paul
Member
Posts: 8
|
Post by paul on Mar 28, 2007 18:39:30 GMT
As a newcomer to this hobby I'd like to get some idea of the sort of accuracy I should be working to for building simple (stationary) engines.
I realise some parts will be good enough if they are a working fit but some obviously need to be made to much closer tolerances. I've just finished making the slide valve for my first 'proper' engine (http://modeleng.proboards20.com/index.cgi?board=statgeneral&action=display&thread=1173547109) and the plans call for 11x11x6mm with a 'steam cavity' of 6x8x1.5mm. My finished part is 11.1x10.8x6 with the cavity pretty much bang on. The control surface is nice and flat and the slot for the valve rod is an ok fit on a 2mm shaft.
Is this level of accuracy likely to be good enough? I don't want to go making all the parts only to find none of them are good enough!
|
|
|
Post by havoc on Mar 28, 2007 19:08:05 GMT
Well a big part of engineering is knowing where tolerance matters and where it can be "about right".
Obvious parts where accuracy matters are bearings, the alignment of bearings, seals working under pressure and pistons. Also parts of motion gear that have to provide timings.
But if the inside of the steam cavity is a tenth off...I don't think that matters. On the other hand, the place of the ports (relative to each other) and the faces of the valve that have to seal and the dimensions of the valve that make the timing work: those are where it matters.
|
|
|
Post by ron on Mar 28, 2007 19:45:21 GMT
Paul I think you have found one of the problems of model engineering, the drawings normally just give a size, whereas in real engineering the drawings give a size and a tolerance, if the drawing just gives a size it then relies on the experience of the modeller to decide what the tolerances are. If in doubt make it more accurate than it needs to be, it won't do any harm and it's good practice. Ron
|
|
SteveW
Elder Statesman
Posts: 1,399
|
Post by SteveW on Mar 28, 2007 20:41:45 GMT
Paul,
It occurs to add that surface finish goes hand in glove with accuracy. Essentially everything you do on a lathe is to a large extent a process of thread cutting, it just depends on the thread pitch/depth being cut.
|
|
Myford Matt
Statesman
There are two ways to run a railway, the Great Western way, and the wrong way.
Posts: 621
|
Post by Myford Matt on Mar 28, 2007 22:22:14 GMT
I have found that a good way to gauge the required quality or tolerance for a process or part is to read books or magazine serialisations of model construction. Often it's not just a case of machining to a certain size, but machining parts to have a particular relationship. It probably doesn't matter within a thou or two what the diameter of a cylinder is, more important is the tolerance of the cylinder to its piston, plus alignment and all the other things you have to take into account. This is where the experience of authors like LBSC and Martin Evans can be very helpful. LBSC in particular, is good at saying this matters and that doesn't.
MM
|
|
|
Post by baggo on Mar 29, 2007 1:01:54 GMT
Paul,
the critical dimensions on a slide valve are the length of the exhaust cavity and the overall length of the valve. These are the dimensions that determine the valve timing i.e. the position of the valve when it allows steam to enter the cylinder and when it allows the steam to escape at the end of the stroke. Deviating from these dimensions will affect the timing, possibly with detrimental results if the errors are significant. The width of the valve is not critical so long as the valve is wide enough to span the ports. It is usual to make the valve to suit the ports, any errors in the port positions/widths can be allowed for by altering the dimensions of the valve. In your case, assuming the length of the valve should be 11mm and you have made it 11.1mm, the 'lap' of the valve will be increased by 0.05mm which probably would not make a noticeable difference to the performance.
John
|
|
paul
Member
Posts: 8
|
Post by paul on Apr 3, 2007 21:39:36 GMT
Thanks for the guidance gents. I've been giving this some thought while making the valve rod fork for my current project and I can see that, within reason, it matters not a jot what the external dimensions of this part are so long as it connects to the rod without 'slop', has smooth and well aligned holes/bushes for the pin and sufficient depth of slot to accommodate the eccentric rod. As I'm re-scaling the plans to 50% 'on the hoof' I've had to take some liberties with sizes to ensure physical strength etc but it seems to be going ok (even if I do seem to have spent the last 4 days filing one very small piece of steel!). The two parts I made that 'went wrong' (I'm sure there's a term for this?) are currently hiding somewhere in my garage in fear of my lump hammer!!
|
|
Myford Matt
Statesman
There are two ways to run a railway, the Great Western way, and the wrong way.
Posts: 621
|
Post by Myford Matt on Apr 4, 2007 0:01:03 GMT
Re: went wrong, reminds me of a bit of spin Intel used to use. Things that went wrong pre-chip production were always called 'corrections', after prodcution had started and a bug was spotted they turned into 'improvements'.
MM
|
|
dscott
Elder Statesman
Posts: 2,438
|
Post by dscott on Apr 4, 2007 20:44:58 GMT
Several years ago I made a part that had the rivet holes in the wrong place, this was followed by another piece, then another. Then the wonderful thought crossed my mind about making a scrap truck to carry all these interesting pieces made to scale to the scrap yard. Its when you start constructing your second truck you need to worry about accuracy.
David.
|
|
paul
Member
Posts: 8
|
Post by paul on Apr 4, 2007 21:04:25 GMT
;D
|
|