Wanted: a simple but unusual part...

Wanted: a simple but unusual part...

Author
Discussion

Pigeon

Original Poster:

18,535 posts

251 months

Friday 9th September 2005
quotequote all
I am looking for something which I am sure must exist as a component of some reasonably common assembly, but racking my brains fails to produce any likely candidates...

It would be a hardened steel tube... about 100-150mm long, 20-30mm diameter, 2mm or so wall thickness, and with a smooth, approaching polished, finish - like a gudgeon pin - on the INSIDE surface.

Apart from the two holes in the ends :) there must be either no holes in it, or one hole reasonably close to half way along of no more than 5mm or so diameter. (Three such holes close to the 1/4, 1/2 and 3/4 points would also be acceptable, but not very likely I don't think.)

It doesn't particularly matter if the tube is attached to some other item as long as the other item is not sufficiently complex as to make the whole thing stupidly expensive :)

If anyone can give me any hints as to what item of scrap machinery I can dismantle to obtain such a tube it would be very useful.

What's it for? I want to make a rotary valve for adjusting the phase and duration of the opening of hydraulically-actuated poppet valves. I reckon that if I can get this tube with the nice finish on the inside ready-made out of something, I can get a modelling-type lathe and make all the other bits myself.

GreenV8S

30,407 posts

289 months

Saturday 10th September 2005
quotequote all
Perhaps the outside of a small hydraulic ram would give you an adequate finish - can you find anywhere that is scrapping hydraulic equipment? If you get stuck I know a jobbing engineer in the midlands who specialises in hydraulics, but it isn't cheap stuff. If you're after a really cheap solution you might be better off going down your local rubbish tip.

Pigeon

Original Poster:

18,535 posts

251 months

Saturday 10th September 2005
quotequote all
You know, I think you might just have it? Hydraulic ram, small... very small... or something like one... Motorcycle steering damper. Gas strut. Perhaps even a washing machine drum suspension damper, though they might be a bit too crude. Many thanks, a useful bit of inspiration there.

GreenV8S

30,407 posts

289 months

Saturday 10th September 2005
quotequote all
At that length you might even be able to get away using the outside from a suspension strut or coilover damper?

dilbert

7,741 posts

236 months

Saturday 10th September 2005
quotequote all
If it's only 100mm long, you could easily machine it to your requirement on most 3 1/8" "modellers lathes".

You would have to machine it soft, and get it hardened, but there are pelnty of people around that can do that for you.

If the part has to be ground on the inner face, then you need someone with a cylindrical grinder, but again, it's not that tough to find someone.

Depending on the precision that you're looking to achieve in the bore, you can do many cylindrical grinding operations in the "modellers lathe".

Edited to add;

What material is the "other" component likely to be made from?





>> Edited by dilbert on Saturday 10th September 10:38

GreenV8S

30,407 posts

289 months

Saturday 10th September 2005
quotequote all
Pigeon said:
What's it for? I want to make a rotary valve for adjusting the phase and duration of the opening of hydraulically-actuated poppet valves.


I'm intrigued, and it would be interesting to find out more about your invention when you are ready to 'go public'.

dilbert

7,741 posts

236 months

Saturday 10th September 2005
quotequote all
I'm just reading this again. Wouldn't it be better to have an electronic angle sensor to work out when the valve events should occur?

That way, adjustments are touch of a button, rather than re~machining, or at best spanner operations.

Pigeon

Original Poster:

18,535 posts

251 months

Saturday 10th September 2005
quotequote all
Ah, that's the point of it - avoiding re-machining during the experimental phase, with the added bonus of variable valve timing during operation in service.

It's for my supercharged two-stroke diesel project, which is looking as if I ought to be able to make a bit of progress now. Since the scavenging process is pretty much driven by the incoming fresh charge rather than the pumping action of the piston as it is on a four-stroke and a non-supercharged two-stroke, the time it takes is fairly constant, so with fixed timing there's a compromise between not wasting charge at low revs (it may be only air, but it's carrying energy from the supercharger) and achieving full boost at higher revs. (This is not an issue on well-known examples like the Deltic, but such engines really only use the blower for scavenging and achieve only a minimal degree of supercharge.)

It is of course possible to get around the problem with exhaust pulse tuning, which is what the Zoche aero diesel does - but that has the advantage of constant-speed operation. Also, Phase Two of the experiment will involve using a turbocharger to unload the blower once it can extract sufficient energy from the exhaust. That not only ruins pulse tuning, but creates backpressure, which is useful for achieving boost but will call for retimed valve events.

I don't mind going into detail as it's a project I'm undertaking purely for personal satisfaction, so the only likely adverse consequence of talking about it is the embarrassment if it subsequently doesn't work

The idea behind the adjustable valve is fairly simple if a little hard to describe without pictures; it bears some resemblance to the spiral-grooved-piston method of regulation used in diesel injection pumps. Inside the tube is a close-fitting rotating steel shaft driven at crankshaft speed; bits of it are milled away in the middle to leave a reduced-diameter section separated into three parts by two ridges which resemble one cycle of a sawtooth waveform wrapped around the shaft. Two holes in the tube are connected alternately to the middle one of the three parts and one of the outer parts as the shaft rotates and the ridges pass under the holes; a third hole in between the two supplies high-pressure oil to the middle part, while the outer two are connected to drains. By rotating the tube the phase of the switching events relative to the crankshaft can be altered, and sliding it axially alters the duration.

The poppet valves themselves will be actuated by little double-acting hydraulic rams, which I may be able to make by butchering hydraulic tappets or the guts of a power steering pump, or failing that make from scratch.

The intention is to run the apparatus at low pressures by normal hydraulic standards - a hundred or two psi rather than hundreds of bar - and correspondingly higher flow rates, to avoid the need for the fantastically precise machining normally found in hydraulic apparatus. The oil pressure will be supplied by an ordinary lubricating-oil pump, fed from the sump oil, with a check valve to ensure it retains its prime with the engine off. The pump and valve will be mounted on the head to minimise the length of the pipe runs, and the "spent" oil discharged into the rocker box.

It seems to me that the main reasons such systems are not more widely used are speed of operation and complexity. Here, speed is not so much of a problem as it's a diesel so won't be revving very high, and complexity is a relative term - for experimental purposes, I reckon it's easier to make one adjustable apparatus than a whole bunch of slightly different camshafts (and I don't even know if Lister-Petter would be into supplying me with camshaft blanks, especially for an out-of-production engine )

GreenV8S

30,407 posts

289 months

Saturday 10th September 2005
quotequote all
Sounds interesting, although I would have thought that the inherent inertia in hydraulic systems would make high frequency operation a bit of a challenge! Did you consider connecting the valves mechanically to the 'wobbly groove' wave form, to make a sort of 'push/pull' cam?

MR2Mike

20,143 posts

260 months

Saturday 10th September 2005
quotequote all
GreenV8S said:
Sounds interesting, although I would have thought that the inherent inertia in hydraulic systems would make high frequency operation a bit of a challenge!


It's a diesel so I wouldn't have thought it would be doing silly RPM.

dilbert

7,741 posts

236 months

Saturday 10th September 2005
quotequote all
That's quite an interesting idea. I'm still thinking about the spiral system, and the variable duration, phase.

Something seems "not quite right", but I'm sure you've thought about it more than I. I'm guessing that the spiral grooves must be tapered in angular width, along the axis.

I think I'd be concerned about running the whole thing at a lowish pressure. I think that's more to do with the size of the control valve and the flow rate, rather than the ability to move the servo valve.

On the other hand, if you think you can manufacture the scroll, on a small lathe, I don't think you'll have too much difficulty manufacturing the tube.

Given the nature of the project, I'd opt for a bigger bore, say 2", in the tube, go for a 0.25" wall thickness and make the tube and the scroll from bronze or ali. You can easily get O-rings that you can make to whatever length you like, and you can easily machine bronze to the right finish and fit to seal an O-ring up to 200 psi and above.

The interesting bit is how to machine the groove for the O-ring. I'd say the only realistic way is CNC!




>> Edited by dilbert on Saturday 10th September 18:43

Pigeon

Original Poster:

18,535 posts

251 months

Saturday 10th September 2005
quotequote all
Correct, I'm not planning on any fancy injection equipment so I won't be busting the "standard diesel limit" of 4500rpm. (Indeed, the highest rated speed quoted for an AD1 is 3600rpm, and I don't plan on exceeding that with the standard piston, though they are heavily built and will take 4500rpm with no oil circulation for a few hours )

In terms of reciprocating metal the inertia should end up less than a conventional OHC design with bucket tappets, and the inertia due to the oil will be minimised by mounting the rotary valve, with a small accumulator on the feed, on the head, as close as possible to the valve actuators.

I've run through various purely mechanical ideas, but while it is easy to vary phase, it's harder to vary duration. Everything I've come up with has involved disadvantages such as awkward follower designs, three-dimensional cams or equivalent, control inputs that have to resist operating forces that vary or even change direction through the cycle, excessive peak forces, overly-complex linkages... The hydraulic design seems to win out on smoothness of operation, resistance to wear, maintenance of performance under wear, ease of control, and simplicity of manufacture - apart from the "waveforms" on the rotating shaft, it's all just turning and drilling of pretty simple shapes.

Pigeon

Original Poster:

18,535 posts

251 months

Saturday 10th September 2005
quotequote all
Ah, sorry dilbert - your reply appeared while I was typing. Here is a quick and crappy sketch of the arrangement rolled out flat:



The shaft is cut away in the working area to leave the black wavy lines sticking up as ridges of the full diameter, and the area between the ridges then receives high pressure oil from the central port. As the shaft rotates the ridges pass under the ports and reverse the pressure difference between the ports. Vertical displacement of the ports relative to the shaft provides the change in duration, from the shape of the ridges, and horizontal (ie. rotational) displacement changes phase.

I reckon that producing the ridges is just about possible though time consuming and involving a lot of fiddling about; the inclined sections could be tricky, but as long as the edge is straight it doesn't matter if the bottom isn't, and ultimate accuracy is not essential since it's a variable device and there's a calibration step involved anyway. Putting O-ring grooves down the centre of the ridges would indeed be a lot harder!

It seems to me that while the ridges are likely to be a pain in the arse, they are still perhaps easier than machining the inside of the tube because all the operations are carried out at close quarters and both tool and workpiece can be well supported. I would expect a modelling lathe to be somewhat deficient in the rigidity department compared to the usual massive cast-iron monster, which would make consistent operations down the inside of a long narrow tube a bit difficult.

dilbert

7,741 posts

236 months

Saturday 10th September 2005
quotequote all
In respect of the actual tool, I'd not bother with anything smaller than 3.125" centre height. If you get a machine that is that sort of size, it will be fine for boring operations.

A lathe is an ideal jig borer. If you get a boring head that fits in the lathe spindle, and a vertical slide, you can mount the job on the vertical slide, and use the leadscrew to get a really good finish.

If you try to make stuff in steel, getting a nice finish can be tough, but if you use cast iron, bronze, brass an ally you can get really superb results.

......

I see what you're saying about the profile of the valve now. I'd previously imagined something quite different. I was thinking along the lines of a screw, but what you've drawn ought to be considerably simpler.

Normal screws on the lathe are easy. Regular patterns are a bit more difficult. The worst is when you are trying to do a screw where the pitch increases or decreases. That really is the realm of CNC, and what I originally thought you were trying to do.

I'm assuming that you've essentially flattened out the surface of the inner rotor in your drawing.

......

There is a thing called the Quorn. It's a tool and cutter grinder. You can find stuff about it on the web. Basically it's a machine aimed at the model engineer, for sharpening end mills, slot drills, slitting saws and such. The point is that it has a grinding head which is adjustable height. The head is mounted on a post, and rather than just having clamp screw, it has a nice mechanism for controlling the head height.

Rather than just undoing the clamp screw and then having to support the weight of the head with your hand, the post has a fairly fast screw on it. The idea is that you undo the clamp and then wind a knurled ring up and down, to get a fine adjustment whilst you eye the grinding wheel to the correct height.

Because it's aimed at the model engineer, it's been designed to be made on the modellers lathe. The really cunning part is the method of making the post with the screw on it.

The grinder has to have some means of mounting the grinding wheel, and the design is such that the actual spindle of the grinder is a tube, which is mounted on the post that I was talking about by means of a bracket. The tube has a taper in one end and a pulley on the other, much like a lathe. The spindle has integral bearings, but overall it is a round bar.

The point of this is that it can be mounted in the toolpost of the lathe, by means of an alternative bracket.

So, back to the post.

What you do, is put the post up between centres in the lathe, and put a slot drill into the grinding spindle, which is mounted in the lathe tool post. You have to jury rig some means of getting the drive to the grinding/milling spindle in the toolpost, but it's not that difficult.

You are then in a position to set up the lathe like you would for screwcutting, where the saddle position on the bed is related to the rotation of the lathe spindle, by the leadscrew. You run the lathe spindle very slowly indeed, and the slot drill comparitively fast. By this means you can cut a square sided fast thread all the way along the post.

So how does this relate to your valve?

......

I guess my point is that using a similar scheme, you could produce the deep cutaway and the o-ring groove in your valve.

You'd need to make a cam that defines the profile, but you would only need one, for all of the complex operations.

Using a milling head in the lathe, you put the cam on the same mandrel, as the valve core, and have a follower or stylus that is essentially a small ball race on the end of a stick. Make a small bracket to attach it to the saddle of the lathe, and run the lathe very slowly. As the valve rotates the follower causes the saddle to move to and fro along the bed, cutting your profile.

The cam it's self is really simple to make. It's just a disc with a hole in it. Looking at one face of the disc, it's thicker one side of the hole than the other.

Obvously, cutting the cutaway, you'd use a smallish slotdrill, and for the o-ring groove a ball nosed end mill.

phew


>> Edited by dilbert on Saturday 10th September 22:18

Pigeon

Original Poster:

18,535 posts

251 months

Sunday 11th September 2005
quotequote all
So... essentially like screw cutting, only with the cam and follower mechanism taking the place of the leadscrew, and a milling cutter mounted on the toolpost? That's cool, I ought to be able to rig up something like that.

As far as the actual lathe goes I'm a bit limited in that I have to be able to get it up two floors of stairs and in through some rather serpentine doorways as I live in a flat and don't have a garage. (Yes, the AD1 is in the flat, and testing will be dependent on the neighbours being out...) Avoiding doing myself an injury during the above manoeuvres would also be desirable. I've found a few possible candidates through Google/ebay, but the pictures are all too small to get a decent impression, and I think it would be best to go and see it in the metal first.

dilbert

7,741 posts

236 months

Sunday 11th September 2005
quotequote all
Well, I guess it depends alot on the building that you're in.

Here a home I live in a cheaply constructed 70's two up two down. I have to say that I'd not trust the floors with much more than a bathfull of water in terms of weight.

If your flats are basically concrete in construction, you'll probably be allright. Timber floors can be OK, they just have to be well built.

The attached link shows the sort of thing I'd recommend. If you go for a really small lathe, I think you'll be dissapointed with the sort of results you can achieve.

www.gandmtools.co.uk/cat_leaf.php?id=2680

The other thing to think about, if this is going to be in a living space, is that this sort of kit can make one heck of a mess of your decorations! The combination of swarf and oil just gets everywhere.

The pictured machine is about 4'6" wide and 2' deep, so it will go through a normal doorway. You will need someone to help you lift it.

The seven usually has a single phase motor, at about 1/8 - 1/4 horsepower, and that's adequate for most stuff. If you're new to this sort of thing, a lowish power motor is good from a safety perspective anyway.

The thing about the seven, is that it has a vast range of fixtures and fittings, that are designed for it, and allow you to do milling, and boring, all manner of operations, that would normally require a completely different machine.

Have fun!

>> Edited by dilbert on Sunday 11th September 10:33

GreenV8S

30,407 posts

289 months

Sunday 11th September 2005
quotequote all
Seems to me that 'cam' is going to be rather awkward to make and you might end up having several goes to get it right. Have you priced it up as a CNC job? Something like emachineshop would be convenient, if you could find one in this country?

dilbert

7,741 posts

236 months

Sunday 11th September 2005
quotequote all
I have to say that if Mr. Pigeon is intent on just making this assembly, it's probably going to be better to do some drawings, and get the thing made by someone else.

The only thing I would say is that I have a feeling he's got intentions beyond just this job. What he's trying to do isn't beyond the bounds, but I think it is if he goes for a machine smaller than the one I posted a picture of.

Even if you don't go for novel ideas, just making small simple engines that work is really satisfying.

Pigeon

Original Poster:

18,535 posts

251 months

Sunday 11th September 2005
quotequote all
You are quite right about the further intentions bit... there are many times I've thought "If only I had a lathe, I could make one of these", but for odd single items there are other ways to get the job done. The diesel project involves enough machining work that the cost of getting someone else to do the jobs becomes comparable to the cost of getting a lathe and doing it myself, which has the further advantage that I can then do all the other odd single items myself. Also, it's more fun.

I'm not totally new to lathe work but I am distinctly rusty not having done any for many years. I do however recall that with care and a methodical approach, while operations may have been slow they were never what I would regard as difficult (if that makes sense).

My floor is timber, but it's old and solid timber... that little Myford weighs what, around 200lb? That's similar to the weight of me carrying a big heavy thing, which doesn't cause the floor to deflect noticeably, plus I'd spread the load by standing it on a couple of big baulks of timber in any case. So we should be OK on that score. Not too worried about mess either; my idea of interior design is hardwearing floor coverings, plain white walls and multiple fluorescent lighting fittings; as long as I can keep the swarf away from the pigeons and the electronics, and the PCB etching chemicals away from the lathe - which shouldn't be too hard - I'll be happy, and the thought of having to replace a few square metres of lino when I eventually move out doesn't exactly fill me with trepidation

Many thanks for the G&M Tools link - that's a very useful site.