Flow Numbers for 450 head

[JLB943]

Just wondering has anyone looked at the volume flow rate (cfm) for 450 cylinder heads at different valve lifts. I was wondering what the values roughly are at different valve lifts and what kind of improvement is seen through porting. I understand that flow numbers are not everything and flow velocity and cross sectional area come into play but just curious. In drag racing heads everyone always posts their flow numbers but I have not seen similar in mx heads.

[moto867]

i just flow tested an 09 crf450 head. I have my notes in the shop, but at .400" intake lift it flowed about 142 cfm. I didnt correct for leakage yet. I'll look at my notes tomorrow and calculate the correction factors and post my results at valve lifts in increments of .100". by the way this head is not stock its my test head i use to develope port profiles so its already been ported.

[moto867]

09 crf450

.100" lift 79.73 cfm

.200" 122.468 cfm

.300" 143.764 cfm

.400" 146.608 cfm

These are corrected #'s. Tests done on a superflow sf110 at 10" wc. I'll have to double check my # at .300 lift, i may not have readjusted my test pressure to 10". The head also has no valve seals so i may have had a little leakage but not enough to make a diff.

[barcode]

What size bore did you flow with? Stock crf?

[moto867]

yes, 96mm.

[moto867]

one thing to remember about flow benches they are like dynos in that #s from one might not be like another. I use a sf110 at 10" wc. if someone flows the same head with a sf300 or 600 at 28" wc the #s will be way different. it does not matter what the cfm is from one bench to another, what matters is the % of gain from stock to the finished ;product. As mentioned before velocity is very important, if it wasnt all anyone would have to do is hog out every port as big as possible

[JLB943]

The numbers look different than expected but probably because of the 10" instead of 28".

Did you have a baseline from the stock head? How much larger did you make the flow area and were you able to keep the velocity near the same at the increase area?

[moto867]

at 28" the max lift would be over 200 cfm. Like i said, for comparison sake it doesnt matter. I dont have the stock #s but i have another 09 head coming in soon so i will post them when i get it. i ended up with only slightly higher cfm but the velocity actually increased. I did some filling on the port floor with epoxy and reshaping. i gained quite a few cfm at lower lift with a five angle seat mod. I gained some at high lift also but not as much as at low lift. When i do the other head i will post before and after #s.

[grayracer513]

If velocity were everything, you'd make the intake smaller, not larger. But there is a lot more to the picture. Maximizing bulk throughput is the ultimate goal, even when that can occasionally mean reducing peak flow at max lift in order to improve it at a lower lift.

The airflow within the ports is often mistakenly seen as a a steady flow, but of course it isn't. It's a series of pulses that occur in such a rapid succession that they are almost effectively a steady stream, but they aren't, and all of their behaviors can't be explained or analyzed by considering them only in that light. The intake stream is driven only by the pressure differential between the combustion camber and the atmosphere, and to a lesser extent by its own inertia. Also, flow at or near max lift is only a factor for 10% of the lift cycle, depending on how wide a window you want to consider. The tendency is to look at a port and visualize a stream of air flowing through it, but the fact is that the air has to flow around the head and stem of a valve and over the edges of a valve seat while the valve is not more than about 10mm off of the seat. The shape of the port has to contribute to that.

In 5 valve heads, the configuration of the port as it branches into three is extraordinarily critical to what the air/fuel charge does as it enters the cylinder.

Given this, there's a lot more to think about in creating an effective high performance port, in or out, and the right approach is not always as obvious as many people think. Even the shape of the piston crown comes into the matter eventually.

[JLB943]

The pulses is something I never thought about but it makes sense the pressure differential only occurs during the intake stroke for intake flow etc. With naturally aspirated automotive engines the generally play with the intake,valve angle and port location so the air has almost a straight path into the chamber, no sharp turns to create losses. Cam timing can be played with to essentially superchage an engine to increase VE.

I guess depending on the valve events the area under the curve for opening of the valve would define more than at max lift which is only achieved a short amount of time. I figured the four stroke engines in MX bike have to be pretty efficient to put out the amount of hp they do for a single cylinder and was curious about the actual flow. 60 hp out of .5 liter figure 8 cylinder 480hp out of 4 liter.

Pulled this from Wikipedia, sums it up pretty good.

There are several standard ways to improve volumetric efficiency. A common approach for manufacturers is to use larger valves or multiple valves. Larger valves increase flow but weigh more. Multi-valve engines combine two or more smaller valves with areas greater than a single, large valve while having less weight, but with added complexity. Carefully streamlining the ports increases flow capability. This is referred to as porting and is done with the aid of an air flow bench for testing. Another major aspect of design is to use a crossflow cylinder head, which has become the standard configuration in modern engines.

Many high performance cars use carefully arranged air intakes and tuned exhaust systems to push air into and out of the cylinders, making use of the resonance of the system. Two-stroke engines take this concept even further with expansion chambers that return the escaping air-fuel mixture back to the cylinder. A more modern technique, variable valve timing, attempts to address changes in volumetric efficiency with changes in speed of the engine: at higher speeds the engine needs the valves open for a greater percentage of the cycle time to move the charge in and out of the engine.

Volumetric efficiencies above 100% can be reached by using forced induction such as supercharging or turbocharging. With proper tuning, volumetric efficiencies above 100% can also be reached by naturally aspirated engines. The limit for naturally aspirated engines is about 137%;^[1] these engines are typically of a DOHC layout with four valves per cylinder.

[moto867]

there are several different bikes out there where i do end up with a smaller intake port. Low valve lift flow is largly dependent on seat shape and the area just above and below the seat. After about .250" valve lift the port shape and volume starts to have an influence on the overall flow. Air can be thought of as acting like water when it flows through a port. it is hard to get it to make sharp turns. Thats why it doesnt like to follow the short side radius and alot of the time it ends up going straight past the valve and out the long side radius. What i like to do on some intakes, not all, is build up the short side radius with epoxy and reshape the port to make the radius bigger thus easier for the air fuel to follow. The fact that the valve opens and closes is why velocity is important. Air is springy and compressable. With a high velocity port , when the valve closes the air rushing in compresses and creates a high pressure area behind the valve. Just in time for the valve to start to open again at the critical overla;p poeriod. Which, by the way, is when the velocity of the incoming air fuel charge is highest. this is due to the scavanging effect of the exhaust, providing its tuned correctly. There is a formula to figure out the basic needs of an engines flow in cfm to produce x# of hp at x# of rpm. naturally this is dependant on the whole setup, but if you make the ports too big for the hp that your setup allows then you loose velocity and power. The only ways to get it back would be to increase engine size or rpm or both.

[kx450f63]

Fellows... talk to me about "claying out a head" I am just curious as I have never instituted the use of a flow bench. I have heard guys who are known for making good horse power use the phrase and assume they are filling areas in the intake tract with clay and testing for gains, am I in the ball park?

[grayracer513]

Yes, you are. The top tuners who do this kind of work for a living will sometimes "ruin" 2 or 3 or more heads in experimenting with different variations. Some guys contact people at service departments and machine shops for heads that have damage beyond practical repairability to use for the purpose. They can end up being hacked up, hogged out, welded, clayed, epoxied, or almost anything else in the search for more power.

[moto867]

thats for sure. Thats why i save all the junk heads i replace! I test with those to get the final port shape that way i dont scrap any good heads

[mystery]

This is the stock intake on 08 crf tested at 28'' I have tested a few and they are all very close in stock form.

0.5 43

0.1 86

0.15 135

0.2 178

0.25 202

0.3 209

0.35 213

0.4 214

exh (no pipe)

0.5 23

0.1 65

0.15 102

0.2 127

0.25 141

0.3 148

0.35 150

0.4 152

0.425 152

I have my own flow bench but I test at very high pressures 40''+ then back calculate to the standard 28''.

here is some information on going back and forth on test pressure calc's. or the formula =SQRT(28/(test pressure))

For reference only:

As for velocity I like to keep it around 320fps MAX at the mcsa. The short turn around 400fps.tested at .87*max cam lift (this is a good estimate of max piston demand 70-75 deg) I use J and straight pitot's @28'' to check velocity.

you may want to look up topics like port choke and discharge coefficient

if you would like I can supply some of this information

I hope this helps

I attached a little CFD sims I did for some of the topics others have brought up. This is just for visual ?

BS engineering

Edited January 31, 2013 by mystery

[mystery]

Made some port molds today. This is a stock crf450r

I have a head that flows 365CFM and 315 fps @MCSA I may post that mold in the future.

Enjoy

[Dug the Thug]

This is the stock intake on 08 crf tested at 28'' I have tested a few and they are all very close in stock form.

0.5 43

0.1 86

0.15 135

0.2 178

0.25 202

0.3 209

0.35 213

0.4 214

exh (no pipe)

0.5 23

0.1 65

0.15 102

0.2 127

0.25 141

0.3 148

0.35 150

0.4 152

0.425 152

I have my own flow bench but I test at very high pressures 40''+ then back calculate to the standard 28''.

here is some information on going back and forth on test pressure calc's. or the formula =SQRT(28/(test pressure))

For reference only:

As for velocity I like to keep it around 320fps MAX at the mcsa. The short turn around 400fps.tested at .87*max cam lift (this is a good estimate of max piston demand 70-75 deg) I use J and straight pitot's @28'' to check velocity.

you may want to look up topics like port choke and discharge coefficient

if you would like I can supply some of this information

I hope this helps

I attached a little CFD sims I did for some of the topics others have brought up. This is just for visual ?

BS engineering

Wouldn't you want the velocity to be just under or at sonic choking?

[HeadTrauma]

320ft/s is just a number to shoot for on the flow bench. Peak port velocity goes much higher on an engine because flow benches don't replicate operational conditions.

I don't know what the port cross-sectional area is for a CRF450, so I assumed it was equal to the intake valve area at 2.355in². With a 3.78" bore and 2.445" stroke, average port velocity is already approaching 400ft/s while average piston speed is only ~82ft/s at 12,000rpm. Peak piston speed is substantially higher at 134ft/s which means that peak port velocity is too.

Edited March 2, 2013 by HeadTrauma

[barcode]

Mystery

Is that dip before the STR help flow around the turn.

How deep is the dip?

Edited March 2, 2013 by barcode

[numbhands]

Mystery

Is that dip before the STR help flow around the turn.

How deep is the dip?

I think thats the valve guide boss isnt it? helps to guide the air flow round the valve guides.