Williams ws. Rankin

Hello Howard & Everyone

I sincerely believe that the valve gear design for my engine will easily rev as high as any steam engine will want to go. Zero to whatever cutoff one wants. Because of the time it takes to open and close a valve, at very low cutoffs some throttling is happening through the valve of course. I gave up on the hydraulic action as it was looking too slow. Really nifty but too slow.

When it comes down to it I am willing to go over the design with Tom as we are already sharing the variable clearance idea. I would love to get it tested on one of the Williams engines.

The compound Williams is comming along very nicely. The efficiency of the Rankine part looks like it is going to be only a slight increase from full tilt to low cutoff. When engine friction is taken into account the efficiency will probably be flat.

I consider the Williams cycle to be a type of Rankine cycle for my two cents worth, consisting of a cycle that recompresses the residual steam to inlet conditions in order to eliminate clearance losses. As Andy said that can result in a major increase in efficiency over that of an engine that has any free expansion of the steam at admission.

The effect of the residual steam being recompressed to a higher temperature than the inlet steam is in need of further investigation I think. Here only good physical tests of residual steam conditions and also recompression conditions will tell us what we are really dealing with. I think this effect would have quite varying results depending upon how the recompressed steam is handled, by being pushed back into the steam chest port area or by being pushed into a variable clearance within the cylinder. In the latter case the residual recompressed steam would have a chance to expand from its higher than inlet temperature without first completely mixing with the lower temperature inlet steam. There should be a slight gain in doing it that way but if the recompression temperatures are really not that high than it becomes moot.

On a test stand recompression temperatures can be varied by giving a shot of cooler or hotter steam into an exhaust port during exhaust thus heating or cooling the residual steam before recompression. This should prove out something about the temperature effect.


Wishing everyone well, and it's good to be back ---------- Bill G.

Bill, If you want to send Tom your valve doring he would be very glad to look at them. Sorry it took me so long to answer. Winter is coming and I had to get some stuff done.

Howard


--


--

Hello Howard,

Winter hit here last night after a rainy fall day. I had to get the heater in the basement lit tonight. I tore out the old octopus furnace quite a few years ago to be able to remodel the basement, and am using two gas space heaters which are actually pretty efficient with their long flues to the stack. I have been working inside most of the year in basements and such doing plumbing and heating boiler installations.

This week I ended up installing a patio door when the cold weather hit. Inside in the summer and outside in the winter, kinda backwards for Minnesota.

I'll get some drawings done up for the valve and cam mechanism and give Tom a call. Our previous agreement was to put both of our names on the patents, if any, $$$, for the variable clearance mechanism. Same for this seems OK with me. I am quite confident that this configuration for valve timeing will do anything a steam engine needs done. It should be good for up to half the RPM of a four stroke automotive engine as the valve train masses are about similar, little heavier. 4 -5,500 RPM shouldn't be too big a problem.

Other than for steam engines this type of mechanism would allow totally variable valve timeing on IC engines as well, so public details will be curtailed as I don't want to give the competition any advantages. LOL, let them continue playing with their electrics and hydraulics.

This is Fun ------------- Bill G.

Hi Bill,
Glad to hear that you are moving on your project. keeping it simple is so imporant. If a door was invented yesterday the new bread of engineers would have all types of electronic relays wires oops solid state proximity sensors ect. When all it needed was door knob.
Gook luck
Harry

I had in mind a latch an'string,,,Ben

Hey Ben,

forgive me if your latch and spring reminds me of the corrliss valve gear on a uniflow engine in a laundry matt as JC describes... clickity-clack on time with rpm and load. Yes im under 40....... actually 33

JW

Hello Harry and Ben, and Jeremy (you young thing),

Yep, keeping it simple. The only hydraulic part is the vane type of device to change the cam timeing. Proven technology I guess. I am thinking that with zero oil pressure when the engine is not running then the timeing would revert to long cut off. Unless someone has a simpler mechanical way to change cam timeing? This is not a sliding cam system.

Question: is it worth much to change the cam advance under running conditions? I was thinking that 5 degrees fixed advance would be OK.

Thanks Everyone -------------- Bill G.

Hi Bill
That is the way we are changing the timing on the cyclone, except we use water instesd of oil. The timing is varibule 5deg retard to 5deg advance however the durationo of the cam also changes. also the clearance volume changes and is rpm dependent of pump preasure and volume,this also adjultableas to these timing changes. The duration has to vari for enhanced torque and for self starting. Mechanicly it will shift to a reverse timing with a long duration .The design of this mechnisum is complicated and took a lot of design hours however is as simple as a door knob in it's present form. may be a little more work and it will work with a string.
Harry

Hello Harry,

I pretty much designed this cam mechanism with my compound in mind, so no reverse, although I suppose for a regular single stage engine it could be made to reverse. Too complicated to reverse the compound setup as the cylinders are not phased 180 deg to each other like a steeple. Also a tranny will be needed for two speeds and neutral so reverse is there anyway.

It will be fun to compair what I came up with and what you came up with, probably next time we meet. I've been at the drafting table all day putting some of it down on paper. Always some revisions when real dementions are drawn out. I am trying to keep it really compact and close to the top of the heads with inertial weights as low as possible. Steam engines just keep growing taller if we let em.

I want a rough calculation as to the RPM of valve float as it would be set up on the compound. Preliminary drawings still look really good as valve train components are light. I have resigned myself to oil in the "crank box" and am planning on running all roller bearings, so using oil for the hydraulics and other controls should be less of a problem. The "crank box" is isolated from the cylinders with packings around the piston rods and a vacuum on the box so shouldn't get too much oil into the steam. Still gotta lubricate the rings and valve guides though.

I have installed more complicated door knobs I guess.

Best of Days ------------ Bill G.

Hi Bill,
Why would you design a compound engine when a high compression uniflow has proven to have a better water rate? It is also much simpler to design and build ie. the williams. The subject of this thread. Some times you can reach the end of a physics I belive the Dobles did that. The numbers on HCU is also known and should have varibals when in operation. A bump valve has good efficency at a fixed rpm (generator). In automotove the engine would have to idle and have a tranmission,all losses. Jay Carter told me he tried almost every thing to get the mpg up but was at the end. We are all waiting the true efficency numbers on the Williams, as there is some confusion as to it is engine eff. or total eff., the only thing that counts, fuel burn per hp.
Harry

Hi Herry

The reasion for going to a compound design over a very short cutoff uniflow is efficiency. Efficiency over a decent power range.

I have goon over this before. But maybe it was not all that under standable. So let me try again.

To start with efficiency is very dependent on the expansion ratio. Given set inlet steam properties temperature and pressure the efficiency is a strong function of the expansion ratio. The expansion ratio is limited by the initial condition of the steam and the phisical limitation of the engine. So the higher the pressure and temperature the higher the expansion ratio can be and thus more efficient. On the mechanical side we run into initia limitation of how fast we can open and close the inlet valve. The shorter the cutoff the higher the rate of valve travel. We run into phisical limitation opening and closing the valves.

I can't use the IFC steam property formulation to calculate expansions from the critical point to atomospheric. The temperature range is limited to 1472F The entropy of saturated vapor at 1 ATM is 1.7568. 3000 PSIA at max temp range of 1472 had an entropy of 1.6798. The temperature would have to be someware above 1472F to have s=1.7568. I am going to use 1500 PSIA as an example.

Looking at the theoritical limits here of expanding to 1 ATM saturated steam the points are:

P: 1500.00 14.6960
T: 1445.75 212.000
V: 0.74069 26.7988
H: 1751.69 1150.48
S: 1.75680 1.75680

Expansion ratio = 26.7988/0.74069 = 36.18

The full expansion implies a specific relation netween cutoff and clearance to get that expansion ratio.

If we could obtain a 2% clearance the cutoff to effect this expansion ratio would have to be 0.819% If we could go down to 0% clearance we would need a 2.76% cutoff. At 0% cutoff the clearance would be 2.843% At 1800 RPM and the 2% clearance the inlet valve must open and close in 0.000273 seconds.

In a real engine we are going to back off from these extreams and allow for some margin. Expansion would be a little less so as not to over expand below exhaust for example pressure.

But you still have power control range limitations with throttling at close to the max full expansion ratio. You have no room to throttle without over expanding.

When you look at the single stage engine operating so close to it's max efficiency potential you are limiting the power range. You have very little clearance and cutoff range to effect power control.

So yes you run up against a wall with that old technology. But trying to utilize the compression idea pionered by the williams you still run into that wall. As long as you use throttling you are going to have to sacrifice efficiency or power range for the other.

But when you try and use variable clearance cutoff to control you run against a wall with the valve open-close time.

The compound engine solves the time problem by dividing up the overall expansion so thet each stage can have less expansion. But with throttling you still have the power range problems. So that is a problem in both cases. But a compound engine does solve the valve speed problem as they have more time to open and close.

The compond does introduce losses by it's nature. In a real engine there must be some pressure drop fromn endof expansion to exhaust pressure. and in a compound that means between each stage as well. Energy is lost to this throttling process.

The compound is said to have an advantage for high expansions by dividing the expansion of some number of stages and thus each stage has less of a temperature change during expansion. The reduces the thermal loss. But with new coating technology and at higher RPM it is a questionable adantage today.

A high compression compound using cutoff and clearance control can be made that will operate over a huge power range at it's best efficiency bu utilizing clearance cutoff control the amount of steam going throw the engine.

I can't see where your very high pressure can be fully utilized in a single stage. It is phisicly impossable to have small enough clearance and cutoff to get anyware near full expansion. But it would give a reasionable power range using throttling. And I can see were it could have better efficiency the what has been in the past and with a decent power range.

Andy

Hello Harry,

I believe that the Williams HCU has proven high efficiency when the inlet parameters of pressure and temperature are within the say 800 - 900 F range. I think the Williams pushed as high as a single stage can efficiently go.

When the inlet conditions are higher the expansion ratio has to be higher also, requiring such a low cut off that there is little power output. The Williams engine is only doing it's thing at low cut off where the expansion ratios are high.

There are two main areas where power and efficiency may be gained. The first of course is with higer temperatures and pressures, but this is not of much use when the expansion ratios are so low that little advantage is taken of it.

For instance with an HCU operating at 1000 lbsa and 1000 deg F inlet and expanding down to 30 psia and 340 deg F. A drop of about 300 BTU, an efficiency of 22% and an overall expansion ratio of 18.9

Now lets raise the inlet conditions to 1800 psia and 1200 deg F. This to stay on the same entropy line. (pressure has to go up fast when we raise the temperatures) Now expanding down to 30 psia and 340 deg F. A drop of 395 BTU, an efficiency of 29% (rounded) and an overall expansion ratio of 30. That would be approaching zero cutoff as the minimum clearance space alone is about there. So little if any power output.

Ok then, if we go back to the 18.9 expansion ratio, then exhaust conditions are roughly 53 psia and 440 deg F. A drop of 346 BTU and an efficiency of 24%. That's a lot of work to gain 2% So much of what was gained by increasing the pressures and temps was lost with the higher exhaust conditions.

If the old Williams engines were updated with modern coatings and designs then the inlet pressure of 1800 could be reduced to 1000 and an 18.9 expansion ratio would yeild around 31 % But to get that expansion ratio we would again be running such a low cut off that the power output of the engine would be very low and the efficiency much eaten up by engine friction.

What I believe we are seeing is that even though the Williams cycle has shown us the way to get the efficiency out of one stage we have found the limits of single stage engines with usable high efficiency power output. To get the job done is gonna take more expansion than a single stage is going to provide.


Now the compound Williams has a minimum expansion ratio at full tilt of 20 to 1. with a 30% maximum cut off. Inlet is 1000 psia and 1200 deg. Exhaust is 23 psia at 303 deg. A drop of 425 BTU and an efficiency of 29% So we can see what a good compounds expansion ratio will do. Since this is at maximum output only a small percentage will be eaten up by engine friction.

At 10% power output (roughly 3% cut off, 20 HP) the expansion ratio is 200 to 1. This puts exhaust into the condenser at (looking at a chart) about 120 degrees 1 1/2 psia and 90 % quality about 1010 BTU. A BTU drop of 607 and an efficiency of 42%. I have outlined before how the exhaust pressure can be taylored to follow the condenser pressure in the saturated area. This after the system has stabilized a bit. Up here in Minnesota a condenser operating at 120 degrees or less isn't unreasonable.

Now as I mentioned previously this is a Williams cycle compound. That means full recompression to inlet pressure in the first stage, and full recompression to first stage exhaust pressure in the second stage. No recievers and no free expansion between stages or anywhere else. The inlet valving and all ports and timing are so open that there should be very little pumping loss. The biggest bugaboo may be in the piston ring friction at low horsepower output (a lot of displacement) but even that is so variable with modern rings, coatings and lubricants.

This is not Grandpa's compound.

Well the Holidays are starting this week. I hope You and everyone has a good Thanks Giving. Tell Frankie I said Hello.

Best of Holidays to Everyone ----------- Bill G.

Andy,

At very low cutoffs we are operating around TDC anyway and there is very little steam to go through the valve. It doesn't have to open all the way so accelerations can be kept much lower. Just give it a bonk and it should work fine.

Also, with recompression the valve can have some advance to get it opening before it needs to be closed for the same cut off. What kind of a cam mechanism are you contemplating? Not classified like mine I hope.

Wishing You well ---------- Bill G.

Hi
Andy and Bill
The engine is not the only thing to be concidered again I am disturbed by eff numbers battered around. the only one that matters is the fuel rate and this is with the total package not just the engine. In a steam engine the fuel is water and it is water rate that counts. I am using 3200psi to incress boiler eff as it eliminates nuculite boiling and controles are effictive and the high pressure incress the power per cu in of the eng. the boiler can operate at 1200deg without burn out. It has no throttel as the eng valve is the throttle as the temp and press remain constant. there is a reheat at the cylinder that increases the cyl temp at induction. If you try to bring the temp to 120f your condencer will be the size of a house. Trying to get expansions that low is like passing the point of demining returns. We chose to use heat regeneration and the cal number is 39% for regeneration(not engine) the real world will be about 25%. But you have to look at the total package. Water lube was the only way we could operate at these temps. And it was not easy. We have engines running and there is a big gap to running and a theory. There are so many problems. I'll give you all the help I can.
Harry

Thanks Harry,

The condenser has to only be sized to the engine for full throttle, something that would have to be done anyway. I figured 90 degree outside air. The condenser temperature can then drop as the amount of incomming steam drops following the engine output. I believe that when I ran the numbers that there was even some additional condenser capacity available at low power outputs. The second stage can also recompress the exhaust steam a bit if it has been over expanded but I'm not sure that is necessary yet.

I agree about the diminishing returns, and engine friction is what will do the efficiency eating. That is why ring friction is my biggest concern. Remember though that the second stage is also the compressor for the Joule cycle, which will add to the efficiency at low power outputs much like a higher pressure reheat cycle would.

I agree that heat regeneration is also a good way to go with a single stage and your engine attests to that, but heat regeneration works best with a large temperature differential from heat in to heat out otherwise the drive temperature of the exchanger eats up the regenerator efficiency. Better explained on paper with lots of doodles I guess.

The engine is enough to do right now. Peter Brow is designing a compact high efficiency Lamont boiler. I had the priviledge of seeing how his mind works. If it is a doable thing he will do it. Besides my engine may use the 550 degree Lamont water for cooling and heating of cylinder walls and valve guides. I really liked Dobles idea there. Peter didn't think it was necessary at first glance. Don't know yet.

Water lube is something I would like to see for the cylinders and heads eventually and your getting it to work in your engine was a supreme accomplishment. Right now though I am sticking with that Mobile oil Rolly came up with for initial development.

I don't see my design as really complicated, just different. I agree about the efficiency numbers but that is all there is to work with untill test stand time. Again though if the efficiency numbers are not good in theory the engine doesn't stand much of a chance, does it?

I appreciate Your offer of help and don't take it lightly.

Thank You ------------ Bill G.

Good morning Bill,
How can you get eff. numbers without water rate? This is the eff. of a steam engine and it can be closely calculated. The more complicated the system the more adding and subtracting will occour through the cycle. Look again at Professor Stumph's 7 losses. This is basic I know but is a good starting place to analize your system.
Harry

Hi Bill

My point applies to what you are doing. See second part of post.

My engine might be called a compound williams cycle engine. But I prefer to just to call it a high compression rankine cycle. It is a counter flow high compression engine. I am working on a valve design that was being developed for IC engines.

My valves are electrico-magnetic latch type. They can open and close in 4ms. The artical gave 4ms as the time but was unclear as to open close time or just transition time make open-close twice as long.

I got the artical on this type of valve mechanism from Tom when I first joined the SACA. He was Editor then. He said that Peter Berrett and others had looked at these and dismissed them because of their slow timmings. Figure how much of the stroke 4ms is at say 1000 RPM. 13% cutoff min. It really would put a cramp on the upper RPM range trying to use very short cutoff.

But the deal is that 8ms is not all that bad in my compound design where at high RPM the cutoff is greater then 30%. That is the neat thing about clearance - cutoff control. At low RPM low power is needed and the cutoff is very short. As the power is increassed for higher speed the cutoff increasses keeping the valve open - close time in the capabilities of these mechanism.

I came up with an idea to make these valve much faster. Basicly it is a series duel valve. I was thinkin of using cylanderical valves. Because of the throw limitation these would have only a small movement. Anyway it is two cylanderical valves. The passage to the engine cylander is centered in the valve cylander. The valves shuttle back and forth in the cylander. They are powered by spring and latched magneticly. Each valve is normaly centered in it's travle between two springs under compression. Initially a high curent in the coil pulls the valve to one end and the current is then lowered just enough to hold the striker plate in contact with the magnet. You have a closed magnetic circuit. There are two electro magnets on each valve. The magnets interact with a striker plate atached to the valve stem. The striker plate is centered, balanced, between the two electro magnets by springs on each side. Once we have the valved latched to one magnetic or the other by the holding curent. We operate the valve we simply switching the curent from the latched side magnet to the other magnet. The spring on the latched side being under higher compression will drive the valve tword the other side. Initia will cary the valves striker plate close to the other energized magnet where it will latch. The origional design of this valve was a singel puppett valve. My idea uses two of these. Two valves and 4 electromagnets. The valves are cylanderical and operate to cover and uncover a common port to the engine cylander. The valves operate on a 4 stroke cycle. One valves shuttles off the hole opening the port to steam passage. Then the other valve shuttles over the hole to close it. On the next revoloution the roles are reversed. Say we lable these valve A and B. On first revoloution valve A uncovers the port and valve B then covers it. On the next revoloution valve B uncovers the port and valve A covers it.

With that setup the closing valve could chase the opening valve. The close could begin before the opening is complete. The open close time could be greatly reduced though with a reduction of open flow area. With a three stage compound I wont need to resort to this idea. The butt seal one would need between the cylanderical shuttle is the stickler something I hadn't figured out, And wont now need.

With throttling you need some extra pressure so that you have room to throttl. It is the ending pressure that you need to think about. The pressure defferance between the end of expansion and the exhaust. That drop sets you throttling range. If the throttle to the point where the end of expansion pressure is below the exhaust pressure your efficiency goes to crap very quickly. For example the 18.9:1 expansion you used as an example 1000 PSIA expanding down to 30 PSIA. Except I have 1000 PSIA 1000F expanding to 30 PSIA as a 15.99:1 expansion ratio. Any it leaves a little any room for throttling. Throttling down to 550 PSIA gives us a 1.97 power range. Efficience 30.34 to 30.54 percent efficiency over the throttling range. A speed range of 1.4:1 47 MPH to 65 MPH for example.

By going to 2500 PSIA 1072,76F(enthalpy same as 1000 PSIA, 1000F) same expansion ratio, you can get a power range of 5.2:1, and a speed rangeof 2.28:1 31 to 70 MPH for example. Efficiency ranges from 28% to 32%. The lower efficiencies are for throttled low power.

Hope attachment is the right version. This forum software took off when I did atachment and I wasn't finished with the post. Attach is really (attach and post).

I'll repost if it's not right.

Andy

The attachment didn't get changed. See nect post please for correct attachment



Edited 2 time(s). Last edit at 11/22/2005 02:13PM by Andy.

Attachment for last post.

Hello Harry,

For an engine without reheat or extraction (regeneration) the thermal efficiency and the water rate are directly related:

Thermal efficiency =2545 / w(h1 -hf)

where w is the water rate
where h1 is the inlet enthalpy
where hf is the enthalpy of the water feeding the boiler normally taken as 180 BTU.

Then (h1 -hf) is the heat added to the system by the boiler.

So w = 2545 / Ef (h1 -hf)

Lets assume the same temperature of return water to the boiler, normally 212 degrees F.. Thus since the BTU content or enthalpy is calculated from 32 deg. I take the enthalpy of the inlet steam and subtract 180 BTU to get the amount of energy added to the water by the boiler. That is the heat added.

The amount of heat that gets converted to work per pound of steam is the drop in enthalpy during expansion. And it takes 2545 BTU per hour converted to work to produce one Horsepower.

Now in the above example of the HCU operating at 1800 psia and 1200 deg: The inlet enthalpy is 1600 and drops to 1209 a drop of 391 BTU (I misstyped 395)

391/ 1600 -180 = .275 or 27.5 % efficiency

The water rate then is 2545 / 391 = 6.5


The compound inlet enthalpy at 1000 psia and 1200 deg is 1617 and drops to 1192 a drop of 425 BTU.

425/ 1617 -180 = .295 or 29.5 % efficiency

The water rate is 2545 / 425 = 5.98


For the compound operating at 10% output the efficiency was 42% and the water rate is 2545 / 607 = 4.2

Water rate can be figured for a reheat cycle also and would be a valid comparative measurement but the extra heat added by the reheat must be added in. It is generally equivalent to the extra work produced with the reheat cycle. I find that with an engine working with higher temperatures to begin with, fitting in a reheat cycle would be a hard and complicated thing to do.

For my compound the only reheat cycle would start at very high pressues and exhaust into the lamont drum of my boiler at 1000 psia. To be effective we would be at critical pressure or above and at about 820 deg. A drop in enthalpy of around 80 BTU. It would probably have to be a turbine stage. It would push that theoretical 42% efficiency to 45% but the complications would be way too much and in the end would probably lower the overall efficiency instead of raising it.

Well so much for water rate, gotta get back to the cam drawings.

Good Holidays to You and All ------------- Bill G.

Hi

Water rate and efficiency are not exactly the same. Water rate doesn't give you efficiency. There is a relation. But efficiency is the shaft work energy out divided by the thermal energy put into the steam. Water is just a carrier of the heat. Higher water heat content will reduce the water rate and increase efficiency. Higher expansion will also reduce the water rate and increase efficiency. But they produce different relations of water rate to efficiency.

What is important is the heat input and heat rejected. The more efficient the engine the less BTU/HP will be rejected and thus a smaller condenser would be needed. By pushing the efficiency up we reduce boiler and consenser size.

Water rate is more interesting for engines running off a common steam source were it has a direct relation to their efficiency. But in analyzing a close coupled system like we are all wonting to develop. The steam properties are not common and water rate alone does not give you efficiency.

The problem is of course the engine size needed gor greater efficiency. The more efficient the engine the more cu in displacement will be needed per HP. Increasing efficiency will neccessate a larger engine for a given power output while reducing boiler and condenser size. But the decrease in boiler and condesner size does not equal the increased space required for the engine.

Analyzing engines is only as accurate as the mathmatic model used. The Text book Rankine cycle is greatly simplified. And doesn't model the real engine very accurately. It is a very simplified model. My analysis increasses it accuracy by including clearance and compression. But still to get closser a dynamic model is needed. One that includes heat transfer and steam flow through ports, passages and valves.

So do you have any figures on you engine efficieny or fuel consumption as yet. I would be very interested to see how efficient you supper critical engine is doing. I was never able to analyze with pressures as high as you are using. The IFC formulations just don't have enough range. But I suspect one would need a dynamic analysis to get very close anyway. With the temperatures and pressures you are running the heat transfer and flow would have greater effects then the much lower pressure engines. The high pressure you are eusing does have significant pump loss(more work required to pump at higher pressure). But that can be off set by greater engine efficiency utilizing greater expansion ratio aforded by higher pressure. But then you need a variable power range and that would neccessate that you do not use all of the available expansion range. Need room to throttle. Or you use some combanation of throttling and cutoff-clearance controle.

Andy

Hi Bill

That conversion factor is:

2544.433577644 BTU/hr = 1 HP/hr

I don't use water rate in my static analysis. And as you allso illistrated efficiency is not water rate. Water rate comparson is only usefull when you have the same heat added to the steam for the different engines you are comparing. Whan we change H1 in your formule we change the relation of water rate to efficiency. Water rate is not a good messure of system performance when the steam properties are widely different.

The efficiency calculated from the text book Rankine cycle is probably no better then using water rate. Water rate probably telsl you the efficiency as close as using the text book method that doesn't include clearance.

I don't think the idea of friction losses setting any limit on expansion is valid. You are going to have friction no matter what. If at the latter end of expansion the force is less then friction resistance then so what. Most of the friction would be would be over head friction. Friction independent of the forces in power path. Like cam follower friction loads. We are still extracting some power from the heat. That means that we do have some power driving the engine and less rejected heat. Less rejected heat = less condenser. So there is a trade off. You have multipal expanders and initia to carry you through stall points.

Andy

Hi Andy,

Yea, I'm considering throttling only for low RPMs and smooth starting. And as a safety shutoff of course. Very low RPMs will probably use low pressure steam routed to the second stage pistons, forgot the name of that technique, Sniveling little valves or something like that.

Anything but the possibility of two hundred horsepower in reverse.

Do you have a TS chart of your engine under the various running conditions? It's rather amazing that our two compounds are so different from each other.

Speaking of large power ranges at high efficiency, what type of range are they getting with automotive diesels. With that 40% efficiency number being bandied about, what is it over the whole power/RPM curve?

I havn't gotten to the inertia calculations of my valve train yet. I pretty much believe the valve spring can do the job, but if I have to I'll use the same cam to shut it to. One way or another that valve is going to do as it's told. (unlike my cat, boy I can get hard on mechanical stuff)

You are very right about the displacement going up with efficiency, but looking at it along with that increased efficiency and displacement is also an increase in power output. My compound is picking off an extra 100 or more BTUs/lb of useful work that a smaller displacement engine could not, or about a third more power at full throttle. 29% / 22% = 1.31.

Another factor for my design is it's extreame compactness, one would have to go with a non piston rotary design to get close.

Enjoy the week ---------- Bill G.



Edited 1 time(s). Last edit at 11/22/2005 04:24PM by Bill Gatlin.

Hi Bill

The steam engine in theory does have a flat relative efficieny curve throttling down to the point you start over expanding (below exhaust pressure). The efficiency is basicly set by the expansion ratio.

We have reports and analsys of running systems that show efficiency not being improved by increassing the expansion ratio. But then we don't have all the info. Probably going into over expansion at low power and cancling any benifit of the increased expansion ratio. Throttling, you reduce the power range when you increase the expansion ratio. I havn't heard anyone talking about this power range problem. So I suspect these old experianced steamers maybe havn't thought about it when they say more expansion doesn't yeald better efficiency. And from what I have seen it will heart efficiency when you overexpand. It really hurts efficiency to over expand. But efficiency isn't the whole story. You may not see as big a loss in MPG when low power efficiency is reduced and high power efficiency gains. That is the case when cutoff is reduced with throttling. At very short cutoff you reduce the flat power range and at low power efficiency drops off rapidly. But then at low speed much less power is needed. You have a cubic relation of fuel to speed because of the arodynamic drag. Fuel consumption is increassing wih the cube of the speed. We can decrease the low speed efficiency a lot and have a small increase in high speed eficiency and still average about the same MPG over all.

MathCad is a powerfun tool. You can analyze over a range of engine cycles and plot the relations. Nice to look at efficiency plot over a power range for example.

Andy

Hi Bill,
Donot fret so much over diesel eff. THe larger diesels get better numbers than smaller ones. from 30% to 38% the 1000hp cat is at the top. diesel burn rate is hp x.05= gal per hr. THis engine is rated at 2300rpm and this nearly constant however is is slighty better is it is losded to 1800. The new VW is hpX.055= gal per hr. This is still not signicantly better than the past,only acceleration over fuel has been improved with electronics. Hardly any dino sheet will go below 1000rpm. also these powers are without transmissions and other acc. The gas engines range from 20% to 24%. Loading will give the best eff.I feel that overdrive would be benificial on a steam engine also.
Harry

Hello Harry,

That means then that if the big diesel is getting 38% @ 0.05 gal/hr then the VW is getting 34% @ 0.055 gal/hr or 90% of the efficiency of the big diesel engines.

If my Williams compound actually could get 29% at full throttle X 80% boiler efficiency, that is 23%. Right up there with the gas engines.

At cruising 40% X 87% boiler efficiency is 35%, in with the diesels.

The difference with a steam engine is that the efficiency goes up at the lower power outputs.

Something to shoot for anyway.

The head design on the first stage is looking really slick. Screwing it into the cylinder like a pipe plug instead of a bolt system was a major improvement. My thanks to the Stanleys for that idea, and to Kevin and Rolly for mentioning it.

I am trying to keep the heads cool so the valve guides and springs don't overheat. The size of the head means that two steam inlets will be needed through it to the steamchest. There is room for two 1/2 inch diameter holes through the head. Does anyone have a good idea or material to line those 1/2 inch passages to help to insulate the head from the high temperature steam. The passages are about two inchs deep and not much metal is left around them.

Also alternate, compact ways of attaching the inlet pipes to the head other than flanges might be useful.

Appreciate any suggestions.

Thank You ------------ Bill G.

Hi Bill,
Cooling the head could be disaster on water rate. We are cooling the valve stem and seal only.
Harry

Hi,,,Vincent,,and Norton motorcycles both put the valvesprings outside away from the heat,on the end of the stem,,1936--1955 era,,,Charlie Metz did this in one of his new designs of 1905 in Waltham Mass,, also,,, re, steam pipe,,Stanley just threaded in a 3/4 pipe,,,not sure why they used a manifold on the ex tho' ,, Herishoff should get the credit for the head thread,, Rolly ,,What year s was Herishoff designing those beautiful,,[as in ARTFORM ] engines,,,Cheers Ben

Hello Harry and Ben,

The inside of the steam chest is designed to be insulated and the valve shrouded to keep the metal temperatures down, and the steam temperature up. This also improves steam flow around the valve by eliminating the discontinuitys of valve guide, valve stem and valve head.

Building the heads this way as an extention of the cylinders separates the upper head, steamchest, and lower head into three separate components, with the cylinder forming the sides of the steamchest. The insulated sheath around the valve and covering the steamchest walls can then be a piece of stamped stainless steel, with (an asbestos like) layer behind it.

I believe then that this sheath will allow the heads to run relatively cool with the major heat proplem being the two steam entry points through the heads.

Thanks Ben, the idea of a pipe screwed into the head triggered another idea. I want the steam entrys to be making a 90 degree turn as they come out of the heads to keep the overall cylinder height dowm. This means then that the head inlets can be like a pipe welded to the head, and that the " inlet manifold" can be attached to them further away from the head. Maybe there is then enough room to put an inner sheath and insulation inside the pipes and heads.

"Artform" is part of the thinking with a good design. A modern steam engine design does give us that opportunity with the formed insulating sheathing. A mechanic can marvel at the inticacys of the machine it'self while others can gaze upon the smoothe curvatures of the covered engine and appreciate it as a working sculpture. Harrys Cyclone engine does both of these very well.

Happy Thanksgiving ------------ Bill G.

Ben
James Herreshoff invented the mono tube coil boiler in 1873 when they rolled all there own tubing. Their boilers were continually tapered tubing, but for their engines I think the best was the one Nat designed around 1898 the steeple triple expansion engines built in four sizes. They were nearly vibrationless. They had balanced hollow cast steel crankshafts with enclosed crankcase and splash lubrication. They had built there own electric furnaces to cast steel. Nat was 52 years old in 1900 and probably at the peak of his genius. Can you imagine most of the bottom paint we still use today was invented by the Herreshoff’s

Hi Bill,
Thanks for the kind words on the cyclone.
We have done a lot of testing on ceramic coatings and most were disapointing at best and very expensive. However there is one material that worked reasonably well and cost about $175.00 a gallon which is cheep for ceramic standards.
SEALMET ZYP Coatings'Inc
120 Valley Court
Oak Ridge, TN 865 482-5715 fax 865 482-1281
ceramics are best for shielding radiant heat and will do little for conductive heat insulation. the lower the density the better for insulation. example a 1/8" styrofoam coffee cup is far better than a 1/2" ceramic cup. The expensilve coatings that claim wonders for ice do not have water in the inviroment where it will desolve even when baked. this is our experiance.
Jerry Peoples wrote some good papers on high compression uniflow that can be purchased at the SACA library. Interesting reading. I still agree with Howard that this is the only modern way to go.
Harry

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