This is the manifold on the side of the crosslide that delivers air and oil to lubricate the crosslide and operate the tool turret. Caked in 35 years of cutting oil and swarf. To give you a general idea of how much cleaning is required.
Z Axis ballscrew support bearings. All of the bearing nuts have been animalised over the years and are now obsolete and unavailable from Hardinge. They will all have to me made.
Cross contamination of oil and grease – not good.
Dovetails look to be fine.
General filth underneath the carriage. No obvious gouges or signs of major were despite the carriage lube line not being connected.
There’s more woodwork involved that you might image when rebuilding a CNC lathe.X and Z Ballscrews were packed up in purpose made boxes to go away and be checked, cleaned and re-balled.
I have nothing but praise for 1A Ballscrew Repairs Limited in the way that they dealt with the whole process. Totally efficiency and the bill was far less than I imagined it would be.
Thank you Dave.
Formally EMC2, Now “rebranded” as Linux CNC
Early days ..... The plan is / was to use Heidenhain LS405 linear encoders with a 10 micron grating pitch for closed loop feedback on both axis and an angle encoder on the spindle. Signals would be interpolated by a factor yet to be finalised but the interpolator chip can do anything from 8 to 8000+ times interpolation. I hope to get to sub micron positioning / jogging. I purchased a MESA 5i25 and 7i77 and installed ver 2.5 on an old PC. I managed to get the servo drives operational, with feedback from the encoders that are built into the servos. The Heidenhain encoders can also be read through IChaus interpolation IC / development board. I also have some other Heidenhain rotary / angle encoders connected and being read for the MPG and spindle.
I had planned on trying to use the spindle for C axis work on smaller jobs like the conrod bolts. It seems I’ve drawn a blank from the Linux CNC community on how this might be achieved.
I haven’t decided yet whether to persevere with LinuxCNC and its unfathomable configuration files or sell one of my organs or limbs on the black market to fund the purchase of a modern industrial control. It’s not entirely clear which of the two options involves less pain!
MESA 5i25 and 7i77 with ICHaus iC-NQC interpolator
ICHaus IC NQC software for set up and diagnostics of the interpolator chip eprom.
Encoder scale mounting on the side of the crosslide
Mounting for X axis scanning unit to carriage
Drilling the carriage for the mounting holes for the X encoder mount The encoder mount also serves as the cover for the cushion compartment.
X Axis encoder test fit
X Axis encoder test fit
Most of the bearing nuts had been so badly knarled, they could not be removed even with the proper tool. They will all have to be remade as they are now obsolete and not supplied by Hardinge.
New bearing nuts starting out with the centres roughed / drilled out in a ½ inch plate.
First side completed. This nut has an oil seal that sits in the centre.
Finding the centre of the Hardinge 5C Pneumatic collet block
Turned out that the lathe came with a 16C to 5C Hardinge spindle adaptor and quite a lot of 5C stuff including an expanding 5C collet. This was ideal for holding the bearing nuts by their centre bore and on the already machined face so that the outer contour and top face could be cut followed by threadmilling.
After finding the centre of the collet block the soft pad of the expanding collet was machined to the same diameter of bore of the nut +0.4mm
Nut on expanding collet, faced and contoured prior to thread milling
As the only time I’m likely to use a 18 TPI, 20 TPI and 16 TPI threadmill is on these bearing nuts, it makes for expensive nuts!
New nut in Bearing support block for the left hand side of Z axis.
Face pin drivers also had to be made for each size of nut.
New hex nut for the pulley end of the Z axis ballscrew.
I think it took nearly a month of weekends to make all the nuts and drivers. The drivers all share a 20mm hex spigot that is doweled and bolted into each driver. So much for making engine parts!
Things have been arriving. Hardinge were able to supply new way wipers for the bed / carriage but the wipers for the carriage are now obsolete. Googling “VAJ A1” revealed that VAJ A1 is a standard wiper profile manufactured by:
They were supplied as two 500mm lengths that will just need drilling and cutting to length. These were about 13 quid each versus the 70 quid each for the two bits of plastic supplied by Hardinge for the carriage. I love a bargain!
Next to tackle was the Servo mount for the X axis. Every HXL conversion you look at seem to come up against the problem that the shafts on the original DC motors were a lot longer than those on newer motors. This opened up a whole can of worms. The shafts on the Omron servos have only about 25mm of usable 14mm keyed shaft instead of the 45-50mm on the ½ inch pinned shafts on the original motors.
This meant a complete re-think on the way the motors were mounted. I wanted to do the minimum amount of machining / modification possible to the casting. In the end I elected to turn the motor though 180 degrees so the motor is pointing out of the back of the casting rather than sitting at the rear of the casting under the cross slide.
X Axis Servo mount block
Trial fit in casting before machining the inside of the casting to create flat surface for the block to sit on.
Plate for servo
Complete unit ready to fit. Belt tensioning is achieved by moving the entire unit in the original bolt slots in the casting.
Plenty of scope for a disaster – hence lack of pics!
So far, unbelievably, the pulleys have been one of the biggest problems. I didn’t want to change the gear ratio but it seems that its almost impossible to source a 15 tooth, steel, flanged pulley for use with a 3/4 inch belt. Its even more impossible to find one with a 14mm bore and a 5mm keyway. I spent weeks looking at this and talking to suppliers. In the end I managed to get my hands on a set of used pulleys. They were in nearly new condition compared to the ones that came off my machine. The next problem was how to bore them and that complete with the boss, they were 40mm long and too long for the space / shaft available.
Some 6082 bar was bored until the pulleys were an interference fit and then driven lightly into the bores. The pulleys were then bored to 14mm to fit the shafts on the servos. The boss that was either side of the flange has now been machined away so give a pulley with an overall length of about 28mm I’m assured that loctite will be fine to secure the pulleys to the servo shafts!
Its all a bit tight in there ...... but it’s lovely when all those evenings of measuring and drawing pay off
Belt, both pulleys, ballscrew etc ..... all seems well.
Belt roughly tensioned and motor driving the ballscrew under power ..... some movement at last.
Z axis servo mount, taking shape.
Z axis servo mount, taking shape.
Mounting block and plate, test fit on casting (partially bead blasted to remove paint)
Z Motor mount
More Z Motor mount
Z Motor mount - parts
One thing that had been worrying me for a while was how I was going to get the carriage back on the bed without damaging the Teflon coating. Unless it went on level, at the right height and square, there is a good chance of knackering something. More woodwork to the rescue. This Heath Robinson contraption enabled me to jack the carriage to the exact height and level it in relation to the bed. In the end it glided effortlessly off the well oiled clean bin bag on the top, straight onto the bed
Contact surfaces under the carriage showing oil ways
A little shove and .... voila Well worth the effort to make the “transfer platform”
Klubering and fitting new bearings to the X axis ballscrew support. I decided to fit new bearings all round for the ballscrews.
One thing i found, it pays to shop around. I sent out an email late one afternoon to 5-6 bearing suppliers. The varying costs that came back were staggering. I had been on the lookout for them on EBay for months, but they are rare and even on EBay fetching a fortune and predominantly located in the US
In the end, Bearing King came up trumps for price and as it turns out, for service as well.
Thoroughly recommended and if I end up replacing the spindle bearings, I suspect I know who will be supplying them.
I’d like to say a big thanks to Mark at Bearing King.
Posh new wipers
Z axis encoder.
I haven’t done anything about mounting the scale yet, but the scanning unit mount is complete. The will be a bar underneath the encoder with dogs to engage the limit switch mounted on the bottom of the scanning unit mount.