Stephenson's Link Hookup Question

[SteamGuy]

I know from talking with a Stanley guy that he starts his car with a late cutoff, and then as he gets going, shortens the valve travel via the Stepenson's link to get an earlier cutoff. He says the shorter cutoff at higher speeds is not for efficiency, but due to a minimal sized exhaust port and pipes, and the shorter cutoff releases the backpressure from the engine after release, and allows the engine to make more power.

I have read that for locomotives, shortening the cutoff was commonly used for efficiency.

Then to really confuse things, I found a spreadsheet someone made on a Fitchburg Steam Engine, and if I am reading the sheet correctly, it will produce anywhere from 50 hp to 350 hp, all with the same 14" bore and 16" stroke.

And looking at the cutoff table, this engine uses 28 lbs. of steam per IHP per hour for cutoffs ranging from 5% to 80% (constant 28 lbs for any cutoff).
The only things that change with cutoff are the expansion ration, the mean effective pressure, the horsepower and the steam consumption.

The steam consumption seems to vary almost linearly with the horsepower.

So my idea of using cutoff to increase efficiency just went out the window.

So I have to guess that if you are running at a constant load, say 200 hp, and you can produce 200 hp with a 25% cutoff, but instead you are using an 80% cutoff, then you are wasting steam by not using the lesser cutoff.

So in summary, it seems like you can use a Steverson's link to adjust cutoff for several reasons, which would be a change in power produced if the power produced is varying, or a change in efficiency if the power produced is constant.

And last but not least, after reading an article by Don Ashton, apparently you can design the suspension point of a Stevenson's link to give a minimum amout of link slip, and also symmetrical valve travel in both full forward and full reverse link.

So others have said that Don's method is ok for maybe a locomotive (but even a locomotive generally runs forward most of the time), but for a marine engine, the linkage is optimized for running forward only, and when a marine engine with a Stevenson's link is operated, it is operated either in full forward or reverse link, but never a partial link like is done with a Stanley car.

I was said on the other post that the lp cylinder of a compound does not need cutoff, so that seems to say that the hp cylinder is not operated at some cutoff point either, but instead operated at a late cutoff, maybe 70%, all the time, since I think the link lever for the hp and lp cylinders are common, and changing the cutoff on the hp would change the cutoff on the lp?

Can anyone verify these ideas?
Thanks

[fredrosse]

"I know from talking with a Stanley guy that he starts his car with a late cutoff, and then as he gets going, shortens the valve travel via the Stepenson's link to get an earlier cutoff. He says the shorter cutoff at higher speeds is not for efficiency, but due to a minimal sized exhaust port and pipes, and the shorter cutoff releases the backpressure from the engine after release, and allows the engine to make more power."

With any fixed inlet steam pressure, I don't think that statement could be true. Longer cutoff will always give higher output power, at any operating speed, and/or at any exhaust pressure buildup due to restricted exhaust.

An easier way to understand steam engine performance at part loads is with the "Willians Line". Documented in the late 1800's, it was shown that virtually all steam engines with cutoff control had a very nearly linear steam flowrate vs power output, up to the point of the engine's "best efficiency point" (BEP). For throttling control of engine output, the "Willians Line" was somewhat curved, showing higher steam consumption (higher than an equivalent cutoff control engine) at loads below 100% of BEP.

The attached graphs show the Willians Line, and the resultant steam consumption per horsepower, for a cutoff control engine. The later graph compares the performance of cutoff control vs throttling control. Note that at 100% of BEP load, both engines have the same cutoff and no throttling, so they show the same steam consumption and efficiency. At low loads, linking up (reducing cutoff) will always give better economy than throttling control, although the difference will be small at relatively high loads.

[fredrosse]

"but for a marine engine, the linkage is optimized for running forward only, and when a marine engine with a Stevenson's link is operated, it is operated either in full forward or reverse link, but never a partial link like is done with a Stanley car."

Steamships would often run continuously in a "linked up" cutoff setting. sometimes for economy, and when in a convoy, running to allow the slowest ship to keep up with the group. I shorten the cutoff on the Margaret S. often for exactly the same reasons. When I am just cruising, cutoff is shortened to build level in the boiler, as the engine is using less steam per stroke, but the feed pump always pushes the same amount of water into the boiler on every engine revolution. The boat goes just a little slower, but I'm not into the sidewheeler for speed.

When running with several other steamboats, I may want to slow down to the speed of the slowest boat, so I shorten cutoff as required. Having as many cutoff notches as I do, the engine throttle valve is usually wide open, and all engine control is done with the forward/reverse cutoff notches.

[barts]

Note that the effectiveness of short cut-offs in traditional engines is limited; the practical number of expansions is limited by condensation. Traditional triples were significantly less efficient running linked up; they were designed for a particular cruising speed. Uniflow engines, on the other hand, offered more flexibility, maintaining relatively high efficiencies over a wide load range.

- Bart

[steamboatjack]

Folks
Large triple & quadruple expansion engines usually had means to “adjust” the link position in ahead running by a slot in each reversing arm. This item is made a big deal by Elliott Bay in their website but in fact is of limited use unless indicator cards (or their electronic equivalent) are used. Cards were taken when running “full away” and the Chief would maybe adjust one or more slots, but this was done primarily to balance the engine power between the cylinders, an important point on large engines. No other means to “link up” were provided. As other posts have suggested a launch should generally control the boiler to produce the steam required for a given speed and conditions (when at cruising speed) with the engine running with the throttle open. If the balance between the hull, boiler and prop are about correct then this should be OK. With a non compound engine using Stephenson gear the provision of a second notch on the reversing lever quadrant could be an advantage but this would be difficult to quantify.
A point here concerns “lead” I would advocate for launch engines up to 15-20 IHP that the lead at full gear is set at minimum (just a crack open). Especially if using “open rods” as the lead will increase should you use linking up. Many smaller engines use “crossed” rods as the weight shaft is at the back of the engine and the suspension point is near the front, in this case the lead would reduce if you link up.
Centre suspension give you a choice but that does not mean things are a mirror image, I had better stop here! As this subject could go on and on.
One more point I would be very careful using the term “efficiency” this purely means the ratio of input to output. Should you reduce the amount of steam being consumed and therefore the amount of work being done this is a saving but is not necessarily an increase in efficiency.
I have corresponded with Don Ashton regarding Stephenson gear and its theoretical possibilities and maybe able to advise on technical questions, Also I would recommend Dons website, particularly the software of the late Prof Bill Hall which I find useful.

Regards
Jack

[SteamGuy]

Jack-

Interesting information.
If I understand you correctly, for a large marine compound engine, the gear could be adusted independently on each cylinder as the engine was running to balance the load between the cylinders? but I guess all the gear had to be connected to a common shaft for reversing?

I would think you would always want to increase the lead as the engine runs faster, much as is done on the ignition system of an automobile. I am not sure why crossed rods would be used in any circumstance, but obviously they were in some applications.

As far as efficiency, it seems like most of the gains seem to come from preventing the transfer of heat (energy) from the incoming hot steam and the outgoing cooler exhaust steam. Engines that are much more efficient include the Corliss and the Uniflow which both I believe separate the incoming steam from the outgoing exhaust, with great inprovement in efficiency.

Condensation seems to be the enemy when runing steam engines, and I have also read that by superheating the steam, you can increase the temperature of the steam, and thus add more energy to each pound of steam entering the engine, and by avoiding condensation, extract more energy out of a given pound of steam, thus the increase in efficiency using a superheater.

As mentioned above, it seems like the efficiency of an engine could be increased (amount of steam used per horsepower) if you were operating at a fixed load, and you were able to changed from a late cutoff to an early cutoff and still produce the same horsepower (this is a possiblility). Certainly operating at as early a cutoff as possible for any given fixed load with reduce the steam used to a minimum, since you would maximize the expansion-after-cutoff phase of the power extraction from the steam.

An example would be a locomotive, operating at say 50 mph. The gear could be adjusted for an 80% cutoff, and the locomotive would operate just fine, but generally I think the gear was adjusted to give as early a cutoff as possible while still pulling the load at 50 mph. I think this was common practice.

I have read much of Don Ashton's material regarding the Stephenson's link several times, but don't have a complete understanding of all of what he is saying. I do understand trying to minimize link slip, and I also understand trying to equalize events for a range of positions of the link. Don also touches on using the link design to correct for the angularity of the rod, which is quite interesting as the imbalance between the instroke power and outstroke power can be significant due to the non-linear movement of the piston with regards to crankshaft rotation.

Since I have not operated a compound engine, it is good to hear from those who have to see how it was really done, and for what reasons.

[gondolier88]

Try to get hold of Audels Mechanic's Guide No.1, 1914 ed. It has everything you'll ever need to know about lead, lap, port sizing, cutoff and expansion ratios, cylinder dimensions, valve gear and operation of marine engines- as Jack says this subject just keeps on going.

Every single steam application is different- even in a marine setting where there are 'no hills to contend with', there are still a multitude of variables- overall ideal effiency of the plant, safe operating limits of plant- ie. the capability to superheat or not as the case may be, what the plant is used in- a fast 50ft launch with super-high pitch prop and high rev's, or a low rev. high torque big diameter prop on a tug for example- and take into consideration the fact that the 50ft launch with 80nhp may acheive a decent cruising speed, say 10 knots at 25% power output, yet the tug with all it's horsepower and high efficiency may struggle to cruise with 25% output at anything more than 7 or 8 knots, despite having 3 or four times the power available.

I can see why you would assume that lead should increase with speed, but the opposite is true in reality- the higher the speed it isn't so much a problem of getting steam to go into the cylinder, more a problem getting the exhaust steam out- any that is left becomes the compression for the pre-dead centre position and also creates a near perfect entry pressure for the incoming steam- the higher pressures involved in the compression stage and that of the steam supply create a hotter engine and hence higher efficiency. I assume the reason that you assume lead should increase with revs is that you would require more steam- taking an example of an engine in shortest cut-off full ahead, using the throttle valve as speed control, as revs are increased then higher pressures are experienced in the valve chest, hence higher temperatures, hence more heat input, hence more power output for the same given lead and cut-off.

As far as Stephensons link is concerned it is very difficult to achieve equal valve movements on all valve gear positions, and as you rightly say the engine would be set up to be optimum in certain cut-off positions to the detriment of others. For real equilibrium through all valve gear positions then a radial valve gear holds far more potential, or as you say a Uniflow or Corliss type arrangement.

Greg

[SteamGuy]

Greg-

I have read the Audel's Guide, and reviewed it recently, and while it provides very good basic information, it Don Ashton seems to have taken the understanding of valve gear to a whole new level, and corrected many misconceptions, some of which can be found in Audel's, such as neglecting the angularity of the valve rod.

Audel's uses methods which are approximations, and I believe Don Ashton uses exact calculations, and there are significant differences between the two.

I will have to re-read the section about the pros and cons of crossed vs open rods in order to make a more informed statement about what is good/bad about either.
As I recall, one benefit of open rods and increased lead with hookup is that at high piston speed, the piston may outpace the steam entrance to the cylinder. I will research more and post here.

When I initially started studying valve gear, I assumed it was all rather rudimentary, and it is more of a tar baby in that the more I study valve gear, it seems the less I understand, and the more complex things seem to get.

I have made some progress in understanding valve gear operation (I think).

For the sake of simplicity, I generally try to limit my serious study of valve gear topics to those pertaining to simple D-valves, open rods, and Stephenson's, although I do make comparisions to other valve gear/ engine types for the sake of efficiency discussions. I figure I would be lucky to understand even a single valve gear type like Stephenson's, much less all the rest. I have read "Valves and Valve Gear Design" by W.E. Dalby, 1906, as well as "Valve Gears for Steam Engines" by Cecil H. Peabody, and "Verbal Notes and Sketches for Marine Engineer Officers, A Manual
of Steam Engineering Practice - Vol. 1 & 2" by J.W.M. Sothern.

Obviously, reading the above books had not made me a valve gear expert, but it has raised some interesting if not perplexing questions about valve gear in my mind.

I can't put my hand on my copy of the paperback book by Don Ashton about valve gear design, but he does discuss the methods used by the designers of old, and talks about the shortcomings of some of the methods they used. He also uses modern computer techniques for calculations. I will post some quotes from his book when I find it.

[johngriffiths]

Hi Pat J, Crossed rods are used with piston valves having inside admission such as on the Leak compound in my boat.

All the best

John

[SteamGuy]

John-

Thanks, I vaguely remember something about crossed rods and inside admission now.

I will go back and review that.

Fred-

You can't dismiss any restriction in a steam engine, all will limit the power produced.
You can have too small a steam supply line, too small of ports and passages, too small an exhaust pipe, all can greatly limit the engine power.

So if a Stanley engine cannot effectively release the exhaust at a late cutoff, then the flow of steam into that engine, and the power produced is limited by the restriction. Steam will not flow into an engine if it cannot get out of an engine.

Woud you not agree?

The situation would be that you are forcing 500 lbs into one side of the cylinder, but due to the exhaust restriction, you are building up 400 lbs. on the opposite side of the cylinder (at late cutoff), so you effectively only have 100 lbs acting on the cylinder.