Guess my fren's HP figures

[f8.]

Originally posted by Jun9970@Jul 21 2004, 11:41 AM
Lets say port size = x
Boost Pressure = y

Cam, intake manifold, intercooler = non variable (hence not in equation)

z = time

For simplicity sake, lets say

The amount of air flowed = xyz

Hence, if the ports r double in size,

won't it equal to 2xyz

n therefore, mean the bigger the port, more air can flow over a period of time?

firstly, dolly, thank you :D

hi jun9970,
when you say amount of air flowed=xyz, i wasn't quite able to understand you. is x the diameter of the port? if so, i can't see it still how xyz is amount of air flowed.

anyway, lets say amount of air is the mass of air, because that is independent of pressure then unlike volume. and this amount i call n.

i mentioned previously in some old old post, quoting ideal gas law from form 5 spm physics, that

PV=nRT

where P is pressure, V is volume, n is number of moles of the air molecules, R is the ideal gas constant and T is absolute temperature.

our issue here is slightly complicated by the time dimension cos we are dealing with flow rates and velocity. anyway, in this case, the only time varying variables are P, V and n. T of course changes over large scale time, but in the small scale time during valve open and shutting its P and n that varies.

so fluctuating V in time, is your CFM then.

man answering this is going to take all day to explain properly. and i haven't even got to my point yet. i think i'll come back to answer this better another time.

all i want to say is, if bigger is indeed better, lets forget having exhaust pipes then. just leave the exhaust ports exposed. will you get more power? surely there is less restriction than feeding it into a pipe.

just bear in mind air is a fluid. and moving fluid has momentum energy because it has mass and thus inertia. once you establish a high flow of gas, it will tend to want to continue rather than stop. and because your engine releases gas and takes in gas in pulses, it really helps when during the interval between pulses, there is still remnant gas flow rate from the previous pulse before the next one comes. this helps to smooth out the pulsing a bit, so that the gas doesn't have to be accelerated and deccelerated each time so much. and very high gas flow velocities help. of course, as with everything, too much of it is bad. in this case, when the flow starts becoming turbulent, its bad. so the trick is to size the ports big enough for least restriction yet small enough for high velocities.

and i think the VR4 head ports are a tad too big for a 2L doing 300-400hp, in which case the evo heads might give better response, promote faster burn rates which will give more power with less ignition advance, promote better mix and actually end up squeezing more air in due to higher air flow velocities.

go speak to any racing motorcycle tuner and they will tell you all about the beauty of small ports.

 

[coldfish]

Originally posted by Joeker+Jul 21 2004, 05:58 PM-->

QUOTE (Joeker @ Jul 21 2004, 05:58 PM)

--QuoteBegin-coldfish

@Jul 21 2004, 11:52 AM ermm.. smaller ports, u get higher air flow velocity but end of the day, you still lose out in volumetric efficiency. think 1.5" exhaust piping vs 2.5" exhaust piping. exagerated since the port size difference cannot be that big. but the turbine is capable of pushing while compressing much more air into the ports. despite higher air flow velocities, smaller ports will not give you much more compressed air but just restrict the volumetric flowrate.

Err...I thought V/E decreases with port size increament? :huh: hence the larger the ports you run, the lower the V/E drops. [/b][/quote] your V/E should be "air velocity". volumetric efficiency is how much air u can flow thru, not the speed of air. with respect to time, it becomes volumetric flowrate. usually in CFM (cubic feet/meter), m³/sec, gallon/sec, etc. imagine u have 2 barrels of water suspended in air (top vented) both containing same volume of water. one with a 2cm hole at the bottom, the other with a 3cm hole. which one drain out faster? ;)

 

[f8.]

hey interesting!

ok bucket a-2cm hole.

bucket b-3cm hole.

if we somehow can block the hole, then open it for only 60ms, corresponding to opening time for exactly the duration of intake stroke on 4 stroke engine at 8000rpm. and further, we make the base of bucket 10cm thick. but somehow the opening valve is at the top. and repeat opening it 100 times. collecting the water that flow out. which bucket do you think will give more water out in the end? actually, if the base is very very thick, which bucket will have water reaching the furthest downstream for the same time?

i don't know the answer. but certainly it feels to me bucket a will give the water out with more pressure. and this might not relate so well to engines because water, although a fluid, is not compressible like the air fluid.

 

[coldfish]

hi f8.

Volumetric flowrate [m³/sec]= Velocity of air [m/sec] x cross sectional area[m²]

multiply volumetric flowrate by time and u get the volume of air flowed. check the units, that's how u validate if the equation is correct. JUN9970 was right, he just simplified the port size to a single variable to make it easier to understand.

cross sectional area of a tube/round port is pi*r² (where r is radius)
so ur volumetric flowrate is directly proportional to the r squared. so the increase in r is very significant to the flowrate. or, if cross sectional area of port is doubled, flowrate doubles. what's so hard to understand?

the PV=nRT ideal gas law only works in low density, high temperature (cos in low temp, high pressure, gas liquefies) with assumptions like:

1. homogeneous state/ constant equilibrium (no variable temperature, pressure or density throughout the system)
2. ideal gas
3. etc depending on situation

if u can quote PV=nRT, i'm sure u know all the above. it just doesn't work for an internal combustion engine regardless of type. to model differential equations for an engine, you need the Fourier transform with everything wrt to time and isolated. and still, with a lot of ideal assumptions. end of the day, bigger ports still flow better than smaller ports because the volumetric flowrate still holds true in Fourier.

if you really insist smaller port flows faster, yes it flows faster but not more air, less usually. engine intake ports were designed to have a smooth constriction slowly becoming smaller. venturi effect. it helps speed up the mixture of 2 fluids in things like bunsen burner, bulb atomizer (perfume stuff) and intake manifold. in engine intake ports, it doesn't really help the gas to move faster by much but it encourages fuel and air to mix better for more efficient combustion (please dun doubt, my professors used to design engines for a living before they started teaching). no matter how big we pnp the head, the port's design is such that it gradually and smoothly becomes smaller. but we give it more potential for higher air flow speed, and increase volumetric efficiency. and no sohai in this world would pnp the head to have constant port size all the way thru the combustion chamber. if someone does, go shoot the idiot.

i burned my thermodynamics text book after i finished the course. u can check yours. both my fluid mechanics and heat & mass transfer textbooks were written by my professors (Prof Robert Fox - fluids, Prof Frank Incropera - H&MT) at purdue university. highly recommended for ur reference. the books are from Wiley & Sons.

have a good weekend.

 

[coldfish]

Originally posted by f8.@Jul 23 2004, 02:13 AM
hey interesting!

ok bucket a-2cm hole.

bucket b-3cm hole.

if we somehow can block the hole, then open it for only 60ms, corresponding to opening time for exactly the duration of intake stroke on 4 stroke engine at 8000rpm. and further, we make the base of bucket 10cm thick. but somehow the opening valve is at the top. and repeat opening it 100 times. collecting the water that flow out. which bucket do you think will give more water out in the end? actually, if the base is very very thick, which bucket will have water reaching the furthest downstream for the same time?

i don't know the answer. but certainly it feels to me bucket a will give the water out with more pressure. and this might not relate so well to engines because water, although a fluid, is not compressible like the air fluid.

the valve can be anywhere. the open top simulates equal boost pressure from the turbine (both barrels of water under atmospheric pressure). i don't know what made u think if any one bucket has higher pressure in it. this is an unbiased analogy to compare the port size effect. you can have the base 10meters thick if it pleases u.

if u want compressible fluid, imagine 2 giant balloons with all things being equal including volume of air and air pressure except for the opening end. one with a larger opening end than the other. enough la. i like ur resilience but drop it will ya? <_<

 

[f8.]

hi coldfish,

most of what you say make perfect sense. particularly the bits at the top. before i continue, let me make it clear that everything i am talking about here is in relative terms, and not absolute.

the problem with the engine, as i stated before, is that it is not a continuous pump. and further, for an NA engine anyway, you only have atmospheric pressure to try to gush air in to fill the partial vacuum in the cylinder, depending on your VE.

i would agree, on a flow bench, where the flow is continuous and already in steady state, and they have a massive pump to pump air through it, you will find that the larger your port, the higher your CFM figures. nothing difficult to understand about that at all.

my gripe is, how appropriate are CFM figures from a flow bench when meriting performance gains of different heads/port jobs/port diameters to a given engine?

let me illustrate. say we have a port of 1inch diameter. on the flow bench, at say 20degC gas temp, and say 5bar pressure, we flow x cfm. on a 1.5inch diameter port, for the same conditions, say we flow y% more cfm. yes, over a period of time, no doubt the 1.5inch port flowed more air. provided 1.y* x is
now, lets measure things differently. we only apply 0.9bar. and instead of applying that constantly, say we only give a window of a few milliseconds. from the time the pump starts pushing air, to the time we stop it. of course, the cfm figure will change with time because we are now measuring the transient state of it. for both ports, we will get a curve of cfm vs time for that duration.

now, integrate that area under curve for cfm vs time to get total volume of air flowed. it may still work out that the 1.5inch port gives higher values, but this is pump dependent. another pump may give a higher rate rise for the smaller port. likewise on an engine, and if you consider that the valve gradually ramps open and then closes back, rather than being completely shut, completely open for a period of time and completely shut again, you will see how far detached the actual mehanisms of the engine is compared to the grossly over simplified flow bench.

like i said, this is all relative. this whole thing came up from whether for the 4G63, the bigger ports on the VR4 head is better than the smaller ones on the evo head. and i have said so many times before, i am not saying that small is good so make it as small as you can. but relatively, i am inclined to believe the vr4 ports are a tad on the large side. but if we use your argument that bigger port more flow not hard to understand, then we might as well port it as big as we jolly well can, not forgeting to include the taper in size as you mention. then we should get massive power right?

i don't know. but tell me one thing then. why is it the most powerful 4G63's out there use the old vr4 block because its supposedly stronger, but base the head on the later evo series? and have you seen the intercooler pipings on a WRC car? its tiny! yes, the pipe between the turbo and air filter is huge, but thereafter all the way to the throttle body it is tiny. surely a bigger pipe can flow more so i suggest you give them a ring and tell them off.

fact is, on the VR4, the ports are larger. on the later and newer evo, it is smaller. if larger is really the way to go, i think we ought to send the entire engineering team at mitsubishi motor corporation to purdue and read the books by Prof Robert Fox - fluids, Prof Frank Incropera - H&MT from Wiley & Sons.

have a good weekend.

 

[vestax]

well rally using small ports , small tb , small ic pipes cause they don't starve for power , they need the engine to be responsive as possible , limited to a certain amount of power but they have massive torque .

 

[Joeker]

Great stuff F8!! keep em comin'.

 

[Danny]

ok, to summarise in english,

smaller ports = higher velocity but smaller air flow(cfm)=higher torque
bigger ports = lower velocity but bigger air flow(cfm)=lower torque?

 

[Jun9970]

this is good shit man. from ZTH summore. who would haf thought. :blink:

btw my fren (the basis of this topic) dynoed at 295 atw, boosting 1.3 - 1.4 bar :rolleyes: clutch was dying on him, so had 2 stop there n then.

oh yeah, he upgraded the fuel pump (inline bosch)

if rallying is all about response as opposed to HP, then higher velocity and torque is the objective. Clearly then, the intercooler piping all the way to the TB should be small. lets face it, rally cars rn't built to make ultimate HP.

ultimate HP its still about airflow right? so bigger is better?

 

[coldfish]

maybe u are right.

come to think of it, i run a silencer with Ø2" opening for my car. without, it is Ø2.5" straight flow with Ø4" outlet.

with the silencer, the engine revs quickly up the low and mid range but kinda restricted at top end. VERY much quicker low and mid range revving with the silencer. some degree of restriction (or in this case, backpressure) is actually good for all but top end.

without the silencer, the low and mid range climbs slower (no back pressure), but top end doesn't wanna stop. i dyno-ed at DMT to see the difference of having the silencer and without. there was 3HP loss from mid range all the way to the top with the silencer on. the graph almost seemed like an offset from midrange onwards. i also ran richer with the silencer on.

simply put, i can rev up quicker thru the low and mid range with the silencer, but still producing slightly less HP. without the silencer, power measured was stronger, but climbs slower at low and midrange.

the torque curves behaved the same like the power curves. i thought i would get better torque with the silencer on.

maybe this can indirectly attribute to some of the above discussion.

the 1300HP veilside R34 does have big turbines and fairly large intercooler piping. that's from what i can see in the video. there's a lot of saying about small turbines spool up quicker and decreases lag, where as bigger ones spool much slower but boost mad up top. i don't know. i don't have a turbo engine. maybe some boosted engine owners should speak.

anyway, back to main topic. 295atw @1.3-1.4bar? what power at 0.9bar?

 

[Jun9970]

didn't dyno the car at 0.9 bar. started off at 1 bar, n went on from there.

0.1 increase netted approxmately 15HP or thereabouts, was about 250+ at 1 bar.

using an AVCR 2 control the boost.

A/F in the high 10s, although a spike is noticeable when full boost hits at 4500rpm. leans it out temporarily to low 12s for a sec n stabilises

interestingly enuff though, he is running a s-flow 3 inch system with a kakimotor back box

had a 2.5 inch straight thru pipe at home from his NA days but din bring it along for the dyno.

would haf been nice 2 see the difference in curves.

btw does anyone haf a link to an english manual on how 2 preset the differing boost in differing gears for the avcr? after f!@king around with the damn thing for 2 hours, we couldn't get the thing 2 boost constantly as preselected. no traction for the 1st 3 gears.

 

[Danny]

jun, pretty impressive 295hp atw, of coz no match for sdf and ramsing's monsters but still good for the setup u mentioned!

 

[coldfish]

english manuals again? i had to look for the english manuals for DMT as well. they couldn't install the ITC without it.

15whp increment per 0.1bar is eye popping for me. damn NA engines.

 

[seantang]

Your friend should bring his car to another dyno - say Speedworks or Aerotech for a second opinion. Chances are good that a different dyno will produce a different figure. In fact, every dyno produces a different figure.

Unless the dyno is properly calibrated, it's best used to show RELATIVE differences between each dyno session. Eg. get a baseline dyno, and then after some mods, dyno again on the same machine to get a % gain/loss.

The closest you can get to an accurate actual whp number for your car would be to dyno at as many dynos as you can and average it out. But then again, just one crap dyno with poor calibration would probably skew the averages so much that you'd can rubbish the whole exercise.

Another thing is that when I dyno, I never dyno at the establishment that modded my car, or sold me something. I'm not questioning the honesty of dyno operators but it is indeed in their best interests to "influence" the dyno figures. In addition, I would never tell a dyno operator my mods, or my whp expectations nor the whp figures of previous dyno runs.

 

[coldfish]

agreed. your last paragraph makes total sense.

speedworks however, has a broken lambda sensor. still waiting for replacement from Germany after 2 months. and when they dyno'ed my car, the calculation for driveline loss correction factor was only 9%. i only have a type-r flywheel and very heavy wheels and tires. besides, my gearbox is 11 years old. i expected more like 20%. don't quite trust the "power at engine" graph from them. power at wheels however, is quite consistent with that of DMT's. and they would honestly claim that all dyno machines are different and would measure differently. but to each his own, we make our own judgments.

anyone else has more knowledge on effect of wheel and tire weight on dyno's please input. from what i gather on the internet, ppl get 1-3HP diff using lighter wheels. maybe just psychology.

 

[Torment]

i think the purpose of having lighter wheels is not about gaining HPs but rather having better excelaration response and better stability in transient maneuvers. the advantages can't be felt or shown on a dyno right? the HP gain is an added bonus.

 

[Jun9970]

Originally posted by Torment@Jul 29 2004, 03:40 PM
i think the purpose of having lighter wheels is not about gaining HPs but rather having better excelaration response and better stability in transient maneuvers. the advantages can't be felt or shown on a dyno right? the HP gain is an added bonus.

heavier wheels would result in greater transmission losses especially on a dyno. How much though is subject to debate. just imagine, opening a fridge from the handle as compared 2 opening it right at the hinge. opening it right at the hinge requires more effort. its about leverage. the bigger n heavier the rim, the harder the driveshaft has 2 spin the hub which inturn has to spin the wheels of the car. effectively, bigger rims would increase the radius from the end of the wheel to the centre of the wheel (ie hub/driveshaft).

 

[coldfish]

http://www.nissanperformancemag.com/may01/dyno.shtml

check that out.

http://www-unix.oit.umass.edu/~tcroy/watsonengineinertia.htm

that too.

the second one made a bit of sense when it talked about rotating inertia and wheel speed inertia. so wheel size would actually make more difference. now i see why my previous car owner had his wheels about 1cm undersized (i'm talking comparison of tire OD's, not rim size). if radius of gyration comes into play, weight is really less significant compared to the size. i'll keep my specs the same until my 16" tires wear out, swap back to 15's and dyno again after a fresh oil change. then, i'll see if there's any change in power and torque @ wheels.

still, guys at both Speedworks and DMT insist that there is no diff on the power @ wheels because lighter wheels only bring it up to speed at a quicker time, not decrease driveline loss. i prefer to doubt, but we shall see when we get some graphs.

 

[ShaunSG]

wheel and tire weight on dyno's please input. from what i gather on the internet, ppl get 1-3HP diff using lighter wheels

Inertia type dynos will show up a gain because acceleration time of the known-mass rollers will decrease. Loading type dynos will not show up any gain if the power is measured at a fixed engine RPM and torque measured directly followed by calculation of HP. Loading dynos in ramp mods will do acceleration sweeps where the operator can specify for example 3000rpm/sec . The faster the ramp rate (top drag motors run 5000rpm/sec) the less time there is to capture samples, so power is calculated using the engine RPM, known resistance applied, and the time taken. On such runs, the faster the ramp rate, the larger the gain in power shown from lighter wheels will be.

In reality, engine power does not change because the only thing that can change it is change in torque which is dependent on combustion event or engine hardware changing. Wheelpower doesn't change either, only mass has changed and that affects the transient calculation.