Four wheel drive is easy with steam. Provided you go independant suspension all round, you can simply drive each knuckle with a small signle cylinder engine. No differentials - each wheel completely independant. What's the chances of all four pistons finding TDC simultaneously ? Well, barring Sod's Law, virtually zero - so the car will always self start. This idea is crude - and possibly halfbaked, but well within the KISS principle.

So you have a ladder type chassis with four single cylinder locomotive type units - one driving each road wheel.

Can the phorum please tear this idea apart.
best wishes for now
johnny F

The idea is pretty old, I have found a stack of 4WD steam traction engine patents, for example. The Gearless steam automobile was a 2WD version of the same idea.

I have a few issues with the concept. First, the sheer cost and complexity. 4 engine blocks instead of one, 4 cranks and so on. I build experimental and prototype engines for a living and I can pretty much guarantee this will be a lot more expensive than a traditional 4WD layout.

All the steam lines and condensate issues that occur when you go with 4 engines instead of one are going to make for some real plumbing problems.

Another issue is whether the use of separate engines really does the same thing as a differential. The differential splits the torque of a single drive shaft, with multiple engines each one is independant. When going around a corner, the same force is applied to each wheel although the inner ones are going slower than the outer. Is this going to provide the same level of control?

Another issue again is engine efficiency. Usually engines suffer from scale effects as you shrink them. Are four smaller engines going to perform as well as a single engine of quadruple output?

Anyhow, how are you going to provide identical output from each engine? If all have separate throttle linkages you can just bet none of them will respond exactly the same. If you have just one throttle, steam will somehow flow better to some engines than others.

I'm not trying to throw a wet blanket on the idea, but I suspect getting such a thing to run properly might be a lot bigger challenge than it first appears.

Regards,

Ken

Johnny

1) Slow speed power pulses from the four single cylinder engines will create strange handling characteristics, even in a straight line.
2) Was something similar not tried back in the 70's by a Mr Hyde? Also the Yuba tractor of the 50's. Both used multi cylinder engines per wheel.
3) If the 4WD is to be used offroad then some means of locking the axles will be required else all the steam will go to the spinning engine / wheel. In hydraulic systems we use a flow divider but this is a wasteful device and cannot match a single power source and distribution through propshafts and differentials.
4) Hill starting could be hit or miss depending on how many engines were off tdc
5) If the intention was low cost for amateur build I think there are plenty 4WD transmissions in breakers that could be mated to one simple multi cylinder engine at less cost than building four single cylinders.

BUT I wouldn't put you off building such a vehicle. As a simple road car it will work just I don't feel it is commercial. Anyway the more people who are building and experimenting the greater the experience gained.

Brian

Ken hi, and very pleased to meet you.
I really value these comments!!

Appreciate what you say about efficiency versus scale.

Regarding differentials however - I might imagine that provided the road surface traction is greater than the elastic forces of the steam expanding in the cylinders the system will balance out even when going round bends. Does the manometer principle also apply? in that if all four cylinders were fed from a common manifold there would be some equalisation of pressures. What I like about this idea is the utter simplicity and crudeness (of the IDEA that is). I can't imagine that you would suffer from any sort of wind up, but yes, I suppose I can imagine some issue regarding control. Is it possible that good steering and suspension geometry could alleviate these problems ?

And doesn't an ordinary differential just switch off one wheel when in trouble - a one wheel drive car ?? - sounds pretty crude to me, but we all accept it.

And is identical output necassary? Does each cylinder in, say, a stock IC engine behave exactly the same ? - or does not the crankshaft joining them together just take the strain of the differences. In my system the road surface would do this.

Appreciated also the plumbing issues - but are these major issues i.e. that cannot be sorted out with good planning / design, manufacturing techniques??

By the way, I am contemplating a camper van - a small coachbuilt bus. From initial sketches it looks like the four cylinders will actually be situated quite close together.

Can the Phorum please continue to pick holes in the idea - go ahead - tear it apart.

Appreciated

JohnnyF
UK

Individual 10 hp engines for each rear wheel were used in the race car "Woggle bug" in about 1905. The race car was called that because at speed, the individual engines for each rear wheel would make the car woggle down the track. With a differential, the steam cars never had this peculiar problem. In spite of it's "Woggle", the race car still set many, many records.

Hi Johnny:

Nice to make your acquaintance as well!

The thing about differentials is that (I assume) they split the force as well as the speed. My take is that we should think of it as a leverage problem, if you increase the force you reduce the travel, and vice versa. So, when a car goes around the corner the outer wheel may be going faster, but with less torque. Since power is a factor of torque and speed, the power on the wheels should be about equal. Put an independant engine on each wheel and all will have the same torque, but differing speeds. The power on each wheel will be different. Seems to me most concepts I have seen for vehicles with independantly powered wheels usually incorporate some kind of scheme to change output in turns, often some kind of throttle interconnect with the steering.

I don't think the manometer comparison is necessarily valid, when we see pictures of systems having equal pressure all around the interior, it is typically assumed to be essentially a static system. When you have a dynamic system, flow resistance starts to have a large impact. Ever see one of those hoses with all the small holes in them for watering shrubs and the like? If the water pressure is a bit low the volume coming out the holes diminishes as you get further from the spigot. This can be allieviated by making the system more static, that is by using a bigger pipe so the flow is slower. Problems with that are that the bigger pipe has to be thicker to hold the same pressure, the slower flow promotes greater heat loss, as does the larger overall dimensions. Much more insulation is needed to offset these losses. The plumbing issues get bigger.

The 'Woggle Bug' brings up some other issues. I specialize in engine balance for GM, so maybe I see the problem narrowly in my field of expertise. My take is that the Stanley style engine is poorly balanced and has primary and secondary shaking unbalances and rocking couples as well. Constructive and destructive interaction of these forces as the two engine car progresses will create a chaotic set of shakes and couples that are transmitted to the vehicle. Possibly the only engine design that is even worse from a balance standpoint than the 90 degree inline 2 cylinder DA engine used in most steam automobiles is the single cylinder. In 4 cylinder engines the angles of the cylinder bores and phase relationship of the crankpins have been established so that the unbalance forces largely or even totally work one against another, and therefore cancel out. By contrast, 4 single cylinder engines free to turn independantly, will from moment to moment alternately cancel one another and reinforce one another. This will lead to moments of moderate vibration alternating with moments of large vibration and moments of little vibration, and no way of predicting when these will occur. Of course, unequal torque variations and unequal timing of power pulses are going to add their own disharmony to the vehicle, added onto the balance situation it might be quite disconcerting.

Anyhow, that's my two cents worth.

Regards,

Ken

Hello again Ken, Brian, SSSteamer, and everybody else,
thank you All for these valued comments. I'm still bemused by the crudeness of the concept, so I'm going to labour this till the death, or alternatively, the birth of Wogglebug II (funny coincidence in that I was contemplating using the front panel and windscreen of a VW camper(Bug??)).

Before I concede that some form of differential coupling is essential in at least one axle how about this: employ some system of biasing the throttle towards each side of the vehicle according to the steering direction of the front wheels, in other words: two throttles somehow operating in sympathy with the vehicle steering.

Incidentally, looking on the net at pictures of Wogglebug, it had spindly wheels and all that sleek design for speed. Perhaps something with more beam, proper wheels and rigid construction chassis might be controlable to an acceptable level - perhaps ??

But yes: the engines need to be balanced to the best possible standard. You are in the game of balancing engines Ken. What is the best result I can hope for in balancing a single cylinder engine, and how do I go about achieving this ? And can it ever be, in your opinion, reduced to achieve an acceptably low level of vibration etc.

ok for now, I like this forum
Johnnyf UK

apologies Ken, you already said about the throttle biasing in your, good and comprehensive, reply.
Re: balancing. I understand that we look to balance a factor of the reciprocating weight, and, unless we employ countershafts etc we will never attain true balance in a single cylinder engine. Accepted.
What RPM are we talking about in my application? say, 500 - 700 RPM - depending on roadwheel circumference? OK Ken, so what balance factor should I aim for, in your worthy opinion ?

Re: unpredictable power pulsation. Well, there is scope in my project for smaller wheels at the rear and larger on the front (or vice-versa). In this case you could employ engines with differing strokes. It seems to add more chaos, but it also ensures (perhaps) overlapping of power phases. Isn't there chaos in all systems?

The more I think about the control problems resulting from lack of differential coupling the more i see that the trick is in throttling the engines in harmony. Appreciated that the manometer principle doesn't apply in a dynamic system - but if the cylinder pressures could be monitored, and throttles adjusted accordingly in order to balance them evenly, would this not achieve an even distribution of power ? Yes - a real challenge (well, at least I'm comfortable with electronics)

Brian, you say that hill starting could be hit or miss. Provided the vehicle isn't jammed up against a kerb, or stuck in a ditch, won't it simply roll downwards, allowing the engines to start ? Will not the differing engine strokes also guarantee starting ?

And if the free wheel does zap all the steam there are gizmo ways of dealing with this : the aforementioned cylinder pressure balancing system, or that fancy traction control (ABS in reverse) which puts the brakes on when a wheel spins.

Food for thought ??

Have a nice day All
Johnny

Johnny

Throttles cause wire drawing! More throttles less economy.
With a central manifold and equal length / resistance pipework there should be no issue for a camper van. Sportscar or SUV yes but not a van!
Also consider the effect of cooling in the pipework that will cause variation in expansion.
Why not build a model with some Stuart engines first, could even be remote controlled.
Even better a go-kart! I have often thought if we tried out all these ideas as karts and raced them it would generate healthy competition and see some progress at moderate investment.

Brian

Hi Johnny:

When balancing an engine we balance for 1/2 the reciprocating mass plus all of the rotating mass. I wrote an article for "The Steam Automobile" a few years ago that went into detail as to why this is so and examines the fundamental concepts of engine balance. I can e-mail you a copy if you desire, it is currently in Microsoft Word though I could likely convert it to another format. It would certainly be more understandable than covering the topic in this venue if for no other reason than the illustrations.

The above statement is generally true except when balancing symmetrical cranks (such as a typical inline or boxer 4 cylinder or an inline 6) in which case the engine layout is such that the forces involved totally cancel themselves out if the crank itself is balanced to zero. It also doesn't hold true if we intend to use 2 balance shafts to cancel out primary unbalance forces, then we just balance for rotating forces.

In a single cylinder engine without active auxiliary balancing devices, the best we can do is reduce the peak primary unbalance by 50%, and we can't do anything about the secondary unbalance force. Figure we can, maybe, reduce the shaking by about 40% give or take a bid depending on stroke and conn rod length. Those little 5 HP one cylinder engines on lawnmowers are balanced about as well as they can be, and they shake like the devil. There are single cylinder engines built to test concepts in automotive engines, these may use either 2 dummy pistons or up to 4 balance shafts to let them run freely without shaking apart.

Realistically, I can't see a 500-700 rpm engine in a car. The tires on my truck aren't huge, but they are about 2 feet in diameter. That works out to about 840 revolutions per mile or 840 rpm at 60 mph. Figure an 80 mph top speed and that's about 1120 rpm. Of course, this will go down as tire size goes up. Still and all, even Saint Abner was running about 1,800 as I recall, way back when grandpa was still a pup. Admittedly, the unbalance goes up with the square of rpm, but the power goes down as the product of rpm. You'd need either really huge cylinders, massively inefficient long cutoff or extremely high pressure to move a car at highway speeds with a 600 rpm engine. While the low rpm would cut imbalance impressively, you'd need massive pistons and conn rods to get the power you need, so the improvement isn't what you'd expect. The bigger powerplant would add more vehicle weight which would necessitate a heavier structure soooo... Anyhow, realistically I think you need to consider some kind of gearing on each engine to let it wind up a bit so as to keep engine size down. Added advantages are that thermal losses and ring leakage is likely to be less as rpm goes up. Downside is that a stepdown gear on each engine is starting to get you closer and closer to a conventional drive system in parts count.

I'll admit old time steam cars were kind of flimsy, but they weren't terribly powerful either. A modern car will be beefier, but the bigger powerplant will also shake more. Anyhow, besides being a bit brute force, adding extra mass to dampen vibration detracts from vehicle desirability as far as economy, cost and handling go. Also, it isn't like the shaking goes away, you are just masking it. The odds that something or other is being stressed or shaken loose is still good.

Actually, most of the powertrain forces acting on a car aren't chaotic. The piston power pulses come at predictable intervals, as do any uncancelled unbalance forces. The crankshaft, being the heart of the engine, keeps these things in synch.

If I were to attack the idea of differential power to a split powertrain, I'd consider avoiding all the electronic gimmickery with some careful mechanical engineering. I'd try something on the order of a cam connected directly to the steering that changes compression on some springs proportional to the output desired for each engine. The springs could be integral to pressure regulator valves, ie, the spring tension is what controls the valve pressure setpoint. Perhaps a better solution would be to use a system similar to an old stationary steam engine governor whereby the governor controlled engine speed by adjusting cutoff. In this instance the steam chest pressure would act against the spring load to adjust cutoff in such a way to maintain the desired pressure going to that engine. It has the advantage of being easily diagnosed, you can look at a mechanism and see if the spring is broken or the cylinder is leaking, usually harder to figure out what those sneaky little electrons are doing inside a chip.

Mind you, I'm not against innovation. I have a hunch a lot of people would call me nuts if they saw the engine admission valve, burner, feed pump and condenser system I am working on for my own project. So feel free to just say "He's a whack job anyhow, what does he know?" and take a stab at your dream.

I REALLY like the go kart idea, makes solid sense to me. My own engine concept is a very large (4 inch bore and stroke) single cylinder engine, even with very short cutoff it would be too lively for all but the hottest go karts. I'm therefore looking at installing it in a sand rail or dune buggy, which is nothing more than an overgrown go kart. I like the low cost and open framework, one fits my budget and the other makes troubleshooting a breeze.

Regards,

Ken

Hi Ken and All,
Ken, I like your style, and thanks for not being too negative - this ain't flogged to death yet.

Yes, I will take a copy of the balancing article. If you know how to work this phorum you can send it to my email (better not to say the address here). MS word is fine.

By now you have realized I have little practical experience with steam, and i am one of these academically challenged sort- especially when it comes to numbers. Well, the latter I have to live with, the former: I have done plenty of research over the last few years, and I am learning all the time. I have recently bought a really good lathe and milling machine, a Deckel FP1 clone - it's fantastic with lots of accessories. I got a TIG welder and am good with it. Have made some lovely patterns and had them cast in SG iron (intention a triple cylinder double acting motorbike engine ((that probably wont work very well but will be good machining practice)). Am currently developing my own steam piston - employing valve ideas etc which I think are good. Some guy on this phorum said 'first do, then brag' so I'll say no more at this time. This will be the basis of my steam project.

Having thought about levers, woggle bug, and chassis geometry, I still have a feeling that this thing can be tuned to work. If the suspension wishbones are suitably triangulated to transfer the reaction loads (from the roadwheels) to a more central location in the vehicle, will not the forces be directed towards some central axis, say, the direstion in which one is aiming to travel. Like, how an autogyro sails over the wind, or a boat sails into the wind, or a windmill turns the wrong way - in all three the axis is fixed so the forces on the blade (or sail) have no choice but to act in the direction of the axis (something like that anyway). Also, whearas wogglebug had just two engines, four (given what I am going to say below under 'ridiculous') may induce some directional stability .


On power pulses: am I right to say that it is a case of power transfer - and that power meets resistance, and when power equals resistance = max efficiency in power transfer. But do not forces act in equal and opposite directions)? and am I right when I say that steam is elastic and will be somewhat forgiving? I guess I'm trying to say that with suitable chassis rigidity these pulses will be absorbed into the chassis via the road traction, and any cylinders that are coincidentally opened to each other (see below on ridiculous), via the steam lines.

And now (drum roll please) ... from the sublime to the ridiculous.
Okay, so I need some gearing. I dont like chains so I will use HTD belt drive. This greatly simplifies spindle design. But here is the ridiculous - here it is, the first one, right now: by using differing drive ratios on each roadwheel spindle to engine crankshaft, the four engines will together never find perfect harmony (I said something like this before when I talked about different size roadwheels). It is then a case of tuning the system with engine inlet orifices, I feel for Brian's opinion on wire drawing, and that my camper van is adequately fed from a single throttle / manifold (if I understood him correctly). The second ridiculous point is already said, that such tuning could acieve a system where certain of the eight chambers (DA engines) are open at the same time. The steam pressure will (surely ???) even out a bit, no matter how dynamic the system is, and reduce overall pulsation effect.

And finally for tonight, how about scotch yoke engines? Although the frictional component may be a little higher there will be no rotational forces that cannot be balanced. The back and forth forces will at least be directed horizontally (i.e. in line with the engine layout, and occasionally opposing. I've been drawing pictures all weekend. Sounds a bit shaky, but any opinions please ??

Okay phorum, I hope i am well understood tonight, I have enjoyed this. Please help me, Boy, I need help .....

Johnny F

Johnny

You would be better off with two engines. One for front and one foe the back.

The torque variance during rotation can be enormus. What is the reaction going to be with 1000 lb-ft of torque on one wheel and zero on the other.

You must know that piston steam engines are low RPM engines. Torque diferential during a revoloution is one of the things I am conserned with in my design. One of the reasions I am looking at using more cylanders. With more cylanders you can get smother rotational torque profile and also use shorter cutoff at low speed. With 4 cylanders (2 DA cylanders) you start getting overlapping cutoff at 90 degrees of cutoff (50% cutoff) with 6 cylander overlap starts at 60 degrees. with 12 cylanders overlap starts at 30 degrees. Assuming the sylanders have equially spaced timming. But the more cylanders you have for a given displacement the smaller they are. Smaller cylanders have higher surface area to volume ratios. Thus with smaller cylanders having more surface area, you potentially have high heat loss. On the other hand at high speed heat loss is minimized and with short cutoff you get better efficiency. I figure there is an optimin balancer between heat loss and the higher expansion expansion that would give a better overall efficiency and smother (vibration free) ride.

Under hard acceleration in an old Stanley you can definately feel the power surges from that torque differential as it takes off. Some of that torque differance was being absorbed by the front end lifting higher off the ground.

Andy

THanks for this Andy,
this is all really usefull stuff ! Perhaps I do, given my inexperience, underestimate the power of steam, and perhaps such a setup might eat tyres. If that Stanley on ebay is anything to go by, little wonder accelleration forces end up where they are not wanted. However, with everyones help I think we are narrowing down on the requirements (remember I'm looking to build a small coachbuiltish camper van sort of thing - although its turning into more of a bus. We'll turn it into a racing car in due course).

1/ step-down gearing, from crankshafts to wheel spindles (HTD belting)
2/ rigid,ladder chassis with wishbones which transfer forces to some central axis/ location
3/ elasticity in power transmission - in addition to my previous opinion on steam behavior,designing a cush drive into each HTD pulley arrangement is not difficult (like rear wheel of motorbike )
4/ less throttles more desirable - tuned at steamchest by inlet orifice experimentation
4/ question of balance between engine size, rpm requirements, forces which cannot be balanced or masked
5/ by using differing drive ratios (see 'ridiculous' above) there is scope to avoid dead centre coincidences - perhaps once every 10,000 wheel rotations etc.
6/ gizmo device to apply brake to runaway wheel

Barring 5/, this is still quite crude, as my original question - just added gearing.
But I think engine balancing is going to be more of a critical issue than odd power pulsation and torque differentiation: all those odd harmonics might end up cracking chassis members, killing light bulbs - more uncomfortable travelling experience
I asked previously about using scotch yoke engines. Does anyone have anything to say about it ?

see you
johnnyF

Hi Johnny

For referance the car I was in was simular to this one:

[www.stanleysteamers.com]

Andy

Hi Johnny:

I tried to attach the balance article to this post, but the Phorum software won't permit it. Can you send me your e-mail address via private message?

Regards,

Ken



Edited 2 time(s). Last edit at 04/18/2008 07:45PM by Ken.