Williams ws. Rankin

Williams vs Rankin

The Williams engine is proven technology. like it or not the Rankin cycle is not good enough and the F Doble is the best you can get using the Rankin cycle. You will not do much better than 12 pounds with F Doble. you are dumping 1100 BTU to the pound with the Rankin cycle. The Williams cycle dumps perhaps half that. It does that because it recycles a substantial proportion of the steam that is used by the engine.

Harry's amazing engine is distinguished by its hi compression which is saying it is almost running on the Williams cycle. Bill's engine is another example of using the Williams cycle because half of it (the HP side) is running on the Williams cycle. there are some more people utilizing the Williams cycle I am not sure about.

Why do we keep wasting time and money on the fundamentally limited tired old Rankin cycle when we have superior technology in the Williams cycle.


Howard

Howard,

Try to explain why the Williams engine is not using the Rankine cycle.
The Rankine cycle is the overall system cycle, not just the engine.
Dobles got down to a 9 pound water rate with the F at 900°F. And the triple in E-12 got a 7 pound water rate.
JC

JC, I was wondering. What would you say. The double triple got 7.5 not 7 pounds. -Running 2 pipes from boiler to engine an exter 2.5% lost . If it was so good why did he drop it. How many cars got better vthan 12 pounds. Other than williams I can think of 2


--

Howard,

Not a double triple, a three cylinder triple that was financed by Leslie Hills and tested in his first Doble, E-12. On a later test with 50°F higher superheat, they showed a 7 pound water rate. Don't believe everything you read in Walton's book.
I have Abner's eight engineering notebooks, Walton did not when he wrote the book. In fact, no one has seen these books. Much from them is in my Doble book, like the two forced draft vaporizing burners.

Why on earth would one want to run two pipes from the throttle to the engine??

The engine was dropped because the shifting eccentric valve gear would often jam and not go into reverse when starting up. Also, they were broke at the time and Hill's wouldn't put any more money in to the new engine.
It was rough when starting up, because they had not developed a satisfactory automatic receiver charging valve that could be depended on every time.

The White could get 11.5 lbs/hr water rate. Not usually in the car on the road; but on a test dyno.
JC



Edited 1 time(s). Last edit at 10/03/2005 09:45AM by James D. Crank.

Jim,

How are you going to reheat. If you do not have a pipe form the IP cylinder to reheater. and one from reheated to the LP cylinder That is 3 pipes with 2.5% heat loss each.
Plus you have to have a re-heater in the boiler. Compared with a Williams you can hook up with a single pipe just like a Stanley. So, no wonder the tripple didn't go anywhere.
a

Howard,
No reheat, just high superheat.
JC

Jim how hi was the super heat

Howard,
Warren Doble told me they tried to use 850°F and had fast valve ring wear.

My old books describe 6 Rankine cycles. Theere are 3 system cycles and 3 engine cycles. The three cycles in each above are full expansion, Partial expansion (or incomplete expansion), and non expanding. Giving a total of six types of Rankine cycles. In newer books only the full expansion cycle is described. Including compression and clearance now (which was ignored as insignificant at the time they were developed) does make it a new cycle. It's still a Rankine cycle. Clearance, compression and pump energy was simply ignored as insignificant. And it thoes are insignificant at the low pressure and low expansions used when the Rankine cycle was developed. They were recognized and left out of the calculation intentationly to simplify calculations. But with high expansion ratios from high pressures they do become significant and should be included.

In any case, multi-stage engines or high expansion uniflow, one is trying to ring the last bit of energy available to produce work. The real problem in doing so is that it reduces the power range at which the engine operates efficiently. When you are using throttling for power control the range of control is deturmined by pressure drop (end of expansion to exhaust) at full throttle. You can throttle until the end of expansion pressure drops below the exhaust pressure. At which point efficiency drops off rapidly.

In going to higher and higher expansions you are also increassing the signifance of clearance losses.

Andy

Andy,
When I went to college there was ONE Rankine Cycle, full condensing. There were several modifications in the later books; but still called the Rankine Cycle, termed reheat and extraction. Perhaps called something else by the particular author to differentiate them for clarity in your books, but still operating on the Rankine Cycle, as our learned professor drilled into our heads.

I have many engineering books on steam that go back to the 1880's and in the diagrams of the Rankine Cycle, all show an "Expander" either reciprocating or turbine. And all explained as a "Device" that takes steam at a given inlet pressure and temperature, expands it down to a lower pressure and temperature thus producing net work output. Feed pump, vacuum pump and burner blower energy inputs, were always termed a "Loss" to be included in the total output and efficiency calculations.
The basic expander efficicncy was to be calculated for the net output depending on the expander's conversion efficiency. Various expander designs were always considered as efforts to get the maximum output under the given conditions, and as designs to enhance the maximum output, with the minimum losses.

I would love to know the titles of any engineering books that call engine design the Rankine Cycle.
JC

Hi Andy and Jim,

Jim


I ran my F on 850f with no trouble. Running the the tripple on 850 F and 800 to 1000 psi, There is not enough BTUs to get a 7 pound water rate. Maybe 10 but not 7.

If you want to give me more info on the tripple, I will look at it again and re examine my calculations.

Andy

Your Rankin history is very interesting. Guess you are right in low pressure days the compression clearance loses were not so important. The Willams is a zero clearance engine. It has four forward cutoffs. The compression is controlled. It will never exceed the inlet pressure.

I am really enjoying this discussion.

Howard

Howard,
Nothing exists for the Hills triple except for some initial bore and stroke ideas by Abner and one original colored pencil drawing of the cylinders. No cutoff numbers and no clearance numbers either, unfortunately.
The notebook says 7 lb/hr water rate on the factory dyno, what else can I tell you?
JC

Hi Jim

I am not sure how to take your post. But I think we are in agreement.

I would say the Williams engine operates on the Rankine cycle.

In the book I use most of the time. Publishing date 1935 it has lots of cycles. But the ones of interest are thoes for piston engines. It has specifics on calculating full expansion cycles(turbines), Partial expansion cycles(reciprocating) and non-expanding(piston pumps) cycles. Reciprocating engines can operate full expanding though unlikely. It then breaks these down into engine cycles and system cycles. The calculation is almost the same. In calculating an engine cycle only the processes internal to the expansion chamber are considered. Aux losses are not considered. No pump. no condensation or heat recovery. Just the engine processes. The system cycles includes boiler and engine and would include pump loss as well as other heat recoverery messures. The book is ment for a collage level engineering course. It targets selection of devices as well as their design. I really don't know how well it fits with industry reality. The idea of the engine cycle coverage is ment for an understanding of selection of engines. In the chapter introduction. It explanes that when comparing engine performance one is interested in engine efficiency as well as thermal efficiency. In this book engine efficiency is to be figured against a like cycle. The cutoff and pressures are to be the same for the cycle as the engine is to run. Engine efficiency is the actual engine efficiency divided by the equilivant cycle efficiency. 100% engine efficiency means the engines efficieny is at the same level as the paper cycle calculation. A piston engine is compared against a like partial expansion cycle. A duplex or simplex operatoring on a non-expanding cycle is compared against a non-expanding cycle.

The is not a lower level book. All the processes are described in turms of their differential equation equilivants. One need to understand differential equations(applied calculus) to get the theory in this work. Though the derived formulas are whats found in simplified works.

My point was that the books I have describe all the basic processes of the "Williams cycle" in the discussions of the Rankine cycle. But compression for example was not included because it was considered insignificant. And that was true back then. The norm was low pressure and not so vary short cutoff.

Hi Howard

The tripple expansion engine I am working on uses compression. I have a MathCad work up for the static analysis and have did some dynamic simulations using MathConex that comes with the older MathCads. MathConex has some problems with memory leaks and runs into problems after a 1000 or so iterations. Enough to get through a few revoloution of degree steps.

It gets complicated. I have interstage recievers that have set min pressures. Makeup steam to maintain the min pressure is regulated in from the main inlet. The recievers have a mixture of exhaust steam from previous stage and the high heat steam. So some heat is being added here. This is a continious thing. Makeup steam is let in when ever pressure drops below min. The stage displacements are designed so as to use at lease as mush steam or more as the previous stage. With some margin in favor of more. Thus the makeup steam requirement. Obviously the efficiency can not be as good as if done in a single stage. But the problem is that in a single stage you run into problems trying to use very high expansion ratios. For one thing cutoff has to be so short you start having flow problems(wire drawing) at low RPM. The short cutoff also runs into mechanical limitations. It's just verry hard to get a valve to completely open and close in the vary short time neccessary. That is the main reasion I am looking at multi stage expansion. A three stage can get 27:1 expansion with little problem. A 3:1 expansion in each stage gives you a 27:1 expansion and even more of a pressure drop with interstage pressure drops at the end of expansion. I am talking a huge phisical engine for it's normal power output. But Oh Boy can you get lots of torque with lower expansion ratios.

The design problems are many though. Space vs efficiency trade offs. Implemention of continiously variable clearance and cutoff and exhaust closing.

The clearance and cutoff are varied so as to always have the same expansion ending pressure. Always above the reciever min pressure. The exhaust steam with a little makeup steam will maintain interstage reciever pressure between it's set min and the end of expansion pressure. Exhaust close is varied to maintain full compression to just inlet pressure. This is the non-throttling mode. Throttling is used from a stop and at low speed.

I am looking at a couple of engine design to implement the idea. A varable pitch Wable plate engine or duel cranks and 2 pistons per cylander. The duel crank engine varies the crank angles of the two cranks with respect to the main shaft. One advancing and the other semetricaly retarding. One +x degrees while the other a -x degrees with respect to the main shaft. These engine have a variable stroke (or vertial stroke in the duel crank). The clearance increasses as we shotten the stroke and viseversa. We increase the clearance while reducing the displacement. The power of the engine is proportional to the steam going through the engine. As we increase the clearance we decrease the cutoff so as to maintain constant end of expansion pressure. And variable exhaust close makes compression back to just fill the clearance space. At low power we have almost 50% clearance to get a 3:1 expansion. At 3:1 expansion a 50% clearance would mean 0 cutoff and no power out. But as we decrease clearance the cutoff aproaches 33% At 0 clearance we have 33.333...% cutoff to get a 3:1 expansion. I'm not saying just how close the actual design will get to the limits here. Just trying to show how the power range will work.

A really nice thing about this is that at low speed we are using very short cutoff where there is more time for the valves to open and close. But as we increase power for more speed we are increassing cutoff. We still have time to open and close the valve maybe without mechanical problems or wire drawing.

You can see were this tripple wont have the problems of the old ones. It should be able to get a good power range in the non-throttling mode as it never over expands. In fact it has almost constant efficienct across it's power band in the non-throttling mode.

In throttling mode I throttling each stage. That way way there is no problem in switching modes as my interstage recievers are still maintaing their pressures. Cutoff and exhaust is also varied in throttling mode for best efficiency and smothe operations.

For even higher power in non-throttled mode cutoff is increassed and over all expansion is lowered for that gut reanching acceleration.

All computer controled of course.

The boiler is also controled by microprocessor. The computer knowing the RPM and all engine parameters can know just about the steam rate required and thus the fuel rate needed to maintain it. This is all done from tables. Feed back from verious sensors are used to correct the tables maintain correct air fuel ratios etc.

This is very different from old compound engines having floating pressures between stages. As you vary cutoff, steam pressure and temperature a stages steam usage will vary. And with the old fixed cutoff points the inter stage pressure vary all over the place. The last stage of a tripple expander may not even have and pressure to use. Way over expansion at low power.

The problem with to common piston engine is that higher expansion is incompatable with throittling.

At full throttle a given expansion ratio produces a specific end of expansion pressure above the exhaust pressure. Now as you throttle down the inlet. The end of expansion pressure also decreasses and aprocahes the exhaust pressure. So as you try to use higher expansions you are reducing your throttling range or over expanding to pressure lower then the exhaust pressure and reducing the efficieny to a great extent.

In the old cars the boiler pressure was well above the point were the shortest cutoff used would even begine to be a problem. Thus having plenty of room to throttl before the end of expansion pressure drops below exhaust.



Andy




Edited 1 time(s). Last edit at 10/11/2005 03:40PM by Andy.

Jim, the numbers say no. However, if he ran a 29 inch vacuum that would explaint it.

Howard


-









--

.

Hi Andy, Your tripple sounds interesting. I hope you get a two-pound water rate. I like the idea of adding hot steam to the receiver. It does sound like you have a big bear by the tail. I hope it works for you. The Williams did a lot of work on cam design.


Howard

Andy and Jim. I alked to Jerry Peoples and he says the Williams cycle is right and you are wrong.


Howard






--
e.
.

Andy,
Can you figure out what Howard is talking about?
JC

Hi Jim

I said that the Williams cycle is just a Rankine Cycle. In fact as far as I know they never did any work on a cycle. It was an engine. How can any one say thay have some cycle or other with out the processes and methods to calculate it. The Williams never did that.

Andy

Andy,
Right, it is just another engine, and it worked on the Rankine cycle.
So, someone tell me please why the Williams did not work on the Rankine cycle.I would sure like to hear that one.
Their engine systems in the car and bus ran on the Rankine cycle.

As far as I am concerned the Williams was nothing more than a high compression expander and only one part of what consists of a Rankine cycle steam engine system.
JC

Hi Guys

Rankine vs Williams

In internal combustion cycle terms, the Diesel Cycle is just a higher compression version of the Auto Cycle yet we make a big distinction between the two. If we distinguish between the Auto and Diesel Cycle why don't do this for the Williams and the Rankine Cycles. There is more difference between the Williams and Rankine Cycles than there is between the Auto and Diesel Cycles.


I felt that Jerry Peoples could express the differences between the Williams and the Rankine Cycle better than I could. On consulting with him he has sent this explanation of Williams vs Rankine Cycles. I hope this helps to clear up the confusion between the two.


Here is the letter I received from Jerry Peoples on the differences between the Williams and the Rankine Cycle.

______________________________________________________________

There are major differences between the well known classical Rankine Cycle and the little known Williams Cycle.

Four state points are required to characterize the Rankine Cycle:
1.
Pump input
2.
Pump output
3.
Supply conditions
4.
Exhaust conditions

The Williams Cycle requires two additional state points.
1.
Compression state
2.
Final mixing state at cutoff between the compression and admission states

These two additional state points necessitate three additional lines on the T-S diagram.
1.
Residual exhaust to compression
2.
Compression to final mixing at cutoff
3.
Supply condition to final mixing at cutoff

It is also noted, therefore, that the expansion line will be different than the Rankine cycle because the steam temperature at cutoff will be 50F or greater degrees than the supply state.

Any legitimate analyzes of the Williams Cycle must consider the effect of compression on the final mixing temperature and the required admission mas as a result of an effective zero clearance engine.

To my knowledge, all analyzes of the William Cycle have been flawed because they have been based on the Rankine Cycle. Rankine Cycle analyzes techniques do not have the fidelity to evaluate the Williams Cycle. The Williams Cycle is therefore unique and distinctively different from the Rankin Cycle.

Jerry Peoples



Hi Guys

Rankine vs Williams

In internal combustion cycle terms, the Diesel Cycle is just a higher compression version of the Auto Cycle yet we make a big distinction between the two. If we distinguish between the Auto and Diesel Cycle why don't do this for the Williams and the Rankine Cycles. There is more difference between the Williams and Rankine Cycles than there is between the Auto and Diesel Cycles.


I felt that Jerry Peoples could express the differences between the Williams and the Rankine Cycle better than I could. On consulting with him he has sent this explanation of Williams vs Rankine Cycles. I hope this helps to clear up the confusion between the two.

Howard

Here is the letter I received from Jerry Peoples on the differences between the Williams and the Rankine Cycle.

______________________________________________________________

There are major differences between the well known classical Rankine Cycle and the little known Williams Cycle.

Four state points are required to characterize the Rankine Cycle:
1.
Pump input
2.
Pump output
3.
Supply conditions
4.
Exhaust conditions

The Williams Cycle requires two additional state points.
1.
Compression state
2.
Final mixing state at cutoff between the compression and admission states

These two additional state points necessitate three additional lines on the T-S diagram.
1.
Residual exhaust to compression
2.
Compression to final mixing at cutoff
3.
Supply condition to final mixing at cutoff

It is also noted, therefore, that the expansion line will be different than the Rankine cycle because the steam temperature at cutoff will be 50F or greater degrees than the supply state.

Any legitimate analyzes of the Williams Cycle must consider the effect of compression on the final mixing temperature and the required admission mas as a result of an effective zero clearance engine.

To my knowledge, all analyzes of the William Cycle have been flawed because they have been based on the Rankine Cycle. Rankine Cycle analyzes techniques do not have the fidelity to evaluate the Williams Cycle. The Williams Cycle is therefore unique and distinctively different from the Rankin Cycle.

Jerry Peoples

Howard,

The Diesel engine cycle is most certainly NOT the same as the Otto cycle.
The Diesel injects fuel at the peak compression condition, relying on the high compression temperature to provide ignition. The Otto uses a spark plug.
The Diesel uses the amount of fuel injected for control, the Otto uses a throttle valve which introduces variation in the pumping loss.
Any good engineering book will clearly explain the differences and why Dr. Diesel invented this cycle. Remember, he called it a "Rational Heat Engine", and for good reason.

As far as the Rankine is concerned, when one calculates the cycle for a specific engine system, each and every loss, and each and every working fluid condition is taken into account, like high or low compression, exhaust conditions, etc..
It is not some simple calculation of just the basic steps in the cycle. To do it properly, in order to see if the particular system is going to work with high effiency or not, one has to take ALL the variables into consideration.

This is always an interesting excercise to go through; but since nothing is going to be done with it, in terms of new hardware, it remains just an academic excercise.
JC

Howard.

My Steam Cycle Calculator does do residual mixing.

[www.greenhills.net]

With any clearance there is the residule mixing to consider. There is mass in the clearance space that combines with the inlet steam regardless of the amount of compression or even no compression. There is always this mixing of the two different steam masses when there is a non zero clearance.

There is still the two classes of three Rankine cycles. The engine cycles look at only the engine ignoring boiler, pumps, etc. As many devices can be run off one or many boilers at a given site the engine cycles was used to compare engine performance independent of the steam source. The system cycle can include all devices in the cycle including boiler, heat recovery, throttle, engine...

As described in my books. Rated engine efficiency of piston engine is bassed against the equilavent engine cycle. There is more on this in my book that I ignored (just glanced through). It is of little use for automotive application. These books were written with large stationary plant operation in mind. I use system cycles.

Jerry ignores all this. Rightly so for moble application were we are interested in system performance. But it is important background to understanding the intent of these cycles and the comprimises made.

There is no real zero clearance engine. The background on the Rankine cycle in my books do make note of there being no real zero clearance engine.

The zero clearance engine Rankine cycle calculation requires inlet properties, end of expansion pressure state, and exhaust pressure pressure state. In the full expansion cycle the end of expansion pressure and exhaust pressure are equal. In the non-expanding cycle the end of expansion pressure and inlet pressure are the same. The text book cycle are greatly simplified. The discussion on the Rankine cycle in my books does talk about compression, pump losses and other things not included in the Ideal cycle. These were left out to simplify calculations(including compression would have been imnpossable at the time). The engine efficiency spec then convayed these other processes to the Ideal Cycle by comparson if the real engins performance to the Ideal cycle. And of interest is the possability that an engine could have greater then 100% engine efficiency. So stated in my book. (Engine efficiency is actual engine thermal efficiency devided by equalivent ideal cycle efficiency). Point in fact is that some Uniflow engine exuhibited engine efficiencies greater then 100% when compared to a like ideal Rankine cycle. That is compared to a partial expansion Rankine cycle of the same expansion ratio. (Old time before turbines).

My cycle calculator uses the method discribed by Jerry for calculating the cycle. It is a system cycle in that it takes boiler steam properties. Throttles that down for the engine inlet pressure. Figures the properties of the cutoff pont as a mixture of residual steam and the inlet steam reguardless of the amount of compression. Residual steam can be at exhaust pressure. It is expanded to the end of expansion in an isentropic process. Droped to the exhaust pressure as an isoenthalip process. Part is exhausted and some part Compressed (or not) to some pressure. The residual steam or isenthaliply compressed or expanded to the inlat pressure. The cycle is repeated.

You don't have to have any compression and there is still the mixing process. There is no differance between the cycles with or with out compression or expansion other then the end results.

The cycle process are:

1. Throttling Process.
2. Mixing Process. (Residual steam with inlet steam)
3. Isoentropic Expansion Process. (Constant entropy)
4. Isoenthalip Expansion Process(*a). (Constant enthalpy)
5. Seperation Exhaust process.
6. Compression Process.

(*a)There is some argument as to wether process 4 should be constant enthalpy. Ted Pritchard says the residual steam should be computed as an isentropic process. I don't agree. But there is the conservation of energy law that must be obayed. And a constant enthalpy process doesn't agree with the conservation law. But nether does the isentropy process sugested by Ted. So there this a problem either way. In my Mathcad analkysis I switched to figuring the uncompressed steam properties using the energy conservation law. My steam books say that the engines valves are examples of throttling processes. Why I made process 4 a throttling process.

At this time I am not all that confident that we know how to figure the residual steam state. I half agree with Tad that the Throttling process is incorrect. But I do not think it should be an isentropic process. There is evidense to the contrary. Posted on another thread, by Peter h., is charts of a Uniflow test that clearly shows the residual steam temperature going well above the inlet temperature on compression. For that to be the case it's enthalpy content must be higher then as if it were an isentropic expansion to exhaust temperature.

The main point of there being a Williams cycle is refuted by: That in any non-zero clearance positive displacement engine(any real engine) there is mixing of residual steam and inlet steam. Real engine do have compression. Duplex and simplex pumps have no expansion process. They still operatoring on a Rankine cycle. I have published books from around 1930 that have non-expansion Rankine cycles described for these types of pumps.

Expansion or not we have a Rankine cycle.
Compression or not it's a Rankine cycle.

For descision though Williams Cycle is OK. But it is just a Rankine cycle with compression. Like the complex turbine cycles also described in my books. There several of these that have special names. But all are still Rankine cycles.

The Williams engine is just a Uniflow engine that can handle over compression in a unique way. The Uniflow engine was never considered to run on other then a Rankine cycle.

At any rate this is a zero gain excercise.

We have no ideal cycle that can figure engine performance percisely. And to really get efficiency we need to be able to figure engine, no system, preformance accurately. There are trade offs. But the system must be very closely matched to the vehical.

I think high compression is the way to go to get effiency up. But the Williams engine is not the best that can be done.

Efficiency is very much dependent on expansion ratio. In the ideal cycle the best efficiency is gotten when expansion is to exhaust pressure. But in the real world that would be a fixed output engine and we need variable power.

Most everybody is looking at throttled engines. As was the Williams. With throttled engines I found the efficiency to farly constant over the power range set by the cutoff(expansion ratio). At full throttle you expand to well above exhaust pressure and lose a lot energy out the exhaust. As you throttle down you lose less and less out the exhaust but the throttling loss brings the loss total to about the same making for about constant efficiency as you throttle until you expand below the exhaust pressure were the efficiency quicklly goes to crap. The expansion rario sets the efficiency. Your sweet spot is full throttle tp the point you over expand. And as you increass the expansion ratio(shorter cutoff) the available power range is decreased. One has to look at the whole power range. From stop to a reasionable road speed and above. We are designing for variable power.

Take a step back. Power should come from the amount of steam used. We vary power by controling steam usage rate. Throttling restricts the flow. Reducing the quanty of steam that can be used. Cutoff controls the amount of steam admitted into the cylander. Cutoff controls the steam usage rate also. If we could operate at full expansion with no flow losses that would be the best efficiency we can get. But figuring in flow we do not wont to go to exhaust pressure. Some ware above provides good exit exhaust flow. But still the highest expansion at full pressure is were we get the best efficiency. However we would have no room for throttling a power range out of such an engine. It would imidiately go into over expansion at the slitest throttling.

This then is the Patterson cycle. It is different then the Williams cycle therfor I can call it mine As as you clame for the Williams it is different then the text book Rankine. Is different from the Williams engine. Mine The Patterson cycle TIC

In this cycle we have the inlet mass of steam amd the residual mass of steam. The expansion is basicly constant. The end of expansion pressure is constant. Say 5 PSI above exhaust. The power range is acomplished by varing the ratio of residual steam and inlet steam. In theory We can go from 0% inlet steam to 100% inlet steam for a power range from 0 up. But in the real world that would not be possable. But in the Ideal we can. So lets figure this out.

We have an expansion from pressure Pi to Pe and then a drop to Px where Pi is inlet pressure. Pe is the end of expansion pressure and Px is the exhaust pressure. Our displacement volume is V. Clearance and cutoff are a percentage of V. An exhaust close is also needed in this cycle. To get 0 power out the inlet valve would not open at all and there would be no inlet steam. Cutoff = 0. To get to the exhaust pressure we have an expansion ratio r. r is the volume ratio at cutoff to that at the end of expansion.

r = (clerance * V + V)/(clearance * V + cutoff * V)
r = ((clearance + 1) * V)/(clearance + cutoff) * v
r = (clearance + 1)/(clearance + cutoff)

We can see from above that clearance is constraned by r when the cutoff is 0 and likewise cutoff is constraned by r when clearance is 0.

But power can be varied by adjusting clearance and cutoff to give a constant ending pressure. Assumming that compression is maintaned to inlet pressure by varing exhaust closeure. (Uniflow not work on this cycle)

cutoff = (clearance + 1)/r - clearance
clearance = (1 - r * cutoff)/(r - 1)

Figuring residual steam to be fully expanded a per Ted then:

Power = cutoff/(clearance + cutoff)

Power = 1 - r * clearance/(clearance + 1)

Power = 1 when clearance = 0, cutoff = 1/r
Power = 0 when clearance = 1/(r-1), cutoff = 0

We have variable power with no throttling and at a constant efficiency. Far better then the Williams which uses throttling.

The problems: With high expansion ratios clearance and cutoff would be impossable to obtaine. Take an expansion ratio of 25:1. Aproxmate of 1000 PSIA to 15 PSIA expansion. The the max clearance is 1/(r-1) = 4.167% and the max cutoff is 1/r = 4.000% For 1000 PSIA steam at 850F expansion to 20 PSIA r = 24.925.

You see our clearance range 0 to 4.167% and cutoff is 0 to 4% are very small. In reality we need thoes max values to be our min values. A minum of 5% clearance is a possability. But much below that is going to be next to imposable. Not to metion wiredrawing etc.. And to get any power range min clearance has to be much much lower then the max. Really really a problem in implementation.

The exhaust close is varied so as to compress to inlet pressure. No fixed exhaust close as in a uniflow. Exhaust close at low clearance high power would be very close to the end of the exhaust stroke. While at low power with the higher clearance it would early in the stroke. In the compound engine I am working on the HP cylander:

at high power has
cutoff 32.96%
clearance 5.0%
exhaust close 9.58%

at low power has
cutoff 9.73%
clearance 41.39%
exhaust close 79.32%

Exhaust close is percent stroke remaining. The expansion ratio is 2.76:1
Work sheet calculating thoes values is in a state of change and not all working again as yet. Values above will probably change.

This engine is a counter flow with seperate independent inlet and exhaust vlaves.

So Howard, these differance make this a Patterson cycle. Right?

Andy

Jim,

I know the difference between the Rankine and the Diesel Cycle. As I understand the reason you don't like Williams is that you feel they made fraudulent claims. Who did they defraud as they never sold any engines. Doesn't everyone claim their pet project is better than it is. There is a fine line between boasting and claims. As I understand it the Doble brothers were prosecuted for fraud by the state of California. In one of your of your earlier posts you said that the Williams brothers tapped danced very well. So it sounds like the Doble brothers taught them to dance.


Howard



n
n

Howard,
You said that the Diesel cycle is just a high compression IC engine, just like an ordinary gas engine. It certainly is not. Read your own post of 11-31.
Not some comparison between the Diesel and the Rankine cycle; but between the Otto and the Diesel.

High compression steam engines are better than the old ones, no one doubts that.
The less clearance volumes the better.
The reason I am still suspicious of the Williams claims, is because they would not let anyone really test their engines outside of themselves, without their supervision of the tests. That alone raises questions.
I mean a real, fully instrumented test with calibrated instrumentation, by competent people, not wild eyed steam groupies.
Kimmel has an engine and wants to test it. Fine, I reserve judgement until that is done.
The whole point that I continue to make, Howard, is just HOW much better. Let us see real proof, not just claims.

The Williams refusal to open all their files, during the "Clean Air Car" era, was fully justified. Letting the government see all one's data only puts it in public domain. That was one required part of anyone involved in those goverment sponsored developments then. You take their money, you expose ALL your work, and under the contracts, anyone could then use what you did for their own use.
They thought they could make some money on it, which didn't happen.

Abner Doble was up to his eyeballs in stock fraud, not dubious engineering claims.
JC

Howard, Jim

We are getting away from the subject throughing shit here.

Howard, My last post was a bit scarstic about the Patterson cycle. The point is the Williams is a Rankine cycle. If clearance is included in the Rankine cycle then of course so is compression. And there is nothing in any text book that prevents a Rankine cycle including clearance in it's calculation. It is so stated in several of the themodynamic books I own. It could be uncluded. And I do included it. It is vary complicated to do such a cycle. And would have taken a great amount of time to calculate. You have the mixing problem. The calculation of mixing of the residual steam with the fresh inlet steam to arive at the cutoff steam properties requires the residual steam properties and quanty. To compute the compressed residual steam propertie one needs to know the uncompressed residual steam properties. To get the uncompressed residual steam properties one needs the endof expansion steam properties. and that requires that we know the cutoff properties. Which needs the properties we are trying to compute for the compressed residual steam. A catch 22. Only with a series of closer and closer aproxamations can we calculate such a cycle.

The processes I use in my calculation is many passes. The cycle is repeatedly calculated until the enthalpy at the cutoff point stabilizes to some set number of siginificant digits. In other words the enthalpy from the previous pass must be within a small percentage tollorance of the curent value for the iterations to stop. The enthalpy |H[i-1] - H| < 0.000000001 H. They must match to 8 significant digits.

With zero clearance there can be no compression. Impossable to compress any substance down to 0 volume. But in the discription of the Rankine cycle in my books. It is stated the clearance is not needed because it is usually insignificant. But in thoes days we are talking of 150 PSIA being high pressure. Cutoff's were never less then 25%. And compression is basicly insignificant. At 1000 PSIA and 4% cutoff clearance and compression are very significant.

There is no Williams cycle seperate from the Rankine cycle. But we all (I think) understand what is ment by the Williams cycle. And your topic. "Williams ws. Rankine". What I take it to mean in turms of Rankine cycles is "Rankine cycle with clearance VS. Zero Clearance Rankine Cycle"

I don't think the Williams engine was the first to reclame over compressed steam. Most uniflow engine have some way to handle over compression. A seperate exhaust valve for example. In some the inlet valve lifts allowing over compression to force steam back into the steam box. Over compression causes some efficiency losses. It is simply better to never over compress. But in a uniflow in a variable torque app it is impossable to avoid over compression in some casses.

In my anallysis I have the found clearance losses to be very significant at short cutoff. It can cause great losses. The is no doubt in my mind that compression is needed to prevent clearance losses. I also found the clearance losses to be around 15% in many casses using normal operatoring parameters of old engines. !5% is the normal fudge factor sugested in my book. It seams that clearance loss could account for 15% or most of the losses seen in piston engines of the past.

But really I don't think we wont or can elliminate clearance. It is simply neccessary. Have to have it. Expansion tollorances and valve clearances etc. Just something we have to live with on a piston engine. Clearance can eliminate most of maybe all of the loss. But with this knolage we can do much more. I explained in previous posts how we can use variable clearance to modulate the torque output. By control the relation between the residual steam and the inlet steam we in effect contol the torque output of the engine. I know it is not going to be easy. Or as simple as just using old steam technology. But it has the pontential to be the most efficient engine posable operatoring from a given steam source. With a large clearance and very short cutoff we have a small amount of inlet steam going through the engine. Most of the steam going through the cycle is residual steam that contributes nothing to the output. The torque from the residual steam basicly goes into compressing it for a 0 or - work. leaving only the inlet steam to produce output work at the shaft. With a small clearance and larger cutoff we have most of the steam going through the cycle being the inlet steam and very little residual steam is being cycled in the engine. The high percentage of inlet steam produces more torque at the shaft. Be varing the exhaust close so as to always compress to about inlet pressure we elliminate clearance loss. The compression energy is mostly given back on expansion. The added fresh inlet steam mostly goes to producing output torque.

The Williams engine still sufferes the throttling losses. It uses fixed cutoffs and throttling. It is not the best that can be done. The best efficiency would ne close to full expansion and no throttling. That is what my engine idea does. It goes for close to full expansion and varies torque by the ratio of residal steam to inlet steam. And it's steall a rankine cycle.

One must look at the range of torque. Your nowmal piston steam engines efficiency varies all over the place depending on it's torque output. AT long cutoff, slow running or gobs of torque, it is not vary efficient. It gets better as we link up reducing cutoff. We are also reducing the torque range as we link up. The higher the expansion the less the torque range we have. If we were to fully expand the steam we would have no torque range.

That last sentance says it all. "If we were to fully expand the steam we have no torque range." That is with the normal engine. And I think that is significant. It says that no matter what we do. In a variable power application a normal engine can not get the best efficiency and have any power range. It simply has a fixed output and that's it. You have no throttling and expansion to exhaust pressure. There is no room room to reduce power by throttling with out going into over expansion with great efficiency losses incured. Variable clearance gets around that problem by being able to control the quanty of steam going through the engine as opposed to the amount of steam going through the cycle. The expansion ratio can be as high as possable/pratical and you have a wide power band only limited only by mechinical constrants. The amount of time we can open and close a valve is the limiting factor.

Andy

Andy,
Indeed we are, and let us end this now. It serves no purpose.

When someone builds an engine that uses very high compression, no throttling; but power control via cutoff, and unaflow exhaust, then we can go at it again.
Look at all those Skinner Unaflow engines, wide open throttle with varying cutoff depending on the load. One in a local laundry here worked that way and was most instructive to watch as the load changed.
JC

Hi Jim

Actually My idea does not work with a uniflow. I require variable exhaust timming. A valve.

In my compound design each stage is around a 3:1 expansion. Looking at the HP stage

At high power: 5% clearance, 30% cutoff and exhaust closes at 10.2% of return stroke.
At min power: 45% clearance, 3% cutoff and exhaust closes at 92% of return stroke.

High power: 85.8 % fresh inlet steam, 14.2% is residual steam
Low power: 6.5% fresh inlet steam, 93.5% is residual steam

A 10:1 torque range. Around a 3.16 speed range or 20 MPH to 63 MPH if the engine is sized just right.

The other stages are much the same. But the compound design does cause some additional reduction of the power range. But I figure 5% clearance doable. At 20 MPH the engine is running slow and the 3% cutoff looks possable.

This is a strange beast. AT low speed, 20 MPH we have the shortest cutoff and the highway speed we are using long cutoff. The expansion ratio is always around 3:1 though. The short cutoff is at low RPM were we have more time duration with short cutoff.

I used a constant displacement in the above example. The variabvle clearance engine designe us a variable length stroke to get the variable clearance. So at low power we have the shortest cutoff and largest clearance and reduced volume from shortning the stroke. There additional torque from the reduced displacement of the shorter stroke that I didn't include in my example. With the added reduction of 57.9% displacement at low power we get a 4.153 speed range. That around 18MPH to 75MPH speed range.

If the engine is the right size. Matched to the arodynamic drag of the vehical that the maxum efficiency can be had over the above power range.\

It's critical that the engine size be matched to the vehical to get this to work. Below the low power above throttling and longer cutoff is used to to 0 RPM. Aboce the max output above the cutoff is just increassed for more power at the cost of efficiency.

Every thing is speculative. The power range may move etc. Could be that 25 to 100 MPH is better. A lot depends on the driving cycle. Were we wont efficient operation and were we are going to scrifice efficiency.

I figure efficiency can be scrificed at low speed as the MPG is not hurt so much at low power.

The power range above the 30% cutoff above is rether limited. Power increase goes up with cutoff increase at a little less then 1:1. So we get very inefficient in a hurry and steam consumption would increase at the same. But only have room to increase a little over 300% and hit the 100% cutoff limit. I don't intend to push it to that limit. But there isn't a lot of extre head room thee.

There are limits to how much of a power range any engine can have. It can get very complicated to figure all variables. I don't know how flow dynamics will effect limits. I am looking at static ideal cycles that of course don't include flow and other losses. But I know top end power will be reduced. But how much is the question.

I am slowly accumulating meterial to build. I plan to make a test rig first to expermently figure this all out. Starting a new thread on the test rig.

Andy

Hi Andy,

I never said anything about your engine. If it gets better than 6.5 pounds, 500 HP
and weighs 350 lbs or less I will be your biggest booster. It sounds like you are comparing
a paper engine with a real engine. Do you have running hardware? We already have drawings of the Williams engine which were made in 2004 with Skinner gear and variable compression. The Williams people have had several meeting with Hal Fuller of the Skinner engine company. We are concerned that the Skinner valve gear will not work at the high RPM of the Williams Engine.

Howard

Hi Haward

Your right my idea is just paper. Not even paper hardware. Just the cycle processes.

The point though is that there are problems not being addressed by just building engines. Not just in my engine design but all.

1. Power control range.

Power range and efficiency are at odds.

This is important in getting good efficiency. What I am looking at here is how the power control effects efficiency over the control range. How the operatoring conditions effect the power range etc.

With throttling. You are throttling the steam down to some lower pressure to effect power reduction. We all know the basic cycle. Take in boiler pressure and throttle it down. With a fixed cutoff(expansion ratio) We can figure the cycle. The inlet pressure and temperature to the throttle along with expansion ratio set the power range. Or at least we can look at the verious cycles over the throttling range. We can throttle down to the point that we expand below exhaust pressure. Expanding below exhaust pressure results in high losses. Ask Jerry.

When you analyze power control by throttling you find some interesting things. And yes theory here does match reality. One thing that doesn't at first seame right is the throttling produces an almost constant efficiency from the max power output down to the point that you start to over expand the steam(below exhaust pressure). That efficiency is basicly set by the expansion ratio. Steam temperature of course effects efficience. I am looking at a constant temperature into the throttle. An yes steam temperature does have an effect on the range. But very little compared to the expansion ratio. Text books tell you throttling produces a loss. Yes. But as you throttle down you are throwing less heat out the exhaust from the under expansion. The two balance out. As you suffer losses due to throttling there are less losses due to incomplete expansion.

Ignoring flow losses for the moment. The best efficiency is had with a full expansion cycle. But using throttling to control power limits you to less then full expansion to have any power range.

2. Matching power range to vehical.
Once we figure the power range(s) one needs to match the engine size to the vehical so that we are useing the power ranges(s) to the best advantage. For example. Say you figure that a specific engine design has a 4:1 torque range. Or power range at a constant RPM. I use power range as most seem to think they understand that term. Actually it's torque range that is of interest. Or work output form the cycle. With a 4:1 torque range we only have a 2:1 speed range.

This all gets a bit complicated. At long cutoff we can more easly pull down boiler pressure and that reduces our range. So there lots of factors that must all be figured in.


The thing with the compression cycle is that several of us theoriests can't agree on how to figure the heat content of the residual steam.

Ted Prichard says that it should be computed as an isentropic drop just as if it had fully expanded.

Jerry, In his paper on the williams treats it as a throttling process. isoenthalpic process.

I orrigionally used isoenthalpic throttling process as my book said the valves in an engine were examples of throttling processes.

After Ted said it should it should be an isentropic process I did a lot more research to find out just what it should be. I do know that neither process will balance out the conservation of energy law. The actual enthalpy of the residual steam has to be some were inbetween to balance out the enrgy equation. But the throttling process does make some sense. If treat it as a throttling process that the total enthalpy of the residual and exhaust should not have change from that of the steam at the end of expansion. If we take Teds isentropic process. The steam in the cylander would be going through an isentropic expansion as it pushes steam from the cylander. So heat energy is being converted to mechinical energy accelerating the exiting steam. And then the is converted back to heat is the mechinical energy is reclamed in the exhaust deaccelerates. Do this in a simulation is the pressure drops in the cylander and it will not balance. The exhaust steam would have to gane the enthalpy that was lost during the isentropic expansion. The exhaust steam would have to gain considerable temperature for the total enthalpy to balance. There is the clasic example of a throttling process described in most text books. You have two chambers. One high pressure and one a vacume. A valve between. The valve is opened and the pressure is allowed to equlize. The enthalpy remains constant.

Ok. Here is a big question. In all my books they apply the energy conservation law to the full expansion cycle. EnergyIn must equal the WorkOutEnergy plus the Energy rejected in the exhaust. But when thay discuss the partial and non expanding cycle no metion of made of ballance energy.

In the partial expansion we have a full expansion cycle (end of expansion pressure being the low pressure) setting on a constant pressure (non expansion) cycle. The constant pressure part is more like a motor. The energy was produced in the boiler and transfered to the motor in the form of pressure. Do we have heat rejected for that part of the work?

At any reat all we can say for sure is that compression can eliminate the clearance losses. But wether it can do more and reduce the partial expansion losses is a different question.

But in any case even using the throttling process (constant enthalpy) for the pressure drop at the endof expansion to exhaust pressure the efficiency never goes above that of the full expansion cycle. On paper it does improve efficiency above that of the equilivant partial expansion cycle. And when you fugure a non-compression cycle with clearance losses against the same cycle only with compression back to inlet pressure that is great gains.

Take the williams at 4% cutoff. Probably around 4% or 5% clearance. If you didn't use comprssion the losses do the clearance could be as high as 35 to 40 percent. Look what can be gained just eliminating clearance losses.

Andy

All files from this thread

File Name	File Size		Posted by	Date
ThrottlingTurnDown.pdf	58.1 KB	open | download	Andy	11/22/2005	Read message
ThrottlingTurnDown.pdf	62.3 KB	open | download	Andy	11/22/2005	Read message
hydro.jpg	123.5 KB	open | download	HLS	02/14/2006	Read message
P1010001aa.JPG	113.5 KB	open | download	Rolly	03/09/2006	Read message
P1010003aa.JPG	66.8 KB	open | download	Rolly	03/09/2006	Read message
P1010002aa.JPG	65.3 KB	open | download	Rolly	03/09/2006	Read message
SingleExpansion Vs Co.pdf	111.7 KB	open | download	Andy	03/10/2006	Read message
SingleExpansion Vs Co.pdf	111.6 KB	open | download	Andy	03/13/2006	Read message
HLS Vs Compound size.pdf	112.1 KB	open | download	Andy	03/13/2006	Read message
HLS Vs Compound =PD.pdf	111.7 KB	open | download	Andy	03/13/2006	Read message
HLS engine.pdf	64.8 KB	open | download	Andy	03/14/2006	Read message
HLS engine.pdf	73.6 KB	open | download	Andy	03/15/2006	Read message
Rankin.pdf	171.9 KB	open | download	Andy	03/15/2006	Read message
HLS engine.pdf	74.6 KB	open | download	Andy	03/15/2006	Read message
Endiing temp 850.pdf	28.2 KB	open | download	Andy	10/02/2006	Read message
Endiing temp X27.pdf	28.1 KB	open | download	Andy	10/02/2006	Read message
Fickett.JPG	66.9 KB	open | download	frustrated	10/05/2006	Read message
Over Expansion 1.pdf	24.2 KB	open | download	Andy	10/24/2006	Read message
FlowSpeed.pdf	23.7 KB	open | download	Andy	11/14/2006	Read message
Material.pdf	16.9 KB	open | download	Rolly	11/20/2006	Read message
white cliffs project engine.jpg	499.8 KB	open | download	grblake	06/30/2007	Read message
SV pickup.jpg	81 KB	open | download	Rolly	07/05/2007	Read message
112908ab.jpg	82.3 KB	open | download	Jeremy Holmes	11/29/2008	Read message
112908b1.jpg	87.6 KB	open | download	Jeremy Holmes	11/29/2008	Read message
Dieter engine.pdf	294.8 KB	open | download	Rolly	11/30/2008	Read message
Bryan Tractor.JPG	108 KB	open | download	Rolly	12/01/2008	Read message
Bryan Engine photos.jpg	84.6 KB	open | download	Rolly	12/01/2008	Read message
p1010002aa.jpg	36.4 KB	open | download	Rolly	12/02/2008	Read message
tractor1.jpg	136.6 KB	open | download	frustrated	12/02/2008	Read message
tractor2.jpg	111.8 KB	open | download	frustrated	12/02/2008	Read message
Tractor3.jpg	137.9 KB	open | download	frustrated	12/02/2008	Read message
tractor4.jpg	159.5 KB	open | download	frustrated	12/02/2008	Read message
tractor5.jpg	113.6 KB	open | download	frustrated	12/02/2008	Read message
tractor6.jpg	98.1 KB	open | download	frustrated	12/02/2008	Read message
071709a.jpg	77.4 KB	open | download	Jeremy Holmes	07/16/2009	Read message