Donations towards server fund so far this month.

 
£0.00 / £100.00 per month
Page:
Home > Technical Chat > A Series simulation thread.

Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans




On 12th Nov, 2014 adcyork said:
On 12th Nov, 2014 Paul S said:


A reverse engineering exercise such as what is going on here, with a current engine configuration requires the model to be correlated against cylinder pressure (and derived burn data) and pressures/temps from the relevant ducts. This allows the Wiebe model and other parameters to be tweaked to bring the model into a correlated state.

Simply correlating the power and torque curves by tweaking the model is highly inaccurate and only a small part of the picture. If this was the only goal then a number of engines could all well look very close to an A-series engine on paper




I'd rather have sensor data as well but I don't have the cash to spring for a multi sensor setup. However this is not a start from scratch program and has been verified (but not for siamese models I reckon)

You have to tweak a bit but you have to keep it sensible. You can force a super high correlation when you start changing values at will. But the simulation will be rubbish as it looses all predictive capabilities.

I'm currently working to see it I can help somebody out with an odd idea. Luckily he has a dyno at home so that makes it a lot more effective..



That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

Paul do you have data for mach numbers at the valve at about 8k ? one inner and one outer if at all possible

Cheers

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Paul S

User Avatar

8604 Posts
Member #: 573
Formerly Axel

Podland

I don't have Mach numbers.

What are you looking for? Velocity at the valve?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

Hi
velocity at the valve would be good. I get a significant difference between the inner and outers .

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

Paul you where right about my fmep numbers btw . I sorted them
and now i get something like 2.9 to 3 bar at 8000 which is more sensible.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Paul S

User Avatar

8604 Posts
Member #: 573
Formerly Axel

Podland

Jolly good.

I forgot about the valve velocities, so I've just had a look at my latest sim results.

I can't find velocity across the seat, but max. port throat velocities at 8000rpm are as follows:

Outer - 128.6 m/s
Inner - 150.9 m/s

As the outer cylinders run a higher VE you would expect them to run a higher maximum velocity. However, this is not the case as the VE depends on the area under the mass flow curve. As the outer cylinders get a huge helping hand from the inner cylinders, they do not reach as high a peak velocity. Have a look at the mass flow curve on page 1.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."


Paul S

User Avatar

8604 Posts
Member #: 573
Formerly Axel

Podland

Just to clarify: the data is inlet velocities for my latest 998 with a 12G295 head, 33mm inlets and tuned length inlet and exhaust manifolds.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

I get significantly higher velocities at the inners as well

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

If somebody is interested I posted some simulation stuff on Long Outer Branch manifolds on me blog. Its use for turbo's is pretty much zero but hey .. if you want to run a 8000 rpm 998 give me a PM

Oh,
In a few months I'll have access to A-series in cylinder pressure data :) The guy I've been helping with exhaust design has actually bought a TFX setup .

Edited by Sir Yun on 11th Nov, 2015.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Paul S

User Avatar

8604 Posts
Member #: 573
Formerly Axel

Podland

Interesting that the longer outer branch increases the trapped mass/power on the inner cylinders *wink*

With the longer outer branch, you have introduced a pulse at the point when charge robbing ocurrs, when the inner and outer cylinder intake valves are open, but also the outer exhaust valve. That pulse is limiting the charge robbing.

As far as 8000rpm turbo 998s with long outer branches are concerned, you should read my build thread *smiley*

My latest manifold is still 3-2-1 but the outers are only 100mm or so shorter than the centre branch.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans



Maybe SCB would be a better name.

Well AFAIKT you need to keep the Centre tuned for the centre and the outers for the outers and by that logic the outers need to be longer and the inner shorter. Combined length of the outer (Pri+Second) would be about 1000mm the inner slightly under 800mm.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Paul S

User Avatar

8604 Posts
Member #: 573
Formerly Axel

Podland

I'm not sure that I agree with your logic. The data on your blog would suggest otherwise.

Also, lets keep it practical. The longest you can get under the bonnet is around 600mm. I think mine are 550mm centre and 450mm outers.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

These are the power per cylinder # 1&4 for both configs.

So basically the same. You do tend to gain a bit lower down but this cam is 301 degrees at clearance so it does not do much down low at all. RP reckons he is only under 4000 rpm when something is wrong.

As it is NA we can get a lot more length in (He has build a very nifty sequential shift linkage above the floor)

the class is like a Miglia with a lot less restrictions.


Attachments:

Edited by Sir Yun on 13th Nov, 2015.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Paul S

User Avatar

8604 Posts
Member #: 573
Formerly Axel

Podland

Well I'm confused *wink*

This chart shows that the longer outer branch manifold improved the inner cylinders:


Or have I misunderstood you?

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

well it is more the short centre in combo with the longer outers. but I already had SCB for an other config :)

the engine is NA and the shifter linkage is moved up above the floor so it can link up under the floor

Edited by Sir Yun on 15th Nov, 2015.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

I have seen the in-cylinder pressure figures and they are surprisingly within about 1% of the predicted values. Can't share much else but it is pretty nice to see that the simulation seems to work well

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Evoderby

224 Posts
Member #: 9987
Senior Member

Amsterdam

Not bad, especially considering the average physician I know can not even manage to hang a painting on the wall without screwing up big time*wink*

Nice blog, nice train of thought...will definitely need to spend some time to digest all of it in detail.

Two questions/remarks based on the quick diagonal scan I just did:

#2-3 cylinder differ significantly in power output using the classic 3-1 race manifold, yet show almost equal power with the longer outer shorter center or LOSC*smiley* manifold. Haven't got my head around this one yet...are we able to rule out a bug in the sim with emperical cylinder pressure data?

The overall power graph sees the traditional exhausted engine plateau-ing at 6500rpm, with the other still climbing to 8000rpm onwards. However, I can't seem to recall a traditional built mini engine that give or take 1-2% produces steady power between 6000 and 8000rpm...which leads to the question of how realistic the base sim is?

Anyway not trying to criticise your work in any way. Makes a really nice addition to reading Race Engine Technology magazine @ €20,- per issue *smiley* Love the way you're applying a modern approach to further developing our ancient machines!

Edited by Evoderby on 4th May, 2016.


Evoderby

224 Posts
Member #: 9987
Senior Member

Amsterdam

Sorry, double post....

Edited by Evoderby on 4th May, 2016.


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

The 3-1 produces less power on the inner cylinders. We have measured numbers for that. As the main developement focus is a much wackier setup there is no proof for the short centre long outer idea but it is basically standard exhaust theory rigidly applied.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Evoderby

224 Posts
Member #: 9987
Senior Member

Amsterdam

Is your software capable of running stepped header configurations as seen in superbikes a.o.?

Could be a relatively simple experiment on a 3-1 manifold (long outers) with modified centre primary that features a step in diameter along the way to 'fool' #2-3 cilinder into thinking they're seeing a shorter length pipe.

Edited by Evoderby on 5th May, 2016.


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

yes you can do steps and pretty much everything you think of and a few more that you did not.

I think that a AR chamber might be the most promising option. Steps do not seem to do a whole lot.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans




On 13th Nov, 2014 adcyork said:
On 12th Nov, 2014 Paul S said:


The 5 port A series is unique due to the siamese ports. Understanding how the siamese ports affects performance is key to the decision making on modifications. I doubt that there are many engines still out there getting this level of attention that look anything like the A Series in simulation terms.


Precisely, and so correlation of a model from torque and power figures alone is highly inaccurate. Like you say, the dyno's themselves are plagued with inaccuracies through frequent calibration requirement and mechanical variances, so this is why cylinder pressure data from kistler sensors with a high degree of accuracy, accompanied by pressure/temp traces is essential.

Without fast response pressure/temp traces from real world data, it is going to be highly iterative to create any form of valid model of these Siamese port structures.

For example I see that some discussion has focused around modelling the siamese junction as a plenum with three ducts by Sir Yun.

Without any cyl P/duct temp/duct pressure data, how exactly will this modelling philosophy be validated as the best practice?







I just got some data back and I am happy to report that after using in cylinder pressure data to slightly tweak the burn model, the engine model will predict a new configuration quite splendidly within the odd percent.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Turbo Phil

User Avatar

4621 Posts
Member #: 20
My sister is so fit I won't show anyone her picture

Lake District

Some interesting research, well done.

Phil.

WWW.TURBO-MINI.COM


Sir Yun

User Avatar

510 Posts
Member #: 1592
Smart Guy!

mainland europe near ze germans

It is not 100% bullet proof as my new ( and very odd) exhaust design does not want to play ball yet.

That sir, is not rust, it is the progressive mass reduction system

http://aseriesmodifications.wordpress.com/


Barrieri

307 Posts
Member #: 11231
Senior Member

Hi guys I am currently running some simulations, anyone has some tables of lift vs crankshaft rotation for different camshafts ?

Thanks

Home > Technical Chat > A Series simulation thread.
Users viewing this thread: none. (+ 1 Guests) <- Prev   Next ->
To post messages you must be logged in!
Username: Password:
Page: