Making-A-Taper-Gauge

[awemawson]

I've decided that to make progress on the A2-6 to A2-5 adaptor for the Beaver TC20 lathe I need to make a male taper plug gauge to fit the recess in the existing chuck, so that I can test the taper that I create in the adaptor 'as I go'

The plan of attack is to make a taper gauge slightly 'longer in the taper' (really fatter as it's a disk) than the finished article. Then once the existing chuck is removed carefully measure how much fatter it is, and reduce the disk thickness on the surface grinder to 'dead nuts on'.

So I needed a 15 mm or so blank about 110 mm in diameter. Irritatingly I already had a 12 mm thick one that had been destined to be a gear wheel, but 3 mm too thin. The next (and only!) thick plate that I had left was a section of very rough and rusty 20 mm 'road plate', the sort of thing they put over holes to allow traffic still to run.

So major clean up with wire brushes and grinders, followed by marking out, and slicing close to size on the band saw.

That's as far as I can go today as we have three of the grand children and their parent's for the first 'post lockdown' outside meal round the fire pit.

But next step is to turn it circular on the manual lathe, approximate the taper again on the lathe, then mount it between centres on the J&S 1300 EUIR cylindrical grinder and try and remember how to set precise angles on it's swivelling table. (I've not used this beast for literally several years, but I steamed it up last night and it was smooth as silk !)

More to follow - obviously !

[nrml]

Is the final finish on the back plate going to be ground or left as a turned finish?

[awemawson]

Not sure yet - if the turned finish is 'OK' then I'll press on with other things, but if I need to cover my tracks I'll grind it !

(I'm hoping to be able to grind both the male and female tapers though. (I have enough internal grinding spindles for the J&S 1300 to sink a battleship !

[awemawson]

Main task today was to clean up that 'almost circle' by first tickling both sides on the surface grinder before turning it actually circular on the manual lathe. but . . Slow start this morning for several reasons:

- Late arising due to exhaustion from three grandchildren yesterday

- Had to fit the replacement extraction vacuum motor for the big disk sander - replacement arrived yesterday (Bank Holiday so don't knock Hermes !)

- Searched high and low for my box of slip stones to no avail - they've disappeared !

Although the surface grinder magnetic chuck is nicely ground flat and true, I like to gently pass a fine slip stone over it before use to ensure no burrs have been raised since last use - humph, need to order some more !

So pressing on I decided to let the spindle warm up to equilibrium (as you always should with precision grinding) while I checked hydraulic feeds and coolant flow.

. . .  no coolant flow ! - it uses a 'weir style' coolant settling system which is excellent for letting the very fine particles settle out of suspension, but it does (by design) have a very large surface area. This needless to say results in evaporation over time of the water content of the soluble oil coolant. In the end I had to add 6 gallons before the pump head once more was under the surface and delivering coolant.

. . . so no actual grinding yet, hopefully this afternoon but now other duties call.

[awemawson]

I eventually got back into the workshop and managed the first rough grinding of one side of the source plate to give me a flat surface when I mount it on the milling machine to bore the central hole for mounting between centres and subsequent turning operations.

Finish isn't fantastic with a bit of herringboning but I was taking it off at 2 thou a pass - final pass 1 thou, total removed 10 thou

Video still  uploading but it will be here :

[awemawson]

Not much done today as it's Easter Sunday and apparently I have other things to do 

But I did manage to centre up the plate that I've ground flat on the underside, select the taper arbor that I want to use, and drill out the central hole to 21 mm. This gives lots of room for boring to size.

The selected arbor is the type that remains parallel on the bearing surface as the internal taper expands the sleeve. At 1/3 travel it opens to about 24 mm and at mid travel is about 24.5 mm, so I will aim to bore out the hole using a boring head tomorrow to about 24.25 mm .

[awemawson]

I got set up to bore the hole for the turning and grinding mandrel this morning.

I'd forgotten how satisfying boring precision holes is when it all goes smoothly - which for a change it did. Producing nice curly swarf and a hole that admitted the expanding mandrel just as far as I wanted.

To celebrate I cleaned up the second side of the plate, this time having to remove 11 thou to get though the surface muck and bullets.

[NormanV]

Surface grinding is a mystery to me. I once made an iron casting for a 12"x10" surface plate. I took it to a local engineering college and they surface ground it for me. I was puzzled by the fact that they could grind across the 10" width with no appreciable wear on the grinding wheel. How is that?

[awemawson]

As you feed across the work, the theory goes that, the leading edge of the wheel wears (very slightly), but the remainder of the width of the wheel remains at the diameter to which it was dressed so the ground surface becomes flat (or as flat as the grinder!)

Remember that feeds are exceedingly small on finish passes, measured in tenths. I was fairly gulping that plate in 2 thou bites as this isn't the finished thickness of the part yet - I just wanted it flat enough to get a bore true to the surface.

I started turning the plate into a true disk on the manual lathe. Firstly pressing the arbor firmly home using the arbor press, then placing of a 'soft centre' turned in place in the 3 jaw to avoid changing chucks etc.

This wasn't satisfactory, as the intermittent cut was pounding the soft centre and slowly putting it out of true.

So I mounted up a hard centre in the spindle bushing, and used the catch plate. A process rather delayed by seeking the catch plate -WHY was it in the cupboard with all the Bridgeport bits, NOT the cupboard for the Colchester bits  Oh well it only took an hour to find it !

Haven't finished turning it yet as I've been called away for canine supervisory duties 

[awemawson]

I managed to get back out to the workshop for half an hour and have at last finished the intermittent cut phase of the operation. It's now a true cylinder of 110.8 mm diameter.

End target is 106.358 mm so a generous 4 mm to play with - I will approximate to the 7 degree 7 minute 30 second angle on the top slide and taper the side to that but leaving it well over size for final tapering on the cylindrical grinder where the angle can be set far more accurately (I hope! )

Still at least I've got to where I'd hoped for today

[millwright]

I enjoyed watching the surface grinder, it seems to have a hypnotic effect, spent many a happy hour on the Jones and Shipman over the years

John

[David Jupp]

Andrew,

In the past I've seen 'Go, No-go' gauges for 'Grayloc' hubs - the hubs have female tapers (rather like the back of your chuck).

The interesting point is that whilst for smaller sizes the gauge is round, as the sizes get larger the gauges are made from flat plate having curved / tapered ends.  Imagine the plate as a 'thickened diameter' across the hub.  Presumably the plate is mounted in a 4 jaw chuck to turn/grind  the conical faces on its ends.

This approach both saves material, and keeps the larger gauges to a manageable weight to handle easily.

[awemawson]

David I can well see why they'd want to do that - things get unwieldy rapidly as they get bigger !

As I had the 20 mm plate my plan is to fettle the taper to fit the rear of the chuck when I eventually take it off, leaving the taper long and concentrating on getting the correct engagement of the smaller end. This of course is critical as the O ring seals for the hydraulic actuation need to be properly squashed but not too much.

Once that's achieved I can measure how much of the big end protrudes with a depth gauge, and remove a small crescent to that depth as a tell tale for the gauges final use making a recess in the adaptor.

I spent this morning chasing my tail and going down the rabbit hole setting the lathe for the correct taper - it's dead easy at a theoretical level - applying SOHCAHTOA tell me that the sine of my angle times the travel that I can achieve on my top slide gives me the offset. Tap over with a dial gauge and Bob's your uncle. When I came to actually doing it I found that my tailstock wasn't dead nuts 'on' and had to dig out my setting bar and do a few adjustments.

I might actually get something done this afternoon !

[awemawson]

So I did indeed rough out the taper on the lathe this afternoon using the top slide.

I emphasise that this is roughing out, although I can set a pretty accurate angle using trig on the lathe top slide it is at least an order of magnitude worse than I can set on the J&S 1300 cylindrical grinder as after all this is what it's intended for.

My top slide gives me a comfortable travel of 90 mm, which by applying trig gives an in feed at the smaller end of 11.1627 mm. Now 90 mm of top slide travel is 45 turns of the knob. This rapidly becomes a pain - I did three (or maybe four) iterations and got the displacement down to 11.171 mm. Perfectly adequate for a roughing cut - this is just to reduce the amount to grind off. Fairly arbitrarily I just let the taper take half the width of the disk .

Once mounted on the cylindrical grinder I'll take the diameter down to the major dimension of the taper (106.358) before putting on the taper.

The J&S has a table intentionally designed to swivel in a precise manner and is equipped with a racking screw to make the setting rather more precise than whacking the top slide with a soft mallet !

(I borrowed the measuring equipment dedicated to the cylindrical grinder !)

[awemawson]

I did actually skim off a bit more from the disk this morning - still left a generous millimetre for the grinding, talking of which, time to re-commission the J&S 1300 EIUR cylindrical grinder:

Poor thing has been neglected for ages, infact I'm not sure that I've used it in earnest since I moved here - I probably have as there was still coolant in it which remarkably was in good condition. I had sprayed it with Shell Ensis preservative years ago before moving, and the table was still thick with it. A bit of elbow grease and a rag soaked in white spirit and it cleaned up OK. I obviously had missed the swivlling table clamps and tailstock release handle as they showed a bit of corrosion, but that was soon cleaned up.

So far I've proved the hydraulics are OK, and that the controls all work - the fine infeed dial was stuck but that was just dried up oil. The coolant pump works, and the coolant clarifier seems to work. This is the type where the used coolant discharges over the back of the machine through a weir falling onto a paper roll filter that sits on a mesh conveyor belt. As the filter draws particles from the coolant it gets heavier, weighs down the conveyor tripping a microswitch that advances the roll dumping the contaminated bit into a bin. I'd tried poking the switch to prove that it worked (it did) but I've left it pumping coolant round the system as I type this, and it's just tripped of it's own accord 

I've yet to check the alignment of the powered swivelling headstock (MT3 test bar job) and the swivelling table (parallel bar between centres job) but I now have other duties to perform, first lambs having been born this morning.

[awemawson]

Not been able to get a lot done over the past few days due to other commitments, but I HAVE managed to set up the 1300 and do a test grind for parallel.

Just a bit of 32 mm hot rolled bar from the scrap bin, faced and centred at both ends, and set up in the cylindrical grinder to test for parallel grinding. First go produced a difference end to end of 5 thou over the seven and a half inches of the bar.

Now many years ago I produced a crib sheet for removing tapers in the grinding as it gets quite tricky working out what needs rotating and how much. The 'top table' of the grinder swivels on the travelling table, the pivot being in the centre. By knowing how much the part tapers and over what length, and which end is bigger my crib sheet tells me how much to move the end of the table. In this case the 5 thou taper 'should' be eliminated by 10 thou movement which I of course then applied. This has reduced the taper to five tenths of a thou over seven and a half inches. A bit of persistence would get rid of that but it's quite near enough for what I'm doing.

I was pleased with the ground finish considering the grotty but of steel that it was !

May get a bit more time tomorrow.

[awemawson]

Of course I couldn't leave it at that - I had to get that last bit of non-parallelism out. A couple more iterations and it was gone to the limit of my measuring capability 

However while doing it there were ominous growlings from the work head - a bearing not sounding too good. I eventually tracked it down to the front motor bearing on the work head -  the grease had dried out. However getting at it involved rotating the entire work head on it's swivelling base thus loosing my parallel setting 

Now the base is graduated and the graduations are on a good diameter so you can get it back close to where it was but never exactly the same place! OK another round of grinding that bar and I eventually got it back spot on where we were a couple of hours ago. The reason for the dried up grease - one of the grease nipples was blocked and not actually passing grease to the bearing - nipples both changed and the bearing seems no worse for wear.

Now years back I had made a big test bar to aid setting this machine up - just a 3-1/2" bar accurately centred and ground, I've even made a nice wooden box for it, but when I recently got it out of the cupboard I noticed a couple of small spots of corrosion despite having been sprayed with preservative.

So as I type this it is having a thou taken off back to a fair finish (I hope!)

A machine like this grinder has too many degrees of freedom. The work head can swivel, the grinding head can swivel and raise and lower, and of course the main upper table can swivel - all these movements make setting up great fun, and also as you slide the tailstock up and down the ways it will be pointing ever so slightly in a different direction ! Even clamping the upper table results in a very small angular movement - it's tiny but when trying to split tenths of a thou tiny counts !

[WeldingRod]

How are you driving that test bar?

Sent from my SM-G892A using Tapatalk

[awemawson]

There is a 1/4" Whitworth tapped hole in the driving end with a cap screw in it that engages a tab on the rotating bit. (Tab held in a left hand threaded 3/8" Whit bolt that had me puzzled until I remembered it's handedness (original J&S feature)

[awemawson]

Wanting to press on with this gauge today I dug out my metric slip gauges and their accessory set, and made up a 'snap gauge' 5 thou larger than the maximum limit on the fat end of the chuck taper.

This being 106.371 mm I assembled slips to the tune of 106.498 leaving 5 thou for tweakng when I was happy with the measurement method. Sorry about the mixed units, but the grinder is Imperially calibrated.

Once there and happy with the process I dug out an Imperial micrometer of suitable range, re-adjusted the snap gauge to finished size, and also leaving projections to use for zero setting the micrometer zeroed it, so it is only being used as a 'distance to go' meter not absolute measurement.

The parallel section of the gauge was then quite quickly brought to finished size, and this being the upper limit will leave me some wiggle room when finally I remove the chuck and use it's recess as a standard to blue it to.

I've layout blued the gauge, so that when grinding the taper I can see it progressing across the width of the gauge to give me an estimate of depth of the taper. This will be ultimately adjusted to the chuck recess when removed (remember that crane !)

[awemawson]

So having totally failed to be able to accurately measure the J&S cylindrical Grinder table swing radius I thought I'd 'just go for it' using the table angle calibrations and check progress with micrometer blue and adjust accordingly. Now I have removed the chuck I have it's female taper as a 'standard'.

Setting the table to 14.25 degrees the graduations are rather coarse, however the first tentative 'bluing up' showed reasonable (but not perfect) contact (Don't be confused by the layout blue that the parallel section of the gauge is covered in - this is just so that I can see the grinding as it happens) I've never set the table over so far all the time I've had this machine!

One iteration of angle tweaking later and the contact was pretty good so I went on to remove 25 thou of diameter in stages bluing for contact and checking engagement depth at each increment. By recording progress hopefully I can accurately predict an endpoint when the taper is the right length and the smaller face has just come into contact with the inner face of the chuck recess.

Had to stop at this stage due to other commitments but will resume play later !

[awemawson]

I finally got back to finish grinding the taper gauge after many distractions and interruptions.

Continuing to plot progress in a crude chart gave me a bit of confidence that I wouldn't overshoot, and in the end I left a calculated 90 microns between the two faces that squash the O rings, while still giving full taper engagement. As a check I used progressively thinner shims to prove engagement depth, a surprisingly delicate test - you can adjust things so the taper just 'floats' on the shim with no trace of looseness.

I just need to put it on the surface grinder and put a flat on the upper edge as a tell tale for full engagement for the final part.

[awemawson]

And finally - I finished grinding the 'tell tale step' at correct insertion depth this afternoon so hopefully that's the gauge finished.

The grinding set up was more stable than it looks, but I wanted to leave the expanding mandrel in place just in case I need to give it another lick on the taper face. Although the expanding mandrel will re-locate pretty accurately in the bore we are talking fractions of a gnats tender bits so it's asking a bit much to get it back bang on.

Hopefully the next step is to actually use it in earnest !