Tuned manifold length

[MeXX]

The Basics & How it works:

When the inlet valve closes a sound wave starts to travel up trough your inlet port, the manifold and caburator/throttle body to the edge of the velocity stack, as the edge of the stack is reached and the wave goes to free ambient (BTW this is the roar we can hear and love so much) there is a reflexion in form of a sound wave that travels back the whole way down. This happens with sonic speed (which is influenced by the temperature).
We try to tune the length, so that the sound valve reaches the inlet valve when it starts to open so that more air is pushed in.
That is why U can't calculate it without knowing the cam duration.
As everyone knows every wave has a high and a low, so there is a RPM where there is a power gain and a RPM where there is a power loss.
Your usefull (best) power range is between the lower RPM power loss +8% and through the power gain RPM to the upper power loss RPM -8%.
With a well tuned engine you should reach a volumetric efficiency of over 100% this means your engine sucks (gets) more air into the cylinder than its displacement.

BTW it's pretty the same effect on the other side, the exhaust.


MeXX

[MeXX]

To show what I ment here is an examples:

FK 47:

Cam duration 278° @ .50

total duration 720°-278°=442°

Inlet lenght 530 mm from valve seat to edge of the stack.

n = 41,67 * duration * sonic speed / length / no. harmonics
1,5. harmonics (high RPM power loss)
 
n = 41,67*442*346,6/530/1,5=7737 RPM

2. harmonics  (RPM power gain)
 
n = 41,67*442*346,6/530/2=5803 RPM

2,5. harmonics (low RPM power loss)
 
n = 41,67*474*346,6/530/1,5=4640 RPM

4640+8%= 5011 RPM
5803+/-8%= 5338 RPM - 6267 RPM
7737-8%= 7118RPM

This means inlet lenght of 530mm on a FK47 is good from 5000-7200 RPM and best for 5300-6200 RPM

[MeXX]

Here is an other examble

same length different cam:

Web Cam 277

Cam duration 288° @ .50

total duration 720°-288°=432°

Inlet lenght 530 mm from valve seat to edge of the stack.

n = 41,67 * duration * sonic speed / length / no. harmonics

1,5. harmonics (high RPM power loss)
 
n = 41,67*432*346,6/530/1,5=7562 RPM

2. harmonics (RPM power gain)
 
n = 41,67*432*346,6/530/2=5672 RPM

2,5. harmonics (low RPM power loss)
 
n = 41,67*432*346,6/530/1,5=4537 RPM

4537+8%= 4900 RPM
5672+/-8%= 5218 RPM - 6126 RPM
7562-8%= 6957RPM

This means inlet lenght of 530mm on a Web cam 277 is good from 4900-7000 RPM and best from 5200- 6200 RPM

MeXX

[MeXX]

As the theories and reality are two different things and allot of dyno tests showed the RPMs where allways lower than calculated.
The reason is that the sound wave is not reflected on the edge of the velocity stack it leaves the stack more like a bow so if you subtract halve of the diameter of your stack, the calculated and dyno results are pretty close.

So I was ending up with a length of 500mm on my drag mill.


MeXX

[andy198712]

when i looked into this it seemed hard to get the desired length in a closed decklid, depending on which reversion you were trying for... i think it would deffinatly help to atleast try and get close. but then i didnt take in account cam duration i dont think.... i worked mine out by valve seat ID...? will dig it out for comparison although our method seems alot better!

[modnrod]

There's lots of tuning tweaks available to help the power output.

I've always set exhaust total length to bring a strong negative suction to the chamber at overlap just on and after IVO to get the intake airflow started, then used expected airspeed and temp on the intake to get the most cramming into the cylinder just before IVC.
Both exhaust length and intake tuning are obviously rev dependant.
I try to use the 2nd Harmonic for the best boost, but if space is an issue then the 3rd can be used as it's quite a bit shorter.

[Type1/DVK]

What's that 8% correction for?   

[Torben Alstrup]

Adjusting the intake for the 3rd harmonic wave amost always give the best hp and torque output. If rpms get "too high" the intakes should be shorter than practically possible. Then we set it up for 2nd. pulse. With front wheel drive and/or inline engines it is almost never possible to get enough length for the 3rd pulse. So there it is more "natural" to adjust for the 2nd pulse.

I don´t think that I ever have built an engine with 530 mm from valve to stack. Never had the need.

T

[spanners]

How did you fit 530mm in the engine bay?

[modnrod]

[ Attachment: You are not allowed to view attachments ]

[folkevogn]

[ Attachment: You are not allowed to view attachments ]

Love this tread, so much good info!! Thanks for chearing your wisdom!

Why is it a  dipp in 2 order with a 13 inch intakerunner? 3 and 4 order seams to follow a different curve. And how does this chart handle camprofile?

Folke

[modnrod]

The dip in 13" is just coz the "animator suffered a massive heart attack" (sorry! Monty Python moment......... ), it's just they ran out of room. That's also why the number "12000" is written in above the 12" line, that's the upper limit for the 12" runner length.

3rd and 4th harmonics do follow a different line, but if you look at the chart you'll see they all still relate to each other fairly evenly, just not in a direct linear fashion.

The chart is just for intake length from valve to end of stack. You still need to select cam specs, carbs, exhaust, head etc, for the rev range you are tuning for. Ideally if everything in your engine matches up and is complimentary, then choosing the correct length intake will give an extra boost at the revs shown over an intake with no tuning effect (eg, too short, way too long, etc). This extra boost can be worth 5% at the wheels quite easily, so is worth pursuing if you have the tuning time.

The lengths are for standard intake air temps from air around 20*C. Choosing an intake length for 0*C air temp will need 4-5% less length (in cold air the Speed of sound is less, so reflections take longer, so tuning length is shorter), likewise 45*C add 4-5% length. Methanol is VERY cold, so pick a length 5% less.

Better yet, as long as you have the time, opportunity and curiousity, experiment a bit with carb tops of different lengths (stack lengths) and see how it affects your MPH at the end of the track.

It isn't an exact science, but it's easy and fun, so again I hope it helps someone else out.

And Mexx, if you have any more calculations in your little black book you could share, or anyone else, I would love to add them to mine! 

[BeetleBug]

The lengths are for standard intake air temps from air around 20*C. Choosing an intake length for 0*C air temp will need 4-5% less length (in cold air the Speed of sound is less, so reflections take longer, so tuning length is shorter), likewise 45*C add 4-5% length. Methanol is VERY cold, so pick a length 5% less.

Excellent thread, thanks guys!

What do you consider the temperature to be inside our tight engine compartment? I did a test with a closed lid, outside temperature +6, all engine tins in place and I was a bit suprised. And opening the lid at the top did not help at all.

-BB-

[MeXX]

What's that 8% correction for?  

Hi

The 8% is no correction, it's just the range (+/- 8%) from the calculated RPM (frequency) where the effect (the loss 1,5 harmonics and 2,5 harmonics) & (gain 2 harmonics) is very strong.
Thus the usefull RPM range is from RPM 1,5 harmonic +8% where the loss gets less (?!? sounds strange) to RPM 2 harmonics -/+ 8% where the gain gets most (sounds better) to the 2,5 harmonics -8% where loss gets more again.

I hope this helps

MeXX

BTW I was ending up with a length of 500mm and the power gain between 6000-7000 RPM pretty the same as in the graphic from Reply #9 for a 20" manifold  .

[dangerous]

Hi Mexx, great post!
Love the real world examples!

Can you explain further how and why the calculation of duration adds and subtracts those 15 and 51 numbers,
and where did you get them??
Same for the 20 and 56 numbers used on the web grind.
( the 15 and 51 numbers still used on the calc for the Web grind but I realise this is a typo since the answer is correct)

I realise the 540 number is 1.5 crank rotations, but this is 3 turns of the cam to include all events? 

I normally use the inlet (and exhaust) closing crankshaft numbers at .050" in pipemax program,
and those dont seem to match.

Perhaps you can better explain the numbers for me and any others scratching their heads.
Thanks, dave.

[modnrod]

What do you consider the temperature to be inside our tight engine compartment? I did a test with a closed lid, outside temperature +6, all engine tins in place and I was a bit suprised. And opening the lid at the top did not help at all.

-BB-

Hi Mr BB!
The chart assumes the higher underhood temp (of your average US V8 '60s sedan) that ambient temps of 20*C will produce. The chart itself was made as a fairly accurate "cheat-sheet" averaging out factors that only had effects of 5% or so, but does a reasonable job of it.

As an example, on a little dragbike I used to run, the chart suggested a length of 13", as my rev range went from a shift point of 11000 to the rev drop point of 9200-9400. On the track however, I could feel it "nosed over" very slightly before the shift point. I started playing with stack length until I ended up with the best MPH, but it was different from the chart. However, after I removed 5% due to my "engine bay" air temps being much cooler than a US '60s sedan, and then another 5% due to running methanol, the chart suggested intake length was right again.

I used the 2nd harmonic, but in addition I used a launch RPM in the middle of the 3rd harmonic for that length, and I gained a full 1/2 tenth on the 60ft every run. The total ET gain for getting the stack lengths right was 0.25secs and 4mph approx. That's fairly significant on a low-9s bike.

I'm about to add an inch or two to the intake length on my little 250 Piaggio scooter I ride to work down the highway, to get it in line with the CVT revs, just for a giggle!  

[MeXX]

Hi Mexx, great post!
Love the real world examples!

Can you explain further how and why the calculation of duration adds and subtracts those 15 and 51 numbers,
and where did you get them??
Same for the 20 and 56 numbers used on the web grind.
( the 15 and 51 numbers still used on the calc for the Web grind but I realise this is a typo since the answer is correct)

I realise the 540 number is 1.5 crank rotations, but this is 3 turns of the cam to include all events? 

I normally use the inlet (and exhaust) closing crankshaft numbers at .050" in pipemax program,
and those dont seem to match.

Perhaps you can better explain the numbers for me and any others scratching their heads.
Thanks, dave.

Dear Dave

To complete all 4 strokes it takes 2 revolutions of the crank thus 720°

IVO (Intake Valve Opens before TDC)          FK 47 15°, Web Cam 277 20° (@.050)
IVC (Intake Valve Closes after BTC)              FK 47 51°, Web Cam 277 56° (@.050)

The 540° in the formula are 720° (two revolutions) minus 180° (difference between TDC and BTC).

The calculated total duration is the time in crank° the sound wave has to travel up be reflected and travel down again, and to end up at IVO.

MeXX
 

[dangerous]

Thanks Mexx,
I realise why I was confused,
would it not be that 278@ .050" with 108 LC would be 31/67 inlet (& 67/31 exh) ?

and 288 @ .050" would be 36/72 inl and (72/36 for exh).
This assumes cam is dialed straight up.

So effectively you could just get 720 degrees minus 278 (or 288),

or, 720 minus 180 minus 31(or36) minus 67(or72),
which makes duration 442 for the FK47 and 432 for the web 277 ?

 

[MeXX]

Thanks Mexx,
I realise why I was confused,
would it not be that 278@ .050" with 108 LC would be 31/67 inlet (& 67/31 exh) ?

and 288 @ .050" would be 36/72 inl and (72/36 for exh).
This assumes cam is dialed straight up.

So effectively you could just get 720 degrees minus 278 (or 288),

or, 720 minus 180 minus 31(or36) minus 67(or72),
which makes duration 442 for the FK47 and 432 for the web 277 ?

 

Dear Dave

U r right.
Just mixed up some cam numbers,  Sorry.
Have already corrected them.
THX

MeXX

[fish]

Very interesting topic scientifically......to us normal folk this is as Mexx pointed out; about the gargle of your favourite carburettors and the rasp of your exhaust, but there may be a few things to add.

Does the intake sound wave start to compress back towards the valve when encountering a change in cross sectional area, e.g. the step between manifold and Carb

As the piston does not pull mixture into the combustion chamber, this job is left to Atmospheric pressure which is approx. 1/6th of the pressure that exists when the exhaust valve opens. The piston pushes the exhaust gases out, reaches TDC, suddenly stops while the momentum of exhaust gases continues creating a vacuum behind it. At this point of valve overlap which is crucial for an efficient and powerful combustion engine, the air/fuel mix starts being pulled into the chamber before the piston is moved to make space for the incoming charge and the residual gas is cleared.
 
Exhaust systems primarily aid intake flow by their manipulation of the combustion "sound wave". A sound wave creates a disturbance ahead of it and leaves one behind; such "positive" waves bursting from the exhaust port are followed by negative pressures. When the strongly-positive exhaust wave emerges from the end of a pipe, it leaves behind a negative-pressure tail, which then reflects back toward the port. If the length of the pipe is right, the negative wave will arrive back at the exhaust valve as the piston reaches TDC, further assisting in clearing the combustion chamber.

I have outlined this because I believe that a lot more attention needs to be placed on exhaust tuning in conjunction with intake track length than some people think.

Theres also a lot to be said about intake runner shape, convergence of degrees in diameter, finish, intake port velocity.........the hardest thing here is being able to measure and calculate accurately what air is doing.

Regarding valve timing, I think FK-47 33/65 29/69 straight up.

[dangerous]

^^ that is advanced 2 degrees Ben.(which is how Engle normally likes them installed).

[Jon]

According to David Vizard the exhaust can create 500% more suction than the piston for the intake port.
From what I have heard its normal to use the exhaust tuning to remove the 2,5 harmonic effect, to give a flat curve. (bad for drag racing?)

With regards to tuning of intakes and exhausts, the talk about "when the puls reach open air" makes me think, what is "open" air?
At what point does the puls feel it's there? Our intakes grows steadily from the valve and up, but how much and how fast is the question. Imagine the puls traveling up a cone shape, where will it turn back? And how crisp is that pulse going to be, also will it be stronger or weaker?

This, I would think comes in to play with stingers also...  lenght and shape... and is it a good way to turn the second puls...  or is it just a great way avoid problems with the pulses?

BTW love the thread!

[fish]

This is why I brought up the exhaust pulse because there are so many variables to consider on the intake side and where the pulse starts the rarefaction.
A lot of tests have been conducted into this topic and a hell of a lot written about it.

As per MeXX's post, the optimum rpm power window is somewhat limited due to the fixed intake length and corresponds to the measurement. eg. the length, diameter, volume together with valve timing approximately determines that rpm. Maybe he's fast because he has taken the time to ensure the calculations have given him the point where most power is created.

Charts and calculations can only give us the approximate window not taking in consideration port shape, size, finish, valve to wall proximity, in other words, Unsteady Gas Dynamics are tricky to understand.

As I understand it, please correct me if I'm wrong:
Low RPM - Long and Small Diameter
Mid RPM - Short and Larger Diameter
High RPM - Short And Large Diameter

Also tests involving plenum chambers together with Variable Length Intake Manifolds and Variable Pressure Exhaust systems adopted by today's modern car manufacturers  have proven to be a part of the solution of getting the most power output and efficiency @ much broader RPM range.

[modnrod]

The intake length runs from the valve seat to the end of the trumpet/stack/runner. The tuning length obviously isn't a factor at part-throttle cruise. Small variations in diameter, like slight mismatches at carb-mainfold surface have a slight effect, but very small compared to the stack end.

Intake runner shape and volume considerations affect throttle response, cylinder fill, etc, but the intake length tuning seems independant from this. For example, an intake runner with a volume of 150cc and straight pipe and one with a volume of 250cc and tapered pipe still use the same length for the same tuning point.

Compared to matching things like airflow figures to cam lift/duration specs, and matching intake runner taper to piston speed or bore stroke diameter, simply picking an intake length off a chart and knowing the rev range will "be thereabouts" within 500rpm is easy!

[dangerous]

Here are some interesting things I tried a while back. The dyno numbers told me what worked and what did not:










The difference between the old short stacks and the new long ones was very radical as the dyno sheet shows.
Moving the nozzles to the top smoothed the curve, but the length, and resulting low power peak-rpm did not suit my desired application.





^^ the original curve with the short runners and air cleaners was the dark purple (220.8hp). This had run 11.00 @ 121.46mph at 1785lbs
Changing only the intake length with the wider bell gave me the very wavy curve that peaked around 6200 but then tried to return at 7800.(223hp)
Moving the nozzles up top  gave a curve that was pretty fat, at 6700. (219hp)
Shortening the new runners back to 13" gave me the 219 curve that peaked at 220.8 at 8200, still with nozzles high.
But while this was good at 6200 and 8200 was less where it mattered.(prob too short.)
Moving the nozzles back down lost a bit down low , but was better from 6500 up to have a peak of 224 at about 7800(green).
Leaned it off a tad and it gave the final 225.5hp curve.

[modnrod]

Big changes just from length and injector position. Thanks for the dyno runs Dave.

I love your motors mate! I made the stacks for my old Kwaka from 2" PVC heated to shape in the oven! 

[fish]

Dave never fails to factually inform us, thanks mate for your insight, knowledge and willingness to experiment.

[markvo]

So if you want half the rpm for a street car you need double the length? Maybe need to circumvent? Like from the other side  IDA carb then have it just go straight down at higher RPM!