DIY 40 tapper spindle For XQ6232A CNC converted knee mill

[epis]

Hi everyone.

I have more than 2 years XQ6232A milling machine, bought from china fob price was ~4600$ without options, i paid around ~6000$ with options like automatic oil lubrication, digital readouts, and more stuff.. which last year I converted to cnc, with ac servos, and X,Y ball screw, and now I need more spindle speed, currently I have max 1500 RPM 2.2kw ac motor and that is too slow, + that gearbox is very very noisy, and all machine vibrates at max RPM, that leaves poor surface finish.
So I started to think about new spindle, I want 4000-6000 rpm, belt drive, I made research and new cartridge spindle BT40 cost ~1200$ from chine + shipping 250-300$ that's 1500$, + taxes, papers,and if adding pneumatic draw bar that all will go close to 2000 euros, that's a lot of money for such machine. on low cost side there was cheep spindle about 350$, but shipping is the same and it has 3000 rpm max, and at the end will cost more than 600-700euros. + very long shipping, close to two month. at the end I decided to try and make spindle myself.
Already made spindle housing for first spindle shaft design, it is in the picture below it took 6 days, drawing, milling, turning, and supply materials.

Problem is that few days ago I started to rethink my spindle axis design, and learned to make axis inertia calculations, and did some research of other 40 tapper cartridge spindles and they all have bigger inertia values, for example, belt drive cartridge spindle bt40 lowest power range has 8000 kg/mm2 inertia, higher Kw spindles with lower Rpm have >10 000 kg/mm2, but very interesting was 22kw high speed spindle with integrated motor has 20000-25000 kg/mm2 inertia, So I guess spindles with integrated motor have much higher shaft inertia, and I think this is good thing.
My first designed axis with inner bearing spacer has inertia ~3000 kg/mm2 if I add 32 tooth HDT M8 30mm wide pulley I get ~4250 kg/mm2 and that is very low number, (2.2kw 1500 rpm 60hz motor has 5400 kg/mm2 rotor inertia) currently my milling head with all thous gears have very huge inertia , and currently I can cut with indexable endmills like 16mm diameter 2 insert and even 50mm 4 insert cutters steel very well with only 2.2kw motor, i have doubts that if I make that first low inertia shaft that i won't be able to cut with indexable end mills steel, to explain my concerns I included print screen picture of ISCAR power calculator where you can see that 16mm indexable endmill can cut at 4000 rpm steel 2mm wide 10mm deep, and feed 0.15mm/tooth at 0.9kw of power, but at power time variation plot you can see that max power is 7.5 kw and the same is with torque, average is 2.2nm but max is 17.9 nm so how can under powered spindle 2.2kw take such a cut, and answer is inertia, inertia stores kinetic energy and smooth things out, and I made some calculations for 2.2kw inertia number (0.0054kg/m2), and at 1500 rpm it has kinetic energy 66 joules(or Nm) at 3000rpm 266J, at 6000rpm 1058J, Formula is Er=1/2* Inertia*W2 (angular velocity 2square), so faster you rotate more energy you store and I guess it also smooth torque/ power ripples and vibrations when cutting with indexable inserts, Now my milling machine gearbox does that kinetic energy storage function, but will new spindle do that ?

Solid carbide endmills don't have that torque ripple if cutting with full flute depth, so it has very low vibrations, this is my video but I want my new spindle to be able to cut as my old gearboxed spindle, so my conclusion is to make spindle shaft as heavy as I can to increase shaft inertia, that will also damp vibrations, and I think it will also extend bearing life significantly so i redesigned my shaft and got 12145 kg/mm2 inertia that is 4x more than first design, with 32t pulley I got 13345kg/mm2 , and if I add 2.2kw motor shaft inertia, and other 44t pulley inertia I get 23400 kg/mm2, that is something I start to like. it could be close to some 5.5-10 kw lowest end CNC machining center with bt40 spindle inertia numbers (I think so), or direct drive internal motor high power spindles, thous all cut steel very good with indexable insert cutters.
I got so low inertia number in first design because spindle shaft length is just 304mm, compared to cartridge spindle 450-460mm overall length, I can't drill hole so deep , and my mill capacity is already on the edge, I have 17.5mm spade drill with coolant holes, it is 180mm long, I got it at discount Sales long time ago, I will try to make that spindle shaft, and inner tapper I will turn at the end on milling machine, by milling machine with lathe tool vertically, I think I read about that in this forum some spindle building topic long time ago, so I will try that too.

what do you think about my thoughts on this spindle shaft, inertia, other things ?
any comments suggestions are welcome.

I know that everyone in industry want spindle with low inertia, but i think this is not my case, spindle starting and stopping time doesn't concern me. I need speed, long tool life, low vibration, low noise level.

i will update my progress next week when I will start to make shaft.

Similar Threads:

• 6 ATC for BridgePort Knee Mill R8 Spindle
• Looking for some info on disassembling a knee mill spindle.
• Bridgeport Knee Mill Machine Head Spindle Square Tram
• 8x 36 knee mill spindle speed question
• Spindle Speeds for a cnc knee mill

[epis]

Here are few pictures of what's already done. (this week) and how I made spindle housing.
I made few changes to my First spindle shaft design, and made it simpler, removed 7208 double bearings and placed single 6208 bearing and removed second lock nut M40x1.5, I don't know Why I placed two lock nuts, so now I will have only one and adjust 2 angular bearing preload with M50x1.5 nut. I decided to not make that High inertia shaft, because I will have to remake spindle housing to accept bigger 6309 bearing and that is a lot of work, I don't want to make. Now plan is to make this today's designed shaft it has low inertia 5630 kg/mm2 and see how it will cut steel with indexable cutter, and if it will make lot of vibrations, then make try and make inertia rings and double or triple inertia of shaft.

I noticed grammar mistake in topics description can moderator correct word "tapper" to "taper" .

I included picture of current milling machine Shaft, I disassembled the spindle head and lower roller bearing has this nasty tapper in ID, my first Idea was to use this original 40 taper shaft, and change bearings to angular contact, and make new spindle housing for direct belt drive, but that tapered bearing ID made it impossible. also I included picture of spindle housing and you can see bearing pocket is with holes, this because I ****ed up pocket by 0.15mm made it larger, and this is FIX I made with stick welder, now pocket is ok. it took 3 days to make that spindle housing, and I maxed out my 4kw cnc lathe capabilities, even grinded indexable insert TNMG corner radius from 0.8 to sharp to reduce chatter, and cut bearing pockets.

[epis]

Progress update.

drilled hole D16mm with my longest drill 180mm deep.
then bored hole to 25mm diameter 88mm deep, that is 5mm short of 93mm I needed, in order to put ISO 40 toolholder, problem was that my 22mm indexable drill 4D couldn't reach that dept, solution would be cut 5-4mm off iso 40 toolholders, BT 40 don't have that extension, maby later I will fix that, but now I will leave it, will continue next day with other spindle side.
Checked tapper with one of BT40 toolholders, it fits, but I don't now for sure, I don't have that special paint that highlights contact surfaces, Will try to use marker.
made video


also took some pictures.

[epis]

progress update..

Drilled hole from other side 16mm, then turned stock from back side of shaft and decided to cut thread M50x1.5 and ****ed up, made M50x1, mistake in code, forgot to change pitch to 1.5, and needed to rework, so here is shaft with wrong thread.

shaft after I turned welded area to 50mm diameter

And picture of thread and lock Nut screwed on,

I made that thread too tight, it started to jam when screwing on, and when I started to screw off it got harder and harder to screw, I hammered it half way and tired, in process I almost destroyed lockNut,, I think some debris is left in thread that's why it is hard to turn.
continued work, and made Flange for shaft, and Pressed it on by hammer (10kg steel block)

and welded it from back.

next I will turn shaft, all bearing surfaces, and then shaft is ready.
I checked taper cone with simple marker and it marked all the way on bt40 tool holder cone, so I guess it is nice fit, but misalignment when installed will be big 0.1-0.15mm , and will rework cone when installed on mill.
hole I made aligned in middle good, no more than 0.1mm that spade drills is self centering, so even when I started off center (at the start drill chatters), it self centered itself and got good hole. i thing i will need to make draw bar 15.8mm Diameter, 16mm will definitely jam.

[lcvette]

Looking good! Liking the build so far, quite the undertaking!

Sent from my XT1080 using Tapatalk

[epis]

thanks for comment
today's update.
Today was very hard day as it was yesterday, I started by removing jammed LockNut, and couldn't remove by hammering, so I cut it on 2 sides and removed.

After removing II noticed a problem, and as in picture thread was damaged, probably from debris, that's why it started to turn so hard. so another bad luck, thread was once again ****ed up.

I took my welder turned on 90 amps with 3mm stick electrode 6013 and welded it up, used 3 electrodes, previous time I used 2 and had few spots left, unfilled now i added extra layers.

Then I turned shaft flange I welded yesterday

and started Final task turning all bearing surfaces, and remaking thread, and this time I got it right made thread 0.05mm deeper and nut turned on easy with good fit. finally good thread.

bearing surfaces I turned very carefully by 0.02mm pass till got .98-.985 mm tolerance it took lot of time using finishing WNMG 0.4 radius insert

shaft turning is finished, now I have to mill key slot on thread, and taper cone dog place and make and drill holes for screws. and few other parts i din't made

[epis]

Today is holiday, and I am not working.
decided to post some pictures of my miling machine retrofit, I made

ok this is my machine in china before shipping brand new.


here is retrofited picture of X,Y axis servos direct drive X axis ballscrew coupling can bee seen. X axis motor has 1mm gap between Cast iron table very close fit.

machine from other side, z axis is not ball screw retrofitted, just switched ac induction motor to servo.

and electric box, with csmio/IP-M controller.


ballcrews are C7 32mm cheapest ones with double nut and I adjusted prelode myself meany times to get accuracy i forgot what was that.
I bought that machine to make gang tool block and other parts for cnc lathe, with future plans to retrofit. i also considered buying old used milling machine but couldn't find acceptable priced machine.

I did learned a lot about milling machine, and how they are made, and specially how cheapest machines are made, and precision of thous mills are really low, but with linear scales that doesn't matter, and linear scales help a lot even now, because, mach3 after reset losses position, and thous offsets are changing, I found solution by putting zero coordinate on linear scale panel and each time I start or reset machine I go to that zero and reset mach3 ordinates to 000, so I never loose coordinates, I don't have homing switches or other precision coordinate resetting hardware. ( I have limit switches on all axis, those came with machine, actually lot of stuff I reused from original electronics,)

[epis]

update from yesterday. few hour holiday work
Plan was to make bearing spacer, ID50-OD72mm and found out that uncalibrated stock has ID 52mm (I think I ordered for different design in mind, for bearing with ID55mm, then changed to 50mm). I turned rust off and got ID54mm that's way too large, so I decided once again fix that with welder, here is picture how welded sides look. Other options were order correct stock, or make 2mm ID50-OD54 tube and press in, welding was the fastest and cheapest way to rework, i left the peace to cool down, will finish it next week.


This is what happens when are different designs, lot of changes during build process, result is lot of mistakes, this is 4 mistake, first was shaft house bearing pocket re welding, then 2 times thread rework (welding), and now incorrect stock material dimension, and all times welder saves a day, I can't imagine how much extra work I would have to do if I made new part each time I **** up. with welding it is 1-2 hour fix, without, i think days, (ordering new stock material 2 days and then doing all work from beginning 1-2days) so I saved a lot of time. on other side if mistakes were not made at all I would save some 7 hours and would be more productive, but that kind of mistake free work only happens in production runs where all mistakes are bugged out and process is flawless, that is my experience, so making new stuff that i make first time I do make lot of mistakes. it is also lack of experience

[epis]

Progress Update
I continued with spaced I welded previously, turned welded areas to 50mm cleared surface, and got lot of unfilled pockets, and no corners, then I welded second time with smaller electrode 2mm to make corners and this is how it look after second weld

then I turned everything and this is final part.

next I made 8mm spacer and I forgot to cut stock peace out off material I ordered for Outer bearing spacer I made, so I found 150x150x10mm plate and started turning out part and when turned outer corners I broke 2 edges of 2 different inserts, spindle stalled machine hasn't enough torque you can see damaged insert, that's why it is hard to turn large diameter steel with 4kw 4 pole motor without gear transmission, I could take 1mm deep cuts but at the begging I was cutting 2mm deep difference was that first cut was interrupted and when I cut off square edges spindle stalled, and explanation is that spindle has high inertia, and in interrupted cutting even at larger depth off cut inertia stored that rotation energy in non cutting time and released it at cutting, same thing happens when milling with indexable inserts, as i showed at the beginning of the thread, that power curve varies over time, and peak power can be much higher at cutting time, than average, and I experienced that on lathe, when I got full circle that math didn't worked i had 100% catting and belt started to slip, spindle stopped and broke edge of insert, sort off mini crash, and second insert I also broke up. actually it is very interesting to see how inertia averaged cutting torque, and lathe spindle shaft has lot of inertia, this is why I have this thought about making " high inertia" spindle and benefits of that.

here is finished part

[epis]

Update of past 2 days.

I started milling operation, and before milling I cleaned, and oiled my mill and did some maintenance, that took all day.
I needed to mill 6 bolt holes M8 for Outer Cap Ring, here is setup in mill.

and part milled.

Next I milled holes in spindle housing, 6x M8 bolts, I taped by hand

Checked if all bolts screw properly, and how everything fits, and all is OK

and all parts now are made except shaft, and I start with key ways, first for M50 lock-nut anti rotation ring 6mm and 2.5mm deep groove

and key-way for pulley 6mm wide and 3mm deep

and one of the final tasks will be milling drive dog area and drilling holes, then spindle assembly and final taper turning. here is setup for drive-dog milling. milling wil be done tomorrow.

[epis]

shaft is DONE. finished drive dog milling yesterday, was so tired that update is today,
here is pic of milled drive-dog place

pic of 40 taper with thread cut in drive-dog place

and one more pic of shaft , you can notice lot of scratches, and that's all new, after turning surface was smooth, this metal is soft so easy scratched, also there are hammer marks on shaft itself i did that when placed bearings on axis to see how they fit and of course they jammed and don't have special tools to remove bearings from axis, i use hammer, i think that is just cosmetic look, i will clear all bumps with file, if bearing won't slide

found old pulleys and I will place them at first, just need to order correct middle section for axle shaft 40mm OD and motor (don't know motor shaft diameter size).

I am thinking about heat treatment of 40 taper cone, by welder, , or leave everything as is. haven't decided jet.
I also have new Idea and made new drawing of spindle with even longer shaft, and bigger bearings. and lathe spindle nose on outside flange that would be handy to put lathe chuck on mill and make it more universal.
Next update at the end of next week.

[epis]

i am back and today I was making spindle housing Pads.
I made pads from some junk steel I could found, and here are pictures, of stock material with grinded edges


pic after welding bottom view,, I also welded spindle house tube to side block, just few welds.


and Front view


that black thing | covered part is graphite grease, it protects surface of hot metal sparks don't weld to surface, at least most of the time it works. , and grease is cheap,

[epis]

Today I milled spindle housing mounting pads, and drilled 12mm diameter holes for M12 bolts.
I made video how sparks are flaying, because insert edges are worn out,


I welded 4 plates to the bottom for clamping it to table

Part done after, I got accuracy in X axis in the middle round section = 0.05mm on my measuring device, i will try reverse tomorrow and level other side and then again level to get better accuracy.

[epis]

Update.

I made another spindle housing leveling pass with face mill, to get 0.01mm precision here is video.


after spindle houding is done. I started to make Spindle BOX, it is box where spindle housing will be attached to, first I milled Plate and drilled 4 holes M12 thread to screw spindle to it, and prepared peace for welding, also milled all other steel plates to prepare them for welding.


And Friday I welled all day with 4mm electrode at 192 amps DC. and this is welded box picture



That welded box rings when hit with something, I am thinking to fill it up with something. my first option is use small dense steel chips from lathe with density 3-3.5 g/cm3 + if i add fine sand 0.1mm powder to fill the rest of free space I could get density up to> 4g/cm3, and I also will need some liquid, I am thinking about oil to fill rest of the free space.
Other option is to put solid granite bricks and glue them together with epoxy/sand mix i could get something like granite/epoxy, but easier to make and I don't have vibrating table or other vacuum pump to get air out of granite epoxy composite mix like others are doing, so I could try glue granite bricks together.
haven't decided yet which one I will chose, maybe I will try to make small test of steel chips+sand+oil and see how it works, if it Fail I will throw it out and will go with second option. fist option is cheapest one .

[hanermo]

Excellent thread !

I think You are doing very well.
No concerns with moving the, by now, very heavy spindle box (must be 150 kg or so) ?

Mine is bolted, for this reason.

[epis]

Thanks for good question hanermo

Made some weight calculations of all spindle big heavy peaces
here is drawing

Total weight all all you see in drawing is 73kg
out of that steel plate 20mm spindle box weights 49kg
inside volume of that box is 9.5 liters and if Filled with steel chips+sand+oil (density 4g/cm3) could increase spindle weight by additional 38kg
Other peace will be spindle 2.2kw AC 4 pole 380vac -1500 rpm motor that weights about 21kg
Total weight will 141kg and that is without bearings, belt pulleys, and spacer rings, that could be additional 10+ kg, so yes about 150 kg will be total.

haven't thought about moving spindle box. but as it consist of multiple peaces I am considering mounting it on milling machine peace by peace. I don't have crane, or other lifting equipment,
I have few long 2-3m long steel profile tubes, out of which I could weld up frame and buy chain hoist and build a crane, but that would take time, I will try without crane first

I will also bolt whole spindle Box to machine frame same as current swivel head spindle is mounted, by 4 mounting bolts. I will remove swivel head from machine and take measurements next week. and then assemble machine, mill my spindle box to fit machine and then once again disassemble machine and try to install my spindle on machine, so lot of Extra work on assembling and disassembling,

[hanermo]

You now have the spindle outside the box ?

Mine is similar, but a bit wider, 300x20 mm plates on front/back and 200x30mm on sides.
Mass 100 kg.
Mount plate for spindle is 200x200x50.

And the spindle is inside !
This reduces overhang, and makes it hugely more rigid.

Rigidity = free length pwr 3 !
Something to take into account, imo.

[epis]

Update.

Spindle BOX is done, it took 4 days, and 1 day was for disassembling, measuring and assembling swivel spindle head here are pictures.


this is how I measured exact hole location using small milling head and 6mm drill, I don't have touch probes or any other fancy measuring equipment. and thous holes are very inaccurate off by 1-2 mm that is how these low cost china machines are made, same was with ballscrew mounting holes. every hole in different place.

and now milling, here is cool picture on chips after i milled with broken 12mm carbide endmill (tip is broken, i broke it when milling drive dogs)

and finished picture of mounting surface

here I milled slots for M16 bolts i used 50mm face mill, and finished with 16mm indexable mill,

hanermo

You now have the spindle outside the box ?

yes i have it out of the box, and spindle center will be in the same position as old spindle head.

Soon i will put everything together, and make 2.2 kw motor mounting platform, i just need to make hole in spindle housing for bearing preload adjustment,

[epis]

yesterday I finally assembled spindle, and I have made few mistakes.





In one picture you can see I drilled 3 holes and milled flat 16mm slot in spindle housing, I did that to adjust bearing preload and made it incorrectly by 3mm too far from place it needed to be, because spindle housing is by 3mm longer than in my drawing.
second mistake is incorrect inner angular bearing spacer it is by 0.2-0.25 mm shorter than outer spacer ring, and when I tightened prelode screw nothing changed, there were no prelode,so Question is how much I need to reduce inner spacer to make prelode for 7210 bearings ?
in hands I can feel that bearing play is very large, compared to deep grove ball bearings, that don't move at all, so I made prelode based on my experience on deep grove ball bearings, but it looks like it is very different for angular contact bearings, is there any specification where prelode distance is given ?

I hammered/pressed shaft in and it doesn't have any play, I measured runout of 40 taper cone and it was 0.25mm, but runout of outside welded Flange was 0.02mm, so I will definitely need to turn that inner 40 taper cone. I think I will deel with bearing prelode later, i want to see how spindle runs first.

[hanermo]

Keep it coming, this is a great thread !!!

BEST PARTS:
ACTUAL MILLING !!
IN STEEL !!!
HEAVY AND RIGID, for ONCE !!

10/10..