Member postings for Andrew Johnston

Set screw is the correct technical term. But unfortunately sloppy definitions mean that set screw can also be used to mean a grub screw. Machine screws imply a non-hexagon head.

It's almost as bad as describing a milling machine as universal just because it has horizontal/vertical capability.

Andrew

Another consideration is that having a heavy item at one end of the table makes it much harder work to wind the handles, in all axes.

Andrew

I find having multiple items on the table never seems practical, something always seems to be in the way. So I generally fit just the item I need to use; my mill table is 48" long.

Andrew

I'm not sure what the problem is? Edge finding and cutter diameter are separate issues. All my standard carbide cutters are on size, or a maximum of 0.01mm under, on the shanks. For general work I never worry about the cutter diameter not being as stated. if i use old HSS cutters, or a slot or edge distance needs to be accurate (say better than 1 thou) then I'll cut 'n' measure using depth micrometers or gauge blocks and adjust as needed.

In all the years of using a CNC mill I've never used cutter compensation. Parts fit together straight off the mill. Occasionally I'll tweak the CAM program to cut under, or over, by a small amount as required.

Andrew

Specifically spiral flute, as opposed to spiral point.

Andrew

Oh dear, I've been making special tooling for onesies and twosies:

Andrew

Must be careful not to get over-excited.

Andrew

See answers embedded above.

Andrew

Ooops, I think you're correct.

However, that raises a question; if the tap/hob can drive the work why do hobbing machines go to the trouble of actively driving the gear blank?

Andrew

Unfortunately you seem to have mislaid the first step. If the tap is skewed by the helix angle to get straight teeth on the gear how does it drive the gear blank as there will be no tangential force. The tap drives a worm wheel because it creates teeth that are not straight and hence there is a force causing the blank to rotate.

Andrew

Fair point. The teeth on the worm wheel I free hobbed with a tap look triangular. But the wheel is quite small (~1/2" diameter) so any involute curvature will be also be small. The tap was UNF so the pressure angle will be 30°. Since the tap form is triangular the gear teeth will tend to a triangle rather than the more familiar rack form. No doubt gears made by the same method will run together but there will be significant forces pushing them apart.

The small worm wheel shown is for a speed adjuster on a Pickering governor, so I doubt it'll ever get turned:

Andrew

The method mentioned works fine for cutting worm wheels, although spiral flute taps are helpful:

Getting the correct number of teeth is a bit hit and miss. One can free hob a worm wheel, but it needs to be pre-gashed:

For spur gears it's a useless method. First, the work would need to be swivelled to the helix angle of the tap in order to get straight teeth on the gear. Second, the work would need to be rotated at the appropriate rate, same as when using a hob. Third, the tooth form would be non-standard, and the gear would only work with other gears made by the same tap.

It doesn't take that long to cut spur gears by conventional methods, provide one gets on with it rather than pontificating on forums, two gears at a time in this case:

If I want a quick gear, and don't have the appropriate cutter, I simply design it in 3D CAD and let the 4-axis CNC mill get on with it. As was the pinion:

For mating with this internal gear:

Andrew

I think momentum goes down by the fourth power. Consider angular momentum which is:

radius x mass x angular velocity

If we assume a scaled flywheel, and the same rpm, the mass goes down as a cube law and the angular velocity (radians per second) stays the same. But the radius goes down linearly. So overall the angular momentum scales roughly as the fourth power.

I'm pretty sure my flywheels are close to 1/3 scale, the rim looks pretty thin although they seem to be roughly scale compared to pictures of the full size engines.

Ramon: I agree that the medium behaviour doesn't scale. That could explain why an engine that is significantly smaller than full size, say a tenth or less, might have a problem with scale ports/pipes.

Andrew

That's interesting. I'm in the process of re-designing the steam ports , slide valves and ultimately the valve gear for my one third scale traction engines. The port sizes are mostly copied from full size, scaled as per the model engine. I didn't think the ports needed enlarging. My reasoning went along the lines of the port areas have gone down by a factor of nine but the volumes are down by a factor of twenty seven. So the flows are reduced well below the reduction in area. Calculations for the model engines seem to indicate flow rates and Reynolds numbers well below what I would expect for a full size engine.

Of course I could have got my reasoning wrong - what does the team think?

Andrew

Presumably because the volume of the rim, and hence the mass, has gone down by a cube law. However, for a model, probably running off load, I doubt it will be significant.

Andrew

Not so, if linear dimensions are halved then areas are reduced to a quarter, and volumes are reduced to an eighth, of the original. Basically an inverse square law for areas and an inverse cube law for volumes.

Andrew

That's what I did when I needed to machine an enclosed spigot:

I'm not convinced a roller box would be suitable. They're designed to machine long parts to diameter, not short spigots. Herbert did sell one indended to machine to a shoulder, but I expect they're in rocking horse territory. If I was making the part on the repetition lathe I'd use a knife tool and just set stops for diameter and depth. For a lot of parts I'd machine in two stages, a plunge cut to rough out and then a finishing cut on diameter and shoulder.

Andrew

Me too, after that I followed the manufacturer instructions.

Andrew

Not according to Clarkson - see Section 4a:

Andrew

No idea where your equation came from, but I'm not convinced it's appropriate. The size of the injector is determined by the amount of steam used by the engine. See the first post in this thread for calculations. If the boiler can't provide that steam, or can't accept the calculated injector flow without losing pressure, then the boiler is too small. It is normal to size the injector slightly above the nominal flow rate needed, so it can be used intermittently.

Andrew