Carb-intake manifold-intake port matching

Status
Not open for further replies.
needing said:
Practically, use a flow meter and look for greater than 340 metres per second.
Click to expand...

Lets see now, are flow meters really that sensitive ?
At 6000 rpm, thats 100 revs per sec, so 50 firing cycles per second. (per cylinder)
10 milliseconds per rev, so 5 milliseconds per intake pulse.
Give or take some intertia effects and residual flow time.

What sort of meter can measure that ??
Anything we can afford ?

And would it record peak velocity, or just an average....

P.S. Combustion pressures within engines are also similarly notoriously difficult to measure.
The firing time is so quick, except at idle, very few instruments have sufficient time to actually record it.
By the time the pressure pulse gets to something that can try to measure it, its over..
 
Many years ago a friend of mine was a Senior Research Scientist at the Aeronautical Research Laboratories in Melbourne. He operated the Transonic Wind Tunnel using a Digital Equipment PDPi computer, which was the reason I visited him several times. It was about the time when Bill Gates was getting started. At very high wind tunnel velocities, what you might expect is not necessarily what you get. I suspect flow benches might sell you a pup. The only way I would use one would be to make a change to the port, measure the flow then measure the changes in torque and horsepower, then possibly establish a correlation. I think I'd still be fooling myself at high engine speeds.
 
ugh if hobot has to clarify funny flow fhysics. Speed of sound varies with temp and pressure but near sea level riding temps is ~340 m/s or 11-1200 ft/s. Comnoz has measured good ports in mid 450 f/s though may exceed that some spots for an instant its still not significant velocity to encounter sonic choke - which does not really choke down per say just no more increase no matter how hard pulled on or blown it. Stepped can affect sonic reverb pressure waves but the flow boosting step/lip is not placed where that is a factor either. Peel calcs out to possible hit 700 f/s for a while.

Port flow can be monitored about anywhere along its path by tiny holes put in with a meter or water tube attached. Low reading mean good flow speed along axis while higher reading means stagnation back wash etc. One the references I left had done this and filled in every area that showed low pressure til significantly increased velocity over standard head but did not make as much power on dyno nor tracts as standard head. This is example IS NOT Sonic Choke on getting maximum speed for same flow as standard head but a case of too small a tube for the cmf required.

Good fuel suspension and/or sonic pressure crests helping to pack mix in may give less than 10% increase, increasing flow volume by streamlining steps &/or porting, proper *slight* tapper and velocity stacks can give over a dozen % better. Ya may need dyno and stop watch to tell improvement on textured surfaces and sonic wave tuning but anyone crude seat of the pants and grin stretching will know if steps working or not.
 
I might be a bit pragmatic, however I take note of what actually works in old bikes - things the factory did which were improvements. e.g.
the fastest 350 single in the late 50s was the 7R AJS
the Aermacchi 350 of the mid 60s was faster and had a tapered inlet port.
Gileras Quattro had centre plugged heads with the spark firing through a slot, also had a 7 speed gearbox with a fairly long stroke motor.
Nortons had squish band heads on their manxes, sometimes at two levels in the combustion chamber.
An average four valve Jawa is usually marginally faster then the fastest two valve Jawa.
In the 30s Maseratis used a mixture of benzene and methanol as fuel - toluene is similar to benzene, much but less toxic

I am a scientist, however what works - works !
 
hobot said:
Aw slick intake path areas that sling fuel droplets out of the main flow d/t mass not swinging the bends as sharply as lighter air molecules does not happen in the straighter parts or before bends.
Click to expand...

the stream tubes in a converging Venturi section are anything but straight! There is a 20 -22 degree wall taper!

the BSA engineers who intentionally put in a Venturi to frustrate the guys who were fairing out their intentional steps, could not have been thinking boundary layer trip.

A fuel dispersal scheme AND a boundary layer trip might provide a double whammy.

I am no longer inclined to squeeze every last bit of power out of my Norton, but for those who do, I suggest they keep both these mechanisms in mind and experiment accordingly. And, as you said, what worked for BSA may not apply to some other engine, but that is what makes experimentation fun!

Slick
 
acotrel said:
Many years ago a friend of mine was a Senior Research Scientist at the Aeronautical Research Laboratories in Melbourne. He operated the Transonic Wind Tunnel using a Digital Equipment PDPi computer, which was the reason I visited him several times. .
Click to expand...

And he applied it to inlet ports ???
Airplanes are rather a different kettle'o'fish...

acotrel said:
using a Digital Equipment PDPi computer, which was the reason I visited him several times. It was about the time when Bill Gates was getting started. .
Click to expand...

1st computer I ever saw was a pdp8.
No screen, no keyboard = big calculator= no big deal.
Same as Bill Gates basic !
A programmable calculator was $millions less expensive...

Fortran, Pascal and Cobol in the same boat, for another decade or so !?

P.S. Someone I know of acquired a pdp11 for scrap value ($20 ?), still had the programming switches.
 
Rohan said:
needing said:
Practically, use a flow meter and look for greater than 340 metres per second.
Click to expand...

Lets see now, are flow meters really that sensitive ?
At 6000 rpm, thats 100 revs per sec, so 50 firing cycles per second. (per cylinder)
10 milliseconds per rev, so 5 milliseconds per intake pulse.
Give or take some intertia effects and residual flow time.

What sort of meter can measure that ??
Anything we can afford ?

And would it record peak velocity, or just an average....

P.S. Combustion pressures within engines are also similarly notoriously difficult to measure.
The firing time is so quick, except at idle, very few instruments have sufficient time to actually record it.
By the time the pressure pulse gets to something that can try to measure it, its over..
Click to expand...

Maybe Capt. Obvious will let us use his/hers. I am sure he/she has it covered as usual, with all of the depths of experiance that Google can offer.

Thank you, Rohan, for a relality check. Although oblivious to most cought up in the frenzy, I appreciate it.

This is a classic case of Access Norton Analitical Lunacy "ANAL". The poor guy just wants to know if matching his manifold would be advantagous or not. Page 5 and counting.
Duh.
 
pete.v said:
Page 5 and counting.
Duh.
Click to expand...

Typical Access, when no-one actually knows.
The less folks know, the longer the replies go on.... ??

It does often cover some interesting ground though.
And as long as the OP gets an answer, of sorts... ?
 
I thought the Roland Pike BSA Goldstar finding was directly related to the OP's question and also answered his question.
It is also an interesting bit of info to many apparently, and that might be because it flies in the face of conventional wisdom, which is, always make the inlet tract as smooth as possible everywhere. Somehow Pete has managed to take umbrage at the discussion. That also flies in the face of conventional wisdom, but not in such a good way. :roll:

Glen
 
The BSA reference proved there is nil science involved in their accidental turbulence discovery so hard to accept any application to Norton from such dissimilar engines. Read or ignore the googled paper below but may have historical interest to some. Trial and error, mostly error, but some find power gains in non obvious ways. May be evidence of the butterfly effect.

We have to conclude that the differences in the initial velocity field in the cylinder when the inlet valve closes are attributable to the turbulent velocity fluctuations in the inlet manifold. These may have an indirect influence also, since the velocity fluctuations affect the pressure drop across the inlet valve during filling, and may have a small effect on the sound waves in the manifold. The relatively small differences at inlet valve closing attributable to the manifold turbulence are enough to cause considerable variability at TC, due to the
extreme sensitivity of the turbulence to initial conditions.
Click to expand...
http://www.mie.uth.gr/ekp_yliko/flow_in ... linder.pdf
 
Ta hobot.
Documented science regarding turbulence is so refreshing!
Ta.
 
hobot said:
The BSA reference proved there is nil science involved in their accidental turbulence discovery
Click to expand...

The SCIENCE comes from actually being able to measure it on a dyno.
Remember, science is measuring stuff so that its REPEATABLE.

Guessing stuff, or explaining or understanding it is altogether a different subject.
Gravity works, but the mind that understands how it actually works hasn't been born yet ?
 
Ah so if not for the love of science I would of had shorter boring career as drug dealer cutting product in Florida. Amphetamines were my hottest mover. On the other hand if not for psychedelics I might of ended up an industrial trained chemist checking sewage oxygen levels or making pesticide to injure infants and pets. I have a good bit parts left to put past Peel engine back together so in between bickering here and ongoing mechanical failure recoveries at home have pow wowed with comnoz, JSM, TC and Dreer, Canaga, George Kelly, Ron Fratturelli, Bryan Stark plus Big D Texas shop to see if can reproduce at least what Woods and Axtel had in their hey day and test mis matched stepped single and dual carb manifiolds. Sure do not want another motorcycle though so thinking of a foot forward extended fork drift tricycle or ultralight aircraft.
 
pete.v said:
This is a classic case of Access Norton Analitical Lunacy "ANAL". The poor guy just wants to know if matching his manifold would be advantagous or not. Page 5 and counting.
Duh.
Click to expand...
HA HA! Truth......... :lol: :lol: :lol:
 
pete.v said:
Rohan said:
needing said:
Practically, use a flow meter and look for greater than 340 metres per second.
Click to expand...

Lets see now, are flow meters really that sensitive ?
At 6000 rpm, thats 100 revs per sec, so 50 firing cycles per second. (per cylinder)
10 milliseconds per rev, so 5 milliseconds per intake pulse.
Give or take some intertia effects and residual flow time.

What sort of meter can measure that ??
Anything we can afford ?

And would it record peak velocity, or just an average....

P.S. Combustion pressures within engines are also similarly notoriously difficult to measure.
The firing time is so quick, except at idle, very few instruments have sufficient time to actually record it.
By the time the pressure pulse gets to something that can try to measure it, its over..
Click to expand...

Maybe Capt. Obvious will let us use his/hers. I am sure he/she has it covered as usual, with all of the depths of experiance that Google can offer.

Thank you, Rohan, for a relality check. Although oblivious to most cought up in the frenzy, I appreciate it.

This is a classic case of Access Norton Analitical Lunacy "ANAL". The poor guy just wants to know if matching his manifold would be advantagous or not. Page 5 and counting.
Duh.
Click to expand...

Thanks. Pretty much got my answer on the first reply, although some of the others are interesting.
 
Its never too late to experiment for yourself (?).

And what is there to lose, other than maybe a bit of performance.
And it may be a gain....
 
Rohan said:
Its never too late to experiment for yourself (?).

And what is there to lose, other than maybe a bit of performance.
And it may be a gain....
Click to expand...
.

That's what I thought when I discovered the mismatches; here's something that I can easily correct and at worst, nothing will change. At best, there could be a small improvement.
 
Rohan said:
acotrel said:
Many years ago a friend of mine was a Senior Research Scientist at the Aeronautical Research Laboratories in Melbourne. He operated the Transonic Wind Tunnel using a Digital Equipment PDPi computer, which was the reason I visited him several times. .
Click to expand...

And he applied it to inlet ports ???
Airplanes are rather a different kettle'o'fish...

acotrel said:
using a Digital Equipment PDPi computer, which was the reason I visited him several times. It was about the time when Bill Gates was getting started. .
Click to expand...

1st computer I ever saw was a pdp8.
No screen, no keyboard = big calculator= no big deal.
Same as Bill Gates basic !
A programmable calculator was $millions less expensive...

I was using a PDP8L up against a direct reading emission spectrometer for metals analysis. We used Focal language for the scientific algorithms, and had to rewrite it in machine code and recomplile it when we extended the memory of the computer from 4K to 8K. After about 5 years we had the spectrometer self calibrating. The SRS on the wind tunnel was monitoring the pitot tubes with the PDP8i. I bought one of his DECtape units which we used the same way as a hard drive. In the end we did chemical analysis which used to take over a week on the bench, in 17 seconds after a 21 minute calibration routine. In those days it was cutting edge, however we don't do much like that in Australia these days. I'm a dinosaur.

Fortran, Pascal and Cobol in the same boat, for another decade or so !?

P.S. Someone I know of acquired a pdp11 for scrap value ($20 ?), still had the programming switches.
Click to expand...
 
Is there a new message in there anywhere Alan ?
Or we seem to be raking over old coals...

And a little off topic, since no-one has broached supersonic flow effects in an carb/inlet throat.
 
Status
Not open for further replies.
Back
Top