Welcome! Log In Create A New Profile Recent Messages


Engineering a Steam Engine for Automotive Use. A top down approach.

Posted by steamerandy 
Engineering a Steam Engine for Automotive Use. A top down approach.
March 04, 2013 07:59PM
I think the biggest challenge is power range requirement.

First is defining the requirements of a road vehicle. Of course there are differences for different vehicle types.

The main requirement is to be able to maintain a reasonable highway speed and have good control-ability from the lowest speed to the highest speed. This determines the power range requirement.

For those not familiar with scientific terms "power", "work", and "force" and there relation to each other. Power is the rate at which work done.

power = work / time ... see also Horsepower

work is a force that results in movement of some distance in the direction of the force applied.

work = force * distance

The power required is determined from the force required to overcome aerodynamic wind resistance, rolling resistance, friction losses and gravity. The force to overcome aerodynamic wind resistance obeys the square law:

force = C Speed2 were C is a constant dependent on the shape of the object moving through the air.

speed = distance / time

using the definitions of work, power and speed we derive the formula for the aerodynamic power requirement:

power = C * speed3

The speed2 law applies to an air flow which in most automobiles prevails above 20 MPH. So if we just look at the power range from 20 MPH to some top speed say 80 MPH where aerodynamics power is the dominant power consumer. The aerodynamic power range requirement is:

PowerRange = (80/20)3 A 64 to 1 power range.

That is a 4:1 speed range requires a 4^3:1 (4*4*4=64):1 power range. a 5:1 speed range requires a 125:1 speed range. Torque is in a direct relation to the vehicular propelling force and thus varies with the square of the speed.

Roiling resistance decreases slightly with speed. Frictional losses is a percentage of the power.

The power to overcome gravity applies when the path is on an incline.

Lets say you wont to build a sports car with 180 MPH top end. That is about the same as a Dodge Stealth which with out use of the clutch or break did 30 MPH idling in 1st gear. 30 MPH to 180 MPH is a 6:1 speed range that would require 216 to 1 plus power range. But one would need to be able to obtain an even lower speed.

The problem is to do this with a completive MPG to current IC designs.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 08, 2013 09:22AM
Hello Andy,

I have a couple of related questions for you. For those models of steam cars whose performance would match today's IC cars, what was the average (if there was such a thing) size of the engine compartments and how does that average compare to the space available for an automotive powerplant today? Also, does there need to be any shielding between a steam powerplant and the road surface to protect against snow, slush, water, stones, and dirt or dust?

One other thought. Talk about steam cars occasionally ellicits comments about fires and explosions. With the increasing demands on automotive electrical systems, automotive fires are now almost commonplace. It could be argued, based on what I've read in the forum, that steam cars are actually safer now than IC cars. In case your interested, here is a link to a video of a truck fire in a village near me. [www.youtube.com]. Not the Christmas present the truck owner wanted, I'm sure.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 09, 2013 05:36PM
Hi Jim

It isn't just the latest cars that catch fire. I have seen it happen with much older models. Most were several years old when they caught fire. I do not know if the incidence rate is any higher. We are in a different world to day with anybody able to post on the internet.

There are several reports of Stanley's catching fire during start up. With their remote fuel shutoff valve the fire was no real problem. Maybe scorched paint on the hood. Some great stories. No Stanley has every had a boiler explosion. The Stanley that exploded at Knots Berry Farm was due to a propane tank, used for lighting the boiler, that was stored under the seat, venting propane that ignited when it reached the pilot flame.

I would say that a modern burner control, with O2 feedback, would not be that much different then an IC car. Electric cars and hybrid electrics have high wattage demands.

For those models of steam cars whose performance would match today's IC cars, what was the average (if there was such a thing) size of the engine compartments and how does that average compare to the space available for an automotive power plant today?

I really do not think there were any old steamers that match the performance of today's IC completely. There is a lot one could say about that. But the point of this thread is to look at the design requirements needed to match todays needs. And there are many different applications with different design requirements. For example a UPS delivery vehicle and a sports car are quite different. Today we have materials and control technology capable of much higher performance. A steam system basically has 3 major components. Boiler, Engine and Condenser. The design needs to account for the space requirements of the system. But sizing these components to match the power and performance can vary. The delivery vehicle needs to be efficient at low speed and many stops. Matches old designs but would need to be a bit more efficient. A sports car is completely different. Efficiency would be good to have for normal driving around town and highway. But at high output or a test drive on the Nurburgring Nordschleife is a different matter. I am looking at such a vehicle as a target. It would be designed to be efficient under say 100 MPH and drop in efficiency at higher speeds.

It is not a simple thing to design a modern automotive steam power plant. The old cars were not efficient. The main reason is that the engines were not designed to be efficient over a wide power range. They were low expansion engines unable to extract a significant percentage from the heat put into steam. But there are some real problems in getting a high expansion piston engine to have a wide power range.

The better the efficiency at higher speeds the smaller the boiler could be. It is really a problem of designing an engine that can have a high expansion ratio over a wide range of power control. More power range the better. This would be a larger engine to produce power at high expansion ratio. But size could be reduced if the RPM range can be increased. This would of course be a big undertaking. Appling modern flow technology to the engine design. Avoiding throttling as much as possible and using cutoff control of power.

I am looking at using cutoff control while maintaining a fairly constant expansion ratio and taking advantage of compression to eliminate clearance loss. I am current working on a valve design that is a cylindrical poppet valve. It seals like a poppet valve. There would be two cylindrical valves that work together. I am trying to design a series like valve action that can get very short duration open to close timing. The idea of the series valves is that one controls opening and the other closing. This cylindrical design would have them alternate roles. And closing could begin during opening effecting a throttling due to a narrow gap between the valves. My engine design is not necessarily the only solution or the best.

If you look at instrumented data from piston engines you see heat losses negating higher expansion ratio efficiency. The uniflow engine is said to over come most of the heat loss problems in a single expansion engine. It also has high compression ratio that may account for the lower heat loss. It has been shown that separate inlet and exhaust valves also improve the saturation. Multi stage engine are another method for improving high expansion heat losses. These old solutions all have drawbacks when trying to get a wide power range. Typically automotive engines have been primarily controlled by throttling. With a multistage compound engine you quickly have the larger volume stages working against you. The pressure drop of the first stages make the additional stages useless. The uniflow engine has a similar problem quickly going into over expansion. And the compression is a problem and needs a relief valve to solve over compression.

But we still need high expansion ratio to get high efficiency. My idea is to simply keep expansion ratio constant (as close as possible) while varying the amount of steam by cutoff. Usually varying cutoff would change the expansion ratio. But actual expansion ratio is a function of clearance, cutoff and displacement volume:

ExpansionRatio = (ClearanceVolume + DisplacementVolume)/(ClearanceVolume + CutoffVolume)

Usually clearance and cutoff are measured as a percentage of displacement so the above becomes

ExpansionRatio = (Clearance + 1)/(Clearance + Cutoff)

From above it can be seen that of one can vary clearance as well as cutoff and an engine could be designed such that it could have a constant expansion ration while varying cutoff. But with a high clearance there could be a huge loss to the clearance volume. The William brothers designed an engine that could vary clearance of a uniflow engine that was clamed to be very efficient. but it used throttling and the variable clearance overcome the over compression problem of a throttled uniflow engine. But this is a different solution. We need to eliminate the clearance loss. That would be accomplished by varying the exhaust timing to maintain compression back to inlet pressure. The question is what kind of expansion ratio does one need to be completive with todays automobiles. Were are looking at cars with a 50 MPG highway rating. That equates to around a 30:1 expansion ratio or higher in a steam engine. That is a vary short cutoff for a single stage. It wouldn't leave much room for cutoff control in a high RPM engine. If we go to multistage expansions. We can get close to equal power from each expansion by having equal expansion ratios in each stage. The expansion ratio of a n stage engine having an equal expansion in each stage is the nth root of the total expansion. Two stages of equal expansion having a total expansion of 30:1 would have each stage having a 5.48:1 expansion. 5.48 is approximately the square root of 30. A 4 stage engine would have each stage having a 2.340347319 to one expansion ratio. 2.344 = 29.982195 The Nth root gets you in the ball park for equal power from each stage. But the engine needs to be analyzed so that each successive stage uses equal or more steam than the previous and at low power it would still need to be throttled for smother off the line acceleration an at very low speed driving. But to get around lower stage losses when throttling a method where each stage has a receiver of some volume of steam held to some min pressure each stage can be throttled so that each stage is operating between set pressures. By having each stage using as much or more steam then the previous with the extra steam being supplied directly from the boiler to the interstage receivers making up the difference. The boiler steam supplies extra heat to the lower stages. Due to a minimum clearance physically possible the cutoff controlled power range is dependent on the number of stages and over all expansion ratio. The cutoff controlled power range is fairly constant efficiency. We can resort to a throttle control for very low power and lower expansion ratios for higher power. Lower expansion would have longer cutoff and thus more steam resulting in higher power.

I have explained this engine design here before. It may or may not be doable. The concept has come out of theory. The implementation is the problem and several would probably have to be built to get a good idea of size to power and RPM range. Tuning of inlet and exhaust passages would be necessary to get higher RPMs. The engine would be designed to run at the high efficiency mode above 20 MPH. Sense little driving is below 20 MPH the lower efficiency would have little effect on over all driving cycle efficiency. The upper end of the high efficiency would depend on the vehicle use. In general it would be above the normal highway speed of the vehicle, The reason the speed range is important to the engine design.

The engine designs I have come up with vary the stroke of the engine to effect the clearance change. This gives a double effect to the steam usage. Sense the amount of steam is being varied by cutoff to effect the engine power. The engine stroke is being reduced to increase clearance and reduce cutoff maintaining the expansion ratio we are getting a reduction in power from both the cutoff reduction and a displacement reduction. It interesting that the cutoff duration time is almost constant when figuring vehicle speed (torque against wind resistance aerodynamic square law).

My engine concept is a complicated design as much or more so then a modern IC engine.

The placement of the major components of a steam system can be placed together or separated. thus there is a bit more design freedom. There isn't necessarily a one to one comparison to an IC engine compartment space.

Also, does there need to be any shielding between a steam power plant and the road surface to protect against snow, slush, water, stones, and dirt or dust?

That would be a yes. Though no more then an IC engine.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 14, 2013 12:08PM
Hi Andy,

Thanks for the answers. I know I was a bit off topic, but I appreciate the response. Now I have another question. I understand the idea of cutoff to reduce the steam admitted into the cylinder, increasing the amount of work done by the steam used. Could the same effect be had by reducing the number of cylinders receiving steam as the load requirements decreased? Or would this introduce more mechanical complexity and would the savings in steam used be offset by the cooling of the unused cylinders, which others have mentioned as a concern with uniflow designs?

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 14, 2013 05:26PM
"The main requirement is to be able to maintain a reasonable highway speed .............."

Sorry we have had this same discussion at least three times before and the idea above is still nonsense.
The main requiremet is power avalible during a passing manouver or when climbing a grade.

The fact a vehicle can pootle along a 60mph on 6 horse power is irrelevant. If the vehicle hits a hill or pulls out to pass and instantly become a moving road block it is not viable.

Any modern steam vehicle, as a bare minimum, must be able to accelerate as well as say any 1.6L engined modern car to pass say a truck and trailer unit on a grade.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 14, 2013 06:58PM
And that's the reason hybrid cars have batteries. I once had an NSU Sport Prinz with a 600 CC motorcycle engine that would do 100 KPH (60 MPH) all day and get 20 KPL (call it 60 MPG), but when it had to accelerate from 100 to 120 it took forever. If you didn't have to pass at speed, or go up a hill in a hurry, the gas auxiliary in a hybrid would do nicely. Hell, my 1919 Stanley would do 90 on the flat if I had the nerve, but I wouldn't want to have to pass a semi at 90 unless I had all the time in the world. And if I put a body on it that weighed 2/3 of the original, and had a cD less than a sheet of plywood it would do 110, but I still wouldn't want to pass a semi on a hill.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 15, 2013 05:47AM
This is also the reason for many turbo installations fitted to smaller engines. The small engine allows for very good fuel economy at cruise speeds while the turbo provides added punch when needed. The 1.4 liter turbo in my 2800 pound plus Sonic delivers a consistent 38-41 mpg at 70 to 75 mph and managed 47.6 on an exceptional day. With turbo, the engine delivers 138 HP and 148 ft-lbs of torque, more than adequate for the weight. The non-turbo 1.8 liter also delivers 138 HP but only about 125 ft-lbs, and also delivers a few less mpg.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 15, 2013 04:45PM
Hi Mark

Off topic part deleted. (Edited by Scott)

This has not been put into a topic of it's own. I only started the thread with the efficient speed range requirement because I think it is the most over looked. Thanks for the passing requirement. That is certainly a major one also. This topic is not meant for the restorer of antique vehicles. It is meant to address the design from scratch issues.

Mark say's minim requirement is to maintain highway speed on an incline and have reserve for passing. But I would say it is equally important to have efficient running in 20 MPH to 35 MPH. We live a small town of around 10,000 population and only get out of town 2 or 3 times a month. So more then 90% of our driving is in the 20 MPH to 35 MPH range.

The reason I start with the speed range is because of some misconceptions. I have seen posts that say high expansion and compounding do not increase efficiency in automobile engines. And that is probably the truth for the ones that were built. It certainly isn't the case with stationary engine or those in ocean liners. A smart person looks for the reason. Some say theory is wrong or doesn't apply etc. I would say that theory is being applied wrong. Generalizations do not always work because the conditions are outside the parameters used to make that generalization. High expansion always increase efficiency if it is completed. But over expansion can occur when the expansion pressure drops below the external exhaust line pressure. And that is more then likely the case with automobile compound engines. A cycle that calculates the mixing of the residual steam with the inlet steam shows quite a loss if efficiency. For example with no compression and full stroke exhaust. You have the clearance space filled with low heat content steam. The fresh inlet steam is admitted and mixes with the residual steam. Remember this is in a throttled down condition. The heat content of the in let steam may be high, but it's density is significantly reduced so the residual steam make up of the mix at extreme over expansion can be as much as 50% in a 10% clearance engine. But even 30% residual steam in the mix will substantially low the initial heat content of the mix. But the important mix is actually figured at cutoff. So lesser residual steam in the mix. But still a significant lower heat content at cutoff. Upon expansion the enthalpy will be lower so the residual steam is lower. It takes several cycles for the cycle to stabilize. But the just of it is the cycle is running at a lower initial temperature. And drops below the external pressure sooner and ends at a lower pressure. That part of the expansion below external pressure is negative work.

That is why it's important to know the power range of the engine. Throttling or other wise. I figure a throttled engines power range to be from full throttle down to the point were expansion is equal to the external exhaust pressure. That may be a bit conservative. A little over expansion may be tolerable.


Edited 6 time(s). Last edit at 03/15/2013 06:16PM by Scott Finegan.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
March 22, 2013 08:26PM
Hi Jim

One thing in engineering is that trade offs have to be made. Seams nothing is ever perfect. That is the reason I started this topic. As for cutting cylinders in and out of use. It may work in the right circumstances. There are really a lot of unanswered questions when it comes to piston steam engines. Development on them basically stopped when steam turbines became main stream. A few kept at it trying to produce a viable steam automobile.

There has been little advancement in the analysis of the Rankine cycle. The theoretical Rankine does not include the effect of clearance or compression. With modern computers and calculators it is now possible to calculate Rankine cycle including compression and clearance. It takes an iteration method to do so. As the steam at cutoff is a mixture of residual steam and inlet steam. The expansion is not starting at the inlet steam's state but at a mixture state that can not be easily figured. For the mixture depends on the state of the residual steam the residual steam may be compressed or not but it's state depends on the residual state at exhaust close which depends on the exhaust state process from the end of expansion state which depends on the initial state at the start of expansion which depends on that mixture state. It's a circular dependency. It can be solved by iteration on modern programmable calculators or computers. Steam property formulation became available in the late 1960. The latest are the IAPWS formulations. But even that level of profection is not prefect. Things are dynamic in an engine. There is heat transfer for example that is not included in the static Rankine cycle. A dynamic cycle includes time and could get very close to a real engine. That is one of my projects. But even with such a tool it would have to be matched against real data. That is the kind of tool that is needed to match an engine design to a vehicle.

Throttling is limited by expansion ratio. If one goes to very high pressure and temperature the overexpansion is still a limit on the power range. If you have inlet steam at 3200 PSIA and 1200 F with a 30:1 expansion ratio the speed range is 1.75:1 between full throttle and going into over expansion. Overexpansion being the end of expansion pressure going below 14.68 PSIA. The Cyclone engine is basically operating at those conditions. They have an auxiliary clearance tube that reduces expansion ratio. They are hush hush on that so I don't know how well it works or how much it increases power range. But it is a viable solution. Power is fuel consumption. And would vary with the cube of speed if the efficiency is constant. A 1.75 speed range is a 5.36 : 1 fuel consumption variation. Lowering expansion at lower speed is a reasonable approach to keeping MPG high.

One can not design a generalized solution for an automotive power plant. There are a lot of factors to be considered.

Efficiency is talked about a lot. But how does it really factor in. The efficiency of a steam engine is closely bound to it's expansion ratio. But as has already been shown over expansion can drag that efficiency into the dirt. Depending primary on efficiency, power comes from the amount of steam per unit time. The amount of steam then is dependent on the engine size and RPM. As an example. Say at 4:1 expansion an engine can get 15% efficiency and at 28:1 it can get 30% efficiency. That is a 7:1 change of the ending volume on expansion. You would get twice the horsepower per pound of steam so only half as much steam is needed to produce the same power so we would need 3.5 times the engine displacement per unit time to produce the same power, I haven't studied condensers enough to have a good idea of how fluid volume effects condenser size. We would have 3.5 times the volume and half the mass to condense. With increased engine RPM and engine displacement combination the engine displacement doesn't have to increase much. The boiler is only having to produce half the amount of steam per power unit produced.

Note. the efficiency numbers above are for example only of the effect efficiency has on sizing components. No calculations were done and values may be quite different. The point is that efficiency effects the amount of steam usage. Higher expansion requires more expansion volume while reducing the amount of steam required. They have counter effects on power plant size requirements. They do not cancel each other. And we do not have the tools to figure the precise engine fit to a vehicle.

Another example say we have a 3:1 high efficiency speed range. An engine having 28:1 expansion ratio. In a sorts car with good aerodynamics we fit that engine to a 40 to 120 MPH range. Plenty of passing power there. 20 mph at the same efficiency would use 1/8 the fuel as at 40 MPH. At that expansion ratio and an aerodynamic efficient vehicle one might get 50 MPG or better. At 20 MPH having 1/8 the fuel requirement, as at 40 MPH, you would get 400 MPG. decreasing expansion ratio at this lower speed halving efficiency would still be 200 MPG. halving again 100 MPG and again 50 MPG. 28:1 expansion ratio in the Cyclone on dyno testing was reported to be 30% or better efficiency if I remember correctly. Now figure a 5:1 speed range, 40 MPH to 200 MPH, and you would have an very efficient 200 MPH hot rod.

Food for thought.

Edited 2 time(s). Last edit at 03/23/2013 04:38PM by steamerandy.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 23, 2013 07:54AM
Hi Andy,

Very interesting stuff here! Steam _car_ stuff, ahhhh. smiling smiley

I like cars. smiling smiley

I have a new steam car concept engine in the works. Currently no plans to build it, and zero "emotional attachment", so brutal and vicious criticism of the idea is no problem for me.

The concept is, IC-like RPMs, fully-throttle-controlled, early-cutoff/high-expansion, fixed cutoff to simplify valve system, zero residual-exhaust recompression (full stroke exhaust) to minimize engine displacement and size/weight per unit developed horsepower, change-ratio power transmission to driven wheels, and some other features which for now I will not mention to minimize the risk of "Fred Sanford heart attacks" among those deeply emotionally invested in very different steam car engine features.

The tradeoff is that almost all the time on the road, this engine would be in "over-expansion". But to compensate, the clearance volume can be reduced to 2% of the displaced volume. With valving giving an approximately 30:1 expansion ratio, for full-expansion of 500 psi steam at maximum load. Blatant -- but possibly approved -- ripoff of [name withheld; SACA member]'s new poppet-valve concept may make this practical with production-IC-proven valve-acceleration loads.

I accept that over-expansion can "drag efficiency into the dirt". But at the same time, lowering expansion ratios under low-load conditions to avoid over-expansion, will also substantially reduce efficiency.

This leads to the crucial question: which would be less efficient; later cutoff. longer admission times, and lower expansion ratios to avoid over-expansion under low-load conditions, or earlier cutoff and higher expansion ratios, with over-expansion at low loads?

BTW, one other feature which I am considering for this engine concept is a reed valve/check-valve immediately downline from the poppet exhaust valve, which prevents backflow of exhaust steam into the cylinder when the exhaust valve opens with an over-expanded vacuum in the cylinder. Thus the hp used developing cylinder vacuum toward the ends of over-expansion power strokes, would be mostly recovered during the early part of exhaust strokes. Basically the high-recompression-engine "steam spring" operating in reverse, with of course lower leakage and friction losses.

I am looking at an RPM range of 700-rpm idle to 5000-rpm redline. And possibly an Aircooled-VW 4-speed manual transaxle.

Do you have a spreadsheet working yet, which can accurately analyze/compare my proposed fixed-cutoff/over-expansion zero-recompression cycle to the variable/low-expansion full-recompression cycle? Despite my many previous sniffs at "theory", I am well aware of the immense theoretical complexities involved. In fact, those confounding complexities are what tend to drive me toward the much-criticized "nuts and bolts", K.I.S.S., "build what works", "Stanleyoid" approach [NOT to be confused with an "exact-Stanley-replica" approach]. Perhaps these questions can provide a new focus to get past that. The counter-flow engine concept in question handles the internal surface/volume ratio, resulting surface-loss, kinetic balance, ring/valve leakage, and mechanical complexity/cost issues which have long concerned me about "clean sheet" and "all modern" steam car engine designs -- though I still have unexamined/unresolved concerns with developed-hp to friction-loss ratios and other issues which impact overall automobile net thermal efficiency -- and overall cost/benefit balance -- in typical low/variable-load and short-trip/warm-up automobile drive cycles.

Steam _cars_ are a "bear" to design. And no, "as long as it runs on steam it's OK", is NOT my approach. Never has been. Many steam-car designs are absolute garbage in my opinion. I have been saying that for years. Note costly "advanced" White, Doble, and other designs which only get the same -- or less -- real-world-driving MPG as Stanleys of similar weight & air drag. And "design/build/lab-test a high-efficiency steam engine then just stick it in a car" is a slam-dunk guaranteed-loser idea IMO.

This is addressed to Andy, but as always, feedback from all comers is most welcome and will be given serious consideration.

As noted, I have zero "emotional investment" in this concept. Drop the theoretical H-bomb on it at will; no bigster for me, because "Plan A" is well underway and looking good so far. This concept, which I have named the "Hi-X" [HIgh eXpansion] engine, is strictly a thought-experiment at this point.

As a side note, I have driven a 1969 VW Beetle with 1600cc & 44 rated max HP since 1978. Much less than my now-in-construction light steam car's 40 actual max HP at the wheels, even under the most ideal conditions. I have driven the Bug on really long/steep mountain grades, in the Laguna Mountains, Anza-Borrego & Mojave deserts, etc, even in devastatingly [for an aircooled VW] hot weather, passed giant-engine slow trucks and Winnebagos with ease whenever necessary, and slowed down and "took a chill pill" as needed. I always got where I was going, with time to spare, and I always enjoyed the trip. I look forward to duplicating those drives in a car whose horsepower _improves_ as the engine heats up. In Ye Olde Bugges, you can feel the horsepower fade away as that mighty flat-four warms beyond a certain point; time to downshift and back off on the gas pedal.

[I just edited out the last scrap of a wild tale about a dangerous too-fast tailgater I ran into once on a twisty mountain road in Wales with lots of blind bends and no shoulder or turn-outs. The bit that I missed in my sleepy-headed pre-post editing sounded rude out of context; sorry. The point of that story was that on many roads, it is safer & more pleasant to slow down even in a car that can go much faster.]


Edited 1 time(s). Last edit at 04/23/2013 05:06PM by Peter Brow.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 23, 2013 06:36PM

I don't think there is any modern car with enough room to put anything but a small wimpy steam plant in. Most are front drive (CHEAP) four cylinder or vee six and increasingly highly boosted to get the most power out of the smallest possible engine (CHEAP), with way too much electronic game room garbage tacked on. If you notice, the sheet metal around the engine is now so tight you have a hell of a time even getting to the engine, cheaper to pull it out to work on it. That is why my car is a 98 Mercedes E-300 turbo Diesel STRAIGHT SIX, their last one.

The only old steamer that had any performance was the late draft boosted Series E Dobles. An old car designed in 1923, 7,000 pounds, 151" wheelbase, lousy brakes and gets around 8-9 mpg.
The Besler Kaiser and Chevy conversions had nice highway speed; but the acceleration of a condensing squishy Stanley. The hot early Stanleys like the H, K, or Vanderbilt cars would blow their door handles off.
All had big under hood space crammed with the boiler and the engine under the car directly on the rear axle all connected with yards of plumbing. BUT; they were 1920 designs and in no way related to what is available at your local dealer, there is no comparison possible. All were basically 19th century technology with really little improvement.
The usual modern car is built as cheap as hell and increasingly equipped with a minimum displacement engine that is highly boosted and all depending on increasingly unreliable and worthless electronic trash to amuse the driver.Touch screens to even get the damned thing to work, what garbage!!

As far as I am concerned, there is no modern steam car around that is worth a damn, Driven several and they are noisy, get lousy mileage and you are lucky if it goes two miles without breaking down.
It depends on what kind of a steam car one wants, a steam ZR-1 Corvette, a big limo or a FIAT 500 or one of those cookie cutter, mid sized, blow molded "family " cars.
If one is after a steamer that sort of relates to the modern car, then some radical departure from what was used then is necessary. Personally, I want a hot sports car, so compromise is necessary and I accept the complication. Debating fuel mileage is pointless as only when accelerating hard from rest or steep hill climbing are the only times you want high power and they are transitory conditions, cruising down the highway uses little power and there when hooked up and in overdrive, most will do this very well.

After working on many new and old steamers I am sort of settling down on a few thing that make sense to me and no, I will not debate any of them.
The car would be a custom mid engine design like a Porsche Boxter, Corvette front and rear complete suspension assemblies and disc brakes.
No injected cylinder oil, water lubrication only.
2,000-2500 psi @ 1200-1300*F at the engine.
Vacuum pulling off the water tank to get rid of entrained air, like the McCullogh car used.
Six cylinder single acting engine, unaflow and poppet valves. Undersquare, water lubed bearings. Maybe an opposed piston layout, two crankshafts in opposite rotation.
Modest speed like 2,500 rpm max, there is no reason to run a steam engine like a gas engine.
Two speed with neutral transmission or transaxle if one is available to take the starting torque. Possibly on the twin crankshaft synchronizing shaft and in unit with the differential, dog clutches.
I still love the Wankel and a local firm will make one with all iron housings and no water jackets. The old objection of high surface to volume ratio hurts the IC version; but not as a steam engine, insulate the whole engine. Seals are now not a problem with steam.
Lamont coil layout,radial inflow, two parallel circuits, extended surface. 5-15 psi air pressure burner.
Direct water mix condensing with the condenser now radiator to air only cooling water and not seeing the heat of vaporization load.

There are other points; but this gives you the way that appeals to me, of course subject to instant change like any good modern steam car design.
Or honestly, another OO White, the best steam car I ever owned, 14 years of reliable service and not one serious problem. So it liked 35-45 mph, well take your thinking back to 1910 and enjoy watching the scenery go by. Wish to God now I had never sold it, or the Stutz too for that matter. The White will always get you there and the control system worked flawlessly all the time. Magnificent car.
Sure the E Doble was faster, when it wanted to keep running; but it is a sad admission that the first thing before setting out was to call your friendly flatbed truck guy to see if he was going to be in town. Not with the White, it just ran and ran and we always got home in a good mood, unlike trying to use that Doble. Glad to be rid of it.

Now, if Cyclone ever gets the Mk-6 into production with some needed changes, I would finance just such a car using the Cyclone.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 24, 2013 07:27AM
Hi Peter

It's really not possible to figure the actual inefficiency of over expansion as yet. Figuring a static engine cycle process by process is all I have working:
Inlet as a constant pressure process .. admittance_work = volume_change * inlet_pressure.
Expansion as an isentropic process .. expansion_work = initial_internal_energy - ending_internal_energy.
Exhaust a constant pressure process .. exhaust_work = exhaust_pressure volume_change
Compression as an isentropic process ...initial_internal_energy - ending_internal_energy
Any clearance makes for a complicated calculation requiring iteration to solve the circular dependencies.

I havn't been using spread sheets for quite a while. I was using Mathcad but lost steam properties library when my old computer crashed. And the site were I got the plugin is gone. I almost have my VisSim steam property plugin working. I have old development tools that the debugger doesn't work on windows7. I have downloaded WinDbg but cannot get it to work on a DLL (dynamic link library) being called by the release version of VisSim that I do not have debug symbols for.

Peter, You said a 30:1 expansion ratio and 2% clearance. That would be a 1.4% cutoff. I think the clearance and cutoff might be a problem. Looking at cutoff for that kind of expansion is what got me looking at a compound engine.

I accept that over-expansion can "drag efficiency into the dirt". But at the same time, lowering expansion ratios under low-load conditions to avoid over-expansion, will also substantially reduce efficiency.

One thing of note is that longer admittance would be needed for self starting from a dead stop. It is not simple to throttle and increase admittance in concert so as to avoid over expansion.

Jim. You make some vary good points. I can not argue with most of them. Though:

I think that 2500 RPM is fairly high for a steam engine. The dyno runs I have found show a exponentially decreasing power decent above 1600 RPM. Though I think that can be solved applying modern flow improvements as done with IC manifolds. But is an area that needs work. My self I would like to push it to 2800 RPM. High RPM is the key to getting a more efficient engine in a smaller space.

My preference for a steam car as a first car would be something like an Arial Atom. Because it is basically a large road going go cart. I expect to be making changes and easy access to the power making components is a priority with me.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 24, 2013 08:26AM
Hi Jim,

OK, no debate on your steam car propulsion system preferences. For all I know, 10-20 years from now I may be building exactly what you prescribe.

I agree completely about the electric & electronic excesses of current/recent IC cars. I now have several years experience with a "feature-loaded" 2001 model car, "don't get me started". In a few years, many of these gizmos will be obsolete, broken, irreparable, and/or useless. It is as if they have forgotten what automobiles are for. Disposeable cars; run it a few years, then throw the entire thing in the trash like a worn-out Bic lighter, and buy a new one.

It is important to note the distinction between internal and external surface/volume ratio. The external surface/volume ratio is no problem; as you note, just add external insulation. The internal surface/volume ratio, however, is a whole different issue, discussed at length in Stumpf's unaflow books, both editions of which are now on my computer hard drive and well-perused. Internal surface/volume ratio involves the surfaces of the cylinder bores, insides of cylinder heads, piston crowns, and steam passages, which are in direct contact with inlet, expanding, exhausting, and compressing steam. These "steamed surfaces" absorb heat out of incoming and compressing steam, thus reducing the pressure and volume of the steam, and then re-release that heat to expanding and exhausting steam. This is what Stumpf calls "surface losses". These surface losses reduce the efficiency of the engine. The higher the _internal_ surface to volume ratio [square inches surface to cubic inches displacement] for a given engine displacement, the less efficient the engine. The long steam passages between valves and cylinders in "classic" steam engines are often implicated in surface-loss heat-waste, while the higher _internal_ surface-to-volume ratios of multi-cylinder steam engines, and the resulting increased "inlet to exhaust" heat/pressure losses, are often ignored. The more cylinders for a given displacement, the higher the internal surface area and the higher the internal surface losses.

This leads me to "spill the beans" on one feature of my "Hi-X" engine concept which I was reluctant to mention.

It has only ONE single-acting cylinder.

Yes, a ONE-cylinder single-acting engine. For a car.

For the same engine displacement, rpm, and cutoff/compression regime, a one-cylinder engine will have far lower internal surface/volume ratio and "Stumpfian" internal-surface losses than a steam engine with 2,3,4,6, or more cylinders.

Perfect inherent balance and absolutely smooth running can be achieved with this one-cylinder engine. The trick is to have 2 reciprocating counterweights exactly opposite of the single power piston. Each reciprocating counterweight has 1/2 the mass of the power piston, and each counterweight is actuated by a connecting rod of exactly half the mass of the power piston rod. The counterweights must have the same stroke as the power piston, and the counterweight connecting-rods must have exactly the same length as the power-piston's connecting rod. The counterweights and their connecting rods must be equally spaced to either side of the power piston's axis.

This design has been successfully used in one-cylinder IC test engines, for the purpose of eliminating destructive vibration from engine test benches.

It allows perfect inherent engine balance without dual crankshafts, extra gears, multiple cylinders, balance shafts, bizarre swash-plate or cam-drives, and other complications.

Such an engine of course needs a flywheel and higher rpms to carry it from the power stroke through the exhaust/compression stroke, and to smooth-out the "power pulses" of the power piston. The weight of the flywheel can be reduced by running it at higher speed than the crankshaft, and driving it via a polymer step-up synchronous (toothed) belt, which is an increasingly common power-transmission method today, due to its gear-comparable mechanical efficiency, silent operation, low cost, light weight, and high durability.

It bears mentioning that multi-cylinder designs gained favor in internal-combustion engine use because, for various reasons, they allowed better cylinder cooling than one-cylinder IC engines. This became apparent when IC cars with greater power output were developed; gradually the early one-cylinder IC-car engines were abandoned in favor of multi-cylinder engines. Of course, for efficient steam engines, what we want is the exact opposite of effective cylinder cooling. In a steam engine, the ideal design for thermal purposes is ONE single-acting cylinder. That gives both the minimum possible internal/steamed-surface to displaced volume ratio, and also the minimum possible ring-length to displaced-volume ratio, to minimize ring leakage.

The tradeoff is that the single-cylinder single-acting steam car engine requires a starter motor, relatively high rpms, idling at stops, and a clutched multi-ratio change-speed [or continuously variable ratio] transmission, akin to those used in internal-combustion cars. And of course the high rpms require poppet valves.

Now, the energy lost in engine idling and starter motors, and the extra cost, volume, and weight of starter motors and multi/variable-ratio transmissions, and some other factors, may indeed make a no-transmission, 2-cylinder, 90-degree-crank, double-acting, self-starting/reversing, direct-drive, and non-idling "classic" steam car engine more efficient and/or less expensive per mile, overall.

It also bears mentioning that "CHEAP" is far from the problem with today's IC car propulsion systems. In fact, microprocessor engine controls, balance shafts, catalytic converters, increasingly-complex multispeed automatic transmissions, turbochargers, superchargers, IC/electric hybrid drivetrains, etc, have tremendously increased the inflation-adjusted cost of today's IC cars. Today's problem is that vehicle propulsion systems often cost the consumer $5000 or more extra, to save $4000 or less worth of fuel over the life of the vehicle. The extra cost is often concealed from consumers through taxpayer-paid subsidies; EG every electric vehicle receives a $7500 government subsidy in the allegedly "free market" USA. With "fracking" and other technological advances vastly increasing economically-recoverable oil reserves, and thereby stabilizing or reducing petroleum costs, the future cost-effectiveness and long-range democratic viability of fuel-economy/emissions subsidies and mandates looks doubtful IMO -- for better or worse. See the March 2013 issue of National Geographic Magazine for more information. Current projections show the USA as a net oil _exporter_ in a few years, and for at least several decades to follow. California alone now has economically-recoverable oil reserves larger than Saudi Arabia. Today's world is quite different from that of the "Energy Crisis" of 1973/4, which permanently traumatized and transformed so many people's thinking. Let's not even mention the latest 15-year-averaged global temperature data, for anyone who actually follows "that math and science stuff".

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 24, 2013 02:05PM
Actually, I could not agree more on the rpm. Go to high speed and you lose one of the basic charms of the steam car, that lazy yet powerful performance. Between the cutoff and the throttle, there really is no reason to go much past 1500 rpm. One of the ongoing arguments I have with Harry. With a steam engine one has total control over the BMEP, unlike an IC engine that needs rpm to develop the desired horsepower.
I have a feeling that your dyno runs showing a drop in horsepower as the rpm goes up might just be due to lousy port design, the engine cannot breath and/or the exhaust is starting to choke because it cannot flow out fast enough or is looking at back pressure, one more reason why one uses a vacuum on the condenser side.. Steam engines need to breath just like IC race engines. The Stanley had horrible porting if you considered the flow paths. Seems that I read somewhere that Fred complained to Stanley about vibration, which turned out to be caused by lousy breathing in the Rockets engine and that was one of the changes they made between 1906 and 1907. Excessive compression? One thing you will see in Beslers second airplane engine and the Chevy conversion is short, smooth and straight in ports and the removal as much as possible of turbulent flow.

CHEAP meant cheaper for the company to make the car. Engine and transmission in one package meant front wheel drive, not necessarily the best way to put the power down, thinner sheet metal and use the body dies as long as possible. They think that the digital revolution is a Band-Aid to cure their problems caused by stupid political meddling and more rumpus room toys are what people want. Note the flood now of governments trying to make people keep they eyes and brains focused on the driving and not on some needless gossip with someone.
This panic about going after high fuel mileage as the panacea for high oil use only hurts the vehicle design. Pure bio fuel oils in the Rankine cycle or Diesel engine not only reduces oil consumption; but are carbon neutral. Yet the abysmal lack of education on the whole matter in government means the savage waste of our tax dollars on such stupidity as hydrogen, alcohol, battery electric and hybrids. Use the right engine, burning the right fuel and things would be better off.
Too bad I cannot attach the entire White Paper I did for Cyclone to this site but it will not take it , I tried to cover all this as I see it. E-mail me and I can send it for you to review and get a laugh over.

The high surface to volume ratio in the Wankel seriously hurt it as an IC engine because of fuel mixture clinging to the surface and getting swept out the exhaust port, thus that afterburner thing that my Mazda RX-3 had.. As a steam engine, providing that the whole engine is well insulated, once warmed up, the problem vanishes. As an IC engine, they had to cool it or the engine would see destructive temperature rise, combustion temperatures being as high as they are. The steam engine version has no such continuing temperature rise. It still fascinates me, just cannot let go of the idea.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 24, 2013 03:29PM
Hi Jim
I do not think there is a total understanding of a power band. The older steam cars were not that smooth as there was a high torque pulse between piston strokes. Only the inertia of the car and the wheels would help this. The MK 5 is designed to operate easily at a top rpm of 3600 rpm it can easily go to 5000 rpm and has. But this is at full throttle with a 2/1 ratio the car would go over 60 mph at 1800 rpm, 120 at full throttle for a small light weight car of course. The older cars were at this rpm and were totaled out and if they went faster you knew it. This ratio would give a 24" wheel over 2,000 # torque at the start with good acceleration. To drop the rpm and achieve the necessary torque and top speed the engine would be somewhat larger. It is sweet to have 12 power pulses per revolution with no shifting. This way there is no gear change necessary.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 24, 2013 05:20PM

Sorry, but after driving and living with steam cars of all kinds for most of my life, I would never give up the operating characteristics the steam car offers the driver.
The power band with a steam engine cannot be equated with what goes on with any IC engine, the throttle and the cutoff percentage takes care of that, along with a two speed rear end. Some way, you desperately need time driving a steam car.

Modest engine speed is just one of the nice things; but there are others of equal charm and benefit. Always I would instantly choose a larger displacement, longer stroke than the bore and slower speeds of operation that keep the engine silent, highly flexible and long lived and those are just two of the reasons.
As with IC vehicle engines, I always choose larger displacement and modest speeds over some hyper, noisy and short lived contraption. I dislike grenade engines.
Think of the recent Cadillac 500 cu. in. V-8 vs. some screaming 3 liter Ferrari and I certainly have lived with both. The Cadillac got better mileage too. The big V-8 needed zero repairs, while that Italian whoopee wagon was always breaking something and needing fussy service.
No, high engine speeds with a steam engine buys one nothing and I will never accept that.
My Mazda RX-3 would happily rev a bit over 10,000 rpm after I got through with it; but I seldom went that high unless some Porsche needed a lesson in who could out drag who.


That is not how the fire killed Wayne Nutting and his wife at Knots Berry Farm. Bill Marsh and I were there the weekend before and I drove the car.
Wayne had a huge propane tank behind his shop to fire his heat treating furnace and also to refill the propane tank that ran his Stanleys pilot. He overfilled the inverted tank until the vent valve was spitting liquid propane, there was no 20% cushion on top as you are supposed to have. The tank sat on the floor in the tonneau right over the cylinders of the engine and that floor got really hot. It was a hot weekend and no wind at all, plus the parade was just creeping along. The safety valve started venting from the overheated tank and the propane filled the tonneau with gas.
Bill Schutzs' wife and kids were riding in the Stanley and she lit a cigarette. The newspaper photo showed a huge ball of flame with only the bottoms of the tires showing. All were badly burned.
Right after, both Fran Duveneck and I got rid of our propane pilots and went back to the original ones.


Edited 5 time(s). Last edit at 04/24/2013 09:41PM by Jim Crank.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 25, 2013 08:45AM
Hi Jim,

I share your enthusiasm and esteem for low-rpm steam car engines. As noted, my high-rpm engine concept is merely a "thought experiment", not something which I currently plan to build. My current low-rpm project continues. Though I did just hear from a SACA member who is working on something very similar to my high-rpm engine concept, and I will be keeping an eye on his results.

The smooth, silent,"effortless power" feel of low-rpm, big-displacement steam car engines is something which really does have to be experienced to be believed. Old literature tells you, but until you feel it in person, it is easy to dismiss the reports from innumerable witnesses over several generations as mere ancient advertising hyperbole or excessive enthusiasm. I had read about and "believed in it" for decades, but when I finally actually experienced it in person, even a "believer" like me was astonished.

Some folks unfortunately have encountered ill-kept "classic steam cars" with innumerable clanks, hisses, jerkiness, and other defects; this "fit the profile" of what they [and, sadly, even the cars' owners] previously/always expected from "worthless old technology", confirmed their prejudices, and "case closed". They will never take it seriously again. "Sure, the pump drive clanks. That's what stupid antique steam cars do. If it was possible to avoid that, then it would have been done 100 years ago. It is an old and therefore inferior and out of production design. What? Turn a couple of screws to adjust the pump-drive bearings, like the Owner's Manual says? What are you, one of those morons who thinks that steam cars can actually run _quiet_ like the old advertising baloney says? Get real man, this is an _antique_ car; it is _supposed_ to run noisy and wasteful! Just like antique gas cars. I _own_ one, so I _know_. You _don't_ own one, so your opinions and speculations don't matter. So _what_ if you rode long distances in somebody else's steam car with silent pumps 2 inches under your feet!"

I, on the other hand, was fortunate enough to ride along in several superbly restored and expertly maintained classic steam cars, which left me with absolutely zero doubts about the luxurious and delightful potential of the technology.

Yes, there are a few IC cars which give a "hint" of the potential of the large-displacement low-RPM car-engine design approach. Older Cadillacs, for those who have experienced them, are good examples. I will skip my usual numerous "Old Cadillac" stories. Suffice it to say that they run with a particular sort of smooth and quiet power -- and that a few well-kept classic steam cars actually run _far_ better.

I got a kick out of your "Italian whoopee wagon" comment.

Andy: Rats. Bad luck. Well, nailing down all of the factors involved is very complex and difficult. I do appreciate your comments outlining the general analytical approach needed. I think that you and I share the same hope for future advanced software which one can input a number of steam power system variables into, and get back a lot of performance and efficiency projections -- which come usefully close to actual test results. My current boiler design is based on some advanced professional engineering spreadsheet analysis, which as far as I am able to calculate, matches results with similar equipment and traditional pencil/paper analysis methods. I strongly suspect that engineering "Gordian Knots" which we are stumped by, will be point-and-click, instant-results, no-problem stuff for ambitious younger steam guys, in the very near future. It sucks being part of the generation(s) which started out with punch-cards and optical-card computer tech. Yeah, my first programming experiences were penciling-in ovals on cards, one card per line of BASIC. 50 lines/cards was considered a huge complicated program, good luck debugging that. Any day now they will put me in a dusty glass case in a museum. "Actual former optic-read paper-card-input/tractor-feed-printer-output computer programmer". Right next to the mannikin display of cavemen hunting wooly mammoths. Groups of schoolkids shuffling by will yawn. Well, maybe we are barely young enough to accomplish something useful/fun anyway! LOL


My goal is the power-to-weight ratio of the hotter early Stanleys -- and beyond. My first build is merely a zippy urban runabout to work out the basic technology and various improvements.

Edited 1 time(s). Last edit at 04/25/2013 09:55AM by Peter Brow.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 25, 2013 01:40PM

I had better clarify what I mean by saying "larger displacement." I don't mean some radical increase, like the gas cars went to back when, a modest increase is all that is wanted.
Like old gas cars, displacement increases went sky high because with the lousy gasoline detonating and low BMEP and rpm they had, if you wanted higher power, you used the boring bar. Cars like the Pierce Arrow 66 or the Stearns 45/90 had six cylinder engines with 5" bore and 7" stroke and displacements like 875 cubic inches. I have driven both and the extreme flexibility in high gear was simply splendid. Or to be more up to date, that 500 cubic inch Cadillac vs. the 3 liter Ferrari, both giving about the same horsepower.

With the steam engine, you have total control over the BMEP, which you do not have in the IC engine, so getting huge starting torque is right there and what horsepower you want is also there without having to have some screeching high rpm destroying the peace and quiet. The resulting extreme flexibility is preserved too.
Take the Cyclone Mk-5, a single acting, six cylinder engine of 2" bore and 2" stroke, running at very high pressure. OK, for a vehicle engine, my thoughts run to an engine of perhaps 2-3/4" bore and 3" stroke, having around 2,000 psi on tap when high power is wanted and some 1500 rpm maximum. Nothing drastic and not needing any 3600 rpm to develop the 300 hp, but more like that Cadillac, slower, quiet and giving wonderful flexibility, one of the major delights in driving a steamer.
Only long driving experiences gives the education, a brief spin around the block is far from what is needed if one wishes to pursue this wonderful engineering challenge.

You are quite right, far too many old steamers, and too many of the modern efforts, pay no attention to how their car is running, they are delighted that it is running at all. To impress, that just is not good enough by a long shot.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 25, 2013 05:10PM
Hi Jim,
It is piston speed that counts. The MK5 top is only 3600rpm not a screamer by IC standards and it is smooth as an electric motor. Check out the U-tube runs. Only the tac showed the rpm as there is no exaust noise no chuga chuga. Only a combustion deep throated hum.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 25, 2013 08:26PM
Sure, piston speed is important; but not all that important compared to how you get there.
A modern car wants to stay below 2500'/ minute, or a highly stressed race engine that can briefly use 4200'/minute between overhauls, because it is worn out and then blows up in a really spectacular fashion.
You can use a normal car engine with a stroke like 3-1/2" or so running at a livable speed and lasting for decades with reasonable care, OR you gan go the Formula One route, with a 3" or so bore and a 1/2" stroke with a speed of 19,000 rpm, lasting for a few hours and costing a few million. (I exaggerate of course; but you see they way these engines differ) The design philosophy between these engines is light years apart.
That is the difference between reasonable speeds and an all out race engine running at the limits of usable technology.

Because the BMEP in a steam engine is totally controllable, depending on throttle opening, the cutoff percentage and the piston area, plus the inlet pressure, things the IC engine cannot immediately manipulate, they are more or less fixed when the engine was made.
Unlike an IC engine, the torque and horsepower in a steam engine is certainly not all that dependent on just the piston speed and rpm and it has a very wide power band from start to its maximum speed. One is a long lived and highly flexible, quiet, while the other has a seriously tiny operating speed range and skirts on disaster starting the first time it runs.

Basing everything on just the piston speed is not the all commanding parameter one should base anything on. There is absolutely no justification running a steam engine fast, when the same performance is had at a speed that insures long ring life, silence and dramatic grand flexibility. My God, E-14 has now gone over 675,000 miles of service with only three ring jobs that I know about. That is steam engine reliability!!
NO, anything smacking of high speed in a steam engine is just the wrong way to go, the IC engine needs it; but steam engines do not.

Edited 1 time(s). Last edit at 04/25/2013 08:33PM by Jim Crank.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 26, 2013 06:46AM
Hi Jim,

Good points on steam engine speed, etc; I agree completely.

Ah yes, the huge early gas car engines. I have heard their cylinders described as "coffee-can-sized". So many fascinating/colorful stories omitted here for time reasons. "So little time, so much to do" [quoth the Nowhere Man, in the light-hearted yet endlessly insightful "Yellow Submarine", now free-view on YouTube].

Right, larger engine displacement _within reason_. Not the over-the-top IC coffee-can cylinders of yore.

Likewise lower rpms, _within reason_. Like 1000-1500 rpm max, not 100-150 rpm and 10x the displacement.

The argument for higher steam engine rpm and piston speed is that with more power strokes and/or piston travel per minute, the engine can be made smaller-displacement, more compact, and lighter weight for a given horsepower output. This is a popular point of view among "modern steam car" guys. For psychological stress-reduction and limited-time reasons, I will not get into that or any other debate. I am merely acknowledging "the other side" of the debate. It does make an interesting thought experiment though, and maybe something can be made of it in the future, with innovative hardware. Maybe for small stationary or other engines, rather than car engines?

Great idea Jim; just say "I won't debate it". I like it!

Where is it written that people _have_ to argue all the time? What's wrong with "agree to disagree"?

For Forum visitors who may be new to/learning steam engines, I want to mention that the practical piston-speed limits of Stanley and most other "classic steam engines" is around 750 feet per minute. Andy has mentioned the simple/easy/correct formula for calculating this, straight from the old steam engineering books. Compare to the 2500-4200 feet per minute gas-engine piston speeds which Jim mentioned. And of course you can stay at the same piston speed, while getting higher RPMs, by using a shorter piston stroke. Likewise, a longer stroke with the same piston speed means lower RPM.

Most new gas cars have overdrive gears to _reduce_ gas-engine RPMs in highway cruising, for better engine durability. Faster is NOT always better -- even with the very latest and most-advanced IC engines/cars.

Maximum piston speed is governed by engine "breathing". I suspect that a really high-piston-speed/high-rpm modern steam engine would need 2 or more exhaust valves [multi-valve IC engines are very common nowadays], or very large cylinder-wall exhaust ports if valveless unaflow exhaust is used. Large steam chest around the inlet valve too, perhaps fitted with muffler-like baffles to break up flow-blocking "standing pressure waves" generated by inlet valve opening/closing. No using small pipe from boiler to inlet valve, if really high rpms are desired. Alas, lots of experimental high-rpm steam engines have had small inlet pipes to valves, and no steam chest. The higher the piston speed/rpm, the larger the steam chest has to be; something the 19th/early-20th century practical steam engine designers well understood. This long-established principle seems to have been overlooked by many modern steam experimenters in the mid/late 20th century.

In the old steam engineering books, piston speed is mainly used to figure out port size. That 750 fpm piston speed limit is what you get with typical slide and piston valves and classic valve actuating mechanisms, with average port sizes and average passage length, shape, and smoothness. For the same size port, you can get higher piston speeds with shorter, straighter, and/or smoother passages between valves and cylinders. Or you can increase piston speed with larger [opening area] ports, assuming that the cylinder/valve passage cross-section area is increased to match.

These piston speed limits also assume "typical engine loads". The same engine can be revved faster if there is less or no load on the engine, due to the lower-density and easier-flowing steam involved. I remember reading articles written by experimenters in the 1960s, 70s, & 80s, who bragged about their experimental steam engines revving up to IC-like rpms with little or no load. Back in the 1980s & 1990s, I was impressed. Today, not so much.

There are lots of other factors to consider too, when selecting/designing steam engine features, dimensions, and characteristics. It gets very complex. Especially for steam automobile engines. One li'l design boo-boo can turn a promising steam engine concept into a great boat anchor, door stop, or steampunk objet d'art. I have personally designed an awful lot of alleged steam power equipment which I fortunately spotted problems with and therefore never built.

I chalk it up to "learning experience". As the old saying goes: "A learning experience is what you get when you didn't get what you wanted". smiling smiley

But actually, learning experiences are now exactly what I want. First-hand learning experiences, that is. Experiences from building and driving an actual steam car.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 26, 2013 11:15AM
Hi Jim
I agree with you on a steam engine compared to an IC engine as they are dependent on RPM to get there power with a lot of gear changes.
What I love about a good steam car is the a total variable cutoff from 35% to 5% with no gear change. It matters not how fast you go by rpm as it is Rev Per Mile. It is the same no mater the speed. The duty cycle of an IC is only about 10% . It is less for our stuff as there is no idle or gear changes. Again ring wear is dependent on piston speed over the mile. The MK 5 would have the ability to turn 3600 rpm but then you would be going 120 mph ,Smooothly..
A Cyclone in a small car would only be turning 1800 at 60 mph, 900 at 30, 450 at 15 mph , 225 at 7 ,and 120 at 4, walking speed and no chug-chugging . No gear change necessary or gear box. A simple machine with far less parts than what an IC engines needs to move a car with less weight and cost, cleaner burning and can use any fuel. You are one of the few who has driven a number of them and realy understands what I am talking about.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 26, 2013 11:21AM

The more power strokes per revolution desire is one more illusion of little value. The four cycle four cylinder gas car has only two per each revolution of the crankshaft, in the steamer it is one per revolution per cylinder with single acting and two if double acting. But, that is not the pacing item, the difference in the PV diagrams is what counts. Fluid torque converters mask the impulses in the gas car.
The gear ratios are chosen to give modest engine speeds and the rest determine what and where the power band is found.

The difference between a tiny cylinder and one more suitable isn't all that much at all, no waste basket pistons either. That is what high pressure and controllable BMEP gives you. Something else real driving experience gives the student.
Only enough cylinders so that the engine in long cutoff will be self starting. Seems that three double acting or six single acting fill the bill quite well. Not to forget that the steam vehicle engine spends almost all its time in short cutoff and running down the road. Only starting from rest and hill climbing want long cutoff and a wide open throttle to give the torque needed, if you are in such a big hurry.

There is one other nice thing about a steamer. I removed the drain valve in my Stanley once and hooked up a pressure gauge. While there was 600 psi in the boiler, just cruising down the road only read about 150 psi in the engine. Even when starting out hard and unhooked it only reached some 300 psi. So with 2,000 pounds in the steam generator, hooked up and cruising down the road, just think of the reserve power you have at your command. Making the engine so small that it needs high pressure to run the car is a gross mistake.
Again, experience driving a steam car is the best possible teacher one can have.

Now, compare the PV diagram of the IC engine against the one from a steam engine and you will see why the steamer even with only two cylinders runs and starts so smoothly. Then couple that with the adjustable cutoff and one sees just another reason why the steam engine is such a good vehicle powerplant. There are a host of other good things too; but we are talking about piston speed and how many cylinders are really needed to do the job. A manual transmission four cylinder gas car isn't all that smooth either from a torque impulse standpoint.

Quite right, piston speed dictates the porting and valve size. Modest speed is very easy to deal with.
If the firm demand is for monster porting, then take a very hard look at the single sleeve valve, nothing else ever came close. The highest power output during WW-II came from frantic engines using that valve. I got one from a Bristol Hercules aircraft engine and the port sizes are simply astonishing.

All this only illustrates the vast difference between a steam vehicle engine and one for an IC powered car, there is now a new world to learn about. The two really have little in common and one must be very familiar with both before plunging in to design a steam vehicle version. Making assumptions based on imagination and scant hands on experience quite often leads to failure and disappointment.
Fun, isn't it.


Edited 4 time(s). Last edit at 04/26/2013 02:27PM by Jim Crank.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 27, 2013 02:50AM
One needs to look at history here. High expansion in the past has proven to
not give the efficiency that was anticipated. I believe the reason for this
should be found. My theory is that power range requirements were not

In a slow running engine the temperature difference between inlet steam and
exhausted steam adversely effects efficiency because of high heat transfer to
and from the cylinder surface and the steam. Compounding, breaking up the the
expansion into stages has proven to be N advantage. However in automobiles
compounding has not worked well. Most of the time the engine is throttled
operating at such a low power that the later stages are producing no power and
are a load loss on the system.

In the Rankine engine cycle the highest efficiency is obtained when the steam
is fully expanded to the exhaust pressure and only enough heat is left in the
exhaust steam to avoid condensation. But not being piratical one can use heat
exchangers to transfer some excess exhaust heat into the feed water as Harry
is doing with the Cyclone engine.

An automobile requires a huge power range. A throttled engine has a near
constant efficiency from wide open down to the point were the expanded steam
pressure is equal to the external pressure in the exhaust lines. Continued
throttling results in an exponentially increasing loss of efficiency.


Your requirements are good examples of top down design. I think you need a lot
more to get to specific details of the engine. The aerodynamics of the
automobile for example, acceleration requirement of 0 to 60 time. Occupancy,
how many passengers. Do you have average MPG or longevity goals?

Thoes big low RPM engines you like so much. What MPG did they get. How many
miles before they needed to be rebuilt. Some Doble's made over 120,000 miles
and more. The Whites have good reports. But don't have any numbers on them.
But as far as MPG most old steamers were doing good to get 15 MPG.

Vintage IC engines required a rebuild around 60,000 miles. Well not to vintage.
1950 to 1970 or so. Modern cars go 200,000 miles.

If the goal is to build a modern steamer, competitive with current ICs or
hybrids, one needs to look at the driving requirements needed today and in the
near future. We need to look at the MPG of driving cycles highway and around
town. To be competitive on MPG one has to use a high expansion ratio 27:1 and
higher. Especially at highway speed. Expansion that high would require a vary
large engine to get the same power of a low expansion if we do not up the RPM.
The high efficiency does mean that a smaller boiler and condenser would be
required. High RPM does not mean it wont be self starting or smoothie running.
The higher RPM is a necessity to get the power into a small space. That high
expansion ratio is bad in a single stage engine. For one the heat exchange in
the cylinder as described above. And second the very short valve open-close time.


Edited 1 time(s). Last edit at 04/27/2013 03:01AM by steamerandy.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 27, 2013 09:06AM
Hi Jim,

I know not this "failure and disappointment" of which you speak. Every time I learn something useful, that is a success and a step forward. Lately I learn more and more from building, and less and less from the internet preaching of self-proclaimed supreme experts and geniuses who never designed/built a single steam car themselves. When my build reaches the "running" stage, the learning will accelerate exponentially. Especially since I am starting from the latest/best (740) Stanleys, which actually run extremely well, rather than from a blank sheet of paper, imagination, theory, or pure self-appointed expert/genius ego.

I have been driving/maintaining Old VWs for 35 years and still don't have the faintest idea how to design/build one. How would 35 years of steam car driving, or 50, or 100, with 10, 20, 30, or more different gas/steam cars to compare, magically teach me how to design/build steam cars? Umm, it wouldn't. Road-proven Stanley engine blueprints and stacks of road-proven burner/boiler/controls blueprints are my guides.

I am glad that my hi-speed-engine recreational thought-experiment provoked your absolutely splendid [and, oddly, utterly un-questioned by anyone] defense of precisely what I have been advocating, designing, and building, for many years. That was not my intention, but the results are pleasing anyway. And interesting.

My steam car construction continues. 1120 rpm at 80 mph max; lower speeds are rpm/mph proportional; no clutch or gearbox, absolutely smooth and silent, 13 moving parts in 2-cylinder 2-valve, 139 cid /2.3 liter engine; any-fuel potential. Well-oiled superheated-steam cylinders which actually work. And that's just for starters. Watch out for bigger engines/boilers in future projects.


Edited 4 time(s). Last edit at 04/27/2013 09:49AM by Peter Brow.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 27, 2013 03:38PM
I always thought that there were so many other design errors and simplistic engineering in the old steam car engines that over expansion fell far down the list. One serious problem that I pay attention to is condensation and re-evaporation in bad porting and really slow speeds, something that Stumph also singled out as one of the really bad efficiency losses in double acting steam engines and that includes the Series E Doble for certain. High superheat helps minimize that. Besler too took pains to minimize this, note his porting in the second airplane engine.

The expansion ratio depends on the initial inlet pressure, the exhaust pressure and the cutoff percentage, not the rpm, it should be independent of that, assuming one has done the best job of insulating the engine. Something around 40% to 5% in a six cylinder single acting engine, maybe with good condenser vacuum one can look at 3% for highway driving. Longer cutoff just makes for smooth operation and gentle puttering around in town. Just see how horrible the efficiency is with an IC engine when running slowly and you won't get all upset with steam engine efficiency across the entire rpm range. I am not concerned about the valve gear, that is no pacing design item.

And, by slow speed, I do not mean the hundreds of rpm like the Stanley or Doble; but in the 1500-2000 rpm range max, that is enough to keep the six cylinder single acting engine down to a respectable size.
A wide power band is there with throttling, cutoff control and a two speed transmission. There if you need it, but really for ordinary driving, just leave it in overdrive, the torque is sure there for accelerating so use it.
Going up to 2,000 psi also helps, because then the BMEP can be much higher, depending on the cutoff of course, so the horsepower will be most acceptable and the torque of the Cyclone Mk-6 will shred the tires, so what more can one ask for. Also, the exhaust pressure would be as low as possible as a vacuum pump is certainly in the plan, so with luck the exhaust will be looking at 20-25" hg vacuum.

Doble E-14 has run a bit over 675,000 miles and E-23 shows 360,000 miles on their odometers. Both get around 8-9 mpg on the average, my White was around 12 mpg; but in those cases, mileage was just not important. At what, some 15-18% efficiency? Good enough for their era; but certainly not for today.
The Cyclone running now at 32% net efficiency will be more than satisfactory and burning carbon neutral fuel is most important now.

The dream car is no super vicious sports car; but actually a nice GT. The car I have always used as my model is the C-4, C-5, C-6 Corvette with the base engine, that is hot enough. A friend of mine has one of the continuation 427SC Cobras with a 500hp crate motor. The car is scary, well up into the instant heart attack level and while great fun, you really need to have a light toe with that baby. Not what I really want, although I do love driving that beast, definitely a white knuckle flight.

Heaven help me, but I still want to wrap up the work George and I started with the Wankel. It is so simple and has a high "piston" area for the size, can easily have a high compression ratio, although a short stroke, a very good expansion ratio if you move the exhaust port, a second stage power recovery turbine is still of high interest, ultra smooth and only two moving parts per chamber.
A local firm has all the OMC CNC machines and programming and would love to make the housings the right way for steam. The Wankel is a charming seductress, once you fall under their charms it is most difficult to turn your back on them.

To be honest about it, I just want to finish the engineering of this steam car to the point where I would be reasonably confident that with great care it just might turn out the way I would like to see it done.
In point of fact, what really interests me now is to install the Mk-6 Cyclone in that 427SC Cobra kit car. This all depends on IF that Cyclone will ever see any production.

Don't you think we have exhausted this subject by now?
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 27, 2013 05:38PM
Hi Jim

If cutoff control were automated:

What speed range would long overlapping cutoff be needed for smoothe running. At 0 MPH for starting obviously. What effect would vehicle weight or an incline have on the RPM were you wouldn't need overlapping cutoff. For hard acceleration, boiler steam generating capacity.

Think about parallel parking on s hill.
Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 27, 2013 06:59PM

Automatic cutoff control would be a good feature I guess, once a satisfactory map was created for rpm vs.cutoff percentage for the specific car, or maybe for large truck use when it is most advisable to keep both hands on the wheel. But; don't we always do slow maneuvering unhooked anyhow? If the engine has a good long cutoff for very slow running without jerking, then automatic control would be nice, one more thing you don't need to think about.
Now some lever or foot pedal could allow the driver to select forward or reverse and let the engine decide what it wants for cutoff. A classic foot pedal does let you keep both hands on the wheel, it would be my choice using the hydraulic clutch system most cars have.

Overlap the cutoff with what? Cutoff control and throttle position should be kept independent anyhow, I think.

Vehicle weight, such as a car or a Peterbilt would more or less dictate how fast the vehicle accelerates. Thus the cutoff vs. rpm mapping comes into play here. Parking on hills with the Stanley, White and Doble was always done unhooked and using the throttle. Not usually a problem, I tried to avoid parallel parking anyhow with the steamers. Not with the White because the line from throttle to engine was extremely short, it is right on the engine so you weren't flailing for the engine drip valve on the Stanley or the handbrake on the Doble to get that long steam line drained before you crashed into the car ahead of or behind you. Plus the White throttle gave you the most delicate control, unlike the other two.

For steam generator output I calculate for full throttle, unhooked, maximum firing and evaporation rate and then add a 20% fudge factor. That is with the draft booster figured in at a 60% increased output rate. If done correctly like the Dobles, it doubles the output rate, so now there is a nice reserve available. If the car and booster are set up right, you reach full speed with the fire still going on or off by the pressure setting.

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 28, 2013 07:20AM
Hi Jim,

Yep, this topic looks exhausted to me. I need to build more and talk less if I'm ever going to get this thing on the road.

Best of luck with your steam GT project. I think that you have the formula for a really amazing steam car there. I look forward to seeing it in action, or at least reading about it. And if a kit version ever becomes available... hmm...

Steam Wankel? Go for it! I have no personal experience with Wankels, but what I have read about them over the years is pretty eye-opening. With steam, it could be downright mind-blowing. Insanely simple, compact, smooth, quiet, and powerful. And it sounds like you have much of the work done already, including an expert shop for the machining. Maybe "Der DampfenWankel" is "meant to be"?

Re: Engineering a Steam Engine for Automotive Use. A top down approach.
April 28, 2013 11:19AM

All you say about the Wankel is quite true. The stinker is that they are not easy to machine; but the inside shape of the housing and the rotor flanks are geometric forms and today with CNC machines, the right programs for each can be generated. Or, there are still quite a few companies working on the engine and have made significant progress. So one is better off having them make the more simple steam version. Never make what you can buy, it's cheaper in the long run.

I lived with a Mazda RX-3 for several years and went for the whole Racing Beat package. It is indeed a fascinating engine. No low down torque yet high horsepower as an IC engine and it revs like an electric motor, as smooth as silk too. A good high BMEP with steam fits right in with the large "piston" area to regain high torque at startup.
There have been some recent dramatic advances in the seals, as they need to keep oil out of the combustion chambers for better control of pollution, one of the serious earlier problems and it is right up the alley for use with steam. That pollution problem and their fuel consumption sort of ruined their chances as a car engine what with the knee jerk politicians running around making rules that almost sunk any chance for a comeback as a car engine.
The identical dinosaur type of thinking that ruined steam cars: "We know it doesn't work so we will not look at it again." As if time and progress never take place. Ah, the closed corporate mind.

This company bought the OMC rights and all the tooling to make the engine here and they are only 60 miles away. No more super expensive Gleason grinders now. The high resolution stepper motors makes it possible today to use a CNC vertical mill to do all the housing machining.
I did contact them and they too are most interested in seeing other uses for the Wankel than just as an IC car engine. Steam from solar and waste heat is one use they are very interested in pursuing, so my inquiry came just at the right time. These are smart cookies too, we speak the same language.

May I remind you that Mazda won the 24 hour LeMans race in 1992 with an unsupercharged 2.6 liter 4 rotor Wankel engine, an outright win over all the highly tooted marques, Audi, Peugeot, Porsche, Ferrari.
A killer endurance race if there ever was one. Winning LeMans really means something, even finishing well up shows you have a good car. Just finishing for that matter.
The R26B develops 700 net bhp (Japanese) at 9000 rpm, its redline, and 449 net lb-ft of torque at 6500 rpm. It's only 39" long and weighs just 396 lb. Later development took the output to just over 900 hp, some reports said 1,000 hp. Pretty impressive for an unblown 2.6 liter engine!!
When torn down for the press after the race it was found to be in perfect condition and could have gone on another 24 hours. So much for seal longevity and engine reliability.

Time will tell. I await the Mk-6 however.

At any rate, I do think we have exhausted what needed to be said about this subject.

Edited 5 time(s). Last edit at 04/29/2013 03:56PM by Jim Crank.
Sorry, you can't reply to this topic. It has been closed.
All files from this thread

File Name File Size   Posted by Date  
Harryandme.jpg 152.8 KB open | download HLS 05/18/2013 Read message
Scroll generator..jpg 78.5 KB open | download Rolly 08/02/2013 Read message