach, and then I imagine I will have time to tear this all apart again. I'm trying to keep myself from spending time on the machine unless absolutely necessary as every hour and dime I spend on the machine and not making parts puts me that much further from my goal, baring failures of course.

BUT- as soon as I can I will pull it and test it though I think the numbers would be un-fair to consider for a comparison to stock. more like "abused stock" now that I managed to seize the dang thing. really wish I had a timer on the spindle because its hundreds and hundred of hours already

sure is dark under the tool now gonna have to figure out some new lights

if money gets loose I'll just buy a whole new spindle one part at a time and throw better bearings in.

congrats on the decrease in tension, very interested to see your motor get sorted out. thats one thing I think I.could justify since its only a matter of time befor mine blows in the middle of a job

[cornbinder23]

I'm thinking about making a heat sink for the quill. im thinking an aluminum round sink with fins pointing out and cut into 3 pieces and held to the quill with a heavy rubber band or two so it could expand with the quill and should draw at least a significant amount of heat out. heck I wouldn't even mind wiring up a couple tiny fans to the x3 power switch and mounting them under the head behind the heat sink.

I look forward to hearing from you on the heat. I will make a heat sink when I can and do some benchmark testing the day befor install so I can see if it really helps. if I can get 20% cooler I can run 20% faster I'm thinking 4000rpm measured at the nose, quill, fixed point on the head, spline and top of gear box (while running) every 15 minutes until the great levels out.


great work on the posting btw. really enjoyed reading this one again

I made a coolant manifold that clamps around the quill and that acts like a big heatsink, there is a decent picture of it on post #38 in Mybuild thread It works very well, and I run it at 6000 rpm for hours on end

JTCUSTOMS

[PriddyShiddy]

I made a coolant manifold that clamps around the quill and that acts like a big heatsink, there is a decent picture of it on post #38 in Mybuild thread It works very well, and I run it at 6000 rpm for hours on end

JTCUSTOMS

how the heck did I miss that in your thread?!?! I've probably read it 5 times and it never clicked.

I'm not sure how effective it would be with my mist system since the liquid is barely moving, but now you have me thinking... ever used a jockey keg? copper tubing coiled up in a cooler full of Ice that the beer runs through which cools it between the keg and the tap.

5 gallon bucket
Pond pump
2' copper tubing
couple press on connectors
20' of 1/4" rubber tubing
5 gallons of water

I already have everything but the copper laying around.

wrap a few times around the spindle with the copper tubing with an aluminum mount, run rubber tubing from each end to the bucket under the table with a 110v wall switch to the pump.

the water will be pretty dang cold but since the system would only be switched on when the head is over heating from running 4000+ rpm for long runs it won't bother me to walk into the house and grab 2 ice cube trays if the water gets hot.

by god that sure seems like a great $50 solution. also gives one of the two pond pumps sitting on my shelf some purpose in life

big machines have water cooled spindles... so can little ones. I'll get started first thing in february and get some benchmark tests to see how effective it is. I think 5000 rpm for an hour and measure the heat ONLY at the head should be effective.

[cornbinder23]

Yep, we call them jockey boxes and if I look hard enough I can probably find 2-3 of them out in the barn from "younger" days

JTCUSTOMS

[cornbinder23]

BTW, are you guys using a DTI to measure end play of the spindle when setting the preload?

JTCUSTOMS

[TaK]

are you guys using a DTI to measure end play of the spindle when setting the preload?

I measured the runout with a dial indicator when I mounted the spindle back on the mill. I wasn't using the dial indicator while actually tightening the preload nuts. Am I missing something?

TK