Please review my ICE thermodynamic model

[Kozy]

I'm looking for some engineering types to have a play around with my thermodynamic model of an internal combustion engine and reply with any thoughts, errors they find, spurious results or anything else which doesn't seem quite right.

http://blackartdynamics.com/Engine/EngineThermodynamics.php

Happy to answer any questions on the underlying maths.

 

[Kozy]

I have been doing some more tweaking with this, now added IAT and exhaust pressure controls for finer tuning of the model.

Any feedback appreciated.

 

[jnnx]

really interesting. as far as I can tell great work :)

I have put there know parameters of my engine (subaru ez30d) and it ended up 100hp (180kw vs 348bhp) stronger than reality (MFR claim).
also, I don't know real VE. maximum what MAF will show is 200g/sec. from my aprox calculation that seems to be around 95% VE (at 6000rpm).
to get real hp output from calculation, I have to drop it to 70% VE. but I don't think that is real

unknown parameters: rod length, piston weight, VE (kind of), exhaust pressure, IVC. AFR and timing taken from engine ECU maps

 

[Kozy]

Not sure how you got it to 348bhp?

 

[Kozy]

I put 89.2mm bore, 80mm stroke, 6 cylinders and 10.7 compression in and it gives me 254bhp at 6600rpm at 85% VE.

 

[jnnx]

if I only change bore,stroke,cylinders, CR, 6600rpm I will get 273hp
as I change AFR to 11.5 and timing to 29 (values I know from my logs and ECU maps -> Rom Rider), it jumps to 327.
VE still at 90
I can lower the VE to fit the result, but I don't think it would be correct

I'm not sure about the correct way to calculate VE, so please correct me if I'm wrong.
what I know from logs:
air pressure 0.95bar
WOT
6031rpm
engine load 1.9 (grams of air per engine revolution)
I don't have temperature logged, but it was 30jan around 5 degrees celsius outside.

then I let wolfram alpha calculate this
"volume of 1.9g of air at 0.95bar 10C"
and it calculate the result as 1625ccm. 1625*2 = 3.25l
that is over 100% VE. I don't think it is necessary correct, but I think value between 70-80 would be more incorrect...

 

[Kozy]

I see the problem, that's an incredibly rich AFR you are running, more than is really scripted for. The AFR input only adjusts how much fuel is added, but at present does not adjust how much is actually burnt, or it's effects on burn speed or temperature. This is one point I do need address in future, but for now the calc seems to give the best results at around a 13-13.5:1 mixture. Adding more fuel will appear to add power, though in reality it won't because only 13.5 is being burnt and the rest is being used to prevent detonation and cool the charge.

I have not used the VE as a calculated figure like you have, merely as a kind of throttle input. Set all the other values, then do the VE last to get the power figures required.

Also, the IAT is the temperature of the air entering the cylinder, this is almost certainly not 5° once it's traveled through a hot engine bay and heat soaked intake system. I think something in the region of 15-20 over ambient would be reasonable when moving.

If we say 13.5 AFR, 25°C IAT, 29° timing and 80% VE it comes out at 249bhp at 6600rpm.

I will need to get that AFR control better suited for sure, that will take some thinking...

 

[jnnx]

As far as I know, engine will produce most power at AFR around 12,5. but with proper ignition tuning.

Will you do something with how the VE is used?
Or will you include calculations based on load / MAF output? guys who work with cars know this from data logging and may be interested...

 

[Kozy]

It will produce the most power at that AFR, however the fuel charge is probably not being entirely burnt, the excess is going towards knock supression to allow more ignition advance. I need to detach the fuel consumed from the fuel burnt, otherwise the energy released and as a result the temperature and pressure produced, is unrealistically high with full power rich mixtures.

I will have to have a think about how to do it, and probably incorporate some control over the burn duration in the process, currently that is only linked to compression ratio and RPM.

I'd be willing to have a look at the VE input and using some other inputs in it's place, but at the moment I wouldn't really know how to approach it, especially with regards to the engine load aspect.

 

[Kozy]

OK made some more adjustments to it. For now it's got a crude lock out at 12.5:1 AFR, fruther increases will not affect the thermodynamics but will affect the BSFC reading. I will improve on this later.

Also added the position of Pmax in crank degrees and a Performance Index number, which is explained in the link.

Another thing to consider is that you are looking at peak power. A VE of 80% might be low to get the rated power output, but peak VE is typically at peak torque, and VE at peak power will be lower.

 

[jnnx]

Another thing to consider is that you are looking at peak power. A VE of 80% might be low to get the rated power output, but peak VE is typically at peak torque, and VE at peak power will be lower.

I don't think this is the case

from my logs, the highest load (1.95 g / rev) is around 5500rpm
load at peak torque (4200rpm) seems to be about the same as at peak power (6600rpm) - 1.9 g / rev
difference I have just noticed is, that at peak torque the manifold pressure is 0.97-0.98bar
at peak power it falls to 0.94-0.95bar

but all this is probably engine specific. this engine is tuned for middle/high rpm, and doesn't have a lot of low end torque

I'm going to look, if it is possible to calculate some VE numbers from load...

 

[Kozy]

Interesting, I do appreciate your input as I do not have access to this sort of data from a real engine.

I'd be interested to see some workings out on how you derive VE from load, load in the way you have described it here is not something I have looked at.

Am I right in thinking that it's basically the reading from the MAF sensor?

 

[xxChrisxx]

80% is a fairly good guesstimate. The air throughput per rev shouldn't be too different between peak torque and peak power.

The (very) rough rule of thumb for MAF to power readings is:
Power in BHP * 0.8 = airflow in g/s
or
1.25*MAF reading = power.

It worked for my Golf.
Stock = 120g/s = 150BHP
Remapped = 150g/s = 187BHP.

I'm not sure how accurate it is for non peak power readings though, or for other cars.

This is not to say that 80% volumetric efficiency is right though, it feels a bit low.

 

[Kozy]

Maybe, but then is anyone really able to say with certainty that it should be higher?

Race cars might manage 100%, and putting 100% will get you to a race car level of power output for a given engine.

I think 80% is reasonable for a road car.