Oh Blimey I bought a CNC Lathe !!!!

[awemawson]

Not quite so easy in a confined space your Johnness 

However I've done an experiment with a short length of 20 mm thick walled plastic conduit and a heat gun, and 'ironed it flat' - it flattens to pretty well exactly the correct width leaving a 2mm gap due to rebound, and slicing out one side with an angle grinder with a 1 mm disk makes a perfectly presentable imitation of the original capping 

Whether I can do a 750 mm length remains to be seen

[awemawson]

So I donned the marigolds and got down to some serious cleaning in the electrical tray this morning.

Firstly I tried the technique on a small 'bodge board' that is attached to the PLC:

[awemawson]

OK that seemed to work 

The 'method' was liberal application of methylated  spirits with a paint brush, accompanied by vigerous scrubbing with the brush, and drying off with a compressed air line. Seems to work so lets attack the rest of the electrical tray:

[tekfab]

OK that seemed to work 

The 'method' was liberal application of methylated  spirits with a paint brush,

Jeesh ! what a waste of good drinking meths !   

Mike

[awemawson]

A fair amount of pushing rags under wires was needed to get the base properly clean - it's still not perfect but a major improvement.

Now to attack the air shelf. Seems to contain an air manifold and air pressure switch, but its all so badly clagged up its hard to tell.

Same technique produced this:

[awemawson]

Well I reckon that was a success ! Can now tell what appeared to be just pipe terminations on the spool valves are actually sintered bronze exhaust noise suppressors. Also there are LABELS 

I might even be able to find my way around all this plumbing if things are labelled 

So what next: Well lets see if we can make up some replacement capping for the stuff missing from the electrical tray:

[awemawson]

  Yes that seemed to work - we're on a roll 

There was no way I could keep 750mm of the plastic conduit hot enough all along its full length to successfully squash it, So I thought I'd try just squeezing it cold in the vice a bit at a time. I expected it to crack - but no it went quite well. Curved like a banana, but straightened up ok by hand after I'd cut out the necessary strip from the back with the angle grinder.

Now need to leave it to completely dry off as meths contains quite a bit of water. Also need to make up the 24v DC supply to go in that empty place next to the transformer.

[awemawson]

I spent a little time last night identifying the thirteen wires left disconnected from where the Power Supply is missing. The two red ones are obvious as they emanate from the 21v o/p of the transformer having passed through a fuse. Two more are obvious as they connect to the 0v and 24v inputs to the PLC. Another two are obvious as they connect to the 0v and 24v of all the inductive proximity sensors.

That leaves a further seven wire unaccounted for. Oddly each one goes to the driven side of a relay coil, and also to the relevant PLC output that is driving it. I can only assume that the original power supply must have incorporated a snubber diode for each of the relay drives. If this is the case then the various air solenoid coils which are also driven by the PLC outputs don't have snubbers!

Anyway as the fellow with the original psu isn't answering emails I thought I'd make one anyway. I bent up a little chassis, picked up the existing holes that mounted the original one, mounted the bridge rectifier and reservoir capacitor, and put a 'din rail' on to take din terminals. Although I have some 4mm ones already they are a bit bulky, so I have some 2.5mm ones on order that hopefully will come tomorrow.

Have a few pictures:

[awemawson]

The 2.5 mm DIN terminals arrived today, so I've assembled the replacement power supply. A little bit of 'solder sculpture' to wire up the seven clamping diodes for the seven relays, but all went to plan. I only had to extend one of the original wires to fit my version of the power supply, and this was only as I wanted the clamping diodes to appear in the same sequence as they are physically laid out in the chassis. (It makes thing so much easier later on in years to come when things fail!)

Once the power supply was installed I disconnected the 230v AC feed to the transformer, and powered it up from a local mains feed. All checked out OK. The PLC seems to be running its program, and responds as I, for instance, bring a lump of metal up to the various proximity sensors.  Can't test it in earnest until I've disabled the various interlocks that would have come from the lathe.

The air plumbing is (slightly) curious. Basically there are two double acting cylinders. One raises and lowers the alloy extrusion that holds the current bar that is being fed. The other advances or retracts the pusher 'lance' that feeds the bar into the lathe. The 'oddity' is that there are three air pipes coming out of the back where the regulators and filters were before the vendor vandalised them. I 'think' that the air enters a filter, regulator, oiler unit, and goes to a pair of spool valves controlling the bar advance / retract. And that a feed is taken from the retract side, via a second regulator, to feed a second pair of spool valves that operate the 'Up/Down' of the extrusion. This would make sense in that it effectively prevents the extrusion moving unless the pusher is fully retracted. There is also a pressure switch on the retract feed telling the PLC that there is or isn't pressure on that side.

Anyway that is the way I'm plumbing it up for testing, though I've had to order another regulator / filter / oiler unit - I had a brand new one but it's a massive 3/4" BSP size whereas all the stuff originally seems to have been 1/4" BSP

Hopefully when that is all fitted I can start testing in earnest.

Have some holiday snaps:

[awemawson]

After an enforced delay due to being invaded by hoards of grand children over the last two weeks at last a bit of progress. Firstly a very nice man emailed me a pneumatic diagram for the bar loader, which shows my initial concept was wrong. Turns out that the third regulator exhausts to atmosphere as an 'over pressure' safety device in case the machine pushes back!

This has allowed me to plumb up the pneumatics

[awemawson]

Having overcome the 'Emergency Stop' link that originally would have come from the lathe I was able to try things out. Turns out that the 'shoe' that pushes the bar runs in an extrusion, and some kind fellow over the years has given it a whack, closing up the channel that the shoe runs in at the far end of its travel. The result was that it jammed and was useless!

I experimented tapping a suitably sized (2" x 0.25") flat steel bar down the extrusion and sure enough could open it enough for the shoe to slide freely, but it would spring back when the bar was removed. Solution: leave a bit of the bar in there, strategically placed so it opens up the channel but doesn't foul the shoe - jobs a good 'un!

Now I can manually operate the machine to pick up a bar in the channel, go through the motions of pushing it into the feeder tube in the lathe spindle, then advance the 'pusher' to feed stock through an opened chuck or collet up to a bar stop.

Whether I'll ever use it is highly debatable 

[awemawson]

One thing I did manage to get done during the onslaught by grandchildren, was to put the lathe cabinet doors back on. This had a rather amusing (in retrospect) knock on result.

I was sitting at the pc with the lathe on in the back ground, and suddenly the noise level went down dramatically. My first thought was something like a servo amp had failed and tripped a breaker. Heart in mouth moment !

Turns out, now the doors are back on the chiller unit is only cooling the cabinet and not the whole of my workshop, and had turned off as it's thermostat said it was cool enough. ...phew !!!!!!

[awemawson]

I came across a rather nice 'Foundry Bench' on ebay a couple of weeks ago made by Emir out of solid beech.  I already have an identical one and know how well made they are. Turned out no one else was interested and I won it for a snip. Seller is a hobbyist machine dealer I've known for over 30 years. What's this got to do with a CNC lathe you may ask. Patience - all will be revealed 

[awemawson]

So I went up to collect it recently with the big Ifor Williams trailer. Looking at his racks of Myford and Colchester chucks and already having a Kitagawa B206 hydraulic chuck on my wants list I posed the question, knowing that this sort of item is way off where he would usually be dealing. No never heard of 'em came the response but we do have this rather nice unusual chuck showing me a box.

[awemawson]

Turns out it is interchangeable with the Kitagawa one I had in mind, was available for a silly price AND came with a large box of top jaws.

All seem to be brand new unused and the chuck is in its original box. I came back a happy bunny.

Now I need to obtain a converter flange from an A6 spindle nose to an A5 chuck. Basically a disk with the appropriate holes in but rather accurately made and balanced.

[Pete.]

That's nice!

[awemawson]

Thanks Pete - it is isn't it !

Today's job: put the bar feeder back together. Simple cleaning and refitting job, sorting out a few stripped threads and here you go - one finished bar feeder. I think I will probably mothball it rather than use it.

[awemawson]

Back to the main lathe after the diversion of the bar feeder!

Today's task: find out why I cannot unclamp the hydraulic collet chuck under program control. Firstly manually there is a two station foot pedal (clamp / unclamp) and some buttons on the control panel to point this at the main or opposite spindles. All works well and does what it should. The pedals are ultimately digital inputs routed via interlocking relays. The hydraulic valves are driven by more relays driven by digital outputs. There is also feedback confirming clamped and unclamped states.

Now by program there are two simple 'M codes' M10 clamps the spindle, M11 unclamps the spindle (or rather didn't !) Much of the relay logic and digital input and outputs are common to both manual and program operation. My main suspects were the five relays involved in the interlocking chain, and sure enough two had rather high contact resistances. Out they came, dismantled contacts cleaned, bench tested and returned to the machine. No change

The other three relays are plug in jobs so it was simple to swap them around and eliminate them. Issue hinges on one digital output not being driven, BUT this same digital output is used when manually opening the collet and works 

It HAS to be a software issue. Quick email to my mole at Traub points me to an obscure 'PLC logical switch' page of set ups on the controller with one clearly labelled 'Enable M11 always' 

Turns out the M11 function is disabled if you have a bar feeder attached . . . . argh

Anyway logical toggle switch duly clicked on the screen and off we go both M10 'close collet' and M11 'open collet' now do what you'd expect. Ah well I can think of worse ways of spending a Bank Holiday

Needed the M11 function as I have ordered a 'bar puller', which is a three fingered claw that sits in a station in the tool turret, and grasps the end of the bar. You then open the collet (M11) move the tool station away from the headstock as far as you need to make the next widget, then close the collet (M10) and pull the claw off the bar.

Have a picture of a bar puller:

[lordedmond]

well done it always pays to RFM but often the info is well hidden in the dark recesses of the setups


Stuart
good day to spend a BH
yes we have had a good day today on the BH  26 Aug  46 years wed

[vtsteam]

Good to see you back on the lathe, Andrew. Looking forward to your first production run of widgets! 

[awemawson]

Stuart, sadly this is not documented. Traub keep these things very close to their chest - it's like getting blood from a stone. I'm absolutely sure that they have internal documentation that would have helped hugely over this resurrection, but they only seem to want to respond to specific points

Congrats on the wedding anniversary - we're coming up to 40 this December

Steve, good to see you back - good holiday?

[lordedmond]

thanks for the Congrats   as with you Lathe keep at it and hang in there for the long term

[vtsteam]

Steve, good to see you back - good holiday?

Yes sir. Went in a boat in an underground lake in a cavern, good friends, etc. fun for 7 year old daughter. 38 hours of driving for me  . Good to be back, too, though.

[awemawson]

So to fit that rather nice chuck, I need a flange to convert from its 140 mm recess to "Din55026 A6" which is what the main spindle of the machine is.

Options were a/ Make one from scratch b/ Buy an A6 backplate and machine it to fit or C/ Buy the correct one from Schunk

To get the precision necessary I drew back from a scratch built one, and the only A6 backplates I could source were cast iron not steel, and the supplier couldn't quote their maximum speed.

So I went (the expensive) option C and got one sent from Schunk in Germany - duly delivered by UPS this luchtime

[awemawson]

Phew - it fits the chuck at least ! Just as well as it cost me somewhat more than the brand new chuck and about 15 sets of brand new jaws 

Now the next issue is that power chuck use a pulling motion on the hollow hydraulic cylinder that operates them, whereas the collet chuck currently fitted uses a pushing motion. 

No problem - there is a reversing valve on the hydraulic manifold specifically for this situation. Ah - a drawback - that also reverses the direction of the 'opposing spindle' hydraulic cylinder.

Quick bit of nagging to my Traub mole reveals that there is another valve on the opposing spindle to reverse the direction back to original - well I think that's what he's saying - sent me reams of info and a broken English explanation (inifinately better than my smattering of German)