Hello Guest it is March 28, 2024, 10:29:50 PM

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - hutchison

Pages: 1 2 3 4 »
1
Show"N"Tell ( Your Machines) / Re: emco pc5 lathe
« on: March 15, 2014, 05:52:23 PM »
i havent found any dc-dc regulator that will cope with 45Vdc +  so i think i'll swap out the ring core to a 30vac out, this should bring the supply into the drives to arund 41- 42vdc, instead of 49.  i am not comfortable with unregulated supply, still hunting for something to regulate 42vdc to perhaps 40.

2
Show"N"Tell ( Your Machines) / Re: emco pc5 lathe
« on: March 13, 2014, 07:43:34 PM »
i stored this machine for 3  years, along with the PC, (hello again) and recently just pulled it out of storage.  Funny thing is nothing worked properly, mach was crashing, spindle not keeping speed.
good news is thanks to forums and members posting i seem to have gotten through some of the initial problems.

First problem, spindle index pulse reading upto 1500rpm then counting half ! after that, so i changed the index debounce to 0.  where as before it was 400/100 .  i changed it to 150/0  in the debounce bounces and it reads well now at any speed.  i also have spindle speed averaging checked, the 's true'  fluctuates a few rpm like 4 or so i think its o.k, its a low inertia drive at high speed so its never going to be digit perfect

Second, actual   spindle speed:  i've spent many attempts trying to cal the spindle, but it always left a dip about 1600-1700 rpm.
i realised that dependant on the number you put into the spindle pulley max, that dip either got worse or better on screen
I cant say how important it is to know the actual rpm of the motor from the decal plate, its been a  few years and i just thought it was 2940,  well, of course all the readings were way off, duh, the motor max was actually 2790rpm. face palm time.   so i tried all kinds of settings 2940 right down to 2000, with different velocities.

after finally putting in the correct resulting speed into pulley max (2511rpm@100% mach out)
i got a fairly linear calibration, with a small dip , heres a shot:



after the cal completed i tried commanding speeds throughout the rev range to see what happened, it was about 20rpm+ plus out on each 200 interval, so here's what i did, i went into the spindle motor tuning and dialled back the spindle velocity a little, about 25 units at a time,  and then rechecked the commanded speeds.  i took about 50 units of the total (it was 1500 before).    and now its fairly close and linear through out the rev range within 5 rpms at bottom end, upto about 2000 its still within 5-10,  get to 2500 its 50rpm out.
 its been a week of late nights, it can make dog meat of you easy!



some settings are here if it helps anyone with a similar set up:
for reference, my index pulse disc is 4" diameter and the slot is 0.460" wide, mounted on the back of the spindle. i'm using a c3 pulse board and sensor.










next problem i have is on the main power supply, i  am too close to 50vdc on my stepper drives , and they  trip / protect themselves regularly when other 1-2kw items switch on the same power supply line (heater)

I appreciate some help here, my PS is as follows;
 240ac into ring core transformer which gives 35ac out, rectifies into dc thru bridge rectifier GBPC15005
I have a 63Vdc 22000uF capacitor before the drives. both drives are attatched to cap.

what is the best way to regulate the drives?








3
Show"N"Tell ( Your Machines) / Re: emco pc5 lathe
« on: September 27, 2010, 06:57:29 PM »
ta, still feels a bit quirky turning the dials and cutting but i wouldn't be without it now! it gives the machine another dimension, good for the odd small manual op.

4
Show"N"Tell ( Your Machines) / Re: emco pc5 lathe
« on: September 26, 2010, 06:52:13 PM »
may aswell pop in a little update while i am here, i made a jog pendant a while back bbbbbbgvxc using a bidirectional LPT2 card, set the bios to 2 way, and a simple parallel interface = 13 I/0's, i can use it semi- manually now or i have some basic auto controls like spindle start/stop feed hold etc,  and spindle speed + /-  and a rotary switch from incremental to constant velocity on the mpeg modes.
some machined ally and formed perspex. yet to detail legend.





5
Show"N"Tell ( Your Machines) / Re: emco pc5 lathe
« on: September 26, 2010, 06:33:18 PM »
its a mill50, no it is not converted but a few people have done so, there is little difference between the 50 and 55 hardware wise, drop
binski a line on cnczone, or try mcgyver on cnczone.nl, he did a conversion with 4th axis also.

6
Show"N"Tell ( Your Machines) / Re: emco pc5 lathe
« on: October 04, 2009, 01:33:01 PM »
cheers Tweakie,  a tin basher friend folded out the gaurd for me from some 18gauge, It's one of them jobs if you attempt it will make a fool of you every time, best left to the pro's!.

i do look forward to making chips, if i ever get the time!,  it just needs a bit of bringing up to date .

7
Show"N"Tell ( Your Machines) / Re: emco pc5 lathe
« on: October 04, 2009, 11:20:36 AM »
The lathe was running well but i had a few accidents with the bed stops whereby when jogging off the limit switch i pushed the wrong button (home!!) after a software reset and homed again into the deadstop!.. crunch...  it Damaged my 'Z' ballscrew, so i stripped it off to take a look.   Now here is the old emco ballscrew, not really impressed I could bend it with my hands easy, but hey it’s only a litte training machine so you can pass on the plastic nut bracket i guess.   There is nothing wrong with an 8mm (5/16) ballscrew but at 508mm long it is definitely a little too flexible.



so i decided to make a steel ballscrew nut bracket, and fit a new ballscrew, the space under the carriage is so tight! i found the design a little testing to fit in a good device to hold the nut in firm without using to difficult a technique and still leaving room for the new screw guard / cover to fit over the top of it.   I came up with a two piece nut which was blocked up on the mill, squared up on the grinder,  then screwcut.



 I wanted to have a firm hold of the ballscrew nut but not too firm that would distort its precision by clamping.    The idea is you simply screw the ballnut into the nut clamp and then bolt on the other half of the bracket so it nips it in place and stops it unscrewing, the clamping dimension is just minus, so it literally may compress it by  1-5 microns.  23,993 - 23,998 as the ballnut is 23,999.  I ground that size on the grinder after leaving some meat on the fitting face.



i decided to go for a 12mm diameter with a 4mm lead, a 16 x 5 I couldn’t make it fit in the constricting space.   A 12 x 5 ballscrew would have fitted also but because of the greater lead it would increase the helix angle and reduce the torque transferred, although it would have increased my rapid speed.  So, a 12 x 5 ballscrew gave me a helix angle of  7.62 degrees, a 12 x 4 comes in at 6.09 degrees.  The original thread of 8 x 2.5 gives 5.71 degrees helix.   12 x 4 was a pretty good compromise,  and I calculated if I just put that in, I should increase my rapid speeds to 1.8m/min  from the existing 1.25m/min max.  



 I had to space the ballscrew off on the X plane by 1.5mm away from the bed casting so the square portion of the ball nut bracket cleared the guard (since it was larger diameter!).   to do this I retained the original brackets and modified them.  I basically got some 1/16th gauge plate and made a spacer for the front supporting nut, and for the motor holding bracket at the back I elongated the original fixing holes by 1.7mm.



The existing ballscrew used two tiny ball single row radial bearings to support the rear of the ballscrew on a 6mm diameter (yes 6mm!)..  so out the window that went and in the new ballscrew design I made that diameter 10mm and searched for a suitable bearing.  I found a nice double row radial with some angular contact ability which  could cope with increased forces radially and also increased thrust forces in both axial directions.  Although not quite an angular contact thrust bearing it would be a much better replacement than the original single row radials.  To mount the bearing again the space was tight but I managed to figure in the 26mm bearing o.d into the original rear mount bracket by boring it out to 25.99 (0.01 for the manufacturer recommended preload! To reduce the axial play to within 10-15 microns) so I bored it out and a tiny bit deeper and made a clamping washer like the original, pressed it in, put the washer over the top applied a holding force to it and peened the edges of the mount block over the bearing using a punch, its just to retain the bearing from moving axially that’s all. Once peened I licked it over on the grinder to remove the high spots.


 I made a few hardened thrust spacers for the ballscrew, one for In front of the double row bearing and one for behind it. The one infront sits up against the ballscrew thread form and provides a contact point between it and the ballbearing inner and the spacer behind the bearing provides contact between the bearing inner race again and the timing pulley which actually drives the ballscrew, so nip up the drive pulley and it loads the ballscrew against the bearing,  then it was locked off with a fine pitch grub screw.   I also increased the threaded diameter on the new ballscrew, and had to re-thread the timing pulley with a new fine pitched thread etc. I had to relieve the bacl end of the bed casting a little to clear the hardened thrust spacer diameter, good old grinder came out to play.





Everything fits as planned, and today I fitted up my new guard over the assembly, its so close if I paint the guard the paint will probably scratch off when the machine traverses, so I’m just going to leave it as it is.  







The ballscrew bracket is one of two, the other is going to another retrofitted pc5 in holland.   the ballscrew itself was precision rolled to c3 tolerance to the size I wanted.  I checked it and I now have an error of 0.02mm over 250mm  and an axial play (backlash you might say) of less than ten microns 0.01, this error (running clearence!)  is from the double row radial bearing (the ballscrew itself is preloaded),  if I increased the loading to maybe 15 microns on the major bearing diameter I would have reduced this axial play, but the manufacturer rec’d  0.01 so I just did what I was told!.      It’s worked out well for me really I’m happy with those dimensional results,   also Instead of a predicted 1.8m/min and probably because of the brand new screw it runs upto 2.1m/min easy!, I guess it has a good friction co-ef right now.  I cant run it any faster because I am only using 2.2nm motors @ 49Vdc and they stall out over that speed, really..  2m/min is well fast enough for this wee machine! I will only ever make tiny parts on it.



 click here for a movie....

http://www.youtube.com/watch?v=1uW3F-qbZNk

The next job for me is to build a small remote pendant box type thing so I can use it properly, it is labour-some using keys to jog set tools and job.


8
General Mach Discussion / Re: leadscrew pitch error
« on: July 02, 2009, 06:18:30 PM »
ok guys thanks alot for the help, i understand by changing the steps per unit , if it was a linear error , would be the perfect solution and a correct one.  hood i have changed all the electronics and motors for new kit.

I did how-ever find the problem this morning,...   i got in the shop at 8, 21 degrees allready , perfect i guess except i knew i was going to cook later!.   And proceeded to  chart all the points of the screw, i thought i would use standards and slips to give me a accurate map etc.   I started out with a 200mm standard  just to see if it was the same as yesterday ,   200.13 it showed..  hmm thats different allready i thought.  , i went straight down to a 100mm = 100.13 ..    thats not looking good i thought i was hoping for half of the error not equal, I had now suspected either i had a completely shot screw on one half or something else not to do with the standards or the lead screw was happening.  I then tried a 20mm slip , guess what..  20.13 (or as near as etc ),  now that really got my goat!!, i wandered around for a while scratching ,  i then tried a 1.3mm slip..  once again 1.43mm displayed in the dro, at that point I was suspecting the error in my screw was not pitch error.

i removed the last 1.3mm slip and jogged the axis upto home, the dro displayed  bang on zero give or take a few microns.   how about that (laughs!) so..  i was excited as i knew i had a good screw  ;D  ..  but mystified why the error went all the way back upto home..   and started looking at my home switch set-up, which is rock solid mounted, and i did use some branded micro switches (saia burgess etc) so trying to limit the can of worms!.   after a while I figured out  why i was getting a bum reading..   well...  the burgess micro switch had a domed front actuator,  and me in my infinte mechanical wisdom also put on a domed face to my rod stop..  nowt wrong with that i thought it would minimise the chance of a chip settling on the stop right....  So the reason i had this funny error..  the two domes were not dead inline, infact one was nearly two mm off centre line to the other,  thus..  you can set zero and repeat zero all day long on a diagonal, put a slip gauge or setting rod between the two, you add that initial error of diagonal contact to the gauge length on the horizontal.  its not rocket science is it! its one of those things that gets you when you think youve come along way, and trip up and fall on your face  ;D

my trouble was solved by swiflty removing the rod stop and facing it flat.  

cheers lads for the help, im going to go and get back in the box   :)


9
Show"N"Tell ( Your Machines) / emco pc5 lathe
« on: July 02, 2009, 04:18:02 AM »
I picked up this emco pc5 lathe a few years back, its my first refit of a lathe, i've been on manual machines for over a decade, not long enough apparently as im still slaving.  I wanted the lathe to learn a bit about code,  it is greener on the other side isn't it?  :D


   heres a picture of it inside before i finished wiring, and the outside, I also knocked up a quick cabinet so i could control both machines from the same interface, its no good having scores of pc's and screens laying around so i split the duty of the control for with a data switch, it would be nice to have machining centre that does both instead!
 
the lathe now has 2.2nm 4.3Amp motors, origonals were 0.5nm. i knocked up some hand dials for the double ended motors, engraved / divided them out and added handles, why didnt no-one tell me the ballscrews are R/H on the emco, pah!, i went to all that trouble and made the dials to make it a bit more useable .. turned the dials as you would....the slides went the wrong way!!  8)  you just dont think of these things.

 it now rapids at 1.25m/min, origonals were 550mm/min. I put in some limit switches on all the axis, made the stops adjustable incase i use different tooling.   and a decent angular race on the spindle nose with a preloading mech instead of the standard deep groove bearing. I made a ground collet chuck, and made a set of matched ground tool holders.  i put in a 550w 3ph motor and modified mount plate, the next size up frame to the origonal and nearly double the standard motor torque, also a belt and timing pulley drive further increasing the torque by a 1.1:1 ratio,  it uses a vfd so i got some more torque available on overload!. its got enough beans i think  ;).    It runs upto 2750rpm standard, i could overspeed it to over 3k but theres not much point really unless i'm drilling tiny holes.

it has all the extra electronic bits that usually go in the cabinet, what more can i say, i just wanted to share it, its just a wee benchtop.

 i'd like to thank arturo from cnc4pc.com for helping me since i started building, and all who support mach, ive probably read your posts at some point or other when up the creek. robin of holland aswell who helped me thru with the elctronic side.  i'm quite happy with my little lathe, it aint a gildemeister! but it's got a place in my workshop, cheers.













10
General Mach Discussion / leadscrew pitch error
« on: July 01, 2009, 05:03:00 PM »
hi all, I recently finished rebuilding an emco5 pc lathe.
I am now working through the configuration of the software and testing the axis and motors / tuning etc.  The traverse I have is 50mm (2") on the X and 300 on the Z (12", for all those yankees  8)

I have a pitch error on my lead screw (the Z axis). I gauged it over 200mm (8") .  I gauged by moving the leadscrew in one direction only and using a 'setting standard' between the carriage and limit switch on the Z, i.e; a  200mm mic setting piece.

I know the mic setting piece is good as i cross referenced it with a couple of different mics, hot day aswell! 27 degrees C' in my workshop, (i started to sweat  >:( )     but that didnt affect the setting piece or mics, theoretically they may have all three items  expanded by the same amount, but I doubt it they all read 200 within 0.01.     The limit switch is also very repeatable to within 0.01mm  ( i tested it over and over with homing (50% rapids), manual jogging 0.01 steps)  i'll also check the error early in the morning when its cool, just to be a fanatic  ;D

I have 0.17mm pitch error over 200mm,  I used to get that same pitch error from 600mm travel off an old 'well used' bridgeport mill, go bridgeport!.

I am using a 2.5mm pitched 8mm dia ballscrew (o.e.m, tiny aint it) , geared with a 2.5:1 timing pulley ratio, and 800 step/rev config in quarter step mode.     it gives me a resolution of 1.2micron/step and a max rapid of 1.25m/min before the motors bum out.

i'm never going to get micron accuracy over 200mm, but 0.05/0.1mm would be more better at least!.

I got two choices i think, first is to put in a new quality ballscew and nut, and new bearings either end of the screw and hope it works as the box says,  or maybe configure mach to account for the pitch error,  i have been reading a few posts about screw mapping but have not yet figured out how to carry out this physically

all I have to go on is a setting standard and blame the lead screw  :)  easy to blame something,

if i could just account for that error i could live with it ,  anyone got any help?, thanks!, adam.



Pages: 1 2 3 4 »