Wet flow Vs dry flow results 3

[NeilRoshier]

I have been reading and asking questions of people again and so far the question has remained un answered (possibly poor questions!?), hence this post. The current range of flow benches are quite advanced (and expensive) but are you able to accurately predict the effect on the wet flow (ie the flow of the intake charge) of something like flow shear at the short turn radius? given that the flow benches I have seen are all dry flow only.
I was wondering if the wet flow would react differently given the added mass of the fuel in suspension. It would seem logical that the fuel in suspension may not follow the same path as the air given the difference in mass and would this difference be more prevalent given larger droplet size in suspension?
My thinking then made me wonder if the valve shrouding that occurs due to the flow shear would then have a detrimental effect on the fuel distribution in the cylinder and then on the burn-resulting power.

 

[SMOKEY44211]

Neil. If you're trying to optimize the A/F ratio for all cylinders on a single carburated muti cylinder engine good luck. Having spent way too many hours trying to accomplish this task myself I can say that its such a dynamic target that if you hit it consider yourself fortunate. Example: reworked the carb, spacer, and manifold configuration to get fairly equal E.G. temps. Switch to a different vaporization temp. fuel and the #'s are all over the boards. Changes in ambient inlet temps. have the same results. Direct port injection eliminates the need to go through all this monkey motion. If you're stuck by a rules restricted racing class to running carb(s), try to duplicate ambient temps. engine running temps. and blend of fuel(vaporization temps. vary with the seasons on pump gasoline) that you expect to be operating with and you'll come pretty close. -------Phil

 

[NeilRoshier]

Phil, thanks for the reply, it is interesting. Unfortunately I am not tackling the problem you describe, though I have spoken to a couple of people who have and I hope that I never have to!
I will be investigating new port shapes in an 8v 4cyl engine soon and will be mapping the valve shrouding due to flow shear. I will also be mapping the effect of post valve modifications on said shrouding. However I am wondering what effect on the wet flow the changes will have, given the increased mass and if that can be predicted by the dry flow results.

 

[SMOKEY44211]

I've given some more thought to the scenerio you described. I don't think you can bench test for any possible changes in mixture strength. Using direct port or better yet direct chamber injection what affects 1 cyl should have the same affect on all of them. All of the state of the art electronic systems I'm aware of use feedback from the 02 sensor(s) to keep the fuel mixture in the preprogramed ratio so even if you make a change that might cause it to go rich or lean the system is going to respond to compensate.--------Phil

 

[franzh]

Having also spent way too many hours at a Superflow bench trying to isolate cylinder to cylinder mixture homogenization issues, I must comment that there cannot be two identical cylinders, primarily due to intake manifold and intake port designs.

This issue is mitigated by using port fuel injection, where the last couple of inches is relatively all that matters, but many vehicles still in use, especially in performance venues, use central located carburetion systems.

When designing an intake manifold, a target RPM is selected and the manifold is optimized for that range. Wet flow systems do indeed behave differently at different airflow rates. Rich systems tend to centrifuge the fuel at short radius turns. Different tricks have been used to keep the fuel in suspension, heat, surface texture, ultrasonics, etc. Intake resonance is critical too, along with runner length, diameter, taper, radius, material, etc.

This is the leading reason for the shift from carburetors to central fuel injection, then to port fuel injection, then to sequential port fuel injection, to minimize the amount of fuel in transit. This has the effect of making all cylinders behave as if they were the same cylinder since the relative plenum length of air is not as critical since fuel is not present.
Franz


eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[AboveRedline]

I believe the question Neil is asking is independent of whether the engine is carbureted or port injected. If anything it might be worse when injected, as the fuel mixture is not as homogeneous as when carbureted.
We've sprayed aerosol Dykem into a flowing port, it will give you a visible trace of where the fuel is going. Not a very exact science obviously, but the information is better than nothing. Also watch the flammables around a flow bench as the motors have brushes with lots of spark activity.
Most people evaluate the wet flow performance by reading the chambers after the dyno pulls. You want to see even coloration of the cylinder. A clean streak means raw fuel is washing the cylinder in that area.
In general a tall smooth transition short side turn will give the best wet flow results.

 

[willeng]

Neil

This may be of interest to you.
An interesting point is made about what was once considered to be fuel wash & clean spots on the piston crowns.

http://www.rehermorrison.com/techTalk/40.htm

 

[NeilRoshier]

thank you everyone for replying. Aboveredline is correct that I am not concerned about the carb/inj element as I am testing valve shrouding due to port flow shear at the short turn radius. As it happens the engine will most likely be injected and have individual port throttles.
However I am interested in investigating the port effeciency issue and reporting on my results (instead of just measuring CFM at peak lift) for a small magazine as a brief investigation in to basic theory. I was concerned that the changes in results in dry flow would not reflect the changes for wet flow and naturally this needs to be noted in the article.
The plan was to investigate 4 different port shapes in theory and test, then valve shapes in each port and then post valve modifications.
Willeng, thanks very much for the article, it was very interesting. I will follow up the author and ask a few questions.