Member postings for mgj

Thanks for that - I have an Argoshield light bottle from BOC. I have had it for about 9 years, and its still well full. 9years at £120 a year....

I reckon I can get one of these little bottles, weld for 5 years and still show a retun on 1 years rental. That Argoshield in the BOC cylinder must be worth its weight in gold!!

Agreed Graham.

But,- if you have to quench, one drops it in vertically to prevent bowing and keep it moving to prevent assymetric cooling..

Personally I'd use mild and case harden, for a couple of reasons.

1. Free cutting mild is a lot easier to cut than gauge plate

and while its not ideal for volume hardening,

2. The surface will still be much harder than tempered gauge plate and

3. Less prone to wearing the die by having a better surface finish

3. Stronger by being less prone to crack propagation

4. Its a lot easier to make a new die block after a few years, than a new expansion link.

For me I'm afraid its a bit like making driving pins out of silver steel. Sound like a good idea but metallurgically a little less than sound.

Edited By mgj on 11/08/2012 22:26:02

Clive - 4mm rad in 4mm pipe is seriously tight. That will be an entertainment. I think you'll need the extra leverage of a bender to do that nicely.

There is a sort of ideal (typical) minimum dia/bend rad ratio. I have forgotten what it is, and someone may be able to help with that - but its not 1:1!!!!

Any way you can get away with it a bit wider than 1:1?

I have a tube bender - Hemingway. Very good, but over complex for the task- but quite excellent in the slightly larger sizes of tube.

If i'm not telling my granny, the operational guts of the thing is a groove, just over 1/2 th dia of thee tube to be bent. That stops it going oval and collapsing. So a lot of my smaller tubes I bend by using a ball nose cutter on a rod of the required bend radius, of about 75% dia depth, and bend the annealed tube around that.

3/16 tube I like the sliding shoe of the bender, but less than that a thumb and a groove does as well!

I agree Kwil. You put me onto the X5 which is just the stuff in a tin. There is no doubt about it it - the cariage slides more easily when lubricated with Ultraglide, rather than SAE30. Its quite noticeable.

If the valve has a wing type plunger, ie an area which sees steam in the boiler and a pop type ring which see steam escaped and provides a larger area, then probalby the area of the ring is too big. ie it lifts at a pressure, and then the increased area keeps it open to a much lower pressure.

The adjustment procedure is to get the valve to lift at the correct pressure- adjust spring compression/preload or rate until that is right. (using a gauge and the air compressor one will scrounge access to) At this stage don't worry about closing.

Once it lifts at the right pressure, then adjust the ring or the area between ring and wall, untill it closes at the right pressure.

I'm just doing exactly this for Metre Maid which has standard ball type safeties, and they buzz and fizz, so its getting new safeties of the pop type.

Google Ross type safeties.

Gets a lot easier with a sharp , preferably 4 facet sharpened drill especialy for the pilot.

If you haven't got access to that, try buying a split point drill for the pilot - once you have a straight hole, the full size will follow it.

2" deep x 8mm. Wouldn't bother with a pilot on a mill - centre drill, lots of coolant and go with a 4 facet/split point. Just don't let the swarf build up (and jam the flutes) - that should keep even an ordinary drill pretty straight.

Why the split point/4 facet geometry - because it has a point on the tip rather than a chisel edge which tends to stop the drill from wandering.

Edited By mgj on 18/07/2012 22:09:40

Are we sure we are not getting confused between repeatability/concenetricity of the chuck, and paralellism/leveling/lack of taper in the bed.

Surely, if you turn a bar in a lathe, even if it is held eccentrically, the result will be a dead true piece, radially, even if it is axially tapered. If then there is a detectable error that is not due to surface imperfections, there is likely to be a problem with bearings, looseness in slides, chuck gripping etc.

If the piece is truly round, but but tapered, then the dumb bell test (leveling, twist - call it what you will) hasn't been done properly, or there is some sort of fault in that direction.

But the two are not the same, and in general terms don't cross reference.

The whole drift of this thread has not in fact been about lathe accuracy, but chuck accuracy, which again, are not the same thing at all. For one to think that a 3 jaw is OK at .001 runout....... They are only built to about .003. You can get an ordinary 3 jaw which meets then normal DIN standard for a very reasonable price. A top grade 3 chuck guaranteed to hold to .001 or less is about £4-500 plus vat in our sizes, and going up, and it will stil only hold to that on one nominated master pinion. Unless one is lucky and you happen on a good one.

.001" when new is darned good for a slef centring 3 jaw. No one ever bought an SC three jaw for accuracy did they? Only for convnienece?

I think I might remember the jaw/pinion number that produced that. A standard SC chuck in good nick by Pratt & Burnerd will only hold concentric to about 3 thou.

A Super Precision SC 3 jaw from Pratt will be under a thou one one jaw only, but with both sets of jaws. It will however vary pinion to pinion with diameter (according to the bit of paper it comes with), so to get the best out of it one needs those test results handy.

So one thou with an ordinary 3 jaw is pretty good. Might also be wise not to abuse it, thopugh up to a point the accuracy of the chuck is not relevant. If something has to be genuinely concentric than one will set up in a 4 jaw, or if using a 3 jaw, you'd turn to size Personally I'd never put something in either a 3 jaw, or even collets and assume it was true without clocking it. A Griptru is handy though. Its a bit quicker to adjust than a 4 jaw.. So on reflection, my super precion 6" for the big lathe was, in hindsight a waste of the extra dosh - it does have hardened slides, sure, but for amateurs like us thats hadly a consideration.

Jon you are right about hammer forging. I believe Mannlicher leave the rather pronounced spiral on - possibly as a trademark. Watching machine gun barrels beign hammer forged at Enfield, again that took what, less than a minute.

Clive one of the advantages of decent quality black powder is its consistency. Its actually a more consistent propellant than smokeless, and its a lot easier ot set it alight consistently. As a consequence a lot of black powder ammo is very consistent indeed. The problem comes with the business of spin - the generally longer bullets are highly spun for stability, and because of the lower velocities the launch angle or line of departure is realativley steep. Combine htat with being heavily overspun for stability downrange and hte thing locks gyroscopically at a rather awkward angle upward angle relative to the trajectory which is tedious. Tends to fly nose up generating slender body lift as a result of the crossflow, and all sorts. So its not the ancient propellant at all which is darned good - its actually the physics of older lower velocity ammunition whihc tended to limit performance.

Not so sure - a lot of time time the rake can be built into the tool holder itself. No the rake angles are not redundant

I use negative rake tips a lot, and they are not designed for use in cast iron, though they are good in that, and excellent in copper and bronze/brass. Even better is the way they send the chips away from you!

Actually one can get tips with all sorts of rakes on them, though its true that the tips most suitable for the smaller lathe are positive rake. (though I have some excellent small Stellram non ISO boring bars in negative rake) However the CCMT tips are expensive because you only get 2 edges, (plus 2) for your money, whereas the negative rake W series give you 6.

For the amateur is it well worth using negative rake tips for parting off.

Quite so, because even in perfectly still air, and on a non spinning earth, the bullet/projectile will still curve in flight(if spun) because the laws of physics say it must, because of cross flow velocity, precession/nutation and the magnus effect. So any single zero point can only refer precisely to a single point in the trajectory. To remove such erros you really need a calibrated graticule, which tank sights have, and you can see the curvature compensation increasing with range The curvature in flight is not very pronounced with small arms.

I don't think there's any doubt that one would normally machine from the breech, mainly because you want any roughness in the cuts to go with the proj, as well as the grain structure. Ordnance barrels are commonly forged rather than drawn of course. But you still need to face off the muzzle etc. So it is to be expected that it might be put into a machine muzzle first, at some stage in the proceedings -even small arms which may be threaded for silencers and flash eliminators.

Droop. I'm not sure about the artillery. In those days one ranged anyway, so that sort of error would be taken into acount, though I'd have thought you'd have needed to offest, since most missions would have been one gun adjusting. You may have found that jump in the slides counteracted the droop! Still they would have known their business, and there must have been limits beyond which one would have to compensate, and inside which you didn't. That was the American 155 presumably.

Tanks of course are the "snipers" of the ordnance world, and are looking for a first round hit on a target no bigger than a metre square at a mile and half or so. Once you had boresighted and shot the gun in at 1100 metres, you set the muzzle reference sight, and the computer could measure the difference between the zero and the current position of the muzzle with respect to the sight and correct accordingly.

Clive, it has all changed from Pas day. Now the gun is gps located. The FOO is gps located, the traget is laser ranged for a fire mission. You know from GPS survey, altitude of both gun and target, and your barometrics are well enough known. So very often they don't range but will set TE and QE off.(Set your spirit levels for those who haven't done it - you offset the levels and then level the bubble to set range and line to target, or you can do it automaticaly off  laser ring gyros) A battery of self propelled 175s will now burst fire a fire mission, loading on the autoloaders/flick rammer. The individual gun dispersion is fed in for each gun, and the whole lot arrives without warning with a calculated spread built in.

Actually its quite wrong to suppose that a barrel needs to be straight for accuracy. All that matters is that it does not move. It needs to be moderately straight in general terms so that the proj goes down cleanly, but beyond that the line of departure is fixed by the line of the muzzle, and so long as that is fixed with respect to the sights all will be well.

The problem really comes when you have something flexible like a ships hull between director/acquisition radar and turret (now we fit strain gauges to measure twist and deflection in the hull and compensate) or in a tank barrel which will cool assymetrically and is very long for its dia. Hence the calibrated mirror and laser projectors to measure where the barrel is pointing with respect to the sight line.

Why would you machine from breech or muzzle. The barrel blank will very often just be a tube. You still have to machine the chamber, and the taper end of the rifling. You also have to finish to length and profile. Length is particularly important for accuracy because length controls shot exit timing, and hence where on the vibration pattern the shot exits. You need for consistency to time it so the barrel is moving slowest on the theoreical sine wave. Profile affects the undamped natural frequency of vibration, so thats important too.

Edited By mgj on 30/06/2012 12:34:46

Sorry.!

Clive you are right about modern manufacturing. A lot of modern barrels are stellite (or more modern equivalent) lined against abrasion and heat. They are using a very thin bore in carbon fibre I think too.

Still I think a lot of match barrels are still cut in the old way, and then polished. In the early days with hexagonal rifling - did they cut that, or form the hex and then hot twist, before final finishing of the outer profile? Modern SAA barrels twist at about 1:8-1:10. How they did it with those long slim bullets and low velocities which must have needed twist rates in the 1:5 mark at a guess, or did they just accept a tumble point at 7-800 yards. What was a Martini Henry sighted to?

You cut it by putting a cutter on a male rod rod or guide and the rod goes down the bore. The cutter then flies up and down the rod turning as it does so. You can use multipoint cutters, or single point cutters and index the guide round, or a combination, whichever led ot less distortion in that process. IIRC they indexed to cut opposite grooves.

Thats fine for a constant twist, but if you need an accelerating twist as in short barreled high velocity aircraft guns, then you attach a cam to the guide, which increases the rate of twist towards the muzzle where acceleration is less. Reduces inertial loads on long high payload projectiles (a cannonshell), which because of their length need a high spin rate to stabilise the projectile, and you want to keep the barrel short for weight purposes

You can broach them - the Americans do this in final sizing which also stretches the innermost layer beyond its elastic limit, Creates a compressive stress which needs to be overcome before the barrel expands, so allowing for a stronger barrel for less wall thickness. We achieve the same effect after cutting using hydraulic pressure, and we call it auto frettage.

Whitworth achieved it latterly by wire winding, so with naval ordnance you are actually looking at a cotton reel or bobbin - a jacket laid over the wire. Another advantage was to act as an insulator - naval guns on big ships fired in broadside - in a crosswind effectively. So stiffness without being stupidly heavy is an advantage, and that construction reduced thermal distortion- and maintained the pointing line in hot barrels.

Nowadays with smaller guns they hammer form a billet over a mandrel usually with a rotary hammer. Roughing out a blank ready for fixtures like sights and breech only takes a few moments - literally.

Edited By mgj on 29/06/2012 19:00:13

OTOH it may just be that he's met that normal problem with boring bars, spring in the tool, and very slight rock of the saddle. Which means that the next cut always takes a touch more out than expected.

In which case, with each cut, cut on the inwards and outwards cuts. I always, when its critical do that twice, and then apply the next cut. Takes a touch longer, but with all the spring in the tooll taken out, increments proceed in an orderly manner, and it is not difficult to hit diameter with precision.

Edited By mgj on 21/06/2012 23:14:14

Locktite 574. Specifically designed as a high temp flange sealant. Its not really a locking compound, - its a thread and flange sealant. and is reasonably easy to undo, but still has antivibration qualiites. (592 I don't know sorry, but being a 5 they are possibly in the same series so could well be very similar)

Sets in a couple of minutes, easy to wipe off don't use too much. Seals everything, replaces all gaskets! Perfect - tool for the job. (I have an absolutely leak free traction engine, and an equally leak free Metre Maid, and no extraneous bits of white to get in the boiler and block injectors etc later!).

About £15 for enough to build and keep 2 engines running for 2 years and more! Available from Polly amongst others.

Why should one not use Loctite on a safety valve plug? You need a spanner to undo it just the same? With respect , there are very many Loctites, and while we normally talk of it as being a thread locker, they make, as we know high temp steam sealants, all the way to low strength nutlockers.

Edited By mgj on 20/06/2012 09:00:37

With respect - no.

I have no huge experience to draw on, but on the clacks on my TE, (4 in all) and on MM (3 working) none are guided particularly. I have seen designs which control the ball up and down and stop it rolling around, - while they may show an advantage in particular applications, the common or garden hole with a ball at the bottom of fairly indeterminate size works pretty well.

As long as the distance from the edge of the hole, or seat to be precise to the wall is just about 1/2 the ball dia, or a touch less, the ball will roll back in automatically.

I did actually knock one up these guided jobs for the TE when I was having injector problems (different issue altoghether) and it showed no advantage or disadvantage.

Personally I'm sure its just a matter of a deep enough seat, into which the ball will drop/roll very neatly. Pretty quickly too - especially as on track or on a TE the while thing is being jiggled about anyway.

Its the same thing as the plug on your bath - that gets dragged into place from some distance, and quite accurately, and for the same reasons.

Well there is no doubt you can get them completely steam tight.

Easy way - write to Polly models and use viton balls, which are fine so long as they cannot rise far enough to block an outlet (dead ends the pump,. and bends conrods ! But they are effective, and seem to last. (also make very good injector steam stop valves with a flat ended doohickey on the handwheel instead of a cone, and less hassle than an O ring.)

Ohter way, with stainless balls. Use a D bit (Tracy tools if you are lazy or make one) to get a niice flat bottomed hole - never had a lot of luck with a slot drill or endmill, but thats me. Centre drill and drill the hole, and I always ream to get a good sharp edge. Take a spare SS ball of the right size, and seat in hole, and then with a brass drift and small hammer, give it one good smart twat to give a seating. (Brass drift in bronze/gunmetal and you don't need to chuck that ball away, but that is the normal advice, which I don't follow).

Some burnish the seat, using a ball superglued into a carrier and then use Brasso/toothpaste and an electric drill. If you are nice and square with the drift and firm with the light hammer, you shouldn't have to. (A guide for the drift can be useful and I have made them for the standard sizes so I do use one) Squeeze the ball into place is another option.

Has this lot been running - not possibly scaled up is it? Or is it new?

Set pump type NRVs for about .030 lift max, injector clacks for quite a lot.( .1 of an inch I use) since oscillation and bounce is not a problem.

Hope it works out.