Crankshaft Modification 6

[mindenm]

Hi all,

I am neither an engine nor automotive engineer but recently using CAD I came up with a quite simple in my opinion addition to a crankshaft. It shows in that CAD model that power and intake strokes will last for 190 degrees of the crankshaft rotation but exhaust and compression ones- for the rest 170 degrees. At the same time a connecting rod deviates for up to 3 degrees from the direction of the piston center line during high pressure in a cylinder from TDC for about 130 degrees rotation of the crank. And only at the end of the power/intake stroke and the beginning of the exhaust/compression stroke that angle goes to about 16-17 degrees.
I would appreciate opinions of engineers in the field if those features would help to improve performance of an engine and if it is advisable to try to build a prototype.
P.S. It is a much simpler addition to the crank then Honda’s EXlink.

 

[BigClive]

Mindenm - I don't know exactly what advantage your mechanism would have over a "conventional" crank mechanism. I think that, generally speaking, for a fixed combustion chamber volume, equal length up and down strokes etc. it does not matter much what mechanism you use to achieve this. Thermodynamically and mechanically speaking I suspect that equal stroke length mechanisms are "independent of path" - it makes no real difference how you achieve it. I have seen a few mechanisms proposed over the years where the conrod is at a better (apparent) angle to the crankshaft when the cylinder pressure is higher - the idea being that the overall torque would be higher. But I think this is not so.
Your mechanism may be useful if it can be used to vary the CR, change the capacity of the cylinder, give a different expansion ratio to the CR etc. but otherwise I can't see any great advantage in your system.

 

[LionelHutz]

I'm not sure anyone has really commented if the extra 190 degrees of power/intake stroke compared to 170 degrees of exhaust/compression stroke would be enough benefit to bother more investigation with the extra complxity. I suspect it's not, unless you do more with the mechanism besides just vary the crank degrees.

By the way, the wheel shifts sides in the channel only at TDC and BDC so that there should be no impact.

This is actually why I would expect it to fail. As a minimum, you should expect to have issues with the reliability of the slide.

 

[140Airpower]

It won't fail if robust enough, but it will wear unevenly and how will it be serviced? A lateral link, actually a pair of links straddling the connecting rod and attaching on the other side would be more serviceable being themselves the equivalent of a pair of con rods. The motion would be an arc, but that should not be a problem.
It remains to show that the advantage of a longer duration (not longer length) expansion stroke justifies the extra mechanism.

 

[mindenm]

Talking about complexity and gains how about this Honda engine?

http://thekneeslider.com/honda-exlink-extended-expansion-linkage-engine/

Low side load on the cylinder and, yes, bigger difference between power and compression strokes. I think intake can be controlled by valves.

 

[Lou Scannon]

The question is, does this concept do anything to increase the net area of the P-V diagram, for the same chemical heat release?
If not, does it do anything to reduce FMEP?

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[140Airpower]

hemi, the area under the curve should increase IF we are talking about an indicator diagram that uses crank degrees as the X coordinate since he is lengthening the crank duration of the power stroke. What raises a question for me is there is no increase in actual expansion. So what then? I think we have to see an indicator diagram from this. Also, I think we need to consider heat rejection to the metal which is sensitive to the duration of the hottest phase of the cycle. If the piston moves more quickly at the start of the power stroke, that should reduce heat loss.
This design could result in small gains. For a stationary or a high duty cycle engine, small gains are big.

 

[Lou Scannon]

140Air, crank degrees, as you know, do not equate 1:1 to volume. In a P-V diagram, volume is what counts, not any surrogate for volume. Now if a cranktrain design change alters the rate of dP/dV resulting in more area under the P-V diagram, there may be something in it.
In an ideal Otto cycle, heat release is instantaneous at TDC. So the piston movement is zero for the entire heat release. Any downward movement of the piston during heat release departs from the ideal cycle, and thereby reduces the effective expansion ratio, which is a primary indicator of cycle efficiency. I think that heat rejection (waste heat) during the heat release process is a secondary effect, that should be considered only after the factors influencing the primary effects have been optimized.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[140Airpower]

hemi, I agree that crank degrees do not equate to volume and in fact in this design it is only the crank degrees to volume ratio that is being varied from the standard design. Total volume change for compression and power stroke are the same.
The actual SI cycle does not see an instantaneous burn and the peak pressure point is after TDC. Piston movement during the burn can affect peak and average temperatures and pressures, the allowable ignition advance as well as total heat rejection. There could be a difference and it would count for or against the efficiency of the design.

mindenm, is the downward piston movement quicker after TDC than the standard design?

 

[140Airpower]

hemi, This is from a Circle Track article ,"It is generally acknowledged that connecting rod geometry, particularly center-to-center length, can have a material influence on a variety of engine conditions. These include specific relationships to valve timing (camshaft design), cylinder pressure history, spark ignition timing requirements and torque output, the latter with respect to the actual shape of torque curves."

Read more:

http://www.circletrack.com/enginetech/ctrp_1011_connecting_rods/viewall.html#ixzz2Jr7fZwQz

The above is with respect to very small differences in the speed of piston motion around TDC that result from small changes in rod length ratio. With mindenm's geometry I expect much larger differences in piston motion from the standard design. I cannot guess how much or even be confident in the direction of change, but I think it needs detailed analysis and prototyping to be sure.

 

[mindenm]

Thanks to everyone for looking into my project.
140Airpower, as you asked I ran both mechanisms thru a cycle by 15 deg. increments. Looks like piston moves slower in this design which, on the other hand, might add some as seen below.
The first attachment in this thread indicates what and how dimensions are labeled/defined.
Results in this attachment are for this mechanism with crank throw = 1.187” and connecting rod = 4.5” taken on 15° increments when compared against a diagram of a typical ICE with a crank throw of 2” and a 6.1” connecting rod which I found on the web.
Stroke is about the same of 4” for both mechanisms.
I assumed cylinder diameter = 4” and compression ratio 8 for both mechanisms.
I assumed force on the piston = 1 for both mechanisms as well.
I picked up dimensions arbitrary with cylinder offset of 5” for the mechanism of the concept.
On the Attachment:
Fr – moving force of torque = 1/cos(Y) * A / B
A and B – arms for torque calculation
Fx – side force on the cylinder = 1* tan(Y)
M – resulting torque = Fr * E
M_2– torque from force = 1 in mechanism with crank throw of 2” and a 6.1” long conrod
H and H_2 – conrod far end (piston) travel from TDC for the respective mechanisms
V and V_2 – change in cylinder working volume for the respective mechanisms
Rate of volume change is different for these two mechanisms as can be seen easily from comparing values of H and H_2. As a result, if I assume the same combustion rate in both cases, then working pressure in cylinders will differ as P = P_2 * (V_2/V)↑1.38. (P and P_2 - assumed pressure in respective cylinders). Then for true comparison
FF – moving force of torque = Fr * (V_2/V)↑1.38 - for the mechanism of the concept
MM – resulting torque = FF * E - for the mechanism of the concept
And that is shown on the plot.

In the side 2 columns are calculations of areas under torque curves for both mechanisms in the power stroke and how they compare – 21% difference. It has to be work done in the power cycle.

P.S. I checked that if a guide/channel is separately moveable the mechanism can be turned into VCR.

 

[mindenm]

140Airpower, hemi, kirrer and ... would you advise to try to build a model/prototype or you think it is not worth it? To build a VCR is definitely above my level even though I can model a mechanism in SolidWorks.
Would appreciate any opinion.
Thanks again

 

[140Airpower]

mindenm, I think your modeling can verify the mechanicals. A working model would be needed to verify the thermodynamics, a tedious and demanding job. But, there is software that can do a good approximation, except it is reputedly very expensive. In any case, it looks to me that the advantage, if any, from the current design would be small.
Converting this to VCR is another approach that can give greater advantages that can be reasonably well estimated without prototyping. There are operational conditions that call for higher or lower compression ratios to maximize economy or to avoid detonation under high load. For this, you don't necessarily need an unsymmetrical stroke duration. You could alleviate some of the high stress aspects.

 

[Lou Scannon]

I agree with 140Air, the next logical step would be thermodynamic analysis. There's no point in building a working model until you have analysis indicating that the design is likely to provide a quantitative advancement on the current state of the art.
As for variable compression ratio, we all know that would be an advancement on the state of the art; you would need to perform mechanical and structural analysis to show that it will function with the desired range and rate of transition, with acceptable power requirement and stress margin. Saab was playing around with real time variable compression a few years ago; I don't know if it is being commercialized yet. Keep in mind, full authority VVT is already commercialized, and provides many of the benefits of VCR, and other benefits as well.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[mindenm]

140Airpower, hemi:
Thanks for your suggestions. I'll try to follow. Meanwhile attached is one approach I am looking into for VCR.

 

[BrianPetersen]

If the length of the power stroke is equal to the length of the compression stroke then I believe any analysis that shows any appreciable difference in IMEP or thermal efficiency or anything of the sort, is flawed.

There was mention of a calculation done at 15 degree crankshaft increments. This is nowhere near sufficient. Try again at 2 degrees then try again at 1 degree.

Changing the duration of the power stroke relative to the compression stroke (but not the length) only changes the amount of time that the stroke takes ... not the total amount of energy delivered. I realize that there can be some secondary effects due to heat transfer, leakage past piston rings which is a function of time, friction against the cylinder walls, etc., but these will pale into insignificance and in the idealized case, their effect is zero.

The Honda ExLink system (interesting, I hadn't seen it before) is different, because the length of the power stroke is physically longer than the compression stroke. This turns pressure that would otherwise be let out the exhaust valve into additional expansion work. That it might result in the compression and expansion strokes taking a slightly different number of crank degrees isn't the point, since in the idealized case it has no effect - the additional expansion of the gases is the point, and the idealized cycle will most certainly be affected. That the arrangement appears to result in less piston side-thrust (biggest source of friction in most engines) is a bonus.

 

[140Airpower]

mindenm, It looks like you shift the static angle of the channel? And that changes the stoke length and the CR. Looks like it will work. Have you examined existing designs for the comparative ranges of variation and the apparent complexities?

 

[140Airpower]

mindenm, This is from Freepatents Onine.

http://www.freepatentsonline.com/6990934.html

The design has similarities and differences to yours.

 

[mindenm]

BrianPetersen,
Thanks for your analysis. I still have a couple of questions. If "the amount of time that the stroke takes" in a cycle is different and the power one is longer, will it add to an engine efficiency eventually? If a piston moves slower then combustion volume increase is slower then and pressure in a cylinder stays longer. And if turbocharge is added in this situation? Honda EXlink reduced side load (friction) in the cylinder but added a complicated mechanism with a lot of additional friction. In my approach a 2" throw crank is replaced with 1 3/16" throw crank and a roller in a channel/guide. Their combined friction path is shorter than for a 2" throw crank thou, I admit, forces are higher. Do you think there should be some advantage in this mechanism with those considerations?
140Airpower,
So far I looked only in kinematics. For 4" stroke engine 4 degs. swing of the roller's path will result in 8 to 18 range in compression ratio. If instead of a swing the path/guide is moved to and away from the crank .25" movement should give the same result. Besides, there is no power needed to move or swing the path/guide. Force from the roller will do the job. Only stops will have to be moved but it can be done when the roller's force is on the other side of the guide.
I checked the existing VCR proposals and looks like this one is a novelty. But in order to check for complexity I think some design should be done but I have no expertise in engine design and kind of reluctant to start in a new for me field.
The Japanese (Honda again) patent one is interesting but different. Thanks for bringing it up.

 

[Lou Scannon]

mindenm,
do you understand P-V diagrams? A key thing to note is that time (or any pseudovariable, such as crank degrees) is not a consideration; only pressure, volume, and the net area. You have to think in these terms when you begin to evaluate any heat engine concept.

"Schiefgehen will, was schiefgehen kann" - das Murphygesetz

 

[BrianPetersen]

For idealized P-V diagrams (ideal engine cycles) the length of time or number of crank degrees that a power stroke takes is completely immaterial, as mentioned above.

There can be secondary real-world effects - some positive, some negative, most are interrelated with other factors.

Lengthening the power stroke in time (making it take longer in time, but over the same volume) lengthens the time that the hot gases are in contact with much cooler cylinder wall and piston surfaces ... you are losing to heat transfer the more time that the power stroke takes.

Piston sealing is, unfortunately, not 100% efficient, either. The longer the power stroke takes, the more pressure leaks out into uselessness, past piston ring end gaps and past imperfect valve seats.

But on the other hand ... there is something to be said for letting the piston dwell near TDC in order to allow combustion to go further to completion while there is still more available expansion left. In other words, what you really want to do is have the piston quickly perform the compression stroke, then sit near TDC for as long as you can get away with but just enough to get combustion substantially complete, then pretty much snap the piston to full expansion as quickly as possible.

Real world limitations on piston acceleration, of course put limits on what can be achieved, even if one devises a mechanism to achieve this.

And then there is the elephant in the room ... detonation. With a premixed-combustion engine (standard spark-ignition Otto cycle), often the combustion process has to be intentionally delayed so that some expansion occurs during combustion, to deliberately reduce cylinder pressure and temperature a little, to preclude detonation. Having the piston dwell near TDC is not consistent with that real-world phenomenon.

And for whatever you can gain by mechanically making a mechanism to dwell the piston near TDC ... could also be achieved by using a fast-burn combustion chamber, and there are lots of ways of doing that.

Study the means by which the effective power stroke can be made longer than the effective compression stroke ... because that really does change the P-V diagram.

And before jumping through hoops in order to achieve this by mechanical means involving the crankshaft and connecting rod ... Plenty of current-production engines achieve this objective simply by varying the valve timing on an otherwise-conventional layout. The P-V diagram doesn't care if you suck in a full cylinder full of air and then lets part of it back out again in order to create a lower effective compression ratio. Notable example in production today (and for years): Toyota Prius. The engines used in Ford hybrids do this, too. The base engine in a Honda Civic does it (sorta) using the i-VTEC mechanism. BMW does it using double-VANOS. Fiat does it using MultiAir. There are others.