Needs Pictures: 0
Picture(s) thanks: 0
Results 76 to 90 of 188
Thread: HM52 CNC Conversion
-
8th May 2010, 11:40 PM #76Dave J Guest
Just to add to the last post, it looks like your ways on the saddle are ground with frosting applied to hold oil.
The oil ways on mine didn't come out the side as can be seen in the photo below, I just found they were so close to the edge that the oil would come out there before anywhere else.
I found a oil way in the ram saddle not drilled at all (2nd photo) and enough crap in the knee ball oilers to stop anything getting through (3rd photo). The last 2 photo's are of some of the other castings I cleaned up.
Dave
-
9th May 2010, 10:47 PM #77Senior Member
- Join Date
- Sep 2007
- Location
- Newcastle NSW
- Age
- 77
- Posts
- 155
Dave,
That looks too pretty to get dirty. If you need to make something you're welcome to come around and use my machine.
Russell
-
10th May 2010, 12:19 AM #78Dave J Guest
Thanks Russell, no need for that, I have other mills but I use this one mostly.
The photo's are of the inside and were done last year when I cleaned it up. I get a lot of enjoyment making and cleaning things up to look factory or better.
I have attached a few pictures of inside the knee, it was the worst casting to do, but came up a lot better than it was.
Also here is a link to a crane I built, 3/4 of the cost to build it was for the $15 stickers to make it look factory.
http://www.woodworkforums.com/f65/ho...photos-113771/
With this mill my main concern was casting sand. I found some on the knee ways and some on the X axis that had started to scratch the ways, also found sand in the head swivel gears etc, so I am glad I went though it. All up I got a bit over a half of a 4ltr ice cream container full of sand out of it.
On the way through I fixed up other things like the X axis nut to low and Y axis nut to high, power feed out of alignment, paint on mating surfaces, etc. I also split the nuts while in there to reduce backlash for now.
All the castings were done in preparation to CNC it like Ray is doing. It gives me a clean base machine to start with. The last thing I want is casting sand in the expensive ball screws that I have bought.
What part of Newcastle are you in, I am out at Holmesville.
Hi Ray,
I also used a air needle scaler to clean up the castings, the type thats like an air chisel with approx 20-30 needles hanging out the end, loud but worked great.
Regards Dave
-
10th May 2010, 03:54 PM #79
Hi Dave, Rod,
Thanks for those pictures Dave, they make for a very handy reference. I'm thinking I'll do something similar, it would make cleaning easier, and it's much better to work on a clean machine.
After much head scratching, we finally got accurate measurements of the ballnut mounting point. After mucking around with various dial gauge setups, we finally got consistent measurements using stacked parallels and feeler gauges. Sometimes the simplest ways are the best... Everything will be referenced from the ground surface of the dovetail slide, and the axis of the leadscrew center at the right hand, because we are keeping the existing dowelled mounting plate on that end, which will be the free end of the ballscrew.
For the other end, we are going to try using the machine as a digitizer, and remove the endplate we want to measure, mount it in the vise, and using a center finder with the DRO to digitize the XY locations of all the holes. Then use those co-ordinates to drill the servo mounting plate. (after replacing the endplate of course)
Meantime, I finished wiring the servo bench test setup, we are closing the loop at the PC running EMC2, the Yaskawa drives are running in velocity mode and have their own PID loop, so tuning is a bit tricky with stacked PID loops. But we finally got it tuned to reasonable values. We had a lot of trouble getting the scaling right, the factory default encoder resolution that was programmed in the drives, didn't match the actual encoder. From what I could tell the encoder output had never been used on these drives. It looked like it had been used in step/dir mode.
Here is the obligatory youtube video...
[ame="http://www.youtube.com/watch?v=y4LohuMTdfo"]YouTube - HM52 CNC conversion. Part 1[/ame]
The EMC2 5i23.ini and hm2-servo.hal files are attached for anyone interested.
Regards
Ray
-
10th May 2010, 05:08 PM #80Dave J Guest
Hi Ray,
That electronic talk went right over my head, lol. All right for you electrical engineers but I will stick to the simple dc servos I think. I understand there workings.
With the ball nuts, if space is tight 4 bolts would be more than enough to hold them. I have seen people including Chich grind/machine the outside of the nuts to fit better.
I am pretty sure I have the matching taper drill for the taper dowel pins if you need to borrow it.
I think it is worth the effort cleaning it and only needs to be done once. My advice would be to strip the machine down to the column and base, then clean everything as you put it back together. Doing this will familiarize you with the mill and it's workings. You can also replace the cooking fat grease to something of better quality in both spindles while your there.
Dave
-
10th May 2010, 07:19 PM #81
Hi Dave,
Yes, DC servo's would be a good way to go, the downside is you need a large DC supply, the advantage of using the Yaskawa AC servo's is they run off 240V AC directly.
I know a lot of people use steppers, which is probably the easiest system to get going. There is a good summary of the advantages and disadvantages of steppers vs servos here... Geckodrive Frequently Asked Questions (scroll down a bit)
What were you thinking of for your control system?
Regards
Ray
-
10th May 2010, 08:15 PM #82Dave J Guest
Hi Ray,
I will be using Mach3 for the control and the DC servo's from Kellings.
I have read most of the stuff on steppers Vs servo's and am going with servo's.
Dave
-
10th May 2010, 09:31 PM #83
Agreed, I like the simplicity of steppers but for speed power and accuracy you can't beat servos. My first 3 axis stepper system was a guitar making machine over 20 years ago, and while it worked fine, (still going as far as I know) I wish I had used servos. It would have been a much faster quieter machine. We also made carpet sampling machines for many years that were multi-axis steppers. The guy's who are making those machines nowadays are using mostly servos. Welcome to Modra Technology | Modra Technology I haven't been involved in that stuff for many years, but it's good to know they are still going strong.
I take it you are going with something like this Page Title should make for a nice fast accurate system.
Regards
Ray
-
10th May 2010, 09:49 PM #84Dave J Guest
Hi Ray,
These are the motors I am going with
1: KL34-180-90 (NEMA 34) Dual Shaft, Front Shaft
1/2" Rear shaft 1/4" 1125 oz-in Peak 90V/40A, $149
Special sale:$139/pcs
http://www.kelinginc.net/KL34-180-90.pdf
As for encoders and drivers I will decide when I come a bit closer fund wise. CNC is moving fast, so there may be new better things available in 12 months or more. From what I have read it will be cheaper to build a power supply here than to have it shipped from overseas.
Have you given any thought to pulling your machine down and going through it. If your going to do it, now would be a good time, incase you find a warranty issue that needs addressing.
Dave
-
10th May 2010, 11:10 PM #85Senior Member
- Join Date
- Sep 2007
- Location
- Newcastle NSW
- Age
- 77
- Posts
- 155
Hello Dave,
I had a look at your jib crane. Great minds must think alike, I have a mini version-its even the same colour, for moving the indexing head on and off the mill table. The indexer lives on the "jib crane" when not in use. I raise and lower the indexer via an eccentric shaft through the chuck/indexer bore (because the table cannot be raised or lowered) Come and have a look.
I live in Rankin Park. PM on its way.
Russell
-
11th May 2010, 09:14 AM #86Most Valued Member
- Join Date
- Jul 2006
- Location
- Athelstone, SA 5076
- Posts
- 4,255
Its all swahili to me as well Dave...i watched the vid and had no idea what he was talking about
gee you guys are ingenius.....like fixing the backlash, pulling the mill apart and getting it right smack on etc etc....I'm reluctant to pull my mill apart just in case it doesnt go back together properly (and cos what Im likely to find ...or not find as the case may be )...but maybe one day... I'll be able to say "I Did It!"..."yahoo"
-
11th May 2010, 09:27 AM #87
Hi Dave,
Yes, I keep looking at it, wondering how I can do exactly that, there are some pretty heavy bits of metal in there that will take a bit of planning to handle. I don't have that magnificient jib crane to work with.
Finalized some more parts of the design, the toothed belt and timing gear selection was more complex than I thought it would be. I am going to go with a Power Grip GT3 belt and a 4:1 reduction ratio with 15 teeth on the servo and 60 teeth on the ballscrew. The servo drive shafts are keyed, and I'm going to see if I can get taper lock bushing.
Regards
Ray
-
11th May 2010, 02:49 PM #88Dave J Guest
Hi Ray,
I didn't have the jib crane up when I pulled the mill down, I bought a 2 ton engine crane to do it and a few other jobs. It makes taking it apart and putting it back together easy.
It would be worth buying if you are going to pull it down, you could always resell it after woods if you had no use for it. You would only loose $50-$100 as they hold there value, and it's cheaper than renting one.
Dave
-
12th May 2010, 06:40 PM #89
The ballnut mounting for the X-axis is now finished.
Squaring up and facing the stock, the finish from the indexable face mill is nice.
Boring the 40mm hole for the ballnut.
The finished part, with a bit of chamfering to clear the bottom of the table dovetail.
The ballnut mounting holes locations were transfer punched using the ballnut as the template
and then tapped for M6, the holes for mounting to the cross slide are tapped M8 and the holes
in the crossslide will be drilled oversize to allow for alignment.
Only design issue I can see at this stage is the grease nipple needs a little more clearance.
Next task is to finalize the design and making of the servo mounting plate.
Then machining of the ballscrew ends will be after that.
Regards
Ray
-
13th May 2010, 01:20 AM #90Dave J Guest
That came out good Ray. Are you machining a square recess into the top of the saddle for this to sit in?
Dave
Similar Threads
-
Hafco Hm52 on the fritz again
By Retromilling in forum METALWORK GENERALReplies: 25Last Post: 12th Aug 2009, 11:14 AM -
CNC conversion
By sonic_racing in forum METALWORK GENERALReplies: 7Last Post: 18th Jan 2009, 03:11 PM