Airflow -port taper

comnoz said:

I have been told the reason for this is -the air goes to a high velocity as it enters the small opening. Then the walls open up so the air will expand to fill the space. This means it has to slow down but since the air has momentum it creates a depression behind it which increases the pressure differential across the inlet.

Click to expand...

aka shock loss

.......................

comnoz said:

If you look at the ram air ports on late model motorcycles you can see they take advantage of this -they have a very small opening at the front of the fairing which expands as it goes back to the airbox.

Car air cleaners often take advantage also by having a small inlet that tapers up in size. Jim

Click to expand...

It's an energy balance where you are swapping out the kinetic energy of the fast moving air with potential energy (pressure) of slower moving air and there's an efficient way of doing it and an inefficient way of doing it. See Bernoulli equations

That is the main thing I would attribute the higher and lower flow to. Accelerating air takes energy. So, going from a larger hole to a smaller hole requires the air to accelerate. On the other hand, going from smaller to larger allows the air to slow down which gives off energy and might “pull” the air behind it.
So, I had not really considered the turbulence aspect but it seems to me that there are two things that make it better for the intake flow to be slowing down as it reaches the valve and combustion chamber.
In a bit of a self defense on my porting job in the “Combat head rebuild” thread. I did not enlarge the port at all on the carby side. I tried to keep a steady taper going all the way to the valve seat taking into consideration the obstruction of the guide and valve stem. If I killed my torque I will find out and be sending my head to Jim next time around.

motorson, take some relief that you got to wait till after Jim learned the lesson he just described. What ya might want to try if the upper power band is your motivator, put a bigger carb on and see what that feels like. If it performs like my early Wovenhaven[sp] head then so so sorry. My only consolation is maybe it won't be such a dog on 20% bigger engine and boost if I can even make that work out useable. Up CR might help draw in quicker. How much would it take to get ya to send it to comnoz to get a flow comparison on yours?

How much air is needed? Roughly, say for an 850 at 7,000rpm?

For any tube total internal resistance increases as the tube lenghthens because of zero flow at the boundary layer. This can be measured with a Pitot tube. Boundary layer being at the inner surface of the tube. A 3 dimensional velocity profile would take on a tapered 'bullet' shape, pointed in the middle of the tube and widening out to the inner circumference. To get an idea of this think of a river - flow is greatest in the centre and slows as it nears the river bank.
Therefore it makes sense, and as Jim has demonstrated, to widen the port from the inlet to outlet to minimise the effects on flow rate from internal resistance.
Ideally you want minimum resistance within the port to achieve a greater volume flow rate, fuel/air mixing through turbulence is done in the cylinder/combustion chamber - not in the port as this will slow done volume flow rate.

During my final year project at Coventry I took measurements on experiments using little fairings fitted to the down stream side of the valve guide, to reduce turbulence and increase volume flow rate.

On anything I've got wrong please supply corrections.

Hi Jim you are a real devill ....! Like Galilée....! so now ,we the common Joe people what must we doe, with our heads, carbs, and inlet stubs , taper or not taper (or must we follow the Dixie Steve way : mismatch the entry !), say a 32mm head with 34mm carbies??

X-file said:

I think the problem here is that the flow test was done with a radius on BOTH ends of the tube.

Click to expand...

I noticed that too.
Jim, why did you do it like that?

comnoz, what you are measuring is not a phenomena within the tapered tube but rather a phenomena at the ends.

At constant CFM, the air is coming out 36% faster on the 30mm diameter side thus realizing a greater shock loss than on the 35mm side. The difference in pressure required to get the same cfm is the potential energy loss (static pressure) where the kinetic energy of the air flow was inefficiently converted to static pressure. If you mated up another taper 30mm to 30mm and ran the test again you would find a cfm value closer to the one associated with a 35mm discharge as it would be performing as an evase and more efficiently converting the kinetic (velocity) energy at the 30mm port to static (potential energy) pressure. That is what is happening.


Fluid flow has different needs to be efficient when designing intake and outlets. A good example is never seeing a 4 stroke exhaust system with the lip shape of a velocity stack. Furthermore most velocity stacks you see are less than optimal and are someone's "vision" of what they should look like and/or how they work. See Prof. Gordon Blair's Four-Stroke Engines

OK, I admit it. I wasted my youth on killing brains cells instead of exercising them. But I am trying to put this together with the tapered intake on RH10 heads. Recent port work and flow demonstrations aside, I thought I understood that a stock RH10 head flowed better in part because of the tapered intake. There is part of this that I am not putting together. Or I have misunderstood the 32mm to 30mm effect. If someone had asked me I would have said this creates an increase in velocity that helps the head breath better.

So help me out here.

Russ

Yep length friction matters and comnoz tube is similar to Norton intake lengths. Blowing gases through a funnel-ing down or a funnel up also depends on a sliding scale of in & out pressure differences on which flows faster and must be accounted for in rockets nozzles operating at sea level or in space, generally narrower longer on launch and wider longer in vacuum, but like us can only make nozzle as wide as the rocket stage below and at some length the extra mass out weighs the efficiency gained. On the other hand longer slug of air has more momentum to keep flowing some into a higher pressure zone, such as keeping valve open a while as piston ascending. Flow in tubes or channel also depends on temperature gradient of outer flow to the central flow, higher temps on walls with colder core flow better than visa versa. Vortex generators and wall riflings can energize the thickening boundary layer in bends by sticking up just enough in faster flow to draw some that energy into the sluggish thick layer to get it on around with the main current. Hydrocarbon molecules weigh more than the air molecules and there is an element of swirl in ports which slings the fuel outward so vortex texture can sling more of the fuel back in the flow too. A good many discoveries are made by pure accident or curiosity to see what happens doing things differently. I had No Expectation of anything but stifled flow with Peels 34mm carb, 32mm manifold into 28.5 head but to get past easy initial run in 1000 miles while waiting for lost head from MAP Cycle missing over a year and didn't realize just how good until I put back in full Combat spec, dual Amals and 32mm hog head. Depending on source and engine - the rule of thumb is 140-150 cfm mixture to make 100 hp of heat burn. I remember comnoz mentioning what little changes in seat contour made on more flow into chamber so little details matter. Andy what were your general findings of internal 'fairings' on flow rates?

Shock loss aka sonic choke, usually only applies to sonic speed and super sonic speed flows in restrictions, which set up pressure 'shock' front that prevents any increase of flow speed beyond sonic speed [at the gas temperature], no matter if increasing the push pressure dramatically, only increases the density passing through the sonic speed bump, which does increase the total mass making it into chamber though.

Going from 32mm to 30mm does increase velocity. The game plan is to do this as efficiently as possible (appropriate taper). What else happens is if the 30mm port is done up properly, the 30mm port allows a larger radius on the floor of the port where it turns in to the valve seat - a good thing. If you can get the velocity in the port up efficiently then it becomes very useful when the flow needs to dilate and slow down as it approaches the 38mm intake valve and somewhat smaller bowl. The trick here is to make that dilation/transition as efficient as possible.

Game plan is to efficiently convert Velocity pressure in the manifold and port into Static pressure at the valve when needed.

With an IC engine, everything is a compromise. One might ask, if a 30mm port is good velocity then 24 or 26mm port is better velocity - well yes for velocity but no for overall performance. With much smaller ports you are hitting a wall with regards to sonic choking (velocities get too high) where the port cannot move enough air.

comnoz has the working combinations for your needs, whatever they may be.

hobot said:

Shock loss aka sonic choke, usually only applies to sonic speed and super sonic speed flows in restrictions, which set up pressure 'shock' front that prevents any increase of flow speed beyond sonic speed [at the gas temperature], no matter if increasing the push pressure dramatically, only increases the density passing through the sonic speed bump, which does increase the total mass making it into chamber though.

Click to expand...

hobot, dear sir, you are full of it. You are not a rocket scientist and shock loss is not sonic choke.

Shock loss occurs all around us; in our cars, in our motorcycles, at our kitchen sink faucet ends and at the end of our garden hoses. Shock loss occurs when a fluid exits an orifice. This is what comnoz is showing with his tapered tube. It has zip to do with the tube in Jim's (comnoz) experiment and everything to do with the relative velocity differences exiting the tube. Higher velocity will result in greater shock loss.

Mark said:

X-file said:

I think the problem here is that the flow test was done with a radius on BOTH ends of the tube.

Click to expand...

I noticed that too.
Jim, why did you do it like that?

Click to expand...

A cylinder head has something of a radius at both ends. One at the carb mouth and the other at the valve entry into the cylinder with a good mulit-angle valve job. Of course the valve end radius is smaller.

The tube was just made to show the effect of the wall taper.

I have a cylinder head that has been bored large and strait to the valve guide to hold a port insert. I have made a lot of different inserts with different tapers and shapes and tested them both on the flowbench and the dyno. This has been where I have done the most research.

Here is the head and a couple different inserts that I have tried. The head has been on my 880 for years up to the big blow-up. Jim

More on Bellmouths.

http://tinyurl.com/nzdrhrn

Great thread, Jim. So far 35 posts in less than 24 hours. Clearly lots of interest in tech threads here.

Ken

lcrken said:

Great thread, Jim. So far 35 posts in less than 24 hours. Clearly lots of interest in tech threads here.

Ken

Click to expand...

No doubt about that. And when it comes to porting there are 1000 different ideas -some good -some bad and I am sure there are some still to be learned. Jim

Thanks for the polite correction dancer between ventilation speed 'shock loss' and higher 'sonic choke'. I'm told everything I eat turns to shit. May your cycle be able to run right out from under, just don't let the road hit your own shiter.

jaydee75 said:

If I remember from college (40 yrs ago), a diverging tube with a seven degree included angle is the most efficient for flow thru a tube. That's why exhaust megaphones are made the way they are. I know Jim is on the right track, I'm just trying to remember the optimum angle.
Jaydee

Click to expand...

I was wondering about this comment. If you think about what a megaphone does in practice, I suggest it is more about amplifying the pressure in the standing wave in the exhaust pipe than increasing flow. I find thinking about the standing wave and mass transfer at the same time confusing - it not as simple as just flowing gas down a pipe. There is a reversal from the open end of the pipe and in the pipe there are low and high pressure areas. The length of the pipe determines where the low pressure point occurs. I don't know whether the low pressure points of both the inlet and exhaust or the high pressure points should occur at the combustion chamber simultaneously, however I believe that is what tuning is about. In the old days we used to use an inlet length in inches of about 28,000 divided by the revs - end of bellmouth to centre of combustion chamber, however I don't know why. I also knew that dropping the gas speed is not good for the torque characteristics at lower revs. When I fitted the 34mm Mk2 Amals to the 850 head (30mm ports) I tapered the ports over the first 25mm simply because I knew that the Aermacchi 350 had tapered ports - nothing scientific about it. The two into one exhaust pipe was also made after thinking about what worked on the z900 kawasakis, as was the trick of advancing the cam to compensate . The whole thing seems to work very well - just dumb luck ?
With bikes I always try to rationalize what has gone before and worked. This probably goes against conventional scientific thinking, and I am a scientist.

So you got my curiosity up about the raduis on the outlet of the flow tube.

Here is the flowtube with the radius on the 30 mm end filled.



Here is the flow with the air entering the 35mm end. 115.4 CFM



Here is the flowtube with the radius on the 35mm end filled.



Here is the flow with the air entering the 30 mm end. 139 CFM

hobot said:

Andy what were your general findings of internal 'fairings' on flow rates?

Click to expand...

Hobot,

Do you mean Al? I can't remember off the top of my head - I'd have to dig out the report. I'm sure they improved flow rate but I would say it was small. The object was to investigate if better air/fuel mixing could be achieved by the fitting of valve guide fairings to increase velocity and volume flow rate, indicated by introducing powdered paint into the inlet to leave a pattern on the static piston crown of a four valve test engine. I'll try and find the report.