De-shrouding YZF valves

[Brad_Aus]

Hi guys,

First time post for me, been reading these forums for some time now. Im from Australia and have some questions for you guys..

Ive been playing with a crf250 08' and a yzf450 08' cylinder head on my flow bench.

What ive found is that the honda showed good gains in flow over the entire lift range by removing the small machining lip left around the valve seats on the blend into the chamber.. However, the same mod performed on the yz head has shown a loss of low lift flow below 5mm on intake and 3.5mm on exhausts.. I thought it would be a good idea to remove this lip on all chambers, however, the yz has proven otherwise..

Anyone else experienced this??

Brad

[Brad_Aus]

Last night I did some more fiddling to try re-gain my low lift flow on this head, I put the valves in the lathe and used a sanding wheel to radius profile the back-cut section, just smoothed over the two sharp angles where they had machined the two faces.. Did NOTHING for my flow.. there you go, dont worry about doing that..

My machinist has a Newen contour BB cnc seat cutter, was hoping to get a nice radius seat profile cut into it, but his newen packed up last week, so im gonna be stuck with a 5 angle from his Serdi.. Im hoping it can make some gains.. ill keep this updated

[grayracer513]

If you want to discuss this with someone who really knows something about it, contact Ron Hamp at Ron Hamp Cycle. He's the only person in captivity to really solve the riddle of the YZ450 head, and has a couple of "fer real" 60+ hp 449cc engines running around to prove it. One of the keys is in the fact that the real world air flow involves not just the head, but the complete combustion chamber, and that the 3 intake ports are of 2 different kinds, are dynamically very different, and have to be treated as such.

[yz454]

The yz likes working the bowl around the guides. Put a 35 deg back cut on the valves,either grind or use a lathe, but you blended the seats so it might be to late.

[KJ790]

If you want to discuss this with someone who really knows something about it, contact Ron Hamp at Ron Hamp Cycle. He's the only person in captivity to really solve the riddle of the YZ450 head, and has a couple of "fer real" 60+ hp 449cc engines running around to prove it. One of the keys is in the fact that the real world air flow involves not just the head, but the complete combustion chamber, and that the 3 intake ports are of 2 different kinds, are dynamically very different, and have to be treated as such.

Agreed, though I don't think Jesse Williams would be a bad person to ask either. Ron tends to be a little hard to get information out of (and for good reason on his end).

[grayracer513]

Ron tends to be a little hard to get information out of (and for good reason on his end).

He usually learns it all by himself, the hard way. 'Can't blame for for not giving everything away.

[Brad_Aus]

The yz likes working the bowl around the guides. Put a 35 deg back cut on the valves,either grind or use a lathe, but you blended the seats so it might be to late.

Have you seen gains in low lift region from this though?? To me, this made a nice improvement at high lift..

So you know, I have made some nice gains on this head, except for the area 4mm and below on the intake, and 3mm and below on the exhaust. Ive lost a few % in this area, but gained 7% from those points onwards..

The velocity is between 360-390fpm on the floors, depending where you measure it, but like you said, around the guides may be a key to making it work, as the airspeed around the centre guide is down to 180-200 and 250-270 on the walls of the divider...

Id be interested to hear if Ron knife edges the dividers or not, I initially did, but after a lot of research it appears that a radius divider is better.. so i raduised them.. However, there was no real change in cfm from sharp or round dividers... In fact, i put a 3.5mm wide blob of blu-tac on the divider faces and only lost 3cfm

[TDW]

He usually learns it all by himself, the hard way. 'Can't blame for for not giving everything away.

He does share some good info over at the honda 250f and 450f forums that I've read, we're lucky he chooses TT to post and make replys to some posts.?

[harrperf]

Have you seen gains in low lift region from this though?? To me, this made a nice improvement at high lift..

So you know, I have made some nice gains on this head, except for the area 4mm and below on the intake, and 3mm and below on the exhaust. Ive lost a few % in this area, but gained 7% from those points onwards..

It would be interested to see how this motor runs....as you have certainly lost quite a bit of very low lift flow, but also gained substantial higher lift flow....

I would venture to say as it is the bike would run well!

The velocity is between 360-390fpm on the floors, depending where you measure it, but like you said, around the guides may be a key to making it work, as the airspeed around the centre guide is down to 180-200 and 250-270 on the walls of the divider...

you must have a nice set up!

Id be interested to hear if Ron knife edges the dividers or not, I initially did, but after a lot of research it appears that a radius divider is better.. so i raduised them.. However, there was no real change in cfm from sharp or round dividers... In fact, i put a 3.5mm wide blob of blu-tac on the divider faces and only lost 3cfm

While I am sure you know this....for those reading on, air flow is an interesting thing! As he has said, it's really not what the air runs into at first that makes the difference (aka the divider can be wide and nasty, or knife edge....hardly any flow difference) but more importantly how the air comes back together that you must focus on.

For instance, and we all see this in nature, ideal aero shape is like a rain drop. SO, you can have a huge front face, but so long as your shape tapers smoothly and not too quickly back to a thin shape, you practically have no drag/pressure drop.

A car shape that looks futuristic and is pointy like a tipped sword at the front and then tapers back is way worse than a car with a brick front shape that tapers to a point in the back...as this way the air doesn't create the "eddy's" that equate to lost flow!

On the exhaust, a knife edge divider should be much more beneficial.

shape B (the tiny round thing) has the same drag as shape A! How crazy right (with air flowing from bottom to top of screen)

[Brad_Aus]

Im hoping it runs real well, I should get it back from the machine shop today, and flow it after work, ill post some new results then..

I built my bench using stuff from flow performance, has a beautiful Pitot tube for measuring velocities, it fits right down in the throats of a yz.. My bench flows 250cfm at 28" currently, but ill be adding some more vac power when I feel the need..

Heres the graphs

let me know your thoughts, Brad

[harrperf]

Looks great actually, I would fully expect that motor to run extremely well.

For the yamaha guys, be happy your stock ports flow as well as they do! The yami is capable of producing very good hp with a stock port....

What did your honda 250 flow (and what year is it) I have stock specs for a 2007 sitting next to me, maybe we can verify each others..

[grayracer513]

What did your honda 250 flow (and what year is it) I have stock specs for a 2007 sitting next to me, maybe we can verify each others..

That's a subject for another thread, in another forum. ?

[Ron Hamp]

I have been sharing information for years on bench testing .throat id is critical to low lift flow as its controled by the valve shape and seat shape blending the seat id out to the port can reduce low lift numbers some times its benficial tomake a radius shaped seat on the long side of the port while reducing the radius on the sides and short side also the yzf benifits from contouring the seat a bit steeper on the center valve reducing the cone of air that is produced around the valve and having less interuption with the outer valves you will find when blocking ports on theses heads that the center port flows more air.The largest gains in flow come from contouring the short side radius on these heads and almost always results in a beter power gain on the dyno .were only flowing @ 28 to 50 " h2o and an engine may see up to 140" we can not predict what the air flow is doing over the short side radius @ the very high velocitsys when you produce flow gains in the port based on the short side radius what you do is extend the rpm before the air starts shearing more like a down draft head and you will usally see your hp contiues to climb on the dyno .I have played with the dividers quite abit and have actually made them ver y blunt with sharp edges what i have found is the beter the countour of the divider the more airflow you gain but what also happens is the fuel tends to adhere to the divider and it forms a wet track that crosses over to the exhaust valves when making it blunt the tracks are reduced or eliminated theresa small loss of flow but the engine almost always requires less jet and the throttle response is better .i have no means of testing bsfc so i cant prove to weather the engine is actually becoming more efficient but thats the therory. i just finished a new yzf head for cnc porting the peak gains in flow were 30 cfm @ 28" h2o the race head that produced over 62 hp had a gain of 55 cfm @ 28" h20 the regular head had produce a gain of hp with the right cams of around 11 hp and the higher flow ing dirt track head was around 15 hp and 6 pounds of torque. There no one can say i dont share .

[grayracer513]

Outstanding, Ron. Thanks.

BTW, good to see you over on this side. Don't be a stranger. ?

[Ron Hamp]

I will check in ocasionally

[Brad_Aus]

What happened to the honda graph? Was I not allowed to post that even for comparative purposes?

Anyway, thanks for the info Ron.. In my short time playing with this head I have discovered a lot of the things you listed. Seat insert I.D being a make or break thing with these heads. Tip for anyone modding their own head, dont enlarge the i.d of the seat, i thought it would be a good idea to radius the pointy section of the bottom cut, which also increased the i.d slightly... Instantly lost 20cfm at low lift.

Im going to make some seat inserts from aluminum for testing purposes and try a range of shapes and sizes, then have the winning profile cut onto my good seats.

Ron, out of interest, what was the increase in port volume to get the 30 and 55cfm gains? I had a 250 and 450 version of this head side by side yesterday, and i believe the 250 has a nicer port because the valves are closer together which makes for a straighter port.

Theres atleast one wall which I was able to fill in with plasticine and achieved a slight gain in flow, I think these ports are massive at the area before the dividers, so all the work has to be concentrated at the seat (low lift) and short side and around the guides for high flow

[Old_604]

I feel like the dumb kid sitting on the back of the class just hoping and praying the teacher doesn't pick me to answer a question.....

Cause I have no idea what you guys are talking about! ;-)

OK...back to eating my buggers....<lol>

[grayracer513]

.... I have no idea what you guys are talking about! ;-)

Port work. Production cylinder heads typically have casting marks, machined steps, misalignment at joints with other parts. A lot of people think they can gain power by smoothing all that out, and sometimes they can. But it doesn't always work that way.

Here, the OP was talking of grinding at the edge of the valve seat where it meets the inside of the intake ports to make them match better. The problems he had illustrate what I have said repeatedly; it's just as likely that you will lose power as gain it by attacking your ports with a grinder unless you have some idea where you're going with it, and some accurate info on where, what, and how much to grind from someone who has tested several layouts already.

[Brad_Aus]

Im hoping to have some new flow data to update this with today.. I popped out the old seats already, and have the first prototype of my venturi seat design.. Not very much scope for changing the profile much, but, i think it will make a dramatic change to the seat flow..

Ive managed to get the seat i.d down to just over 80% of the face i.d.. The general rule is to have the seat i.d between 80-85% of the i.d of the mating face, you can go down to 75% and get even better low lift flow but it tends to hurt the high lift flow some.

As a starting point, the i.d of the standard seat was 90% of the face i.d, which I beleive was equal to the i.d of the port.. So had no venturi effect.

[Brad_Aus]

After 4 attempts, ive got something I think is pretty good, but I wont be completely happy til i'm able to outflow the std seat the entire way through..

My first attempt had the smallest i.d and the sharpest angles, it had killer flow at low lift, but choked up at about 5mm, and was pretty bad up top. see attached pic

This is my current spec, im not sure what to think of the flow at 1-2mm, whether this is important so much?? or is it far outweighed by the gains in the top end?? Id be very interested Ron, to know if your good yz heads are up in low lift flow compared to standard, of have you also traded off some flow in the very low lift range in search of peak CFM??

note, it flows better up top by a fair bit over the original seats