Head flow testing.

comnoz said:

The Axtell head is a nice piece of work but I wouldn't say it is something new.

Click to expand...

I would not say that it is something new either, and I didn't if you go back and carefully read what I actually did say, I said "so maybe he did something that still no one else has done yet", get as good flow and velocity out of a stock Norton head.

I have molds of the intake and exhaust ports of the 30mm Axtell head which Ken Canaga has on lend to you, and there is nothing "D" shaped about any part of it. An Oval "O" is not a "D", at least it was not when I learned the alphabet many decades ago, it does not have any flat sides.

At the point where the valve guide comes through the roof of the port with a caliper you might find less than a sixteenth of an inch oval, most of the port is very round though, simply getting a bit fatter where it goes around the valve guide and stem. the port is wider where it goes around the valve guide and stem, then between the guide and manifold face it is smaller by .100", 1.2" before the guide, 1.3" from the guide to the valve seat.
Axtell did a great job of keeping the port cross section and probably the volume the same from the manifold to the valve seat. The ports absolutely are nothing even close to the D shaped exhaust or the "cobra" shaped intake port of the Fullauto head. Any single cross section of the Axtell ports would roll like a hoop.

If anyone gets a chance to read the section in Smokey Yunik's Speed Secrets book about port shapes, and looks at what Axtell did it is almost as if they were reading each other's minds.

beng said:

comnoz said:

The Axtell head is a nice piece of work but I wouldn't say it is something new.

Click to expand...

I would not say that it is something new either, and I didn't if you go back and carefully read what I actually did say, I said "so maybe he did something that still no one else has done yet", get as good flow and velocity out of a stock Norton head.

I have molds of the intake and exhaust ports of the 30mm Axtell head which Ken Canaga has on lend to you, and there is nothing "D" shaped about any part of it. An Oval "O" is not a "D", at least it was not when I learned the alphabet many decades ago, it does not have any flat sides.

At the point where the valve guide comes through the roof of the port with a caliper you might find less than a sixteenth of an inch oval, most of the port is very round though, simply getting a bit fatter where it goes around the valve guide and stem. the port is wider where it goes around the valve guide and stem, then between the guide and manifold face it is smaller by .100", 1.2" before the guide, 1.3" from the guide to the valve seat.
Axtell did a great job of keeping the port cross section and probably the volume the same from the manifold to the valve seat. The ports absolutely are nothing even close to the D shaped exhaust or the "cobra" shaped intake port of the Fullauto head. Any single cross section of the Axtell ports would roll like a hoop.

If anyone gets a chance to read the section in Smokey Yunik's Speed Secrets book about port shapes, and looks at what Axtell did it is almost as if they were reading each other's minds.

Click to expand...

You are right-it is not a D shape but it is flattened somewhat on the bottom. I did not make a casting but I did use a caliper and made some drawings. It's about as much as you could do on a stock casting without filling and welding or messing up the entry into the bowl.
Of course the Fullauto head's intake is not flat on the bottom either. It is a lot larger radius than the top of the port but it is curved.
Smokey, Axtell and Mondello are all sharp guys who know how to make an engine work. Jim

beng said:

At the point where the valve guide comes through the roof of the port with a caliper you might find less than a sixteenth of an inch oval, most of the port is very round though, simply getting a bit fatter where it goes around the valve guide and stem. the port is wider where it goes around the valve guide and stem, then between the guide and manifold face it is smaller by .100", 1.2" before the guide, 1.3" from the guide to the valve seat.

Click to expand...

What are the valve diameters on this head?

jseng1 said:

beng said:

At the point where the valve guide comes through the roof of the port with a caliper you might find less than a sixteenth of an inch oval, most of the port is very round though, simply getting a bit fatter where it goes around the valve guide and stem. the port is wider where it goes around the valve guide and stem, then between the guide and manifold face it is smaller by .100", 1.2" before the guide, 1.3" from the guide to the valve seat.

Click to expand...

What are the valve diameters on this head?

Click to expand...

Stock. Jim

How big an exhaust vale can be fitted in a CHO head,? - especially if the intake is kept factory size? Is there any issue with bigger valves, particularly the exht, such as weakening head structure somehow. I know pistons might need a bit of relieving or even lips of valves. Whose &/or, which, size biggest exhaust valve would the knowing Nortoneer's seek out? Maney is hard to top in this as far as I know but don't know, if as thin a stem as Deere sourced-ordered products from Kimble/White-Black Diamond valves already installed and tested to tach needle bounce off zero peg disappearing needle rpms for a few seconds of unloaded WOT.
What would ya do if the exhaust side releif was the hindrance not intake flow?

You can easily go 2 or 3mm oversize with the exhaust valve. It will improve the low lift flow.

At higher lift the port will not flow anymore without adding material to the floor. Jim

I moved this from another thread as it seems to fit in here. Maybe it has already been answered!

rvich said:

pierodn said:

About Combat,
Can I mount a Combat head (with his 32 Amals and 32 manifold) on a standard engine?.
What should I be careful?
Just insert a gasket higher (with standard pushrods) or other?.
Ciao.
Piero

Click to expand...

Piero,
The Combats used standard pushrods. It has been recommended for Combat owners to shorten those pushrods to make up for the shaved head. What you are really asking about is how well the head will flow and whether it is a good combination with a standard cam. There is an active thread on head flowing. I am going to copy this message to that thread as I think it fits with that discussion.

Russ

Click to expand...

You can easily go 2 or 3mm oversize with the exhaust valve. It will improve the low lift flow.

At higher lift the port will not flow anymore without adding material to the floor. Jim

Click to expand...

Ok thanx, everything I now know says best to fit as big of valves as can no matter what else about the head, excepting heat transfer issues gets worse on bigger valves. Don't know if our size variations makes much difference or not. I will start thinking of saving for both valve larger someday in my special.

Head interchange, Definitely check AltanticGreen Dave Comeau's site.
I only know that I had Great Success with an early BIRO small port low compression head on a Combat so should work the other way around too.

More to fill in us newbies

http://www.onedirt.com/tech-stories/eng ... w-dynamics

More collected contrary wisdom's >>>>>>>>>>>>>

On 5/27/2013 3:01 PM, bob_foist wrote to mc-eingine lsit:
Hi John...just read a book over winter by David Vizard and that was part of his int tract advice.He advised a straight,parallel section of port 1.5 to 2 valve diameters long measured before valve seat with cross sectional area of 77%-80% of int valve.But.....one of the best,most successful MC engine builders I know,advises filling the outer radius of port between int seat and valve guide in and on top of that,quit using a flowbench years ago,believes they're more trouble than they are worth.So pick your poison....and then use a dyno to sort the flyspecks from the pepper

On 5/28/2013 4:36 AM, Ken Augustine on mc-engine wrote:
> hobot,
>
> Take a small port head, do the Hemmings mod on the intake and that should eat all the clutches you want. I have done that modification and know what it entails but don't remember the exact angle or diameter involved. The part that I don't like about the Hemmings mod is that the inlet valve guide bore needs to be increased to .625" diameter to clean up on the original .500" guide bore when the inlet guide angle is straightened up. In more than just my opinion, the Norton head casting doesn't have enough material to properly support the larger diameter guide and the guide should be made smaller to give a higher caliber fitted ratio (length/dia). Unfortunately, welding the guide bore down inside the valve spring region is very difficult and the necessarily used , porous castings are full of oil. After any welding is done, the guide location needs to be determined, the hole drilled, bored or reamed then roller or ball burnished or the guide will loosen and fall out. The valve seats need to be removed, the odd squish band on the back of the combustion chamber filled, the combustion chamber resphered, head gasket surface recut, valve guides made, installed and fitted to new valves which need to be made, the valve seat bores need to be recut from the guide centers, valve seats and installation tools made then installed, seats cut, ports blended, springs shimmed and installed and pistons be repocketed.
>
> Possibly Hemmings will make one for you or you can find a factory short stroke 750 head which comes that way. You will still need to repocket or find pistons but the original valve train geometry is for a 250cc cylinder from the Model 88. That is why the 750cc head has the odd crescent in the back of the combustion chamber. The 750cc head had to be moved forward or the pushrod tube bores would have encroached on the cylinder bores and the rocker geometry moved with the head but was not corrected until possibly the short stroke 750. Possibly but only possibly.
>
> I have my own angles, locations and linear plus radius dimensions in a head which is almost done but I have $12,000 worth of time in it and am not giving it away. However, it is for sale at that price as I need to do other things and haven't time to build a decent short stroke bottom end for a Norton.
>
> I hope all that helps but don't forget that the Commando chassis is a mess and needs work first or it will wobble or weave more badly than they normally do with the stock power. The Featherbed frame, of course, is worse.
>
> KA

Mc-engine list discussing flow benches for sale and thot this might interest us noob'ies on measuring instrument issues.....ugh.....hobot............

On 5/29/2013 1:17 AM, Ken Augustine wrote:
> I don't know about the SF-120 at all but what I have are two SF-110 units. I have two because I was going to set up a second shop and wanted one there in what may still be a fabrication shop plus one here in my machine shop.
>
> From what I have been told, SF came to be by selling their dynomometers business which either was or became Go-Power and there was a timed non compete moratorium on making dynomometers. They used the money and time to develop their own computer monitored or controlled SF dynos but had the opportunity to sell other related engine test equipment and developed the SF-100 which was inexpensive and all that was required to do accurate, credible engine gas flow research. Flow through orifices is a second order phenomenon so the relationships aren't first order linear but SF bought linear scale analog gauges from Dyer Instrument so there had to be a second order translator which was simply a paper scale with parabolic load lines which were odd to read but worked. The SF-110 was simply the same unit and components with a second order scale substituted on the Dyer differential flow gauge so the scale reading could simply be multiplied directly with the gross orifice flow potential to give direct if uncorrected for temperature, flow readings. This was fast, simple and accurate and all the work I have ever done has been done on that machine. I am not sure what the SF-120 is but I would guess that it is simply a further development of the SF-110 possibly including some automatic data collection. The unit on eBay has Audie Tech ADA but I don't know if that was OEM or after market. Additionally, there seem to be some fixtures and automatic reading functions listed in the ad. At $4265, I wouldn't want to pay for it but wouldn't want to be required to duplicate the capacity. One way or another it seems like a very capable piece of equipment but a machine shop will still be required to make brand and model specific specialized test adapters and other kinds of arcane apparatus. I have two roll cabinets overflowing with such items that I have made over the years.
>
> The limits in the SF-100 and SF-110 are an upper flow capacity of roughly 185 cfm and a total pressure differential of 16" H20. The SF-300 will allow higher test pressure differential including an automotive more standard 28" H20 and higher flow capacity but that can be converted mathematically and I have never found it either necessary or helpful to have that capacity. Additionally, the SF-300 likely has independent cooling of the fans which obviates annoying but achievable inlet and outlet gas temperature recording plus calculation which are required or at least advised on the SF-100 and SF-110. Either way, all of those performance specifications should be determinable from the SF ads.
>
> Beyond that, the subject of gas flow could require a doctorate degree in fluid dynamics, a baccalaureate degree in machine technology and at least access to a machine shop simply to make good measurements without any attention to what makes for otherwise meaningless, plastic motorsport trophies. Hope that helps.
> Ken A

It wasn't asked but not everything tried on the dispassionate flow bench has an effect and much of what has an effect, effects negatively. After that, merely flow testing without changing anything adds nothing at all but knowing what is possible rather than merely wanted, is crucial. There is more to dynamic gas flow in running engines than steady state measurements on a flow bench and more to steady state measurements themselves than the flow bench alone will reveal. Along with that, knowing what is necessary separates the winners from the wieners.

Beyond that, you will have specific failures like porting through to water passages or oil and having valve guides, seats or seal surfaces lose their support when that nasty gas flow impending material is removed. After that, you will need to do ALL of your other work properly, learn tuning and buy that equipment or hire someone like Derek to do that plus ride it to the elevated potential that you will presumably or at least hopefully, by then, have.

KA

I've been reading the posts in this thread and I have a question. If a Norton head is ported to 34mm inlet tracts with no taper, at what revs does max torque occur if a standard 850 cam is used with separate pipes, and spec timings ?
I never think about suction in ports and gas flow, my thoughts are usually about harmonics, gas speeds and mass flow. I believe that once the standing waves are established in inlet and exhaust, we are dealing with flow at sonic speeds. Can a flow bench operate at that speed? ( A friend of mine was a scientist on the trans sonic wind tunnel in an aircraft research establishment) . If the motor is over-ported, the effective rev range may be too high for most purposes, and the loss of midrange can mean the bike is slower on many race circuits. I can see sense in having huge ports for land speed records and big race circuits like Daytona, but even there surely you have to actually race in the in-field ? My thoughts are that I should always tune for maximum torque at 6000 RPM. That is apparently what the other bits of the motor were actually designed for. Moving up the rev range beyond that meets the law of diminishing returns, and becomes very expensive. In any case with all the theories, you end up with a Nourish Weslake engine, why not start with one if the regulations permit that ?

The needed port diameter is most affected by the bore -stroke ratio. A long stroke 750 motor will never reach the needed velocity for inertia induced cylinder filling with a 34mm port. A 850 or larger long stroke motor may get there near the top of its rev range.

Needless to say it will not make a good street or short track motor.

Now a short stroke motor on the other hand may benefit from a 34mm intake. Jim

The highest flow velocity I have read with my flowbench is around 650 fps. That was in the center of two small velocity stacks end to end. My bench is limited to around 185 cfm so I would not be able to get that much velocity with a larger diameter velocity stack.

When I was working on making exhaust ports that flowed I used a 35 gallon air tank with a flange to mount a head. I would pressurize the tank to 100 psi and then open the exhaust valve for a set amount of time and lift. I would measure the pressure drop in the tank to compare the flow of one port to the next. Worked great but it sure was noisy.

Flow testing an exhaust port on a bench is nearly worthless as it is not possible to get nearly enough pressure to create a velocity anywhere near real operating conditions. It does help at low lifts though for valve seat contours. Jim

I recall reading about and seeing pictures (Fayette Taylor) from Rambler or AMC using roots blowers and big receivers to test the valves/ports across various differential pressures. Professor Gordon Blair would use a port flow coefficient map which showed characteristics of the port/valve across a spectrum of differential pressures. For modeling he would use the flow coefficients of the port/valve both breathing in and out as that is what it really does.

comnoz said:

Flow testing an exhaust port on a bench is nearly worthless as it is not possible to get nearly enough pressure to create a velocity anywhere near real operating conditions. It does help at low lifts though for valve seat contours. Jim

Click to expand...

Smokey Yunick knew that running conditions could not be recreated with a flow bench and he did not try to do so. What he looked for was a way a flow bench could be used so that positive results that showed up on it would also show up as increased power in an engine running on a dyno. He wondered why no one tested port flow with water instead of air.
Coincidentally, I know one Norton tuner who did use water flowing through a head so he could see it, and he became well known and successful in Norton circles.

Today there is CFD.

What size ports would be appropriate for a 89mm stroke 920 street fighter with hot cam and good CR?