What did you do Today 2018

I just got X-axis homing working on my S7 CNC conversion (only 8 years after doing it). Needed a precision switch for sensing a suitable X axis position, eventually decided to use the contact height sensor I made for setting tool height on the CNC mill. This sticks down with a strong magnet, so attached to a suitable vertical surface on the saddle, with an adjustable stud in a bracket on the cross slide that contacts it at the designated X home position. There's a built-in function in Mach 3 that drives an axis towards home until an input level from a switch tells it to stop.

Once I'd made the bracket and sorted out some earthing problems, setting up Mach 3 proved easy and it seems to work very well - 3 successive home commands driving the axis to the same indicated position on the DRO to 4 decimal places (though obviously the machine isn't accurate to 0.1 micron!), which at least indicates that the software loop is very accurate. Next to measure the mechanical repeatability.

This should allow me to calibrate a set of tools and populate the tool table, so I can turn accurately to diameter with any tool only having to home the machine at the start of each session rather than adjust the datum for each tool - that's the theory anyway! Also I guess I'll need to make another sensor...

I'll try and post a couple of photos later.

Most limit switches on pukka machines are "just" microswitches, allbeit decent quaity ones. Possibly a quicker solution - and just as repeatable.

Murray

Completed the swing arms for my Vertical cross engine... see workshop progress thread...

George.

Not sure about quicker. I've got some precision microswitches embedded in a Euchner limit switch. They are very nice and have a repeatability of 20 microns, but rather delicate. The Euchner device has a very clever plunger system to make sure the switches themselves don't get smashed if the limit isn't sensed properly, but is too big to easily fit on the lathe. I considered using one of the switches but would need to make a housing as well as a protective actuator. And I'd have needed a similar bracket and actuating rod. So in the end it was easier using something I already have, which is reasonably protected against overtravel.

The silver disc in the middle is a tungsten carbide plate spring-loaded against the underside of the top, which has a couple of mm travel before the spring compresses. Not visible is the 4mm socket for a wander plug, but you can see the plug in the side. On the bottom the steel base has a strong magnet embedded to stick the whole thing down (or on the vertical face of the saddle in the case of the lathe).

That's a mighty serious looking camera! Any details for AP nerds?

I saw radios in the early 70s with the licence gummed inside the case (as a boy I used to help out in the shop and one of the jobs was fitting batteries to customer's radios). I suspect they were bought with the radio and never renewed

Neil

I bought 3 Omron precision switches from that auction site. I beat the poor chap down on the price because I was the only interested party. They do North, South, East and West with 1 um repeatability. I realised that what I actually needed was up down so I had to add an extra lever. Then I realised I would have to remake the entire mill controller and it has all sat on the shelf ever since

j

Neil, I recall, from the same era, that you had to have a separate licence for a car radio. I refused to buy one, because we had a licence to cover radios at home, but my mum was horrified that I was breaking the law, and she bought one for me. Until the need for a licence was removed, I got a renewal reminder every year, so had to keep buying them. My late grandfather had the first radio in my home town of Spalding, Lincs, and my mum could remember listening to the BBC's predecessor, Station 2LO. Listeners had to wear earphones, but several people could listen together if the earphones were put in the bottom of a large pudding basin to reflect the sound.

Finished a cap/top bearing/gland for the trim valve on the Polish tank engine:-

Inside and outside threads were total bastard sizes, O/d random in either Metric or Imperial, but both a perfect fit to a 14 TPI thread gauge. Only way to do them was successive cut & try. For the outside thread that meant taking it out of the chuck, because the valve body casting - which was the only practical gauge - weighs 40 or 50 lb; but if you mark the jaw positions on the holding diameter, it'll go back properly pitched-in for the next cut.

The rotation shaft of the valve is a bit bent, so I needed to put a bit of clearance in the 'ole. I thought about trying to straighten it, but the rotor (1 3/4" diameter) has such a big aperture in it to let the water through that I dursn't try to true it up wiv an 'ammer. I think the packing in the gland will probably seal it anyway. Should see the engine puffing up and down the line in a week or two, I'm told.

I see you can buy Metrol switches online. Down to 0.5 um if you seriously believe you have a machine that precise. The low cost 5um ones start at about £30.

Lots of Youtube warriors like to claim they "work to tenths (of a thou)" ie 2.5um but studiously avoid actually loading their machines when making any kind of visible measurement. My Shizuoka CNC machine claimed 10um work precision when brand new but that must have required meticulous adjustments of the myriad gibs and ballscrews and a following wind. The original NSK ballscrew factory test reports showed initial accuracy of +/-1-2um, which is pretty impressive. Achieving anywhere near 1um final tolerances requires temperature controlled rooms and machines and eye watering capital expense.

Murray

Edited By john carruthers on 14/02/2018 18:52:56

Edited By john carruthers on 14/02/2018 18:54:35

I have a Roland, wax chopping, desk top mill that I have been hacking. It has microswitches for the limit switches and a dinky co-ordinate display on the Z head.

I have a home command to bounce it off the limit switches and set X, Y and Z. Repeated homing starts off okay for the first few times then it starts to creep.

I expected a bit of plus minus but not a creep. Seriously weird but never more than one click difference from one home to the next on a 5um step.

Have now measured the homing repeatability for my system. Best DTI in mag base holder on lathe bed, homed X axis, placed plunger agains end of X-slide to get convenient reading, then moved slide by a varying distance and for each move re-homed. Photos below show DTI face for 4 separate tries.

All well within the same 10 micron window, and probably within 5, so I'm quite pleased. Should do the job OK I think. Now to puzzle out how Mach 3 tool offsets work...

Bought a meter of 5mm od brass tubing, plus a meter of 10mm aluminium rod for stock shelf, not cheap

​George.

George Are you sure you didn't buy a metre of each ?

Roy

John,

Standard practice on machine tools etc is to fit an industrial version of one of these, http://uk.farnell.com/omron-electronic-components/d3v-16-1c5/microswitch-v3-pin-plunger/dp/8774579?mckv=HXNyRh3W_dc|pcrid|78108377589|&gross_price=true&CATCI=pla-131288271909&CAAGID=14989732149&CMP=KNC-GUK-GEN-SHOPPING-OMRON_ELECTRONIC_COMPONENTS&CAGPSPN=pla&gclid=CjwKCAiAn5rUBRA3EiwAUCWb213hek9Rw9KxL1RpZCpmdbYFTvH2VDBWBPPBxnt41c4BUE5tM8vuIRoCnL8QAvD_BwE&CAWELAID=120173390001364401

it sits at the end of your axis in a fixed position and is operated by a linear cam with a rising edge, I assume your positioning is by encoder, the axis is driven to the home position until the switch is operated and stops the axis and the encoder position is recorded, this then becomes your zero position. Do you have an over run switch in case it misses the home position to shut the drive down.

Martin P

.

What I posted on 06-Feb. was intended as a cautionary tale, rather than a request for advice; but in the spirit of 'completion' ... here are some pictures:

.

The first two are just to illustrate that it is a substantial little beast.

The third shows that, as I had surmised, the jam was caused by the thin plastic getting between the cutters.

On disassembly, I had to lever the two cutter units apart ... What you see is the contents of the 'overlap' area, where the cutting takes place.

MichaelG.

Not today but two or three days ago - it being a cold and wet afternoon I decided to dismantle some ancient computer hard drives ready for submission to the scrap metal man.

Depending on the particular model, some of these have black synthetic rubber sleeves round steel pillars. One of the drives in this latest batch had such a buffer sleeve but when I tried to remove it I discovered that it had changed with age to a sticky black glutinous consistency that abandoned its sleeve-like shape the moment it was touched. The resulting messy fingers needed copious applications of meths/IPA/surgical spirit/vodka with kitchen roll and I still had to lose most of the clippable part of one thumb nail!!!!!

I post this as a warning to other would-be HDD recyclers. Has anyone else encountered this pitfall? This type of deterioration isn't universal, some sleeves still behave like rubber. I wonder if it is a cause of hard drive failure? The drives I was dismantling had all their gaskets and seals intact until the moment they came 'under the knife' .

While I'm posting this, I'm surprised that there doesn't seem to be any recycling channel for the Neodymium/Iron/Boron magnets. Comments welcome.

Best regards,

Swarf, Mostly!

Martin, those microswitches appear to have no spec for repeatability, they may be fine as limit switches but not for precision homing. These is no encoder, like most smaller CNC systems I'm using stepper motors. The switch only gets activated when homing, during which the axis is moving very slowly. If it doesn't stop it will hit the end of travel but that doesn't have disastrous consequences, it just means you lose datum and have to set up again. The switch itself is designed so that it can accept a few mm over travel and shouldn't be damaged even if it bottoms. When the switch is activated it tells the Mach3 controller which is doing a homing operation and the controller then stops the axis movement and remembers the position as "home". I did consider using precision micro switches but the ones I have are only spec'd at 20 micron repeatability and would need careful mounting and protection.

John,

That's why I suggested Industrial types, I couldn't find the ones we used for work, why would the switch bottom are you passing over the switch or pushing on the switch.

Martin P