Opposing Piston Engines 10

[Painterman]

Hello to all,

I keep on noticing favorable comments about opposing piston engines, i.e., Junkers and Napier Deltics. If they are so good, then why aren't they used more widely? That is to say, these engines must have a problem or a shortfall. Can anybody say what the problem is? Also, what about their efficiencies?

Any other would be appreciated.
Kind thanks & best regards
Painterman (DFD)

 

[patprimmer]

Bill

I actually meant WW11 but my crappy keyboard keeps dropping letters unless I hit the key real hard, dead middle.

Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[WGJ]

Hitting the keyboard - yes, I hit it dead in the middle with my forehead on some occasions. Usually when I am here, at work.
It has nothing to do with the effectiveness of its primary function though.

Bill

 

[TStaples]

There is also the Ecomotors effort. John Coletti (formerly the head of Ford's SVT group) and Peter Hofbrauer are leading this effort. I have a former employee of mine who is working on this effort in Detroit. Power density is interesting, if nothing else.

http://www.ecomotors.com/technology

-Tony Staples

http://www.tscombustion.com

 

[tbuelna]

TStaples,

Whether you meant to or not, your comment "Power density is interesting, if nothing else" sums up the Ecomotor engine perfectly. The term "power density" sounds great to the average person, but in reality it is not very meaningful with regards to automotive powerplants. More important (in order) are things like production cost, emissions, driveability, SFC, and NVH.

Hofbauer claims he simplified the engine by eliminating the cylinder heads. But at the same time his design needs 6 conrods for 4 pistons, it needs 4 pistons to get the same displacement a conventional engine can achieve with two pistons, it needs a very complex and expensive ported cylinder liner, and (being a two-cycle) it needs an expensive assist device for its turbo to scavenge properly.

It also would not package well in a typical automotive chassis, since it is very wide in the transverse direction.

As for his claims of exceptional "power density", he achieves those numbers by operating at very high BMEP rates (25 bar?). The drawback to doing this is that this engine will be very rough running, have very aggressive torsional characteristics transmitted to the rest of the drivetrain, and will have an unacceptable amount of combustion noise for automotive use.

As for his claims of low fuel consumption, I would be very skeptical of them. It has long been known that one of the inherent flaws of the opposed-piston, uniflow two-stroke configuration is the poor injector spray and combustion chamber shapes that result from the side mounted injector location.

Regards,
Terry

 

[adrag]

That looks very similar to to the FEV OPAC engine.

 

[turbomotor]

That's because Prof. Dr. Hofbauer is the father of both.

To my knowledge, DARPA has suspended their lucrative funding for the FEV effort (someone please correct me if I am wrong). If the above is true, that says enough about the usefulness of the technology to me.

 

[dicer]

Catserveng
<
We had a competitor underbid us on a major repair of two Fairbanks OP engines, and their mechanics were not familiar with the correct timing procedures between the upper and lower cranks, needless to say a VERY expensive mistake. We had the advantage of a former US Navy man who had experience with that engine, and knew the manuals were not very clear on all of the repair procedures.
>
I never though they were different, but an advance on the bottom crank is a good idea, thanks for the info.
So what was the failure on that engine? I would think the drive link to the bottom crank would be the weak point, the bevel gears?

 

[TDIMeister]

I worked with Prof. Hofbauer several years ago, although not in any direct project-related capacity.

I would agree with some of tbuelna's criticisms of the opoc engine regarding the need for a scavenging supercharger; lubrication of ported cylinder liners; less than optimal combustion chamber shape and side-mounted injectors.

What the engine gains in an extra 2 sets of conrods and bearing shells per piston pair it saves in the lack of cylinder heads, with all their associated pieces: camshafts (4 in total for a DOHC HO engine), lifters, valves, springs, etc. (4 of each per cylinder for a contemporary engine). What it does for the overall BOM parts count, think about it...

The packaging in a transverse automobile layout is granted but a moot point as evolutions of the engine have been directed specifically toward trucks, aviation, military and APU applications.

As for the power density being a result of a high BMEP with attendant rough and noisy running, I disagree. I don't expect torsionals and combustion noise to be any higher than contemporary Diesel engines of the same cylinder count and using the same FIE/combustion development technology. The high power density comes about since this a 2-stroke engine, meaning that for the same displacement, RPM and power output as a 4-stroker, you need half the BMEP, or put another way, for the same displacement, RPM and BMEP, you get double the power. Yes, 2-strokes have different major critical orders than 4-strokes, but dealing with them are no different than other successful 2-strokes.

I agree that BSFC is indeterminate; what FMEP is reduced from the lack of valvetrain losses is countered by the added friction of two extra set of conrod small-end bushings
and big-end bearings per piston pair, plus small pistons in relation to the swept volume. On balance I'd say BSFC would be a wash compared to contemporary Diesel engines.

The opoc layout is inherently balanced (though not absolutely completely as claimed, since the L/R ratios of the two length conrods and therefore motion profiles of the inboard and outboard pistons are slightly different. This is however, still better than either inline- and conventional boxer 4-cylinder layouts, the latter of which must still put up with a small yawing moment due to the cylinder centerline offset.

I have no affiliation with the opoc engine or its development company, and I don't believe I have biased what I've said above. Rather, I believe it's a balanced critique based on what is known of the engine with objective engine engineering principles applied.

On a similar note, although not an OP engine in the strict sense of the term, I've seen an BMW boxer engine using an uncoventional cranktrain mechanism referenced in US Patent 5,785,029 that interests me.

 

[tbuelna]

TDIMeister,

Excellent assessment of the OPOC engine. In my opinion, the biggest issue with the OPOC configuration (or any opposed piston 2 stroke) is compression ring durability and scuffing. With a ported liner, the oil control ring travel must always stay below the lower edge of the ports. The compression rings, on the other hand, never travel much beyond the lower edge of the ports. So the amount of "wiping" overlap between the oil control ring and the compression rings is very limited. Thus the compression rings operate on a very poorly lubricated cylinder liner surface environment for most of their travel, with the worst tribological conditions being present at piston TDC where the compression rings also see their highest pressure and thermal loads. This inevitably leads to ring scuffing. And the situation is further compounded by the less favorable heat transfer situation present in a two cycle piston structure.

I worked on the TRC opposed piston engine program. And persistent ring scuffing was one of the problems we never were able to resolve.

Regards,
Terry

 

[unclematt]

I have been looking into OPOC engines for some time, and wonder if the cylinder scuffing issue could be solved with a hypocycloidal crank setup on silcon nitride roller bearings. I even came up with a design that would work. Another approach might be to use a carbon-carbon cylinder and a carbon-carbon piston. Machine them to close enough tolerance and there is no need for rings at all. Silicon nitride might also be used instead of carbon-carbon, which would also solve the lubrication issue. Yes, these components would be more expensive, but I would be willing to go there. It would also power a generator for a series hybrid very efficiently (hopefully). I am honestly tempted to go to

http://www.ecomotors.com

(also known as

http://www.propulsiontech.com/aboutapt.html)

and see about ordering one of their motors for modification and testing. I also thought the OPRE engine was pretty interesting at

http://www.pattakon.com.

 

[PJGD]

I had peripheral involvement with just such an OP engine constructed with no expense or resource spared by one of the US Big-3 circa 1985. It was designed and built with Silicon Nitride liner and ringless pistons with hopes of high efficiency.

The engine ran but not for long enough to prove its objectives. As I recall, on more than one occasion it shattered like glass on the test bed due to some mishap from which a conventional engine could have been reclaimed with a simple rebuild.

PJGD

 

[unclematt]

Thanks for the post, thats good to know. I am curious if that test engine used a hypocycloidal crank setup to reduce friction forces on the cylinder walls? Got any materials to post or additional info on the project you worked on?

 

[TDIMeister]

The definition of an opposed-piston engine is that the combustion chamber is formed by the two pistons in a common cylinder, implying the need of outboard crankshafts. A hypocycloidal cranktrain would not meet this need. You can employ such a cranktrain, but what you would have is in effect a "180-degree Vee" engine with inboard crankshaft and outboard cylinder heads. In fact, all practical realisations of a hypocycloidal cranktrain I have seen are in fact such 180-degree Vee designs -- to be distinguished from Boxer or horizontally-opposed engines, which have non-shared crankpins and bank-offset cylinders.

Balancing such 180°-V engines are the same as inline engines, since both pistons in a cylinder move as one unit. However, a simple 1:2 hypocycloidal cranktrain has a sinusoidal piston motion with no second-order component, therefore eliminating second-order out-of-balance. Therefore, such an engine with 8 pistons and crankpins offset 180-degrees would balance like an inline-4, but in the absence of 2nd-order out-of-balance would actually be perfectly balanced. A 2-cylinder would balance like a single-cylinder engine -- not good.

 

[unclematt]

If you will look at the engine on

http://www.ecomotors.com,

you will see that your statement outbout "outboard crankshafts" is not necessarily true. Their engine is the layout I had in mind when I came up with my hypocycloidal crank design. I also found Pattakon's OPRE design to be interesting, if not entertaining, and wonder if new materials and designs could actually make it work pretty well. For fun I came up with a way to put my hypo crank inside each of the opposed pistons in that design.

Also, counterweights can be employed to offset any/most imbalances in many hypocycloidal designs. Now do I think I know it all when it comes to these setups? Absolutely not, and I welcome your input. However, you can see an example of this at

http://www.wiseman.com

on their hypocycloidal design.

I remember reading some years ago that cardan gears used to create hypocycloidal motion have a problem with tooth degradation when spur gears are used, because one tooth is always taking the shock and load from combustion pulses. I have been looking, but can't seem to find that article. Obviously, helical gears come into mind to spread the load over more than a single tooth, but the resulting gear teeth are usually smaller and weaker, so there are tradeoffs. Helical gears do run quieter and can rotate at higher RPM's, but wonder if there are other issues to deal with as well? Does anyone have any experience or input regarding this?

 

[TDIMeister]

Ecomotors IS the OPOC engine, and as I stated above, I worked in the same company as its inventor Prof. Peter Hofbauer.

The outboard cranktrain as I stated is replaced with the long outer con rod pair connected back to the crankshaft; I can now visualise how you want to realise a hypocycloidal mechanism, and it's an interesting idea. I've been very interested in hypocycloidal mechanisms for a long time, but I wanted to avoid using gears at all cost for much of the same reasons you stated and just transfer the forces and motion by bearings. The second limitation of the hypocycloidal mechanism is that I wanted to avoid requiring a linear guide (the two pistons in the same cylinder bore on either side of the crankshaft serve this function, otherwise it would require a crosshead), so that the mechanism can be implemented in conventional inline, vee and boxer layouts.

 

[unclematt]

Hey TDIMeister,
Sorry if I missed some details in your posts. I wonder if utilizing some helical internal ring gears might work instead of spur gears. I have also been doing research into whether or not carbon-carbon gears might have the required strength and durability, especially if coated with diamond-like-carbon coating to further reduce friction. Carbon-carbon is lighter and stronger than steel, I am just unsure if it could withstand the shocks of combustion.

Moving to the hypocycloidal crank setup would remove the need for the current inner piston connecting rod design, further reduce cylinder/piston friction & wear, allow further increase of stroke length (if desired), and allow the removal of the outer piston rocker setup. The pistons could use straight, relatively thinconneting rods running through the crank connections and rigidly attached to the pistons, so no piston wrist pins either. If you adopt the appropriate hypo design, you could reduce reciprocating mass signifigantly, as well as friction.

I just got solidworks and I'm working to get some of this modeled. I have some 3d stuff I put together in sketchup, and may post some of them.

One thing I am looking at is using silcon nitride bearings, in either a roller needle or ball configuration. In my design they would be spinning very fast and taking some pretty hefty loads and shocks, so I thought using ceramic instead of steel would help shoulder the loads. The need for lubrication is also vastly reduced, as well as temperature stability. Can anyone point out reasons why they would not do well in this context?

 

[TDIMeister]

The problem with ceramics are their well known brittleness. Developments have been made along with better understanding of the fracture mechanics to address this, but they are still of fairly limited use in automotive applications aparts from, say, F1 and other rungs of motorsport.

The lack of fracture toughness is further aggravated when used in applications where shock loads, like from combustion, are involved.

One way to see how using power gear drives have been successfully done in series automotive production is to look at the connection between the rotor and eccentric shaft of a Wankel engine.

I guess the other factor around using carbon- or ceramic components, is still cost. Whether you're a big OEM sourcing for mass production or an engine developer building one-off prototypes, exotic parts will come at a hefty price. The smart engineer knows what is absolutely necessary and what is nice to have but is not a breaker for the design if he doesn't get it.

 

[unclematt]

I think that cost is more to blame for ceramics not being used more widely than their lack of desireable charecteristics. The same goes for using carbon-carbon, it is just too expensive to mass produce for the masses. Although, I did find several companies that produce carbon-carbon pistons for use in competitive 2-stroke motorcycles.

And I agree with your comment about the wankel serving as a good example to follow, that has not occurred to me. So I am looking into that more. If it works for a wankel, it should work for the OPOC too. Rotary engines spin up to pretty high rpms without the gear teeth failing, but I need to compare combustion pulse forces between the two different engines and see if they are comparable. I kinda doubt it. The expect the OPOC would have much higher forces acting on the gear teeth than a rotary, but I hope someone can disprove that. With the compression rations being what they are in the OPOC, I bet it will have much higher gear tooth forces to contend with.

 

[unclematt]

Hey Meister, I did some checking and I think that maybe the gear tooth load thing in cardan gears has been overplayed. Especially if you use cryo treatment to further strengthen the gears, which is what some rotary racers have done to the stationary gear on their engine. They even use spur gearing, not helical, and these gears could be engineered to have 2.1 teeth under load rather than just one. So I am going to focus on other issues unless someone presents evidence why I shouldn't.

I even thought about scavenging the gears out of a rotary motor, but they are not sized correctly for the OPOC arrangement under discussion (diameter is 4 x eccentricty for stationary gear, 6 x e for rotor gear : e=15mm in mazda rotaries). However, their gear tooth design and profile might be used in an OPOC hypo crank since it is a proven design. I still need to gather more data to compare tooth loads to be thorough.

 

[kjoiner]

Hello,

I don't know if you've seen this before but I ran across a site a few years ago about Clen Tomlinson. He built a model of a deltic engine from scratch. It's quite a beautiful piece of work.

http://www.craftsmanshipmuseum.com/Tomlinson.htm

Kyle