Although this really covers every engine, I thought it would be a good tech subject. I also haven't written anything in a while.

Working around diesel engines up to 2000 hp at work all day, I started to think about various engines, sizes, types, and their robustness. Diesel engines are typically very large and very heavy. They are also very tough. Even diesels that dont' make a whole lot of peak horsepower are over built compared to gasoline engines of the same power level. I never really thought about why but my first inclination was the same as everyone else I've ever asked.

When the topic of engine robustness pops up in relation to diesels, the first thing you hear is that they need to be overbuilt because they have higher compression ratios. Sounds logical. I've also heard that they are overbuilt because they are typically designed to last longer. That's fair. It's obviously not due to rpms as these just don't rev that high.

Gasoline engines are typically built stronger to account for forced induction. We understand that. Again, people say it goes back to more air which means a higher effective compression ratio. It's effective and not static compression that matters in terms of engine stress so even a low static compression engine will see a high dynamic compression ratio after boost is introduced.

If we try to draw a correlation between engine size and material the block is made out of, it's hard to make a direct comparison. That's due to the fact that aluminum technology is getting better and better and we are about to use it more and more in place of older and heavier cast iron. It has generally always been understood that larger means stronger and diesels are built stronger yet as well with the same rules applying to strength as they get larger. Bigger needs to be stronger and the real key to why lies at the beginning of this sentence.

Engine strength requirements have more to do with one single thing than anything else. It doesn't mean that there aren't other issues that need to be addressed that also affect this. It just means that there is one primary thing to design around. That issue is torque. To understand why, we must first understand what torque is. Besides being a second rate motorcycle movie, torque is a measure of how much a force acting on an object causes that object to rotate or simply as I like to think of a force of twist.

The more torque you have, the more twisting force you have. This is what your engine block sees. Let's say we have a big diesel engine that makes 250 hp but 500 ft lbs of torque. Let's also say we have a small (relatively speaking of course) gasoline engine that makes 250 hp but 200 ft lbs of torque. Guess which one is bigger, more robust, and heavier? While static compression has a little to do with it, it's the torque that dictates this.

Let's also look at a high revving F1 engine that makes around 900 hp at 19000 rpm but only about 250 ft lbs of torque or so. Maybe not exact but it's low by comparison. Those engines are light. True they don't need to live long but if we were to build a 900 hp diesel engine to the same specs it would self destruct the first time it was turned over.

Now let's bring our beloved rotary into the mix. We don't have an engine block. We have a stack up of parts and this torque gets absorbed by the tension bolts and 4 short dowel pins. Scary thought isn't it!!! Fortunately for us we don't make much torque. RPM and horsepower are irrelevant when it comes to this. When we see high horsepower rotaries breaking housings around the dowel pin lands, this is why. Those high power engines are making high amounts of torque which is greater stress. The solution is to add more dowel pins. This stiffens the engine up from a rotational access standpoint even though you have to remove some material to install them. That's OK though. The stresses are being exerted in different directions from what you would think would tear it apart.

Now I do have to point out (for all the nitpickers) that detonation and other forces inside the engine can have negative effects on rotational stresses on the engine as well. You can have 1 engine detonate (which shoots forces inside the engine way up!) and it'll break a dowel. You can also have a finely tuned engine make twice the horsepower and not break one. In a perfect world we'd have the same situations acting consistently but that's just not always the case.

On the older rotary housings from 86-88, the upper tension bolt location on the front housing on the drivers side was known for breaking easy after about 400 rwhp and sometimes sooner. This happens even if everything is tuned well. This is because the torque that these engines make boosted to these levels are similar in torque and that's the limit for the castings from these years. They're too thin for much more and need to be reinforced.

Now if the thought of twisting your little rotary apart is scary (and it really shouldn't be), just think how scary it would be to have a large V8 built by stacking in the same manner!!!

Just a bit to think about. I'm sure there will be some nitpicky debate but the gist of this is that the primary thing affected robustness of engine design is torque. Everything else is a secondary design criteria behind it when it comes to engine strength. The more torque you have trying to twist your engine apart, the stronger it needs to be.

 

Rotarygod, how does the rotary engine stack up in use as a load bearing member of the chassis. I know in racing piston engines use of the engine block to brace the chassis across the gaping maw of the engine bay.

 

You can not use a rotary as a load bearing member as you can a piston engine block. That has been a hurdle in racing for as long as they've been using them.

 

nice write RG. and i know torque is also more important the hp that most people so love to get. and one thing i would lilke to do in the future is to get the 16X e-shaft swapped out and into my motor.
but good analyze. i wouldn't have paid much details or could've possibly overlooked on the importance of dowel pins in this engine design....extra ones.

and since we're on that subject of torque, it can just as well be worth it to get the whole 16X motor into our 8's. thats on the assumption that its safe to assume that the 16x is goin to be stronger and better. although stated to be built around the renesis... i'm sure there are other key points that make it stronger, other than the 3 oil jetting housing locations which someone pointed out in a thread of the 16X.

nice read. informational.

 

I've been wondering if they could machine the internal "ribs" of the rotors to be angled where the the oil is sprayed into them (towards the center housing) so that when rotating the oil and the rib meets more squarely.

Trochoid Magic, I belive the 16x is going to utilize a larger offset since the 16X housing is visibly larger than the 13B.

 

Horsepower moves you. Torque tries to twist the car apart in the process.

You will not be able to swap a 16X shaft into a Renesis. It's an all or nothing deal which means pick an engine. You can't combine parts from both though.

 

Rotarygod, I am assuming from the job description and pics you posted somewhere one time about what you do, you know a fair bit about metallurgy. How do you feel mazda's current process for housings could be improved? Also, how do you feel about the new aluminum housings for the 16X?

It seems the process BMW is using to produce the 3.0L twin Turbo N54 aluminum/magnesium block would produce better housings than an aluminum alloy...

 

i've always thought that Torque is what "moves" a car as in it's the turning force that gets transfered to the wheel. while Horsepower is power needed to keep you up at that speed? (energy required to move 1kg thing certain distance).

 

/Thread Derailment

Torque is the twisting force and horsepower is the twisting force in motion(through time).

The tire changing anology is the best way to visualize it.

If you are trying to use a tire iron to twist off a tire nut, you are applying torque.

Now lets say you are exerting 100lbs of pressure and your tire irons handle is 1 ft long you are applying 100 ft/lbs of torque. Now lets say the tire nut isn't budging, you are applying 100 ft/lbs of torque, but since its not moving you are creating 0 horsepower.

Now it starts moving, lets say you can turn it 30 times a minutes, but its much easier to turn, lets say you are only using 10 ft/lbs of force. Now you are exerting 10 ft/lbs of torque but making .057 Horsepower.

The mathematical formula for horsepower would be HP=(Tq X RPM)/5252

Since they are tied together you can't just seperate them.

/End Thread Derailment

I never thought of it that way RG. I always wondered why diesels were built up so much. I was always told, and believed, it was due to the higher compression.

Maybe having a flat torque curve is the best way to push the design envelope on a modern engine, or perhaps when you push the design envelope you will end up with a flat torque curve.

 

I'm sure it will turn into a big horsepower/torque debate but at the end of the day ONLY horsepower does work. Can't change that.

We have a test engine at the shop that is a 12 cylinder 2 stroke, quad turbo, diesel. It makes 2000 hp at 1800 rpm. The motor is pretty good sized. If you figure out it's torque there, it's making 5838 ft lbs of torque at the same rpm. Quite a bit.

I was at the OTC (offshore technology conference) in Houston last week. There was a GIANT EMD diesel generator sitting there. These are used for power generation in offshore oil rigs. It was easily the size of a 3 car garage. It was a 20 cylinder 2 stroke supercharged diesel. It made 5400 peak hp but believe me when I tell you, it's WAY WAY WAY larger than the 2.7 times the size of our 2000 hp test engine as the power rating would imply!!! Why? Torque!!! Their engine makes 5400 hp at only 800 rpm which equals 35450 ft lbs of torque!!! Yikes!!! Now that's a block ripping twisting force!

 

I believe the horsepower vs torque debate comes from looking at dyno charts. Everyone wants to know what is the best number to look at.

In my opinion it would be the torque curve, because since there are several plots you are seeing torque over time. Horsepower at that point becomes a useless metric.

I used to work in refineries and the back up power generators they have in some or the refineries are pretty large.

 

That is the key. If there is motion through time, torque is not relevant. Horsepower is.

Higher combustion chamber pressures do affect engine design in the strength department but in other way as the forces being applied aren't in the same plane of geometry to what torque is trying to do. I've started to understand far more about this as I've gotten more involved in internal mud pump design. There are certain things that look like they should be done one way until you realize in which plane the forces at work are acting on them. You need to train yourself to think in 3D in regards to forces rather than 2D. An attack can come from any side and so can stresses. Stresses can be shearing, or compressive. You can have multiple shear stresses in multiple directions in the same location. It all depends. I'm barely scratching the surface but hopefully that gives you some idea as to why things are the way they are.

Needless to say detonation or high combustion chamber stresses aren't trying to twist the block apart. At least not in a piston engine. They are trying to blow past the piston rings below them. They are trying to blow heads off above them. Imagine the forces on the studs! This is the same plane in which combustion forces from comression are exerting themselves and it is not the same plane that is trying to rip the engine apart. Once the force of combustion pushes down on a piston, it applies this force to the crankshaft. The crankshaft converts the up and down motion to rotational motion and it is what is responsible for creating the torque in the same plane as the length of the engine and this is what is being absorbed by the engine block. The rotary is a different animal but I'll deal with it later.


Now you're thinking however expand on that a bit. Remember the problem isn't making power, it's getting it to the ground. This is why we keep seeing more elaborate transmissions that keep the engine within it's powerband for a longer period of time. However the solution is so simple that it's stupid. Think locomotive. These are diesel/electric devices. Don't think in terms of hybrid. That's a stupid term to use anyways. The reason why locomotives use electric motors for movement but diesels for power generation that gets sent to the motors is so they don't have to use transmissions. Can you imagine the nightmare that would be involved in trying to shift a transmission on a train! How big and robust would the transmission have to be? Huge!

An electric motor makes peak torque at 0 rpm. Remember you can make torque anywhere including 0 but not horsepower. However with this massive torque, the instant it starts rotating, it is also making massive horsepower. Think of the diesel engine as the main powerplant just like the engine in a car. However don't think of the motors as mere electric motors. Think of them as fully electric transmissions. Their input isn't mechanical. It's electrical. Now you can run the main engine at a constant rpm where it is most efficient. The electric motor is capable of taking this amount of horsepower that is being produced and can apply it over the ENTIRE useable wheel speed! Currently as you accelerate and your rpms climb, you make more and more power as you go faster until you hit peak power right as you shift. Imagine being able to make peak power the whole time!!! That's faster! This is what electric motors can do. Technically people refer to this as series or serial hybrids. I refer to is as smart. Since it's average power that wins races and not peak power, you can reduce the primary powerplant in size and set it up so that the average amount of power over a certain wheel speed matches that of a conventional transmission oriented setup but you can get it with a much lower peak horsepower engine which means you still get an efficiency increase!

This got off topic and I'd be more than happy to discuss it in depth elsewhere but now hopefully you are starting to see far more potential in those crazy devices that pass you everyday as well as a more creative way to think about them.

 

People can look at it from any standpoint that they want to. Whatever frame of reference makes them confortable. There is more than one way to think about things. Technically speaking though it's horsepower that does work.

Dyno's measure torque being applied and then extrapolates horsepower from it. I'd personally rather see nothing but horsepower plotted leaving torque completely off. To make comparisons easy, I'd also like to see them all plotted on a standardized scale as each other so an overlay is easy.

 

I knew that about electric motors already. Except for the back up power everything in a refinery runs off electricity, even the electric turbines that propel gasses used to make the product(unless the product is a gas, then that too). But good food for thought for others that don't share similar experiences.

As far as stresses on studs it shouldn't be anything a stud couldn't handle. Since the engine doesn't have much leverage against the studs they should be able to handle massive amounts of stress. Now this is only in the frame of car engines mind you.

I've been interested in the hows and whys engines have been built for a few years. The viewpoint you introduced is definitely new to me. The next question would be how would this uniquely relate to the rotary engine.

 

People should start think of electric motors as the ultimate cvt's.

It's not very likely that you'll break a head stud easily. They do of course make stronger ones. The head gasket usually goes way before the studs ever will.

I need to think about my wording a bit as it can easily get confusing and I may inadvertently imply something that I don't mean. When I get it worked out on paper, I'll write it out here.

 

I understand your example, I merely made the mistake of taking it as more than that.

Electric motors have their own pro's and con's. While electric motors are well understood and are quite efficient, I think other technologies (batteries, weight, heat) hold them back in automobiles.

I think gas guzzling engines are here to stay for at least another 30+ years. Theres just too much room for improvement.

 

good topic! i was just reading something about that the other night. on paul lamars website

to go back a bit, the diesel truck motor, is gigantic next to the F1 engine, but the diesel is designed to make its peak hp for a really really long time (300k miles?) while the F1 engine only has to make peak power for like 600miles

f1 engine is a poor choice for this example because its designed to live at full power like the diesel, just the life of the f1 engine is much shorter.

 

I think of electric motor in cars a bit differently. I think of them in the same way as trains. I don't see why batteries need to hold them back. Literally do them the same as trains. No batteries. Straight series. While this isn't going to give you the impossible overnight answer to efficiency that everyone wants, it is a very important stepping stone. No transmission. You can couple a single electric motor to a differential. That's the easiest first step. Next comes hub mounted wheel motors which does away with the differential. This opens the door for electronic traction control and electronic abs. As battery technology improves, integrate them. Now the engine can move away from it's duty as primary power generation to supplemental for batteries. Basically a range extender. Adding batteries allows power storage from regenerative braking. You get the idea. Everyone keeps seeing extreme examples of future tech but never the actual product. When something is too different, no one accepts it. It will probably also be overly expensive. What I propose allows for natural progression without the added complexity and costs associated with young technology. In many ways the vehicle gets simpler and simpler with each advancement.

What's really holding them back is the logic and vision needed to implement it properly. Parallel hybrids such as the Prius are quite frankly a joke and complete waste of time. It's an approach that has so little total potential but added complexity rather than less. That makes no sense.

 

EDIT: I type too slow, apparently i WAS on the right track, RG just beat me to it and stated it FAR better. The picture i have in my head resembles a flintstones era car, while his probably resembles spacecraft from the 32ns century, ROFL


With the flexibility of an electric motor(if thats the right term) and its capability to deliver power/torque the way it does.... forget the batteries, use an internal combustion engine to generate the electricity needed for the motor.

Could/would likely be much smaller and less complex than the motors in our cars now as they could be designed like the afore mentioned diesels that run a constant rpm for large percentages of their lifespan. That would most definately reduce complexity in valving, valve timing controls, intake(like our muti length runners that would not be needed)

I may have missed RG's line of thought completely... but what i have envisioned is an internal combustion generator driving an elctric motor that would take the place of current transmissions. Technically the electric would be the vehicles powerplant, and the combustion engine an accessory(like a fuel pump)

 

I'd like to point out something that might be misleading people. Horsepower is of course a function of torque and revs. More torque means the need for a stronger engine block, or studs in our case. This might lead people to believe that increasing revs would be a way to gain horsepower without needing to strengthen the engine. While it's true that the blocks won't need as much strengthening, more revs brings their own needs. When you increase the revs, the stresses of the bearings go up, necessitating stronger bearings and in piston engines connecting rods also need to be strengthened. While a gross oversimplification, producing more torque to making more power generally requires the strengthening of pieces that are static, while making more revs requires the strengthening of parts that are rotational. Of course certain parts need to be strengthened regardless; connecting rods and valves come to mind. Nonetheless, revs is no free lunch to making more power, but it has generally shown to be a bit more weight effective, but less cost effective!

 

eh, im probably not the person to comment about this, but your definition of torque isnt "entirely correct." it works, but it is not the best definition. sorry, just finished a test on it and...yea for others who are listening, torque is perpendicular distance times the force exerted. imagine a disk with a rope and a weight tied to the edge of the disk. the weight is not at rest besides your hand holding it up. you drop the weight, the weight provides a torque on the disk. i wont get into rotational inertia...

 

I provide disclaimers for way too overly exaggeratingly nitpicky people like you! If you don't state things in an overly simple stupid way, half of the people won't understand it. I'm not saying that everyone is stupid. I am saying that simplest is usually best when you are trying to get a mental image going. Therefore there is nothing wrong with it in the context that I am using it in. If that gets you going, you should see how I use the terms stroke and cycle independently of each other in a tech article as well as how I describe the rotary as the 6 stroke 4 cycle engine that it really is!

 

All true and again why I have to provide disclaimers stating perfectly clearly that there other forces that affect engine design. One thing to think about though when it comes to adding strength in important areas for higher revs is the directional forces that these improvements are trying to account for. Is it the block twist or other areas that need the most attention when it comes to higher rpms? I know most probably haven't thought about it this way before but the answers may surprise you.

 

Your timing is impeccable. I was thinking about torque and HP for the last few days and was going to start a thread on that very topic. I know that was not the real intent of the original post but the subsequent discussion seems to be heading this way. I apologize if this is a complete hijack.

Radius (ex. feet) * Force (ex. lbs) = Torque (ex. ft-lbs)

(Torque * Engine speed) / 5,252 = Horsepower

You speak of the terms as if they are independent but I interpret that horsepower is a result of torque over time. To take HP by itself would seem meaningless since it does not provide a reference point of which speed you are applying it. I do understand how torque in and of itself would affect engine design from a structural perspective and how speed would as well (centripetal force, etc.) but I do not see how HP would matter.

I have been trying to figure out the factors involved in generating HP numbers on a dyno. In So Cal we did a dyno day and people seem to be up in arms about the results. For me, I could care less who's numbers are higher. I am more interested in accuracy and repeatability so that I can measure changes in the performance.

Keeping this in mind, the factors that I can see which are relevant to RWHP would be the tire diameter, the load being applied, the rotational speed, the increments at which it is measured, and the increases in speed in that period of time (acceleration).

In my limited thinking, the only thing that would provide inconsistency in dyno readings that were not real would be the ratio between the tire and the dyno wheel applying load (or force). The distance from the center to the edge (radius) of the dyno wheel would be consistent as the dyno wheel does not change. Everything else is what it is.

If what I am saying is correct, why is there such a vast difference in the way dyno’s measure cars and what configuration actually can be made to a dyno (through the program) that can affect the accuracy besides the tire size?

And why does one dyno read HP/torque differently than another?

 

lol...good deal