|
Post by Deleted on Oct 20, 2009 21:50:54 GMT
Hello All,
This is a question asked by many before and now I ask it. What is the best way of getting into valve gear and its design. I have tried Charlie D's software, this was great but I didn't find enough control over cylinder placement with respect to the driving axle for my application. I also looked at various comments by Messrs Evans and Lawrence but think I may have to get Don Ashton's book. Before I do, can someone please give me a short description of the process of the design process?
Many thanks,
Dave
|
|
cotswold
Part of the e-furniture
Still testing the water
Posts: 307
|
Post by cotswold on Oct 20, 2009 22:11:50 GMT
.... I may have to get Don Ashton's book. Before I do, can someone please give me a short description of the process of the design process? Dave I certainly wouldn't even try to give an introduction better than the one you'll find in Ashton.
|
|
|
Post by goldstar31 on Oct 21, 2009 6:50:37 GMT
I have a letter dated 31.7.97 which includes the comments:-
I still see questions in magazines like 'Model Engineer' asking the same old things- and getting the same silly answers!
Many, many years before that I saw members of my family injured with locomotives. The time span of injury reached far back into the making of 'Sans Pareil' at Shildon.
So don't ask me locomotive questions but I can mention one impeccable source of carefully researched answers.
GS31
|
|
|
Post by Nigel Bennett on Oct 21, 2009 12:03:13 GMT
Well I suppose a general process would be as follows:
1. Determine the sizes and positions of the ports in your cylinder. They need to be large enough to pass the steam to be used, and far enough apart to provide a sensible bearing area for the valve. The ports are probably best determined from cabbaging one of the designs of LBSC/Martin Evans/Don Young/Neville Evans/Doug Heswon most similar in size and disposition to your prototype(take your pick).
2. Determine how much lap and lead you need - read Martin Evans' valve gear book. It's usually based on a proportion of the port width. Again, pinch other folks' designs!
3. Decide on which valve gear you are going to use. Normally it's decided for you by your prototype, of course.
4. From the requirements of your maximum cut-off in full gear (80% is usually enough) and the lap and lead, you should now know how much valve travel you need.
5. Lay out a valve gear to give the necessary movement.
6.Plug all the figures into Dockstader's software to check it all out.
7. Build it
8. Run it.
9. Goto line 1 and repeat as required.....
I can't understand your problem with "cylinder placement with respect to the driving axle" with Dockstader's software - what on earth are you doing to make this a problem? Pray elucidate...
|
|
|
Post by Deleted on Oct 21, 2009 19:16:24 GMT
Thanks for the replies so far. I am taking my design from a prototype. So, inputting the key dimensions into the Dockstader program (he calls them FrameDimension (28 off) and CylinderDimensions (22 off)), I would hope to get a good simulation. Not quite, as the piston and valve are clearly too far horizontally from the cylinder and steam chest for which they are intended to reside. Whilst I can find a way of defining the key horizontal lines WRT the CL of the driver, I cannot say the same for the vertical lines, such as the distance from the cylinder from the CL of the driver - you can however easily define the length of the cylinder, valve etc.
I have probably missed the obvious so will apologise now for that moment when I find the wood in the trees.
Thanks again, Dave
|
|
|
Post by gilesengineer on Oct 21, 2009 22:02:36 GMT
Hi Dave,
If (in Charlie's Program) you go into Edit dimensions, and then cylinder dimensions, you will find the variable 'Cylinder Ctr to Driver ctr' which is the thing I think you're looking for (if I've understood you correctly - and if not, I apologize!).
So far as the inclination of the cylinder is concerned, you have to accept it as horizontal, and as it were 'incline' the positions of the other elements to get the results - effectively the horizontal on the program, becomes an inclined line in reality to whatever degree you need. I hope this makes some sort of sense......
All the best
Giles
|
|
|
Post by GWRdriver on Oct 21, 2009 23:17:56 GMT
I'll be interested in the continuation of this thread as I need to do the same thing, also with inclined cylinders. Dockstader is a good proofing program, once you get the gear designed, but it's the design that has me symied. IIRC there are something like 57 variables, or settings, to plug into Dockstader for the Walschaerts gear. You don't mention the type of valve gear, although I assume with inclined cyls it will be Walschaerts.
|
|
cotswold
Part of the e-furniture
Still testing the water
Posts: 307
|
Post by cotswold on Oct 22, 2009 7:04:47 GMT
I'll be interested in the continuation of this thread as I need to do the same thing, also with inclined cylinders. Dockstader is a good proofing program, once you get the gear designed, but it's the design that has me symied. IIRC there are something like 57 variables, or settings, to plug into Dockstader for the Walshearts gear. You don't mention the type of valve gear, although I assume with inclined cyls it will be Walschearts. To me it seems to be a case of horses for courses. I see Dockstader as a proving and verifying tool but not as a tool for use when designing from scratch. Having failed when I tried to use Dockstader (and I now realise that inclined cylinders was the problem), I put Ashton's formulae into a spread sheet and then tweaked the variables until I was happy. Incidentally having established the theoretical eccentric angles (Stephenson's) I set them between centres on the lathe - an act of faith that has paid off when I listen to my exhaust beats.
|
|
|
Post by GWRdriver on Oct 22, 2009 13:08:28 GMT
I see Dockstader as a proving and verifying tool but not as a tool for use when designing from scratch. Before I began the design process I too assumed the Dockstader programs were "design" programs and quickly learned differently . . . so I was off to study Ashton.
|
|
|
Post by goldstar31 on Oct 22, 2009 15:03:17 GMT
Don is a world authority on tugs as well as having had a full size steam locomotive of his own.
My comments were prompted like Don's!
Note that Don paid for the publishing of his original books.
Prophets not without honour? Sounds like history repeating itself.
Perhaps he should have blown his saxophone louder. Oh, yes, please note.
GS31
|
|
ewal
Part of the e-furniture
Happiness is a good wife & a steam engine.
Posts: 293
|
Post by ewal on Oct 22, 2009 19:13:43 GMT
If I had read this thread before I started building I dont think I would have started. I didnt think it was so complicated .
ewal
|
|
|
Post by GWRdriver on Oct 22, 2009 23:45:24 GMT
Part of the joy and sense of accomplishment is conquering such things. I hope it's like calculus was, . . . many months of beating ones head against the table, followed by an Eureka moment.
|
|
|
Post by Deleted on Oct 23, 2009 0:19:17 GMT
Again, thanks for the comments. Sadly calculus is not studied in many of today's GCSE maths syllabuses, unlike the GCE maths it superseded; and yes, it was a joyful moment when you clicked what the point of calculus was. Thank you Giles for reminding me where the wood was, I tend to miss what's under my nose. The valve gear I am looking at is Stephenson with an indirect valve drive, so the Dockstader should be OK as it also has a valve rocker. The cylinders are inclined but it is not a problem normalising them for the program. Like others, I think I now see it as a proofing and not design tool. Don Ashton has a spreadsheet for the calculations which I think is at his website, so I may have a look at that before getting the book.
One other thing, why would inside admission valves have direct links but outside admission have indirect through a valve rocker; can the same effect be obtained by other means? (this will probably be answered by myself when I get the book!)
Thanks, Dave
|
|
|
Post by houstonceng on Oct 23, 2009 15:25:28 GMT
Quote "One other thing, why would inside admission valves have direct links but outside admission have indirect through a valve rocker; can the same effect be obtained by other means? (this will probably be answered by myself when I get the book!)"
You don't need a rocker just because the valve is outside admission (typically a slide valve) or inside admission (typically a piston valve). You do, however, need rockers to, for example, operate valves where the cylinders and valves are outside the frames when the valve gear is inside. See Walshaerts (sic) and GWR Kings. Some designs use a vertical rocker to lift the motion above the front axle when the eccentrics are on the centre or rear axle.
|
|
|
Post by asquith on Nov 14, 2009 20:26:24 GMT
I thought I’d post a link to Don Ashton’s website:- www.donashton.co.uk/index.htmlI must declare an interest: Don taught me metalwork (Engineering Workshop Theory & Practice) and engineering drawing at school over 40 years ago. He was an inspirational teacher, and is a brilliant engineer and an excellent craftsman. At a time when I was just beginning to be interested in how things were made, Don and another teacher took over from the retiring teacher (who seemed to have regarded the new machine tools as objects to support plastic dust sheets). We were introduced to all the techniques that the workshops could accommodate, from machining and forging to metal forming, and we also saw examples of work Don produced away from school, including a performance-boosting carburettor for his car, for which he made everything including the body casting. Don became a family friend, and he introduced me to many other aspects of engineering. I was amazed at the quality and quantity of work he produced from his tiny garden shed, with just enough room for a Myford lathe and a small drilling machine. I moved away and lost touch for a long time, but I’m pleased to see he’s still a prolific polymath, well-known in the field of wind instruments.
|
|
|
Post by donashton on Nov 16, 2009 11:54:22 GMT
I'm not surprised that people are hazy about valve gear design and expect a simulator to wave a magic wand. First you need to understand the mechanism. Those who can't be bothered certainly should not attempt design. For those interested in this complex subject I have been making the right information available since 1976 and you can download design programs and simulators from my website at www.donashton.co.ukBUT you still have to learn what it's all about for success. All published designs are by amateurs and many full size designs reveal incomplete understanding. The saving grace is that in most cases the wheels will still turn after a fashion, thanks to the flexibility of the steam engine - not very satisfying after nearly 200 years since the Rocket.
|
|
|
Post by Deleted on Nov 22, 2009 0:57:19 GMT
Thank you very much Don for joining the forum and responding to this question. I am having a try at the spreadsheets and have found them useful although no doubt I will probably have a look at your book to get a better feel for the subject. I am working from works drawings with some help from a Martin Evans design in order to adapt to the scale. Although, as you said, all published designs are by amateurs and many full size designs reveal incomplete understanding, I feel confident I am on the right path. Once again, thanks.
|
|
|
Post by drjohn on Nov 22, 2009 2:01:54 GMT
An irrelevant (or irreverent) question - has anyone attempted to drive valve events using stepping motors which would then give total precision to the valve events? DJ
|
|
brozier
Part of the e-furniture
Posts: 335
|
Post by brozier on Nov 22, 2009 14:42:47 GMT
That rings a vague bell, I'm sure I've seen an article on someone who had designed and built an engine run off solenoid valves.
cheers Bryan
|
|
|
Post by donashton on Nov 25, 2009 15:04:16 GMT
DJ Neither irrelevant nor irreverent, but maybe overkill when a properly designed motion can be almost immaculate. Interesting, though. Don
|
|