Welcome! Log In Create A New Profile Recent Messages

Advanced

71000 demo's full throttle and variable cutoff

Posted by Bill Hinote 
71000 demo's full throttle and variable cutoff
January 17, 2015 02:09AM
Hi all:

I know there's a lot of resistance to locomotive-oriented references here--but I think this vid clip demonstrates an interesting operating method and one which I believe has a lot of merit.

Have a look at this youtube clip: www.youtube.com/watch?v=GHuYeYttBb8

--knowing that the "Duke" uses the so-called "British Caprotti" valve gear, which features poppet valves and multiple cams that provide an essentially infinte range of cutoff settings--you can see the "driver" (engineer) just pulls the throttle fully open and then modulates the power by varying the cutoff instead. Interesting to note that he started from a standstill with 65% cutoff (probably made practical by 3 double-acting cylinders)--and then goes down to 15% or even less. I understand that this setup is capable of cutoffs as low as 5% or less.

This has to be the most efficient way to modulate power to a reciprocating steam engine IMO. As a reminder the Dan Gelbart engine appears to respond to a similar approach in the demo video I recently posted a link to.

Bill



Edited 1 time(s). Last edit at 01/21/2015 10:53AM by Bill Hinote.
Re: 71000 demo's full throttle and variable cutoff
January 17, 2015 09:28AM
Bill,

I am relatively new to this forum, but IMHO, this is an excellent post! To me a few of these posts from time to time would be very welcome!

I'll preface to say, to me the Duke is right up there with the highest performing steam locomotives ever constructed.....it is right in the class of Clapelon's phenomenal French compound 4-8-4 and the SAR Class 26 Red Devil.....to me currently the highest performing locomotive in the United Kingdom. Its a one-off experimental as the other two mentioned here but a wonderful exercise in researching how far the technology can be pushed.

A few comments and observations...........

1. First, the exhaust sound of the Duke is very unusual but extremely impressive, especially under high load.....at high speed it starts to resemble the sound of an old American Alco or Baldwin four stroke diesel locomotive climbing a grade.

2. The story of the Duke is amazing..........as its initial design flaw, which rendered it a failure at the time of initial operation was remedied in a restoration from a hulk that was so dismembered at Barry, most sane people would have viewed it as scrap. Now it performs at or maybe even better than its designers originally intended.

3. Its an example of how the poppet valve application can highly perform.

4. To meet its a celebration of mankind's ability to create a sophisticated tool to perform work. This engineer is human "PLC"...but seems quite comfortable and knowledgeable in his skill set. Perhaps the modern steam locomotive is the pinnacle of mechanical complexity before machine automation started to set in.

5. Interesting "micrometer" style cut-off adjustment, note it has two handwheels, assumed to be linked together. Assuming he can adjust the intake and exhaust cut-offs separately?.... but keep them at defined proportion to each other? I assume both wheels are spinning together as he opens the cut-off to 65%....but locked together in proportion.

6. In David Wardale's Red Devil book (note: I'll by any forum member a pint if they have read every page of this book!), David strongly advocated wide open throttle use and adjusting cut-off for speed and economy. Secondly, another point raised in the book,,,due to his development of the gas producer firebox, he purposely keeping the firebox doors open a bit for over-fire combustion air. The drivers in South Africa were so conditioned to keep the firedoors shut, David welded stop bars in the opening so they could not. Later Rick Melvin (of the NKP 765) noted the key is to not keep closing / opening the doors.....but keeping a more consistent minimal cool air feed to the firebox....ie minimizing the chance of quenching the flue sheet with a huge cold air shot. Note on the Duke video, the fireman propped the door open at first but later closed them.

7. No waiting!......notice the engineer stated later in the video they were going 75 MPH.

8. Note the bells ringing in the cab, later in the video, based on the engineer's comments, its appears they are connected to a cab signaling, to repeat the line-side signal position....not just visual cab signals. It also appears he keeps pushing a dead man switch.

8. Note the multiple mugs and canister above the firedoor.....a good supply of breakfast tea would be welcomed in this cab!

Seeing the Duke in operation would qualify for my bucket list! A footplate ride......I could only hope.

Thank you Bill for posting this! Very impressive!

(My apologies, this got me a bit stirred up this morning!)

Regards,

Randy
Re: 71000 demo's full throttle and variable cutoff
January 17, 2015 09:29AM
Hi Bill,

The Capriotti valve gear certainly has long been of interest to me. One problem I've had is that it might be hard to scale it down to automotive size, but that one can probably be handled by looking at the function of all the components and designing other that are relatively beefier and do the same thing, even if they don't look a lot alike.

The REALLY big issue is going to be rpm and cutoff. With pushrods, engines design peak valve accelerations of about 450 gees, with overhead cams that rises up to 800. This sounds like a lot of acceleration, but when you start tossing in automotive rpm you quickly start to find limitations on valve travel and brevity of lift duration. It is possible to go higher but component life expectancy drops, those acceleration figures are for a standard duty IC engine that probably typically sees 2000 to 3000 rpm with occasional excursions up to 6 or 7000.

Fortunately, if Art Gardiner is right, cutoff much below 15% probably isn't effective in the real world (at automotive sized cylinders) because the cylinder MEP is reduced greatly while the friction MEP is essentially unchanged and you reach a point of diminishing returns; if anything that takes some headaches away from the valve gear designer as he doesn't need to wring every last bit of cutoff possible out of the hardware.

Ken
Re: 71000 demo's full throttle and variable cutoff
January 17, 2015 10:38AM
Ken,

In reference to your comments on engine speed, in the IC gaseous fueled engines we represent, observing industry trends, with mean piston velocities resulting from an approximate 8.5 inch stroke, crankshaft speeds have been lowered to 1500rpm, albeit we need a gear increase to run our US 60 hz generator ends (1800 rpm). Historically we were at 1800 RPM and it seems we were approaching valve velocity limits. At 1500 rpm, this is still considered a "high speed" engine in our market, once you go to the next classes we offer, medium speed, 900 rpm and 720 rpm, no gearbox is needed but in regards to valve velocities things change on these slower machines. Slower speeds can be helpful but to get the torque and power, iron content and CAPEX costs start to increase. Its all a balancing act between power density, life expectancy and 1st cost.

Even on the Duke, there may be debate if the complexity of the Caprotti is worth the investment, does it really yield a marked performance benefit over a well designed piston valve? Wardale and his mentor Donte Porta (Porta being a disciple of Chapelon), seem to be advocates of piston valves. For the 5AT, should it come to fruition in the future, in which Wardale did much of the design work, will have piston valves.

On the automotive scale, would it make sense to stay with piston valves or does that put you into lubrication issues and superheat limitations? Wardale put valve liner coolers in the Class 26. Would a tail rod help? Due to the large port openings of a piston valve, are you limited in the percentage of cut-off?......with poppets, especially multiple poppets for admission, sequential openings could help you really reduce cut-off for economy.

Regards,

Randy
Re: 71000 demo's full throttle and variable cutoff
January 19, 2015 11:24AM
NYC S16000 Wrote:

> A few comments and observations...........
>
> 1. First, the exhaust sound of the Duke is very
> unusual but extremely impressive, especially under
> high load.....at high speed it starts to resemble
> the sound of an old American Alco or Baldwin four
> stroke diesel locomotive climbing a grade.

I think you're right here--the sound is distinctive and indicative as well of the great efficiency of the loco.

I would attribute this sound to at least 2 design decisions: 1) The use of 3 double acting cylinders means that the number of exhaust pulses per revolution is increased by 50% over "traditional" steam loco designs using 2 outside double-acting cylinders. and, 2) The "Kylchap" exhaust stack not only promotes the excellent breathing the boiler/firebox was designed for, but augments the "bark" that makes this loco so distinctive.

B.



Edited 1 time(s). Last edit at 01/19/2015 11:27AM by Bill Hinote.
Re: 71000 demo's full throttle and variable cutoff
January 19, 2015 05:44PM
Writing from memory here - the Duke as first built did not meet the expectations of its designers and was not very successful. However when it was came into the preservation to be restored the opportunity was taken to correct errors resulting from the fact that it had not been built to the original drawings, not having the right draught inducing funnel and with too small a firebox. With these put right it turned out to be a brilliant machine.

I certainly sounds superb and quite different to any other loco.

Mike



Edited 1 time(s). Last edit at 01/19/2015 05:45PM by Mike Clark.
Re: 71000 demo's full throttle and variable cutoff
January 20, 2015 06:17PM
Hi Randy,

For an optimum efficiency automotive engine I really can’t envision the use of piston valves. You make a good point regarding lubrication, piston valves require oil injected right into the steam and this raises the problems of oil decomposition and oil contamination of the exhaust steam. From a thermal standpoint, the standard piston valve is a bad design, passing cool exhaust and hot admission steam through the same passages is about the same as just venting some of the steam to the air. At higher speeds, piston valves are going to add significant friction losses while the fairly large piston valve cylinder wall surfaces allow for even more heat radiated away from the engine. At high admission pressures, a piston valve seems almost guaranteed to leak, sealing both constant and high pressure is getting to be asking a lot from piston rings. Actually, any valve that employs two surfaces moving against one another seems problematic (sliding or rotating), it’s hard to beat a seating element that lifts away from the sealing surface because you need essentially no lubrication and the pressure can force the sealing elements into contact. I really can’t envision anything else as good as a poppet valve or something equivalent.

The big issue that afflicts poppet valves is flow resistance and turbulence. Designers really need to consult with developers of IC engines, especially performance vehicles, and learn some lessons about developing really clean ducting. Perhaps redefining “cutoff” is worth examination, as well. IC engine valves are not considered “open” until the valve head is a certain distance off of the seat; while flow is possible at a lesser gap the reality is that the small port area combined with short response time and acceleration of admission mass all conspire to limit flow to something negligible. Steam is much denser and such a gap would be smaller, but an engine employing recompression to admission pressure exhibits no flow across the valve until the piston has traveled far enough back down the cylinder for the differential pressure to begin inducing flow. Since the pressure blocks flow at TDC it also follows that we can begin opening the valve at some point before TDC without invoking any real penalty as far as regulating the mass of admitted steam while reducing valve acceleration and allowing greater overall lift. Of course, IC engines have also been made to “breathe” better by adding more valves to the head. Ideally we would employ overhead camshafts to minimize the valve train forces and safely permit higher valve train accelerations.

Regards,

Ken
Re: 71000 demo's full throttle and variable cutoff
January 21, 2015 12:21PM
Hi all:

I did some more research on the British Caprotti valve gear and found several interesting nuggets:

1. The British modification over the original Caprotti consisted of adding a second camshaft for the exhaust valves; the original didn't have this so exhaust cutoff control was a missing detail.
2. The actual range of cutoffs available for the intakes is from 3 to 85 percent. It is reported that the 71000 loco can actually utilize the 3 percent setting under the right conditions.
3. The valve gear is capable of running up to 100,000 miles without service, due to the mechanism running in an oil bath. Also, the poppet valves didn't need lubrication in the steam so the buildup of hardened carbon deposits in the steam passages and cylinders was largely eliminated (a huge problem with engines equipped with piston valves).
4. Possibly most interesting is that the cutoff for the exhaust works in opposition to the intakes, i.e., as the intake cutoff is shortened the exhaust cutoff is lengthened. Why is this? The only theory I could come up with is that at extremly short intake cutoffs (and high speeds) the need to control stack draft pressure requires that a longer time period for the exhaust event is necessary to suit the stack design--without this the stack backpressure would become excessive and performance would suffer. OTOH at longer intake cutoffs/lower speeds a shorter exhaust cutoff raises the exhaust velocity and again the stack draft benefits. If anyone can think up a better theory please chime in here.

Bill



Edited 1 time(s). Last edit at 01/21/2015 12:38PM by Bill Hinote.
Re: 71000 demo's full throttle and variable cutoff
January 21, 2015 01:01PM
Hi Bill,

It could be that the admission and exhaust valves are linked to the same mechanisms and that retarding one advances the other....and vice versa. Other option might be that they figure very low cutoff only occurs at higher throttle pressure and perhaps they want to increase compression in order to compensate for clearance volume losses.


Ken
Re: 71000 demo's full throttle and variable cutoff
January 21, 2015 01:45PM
frustrated Wrote:
-------------------------------------------------------

> It could be that the admission and exhaust valves
> are linked to the same mechanisms and that
> retarding one advances the other....and vice
> versa.

That appears to be exactly what is happening, the intake and exhaust selector wheels are normally locked together and it's easy to assume they create opposite effects when moved in the same direction.

>Other option might be that they figure
> very low cutoff only occurs at higher throttle
> pressure and perhaps they want to increase
> compression in order to compensate for clearance
> volume losses.

Your'e probably correct here, one review of the design states specifically that there was a concern over the inevitable clearance volume increases when compared to the short steam passages between piston valves and power cylinders.

Bill
Sorry, only registered users may post in this forum.

Click here to login