Cagney valve gear question

[Bill Shields]

Very good.

It is an iterative process .

Given one part, what are the rest?

If you are really good with Fusion or SOLIDWORKS or others, you can build a simulation and look for binds.

Or lay it our in 2D and play with it to look for the same.

Problem area is 99% of the time in the combination link area.

I cannot tell you how many locos I have seen where areas in and around the combination link have been hollowed out or relieved for clearance.

Including some of yours truly. I have a drawer full of parts from 45 years ago that just didn't quite work in full gear...

[Glenn Brooks]

Alas, I have no experience in any of the modern drawings programs. Swore off everything related to CAD when I retired, and now woefully out of date.

I did find a citation to a couple of spreadsheet programs and on line simulators that might provide some answers. They are British and apparently quite proud of their skills in plantar geometry and calculus, so haven’t dived into the rabbit hole as yet.

BTW, I did find bits and pieces for an existing Stephenson’s gear for the Cagney. It came with the boiler and frame, but I had forgotten about it. The only parts I’m really missing are the eccentric centers around which the original straps rotate. Not a big deal to turn out of round stock, or plate, should I decide to keep this valve type on the loco. The eccentric throw will still be the critical dimension, I think.

Glenn

[Bill Shields]

It always is, but easy to calculate. You then need to get the position of the eccentrics relative to each other correct before you do anything else. I typically pin them together using a mandrel and rotary table before I mount them on the axle.

Otherwise, you run around in circles, chasing your behind and never get the timing correct forward and reverse.

There are spreadsheets and on-line simulators, but none really help with clearance calculations unless you are really good with trig.

you have to take into account the daimeter of the pivot pins to determine interference, which has nothing to do with the motion of the valves -> which is what the simulators worry about...

Even the tried and true Langworthy Hudson which I built had a minor bind in full gear in that area.

[Erskine Tramway]

Glenn asked me about the Walschaerts Gear I designed for the 15"-gauge Pacific that we built at Sandley's in the late '70s.

After I did the layout work on 'Mylar', our shop carpenter and I built a wooden mockup to '"prove" it. Here's a picture I took of the mockup. I understand it's on display at the R&GN now.

3-1-2012_001 mail.jpg (461.79 KiB) Viewed 2254 times

Here is a photo of the Pacific in service at Milwaukee County Zoo, with the Valve Gear 'in the flesh' as it were.

20140511_101754_mail.jpg (538.08 KiB) Viewed 2254 times

Mike

[Glenn Brooks]

A few notes to preserve valve gear references and formulas. The use case here is these references and formulas are useful for designing valve gear when no plans or drawings are available, and original parts or components are missing. Such is the case with most all Cagney’s and early day antique locomotives.

references:

1) Locomotive Valves and Valve Gears, J.H. Yoder and G.B. Warren, TEE Publishings, Warwickshire, England, 1917, reprint, 1993.

2) Model Locomotive Valve Gears, John Evans, Model and Allied Publications, England, 1962.

3) Model Engineering, Henry Greenly, Watchmaker Publishing, London, England, 1915, reprint 2010

4) Modern American Locomotive Construction and Operation, Frederick J. Prior, Simmons-Bordman Publishing, Omaha, Nebraska, 1925, reprint 1982.


Definitions:
lead- width the valve opens the steam port at beginning of admission
lap- the amount by which the steam edge of the valve overlaps the steam edge of the cylinder admission port, with valve centered (Yoder, 4)
Valve travel - distance the valve passes from forward dead center to rear dead center, while in full forward gear (Yoder, 50)
Running cut off - distance back from the full forward notch of the reverse gear (usually 1/4 valve travel) necessary to continue forward motion of the locomotive, with least expenditure of steam.


Determining Eccentric Throw (T):

- Measured from axle center to wrist pin bore center.
- Determined with Reverse gear lever in center notch (neutral).

Eccentric Throw (T)= 2(lap + lead)

Determining exhaust port width (WE):

Given cylinder stroke, determine intake port width (IW):

Given, WE = 1/8 cylinder stroke,

then, IW = WE/2

Determining length and radius of link geometry (L) sufficient to provide correct valve timing:

greenly, pg167

IMG_1411.jpeg (92.75 KiB) Viewed 2192 times

Whereas, if:
E = eccentric travel (throw)
V = valve travel
P = radius equal to eccentric rod length
Q = link die center to center

Then, E = PV/Q

[VGC]

If you did not see yet, my program gives excellent results based on main basic dimensions within 3 to 15 minutes, immediately after typing the values into the "Quickstart form".

But the reason I write is that I am still open for supporting someone using my new version of valve gear which in my opinion is superior even over Walschaerts. Because it gives good results even at large valve piston travel and does not need an open loop like the expansion link of Walschaerts VG.
The structure of its single parts will be rather similar to Baker parts but the kinematics skeleton design is similar to Walschaerts.

If you are interested in one of these options, PM me. I can give you access to a running preview of the new type in my program.

[VGC]

You would have the first steam locomotive valve gear in the world that never ever was layouted manually

[Glenn Brooks]

VGC,

Thank you for your post and PM. Your on line calculator is wonderful. I read your conditions of service and apparently will need your permission to develop and save a drawing file for my loco- it is built in 15” gauge.

I am very interested in designing a valve gear with your app, but, at the moment, Iam on light duty recovering from a back injury and may not be able to work around the locomotive - pulling dimensions and measurements for all the components in your data entry forms until the new year. I can PM with additional details.

Regards,
Glenn

[Bill Shields]

You would have the first steam locomotive valve gear in the world that never ever was layouted manually

most likely not.

computer modelling / animation systems have been around for a while (but are admittedly VERY awkward to use for the not-all-the-timers like me).

There are still a few of us who are capable of doing the necessary calculations and can create simulation models accordingly.

I particularly find Southern Valve Gear to be a challenge, mostly due to the lack of a few critical dimensions in their documentation. (which I guess they did not want to share...)

[VGC]

I did not mean that one material valve gear but that new type of valve gear structure which probably never needs to be layouted manually because its "inventor" supplies the according calculation software online.

@Glenn: I will reply via PM soon.