Originally Posted by A400

Do a test with a 2 L and please port it or at least do a extrude hone. I run a hogged out stock big port with the internals restrictions gutted out. I am now running a 2.1L w/a GT30R and would be interested if tests such as I suggested were done and the results were good. The stock (as delivered) #'s wouldnt mean much to me, but if maximized setup indicated large gains I would. Thanks for going to the trouble of making something for the extreme setups - hope it shows large gains when maximized.
Rod

Originally Posted by Badger-5 Email Correspondance

From: JNL Racing <jnlracing@googlemail.com>
Subject: SEM Manifold test results
To: "BILL BROCKBANK" <bill.brockbank@btopenworld.com>
Date: Wednesday, 15 October, 2008, 12:21 AM


Overview

Mathematically the internal runner size on each runner has an identical cross sectional area to the intake mouth of a big port head if we make the assumption it is a perfect rectangle. Obviously we are in a scenario where in real life the intake has radiused sides so there is mathematically scope to open the mouth of the cylinderhead port further than the standard port mouth size without loss of flow. Further experimentation would have to be done to actually work out what the threshold is and as to how much effect the port design has on it.

Testing Method

The flow results of each individual runner Number 1 being closest to the throttle body which in this case cyl number 4 closest to the gearbox. All flow figures have been pulled at 10" on a digital bench calibrated to a superflow 110 with allowances for atmospheric conditions whilst tested. A calculated equivalent using a 1.673 correction factor has also been made for a 28" test pressure result for comparison with existing results from other sources. From experience time and time again the mathematical 10" to 28" conversion actually yields results 1-1.5% lower than an actual tested 28" at flow figures between 250-300cfm. On the individual runners 3 seperate pulls where made at different bench operating temperatures (heat build up over time from extended running of the motors) and an average taken from the results.

CFM Report

Percentage variation between each of the individual runners is a maximum of 1.54% between the highest and lowest. An acceptable variation for high performance mapping purposes when using a single lambda probe to determine the average of 4 cylinders is considered to be 3-4% deviation from minimum to maximum. So the manifold falls well below this safety range.

FPS Report

At 10" test pressure port/runner velocity is considered at choke at 209fps which equates to 350fps at 28". The runner velocities where measured as central as possible to the runner at the cylinderhead flange exit/entry point. Higher velocities where recorded towards the sides however it was difficult to record a constant across the 4 runners for comparison so it was decided to take the lowest velocity found at the centre of the runner as the comparison point. As can be seen all runner velocities far exceed choke point which within a cylinderhead port would be considered detrimental to the power output of the engine (similar to running too much timing advance). However such characteristics in a manifold can be deemed beneficial in improving throttle response at certain points in the rev range.

SEM Manifold Results

Test Pitot SCFM
Intake Pressure FPS
1 10 269 162.3
2 10 282 164.5
3 10 287 164.8
4 10 288 162.4

Test Pitot SCFM
Intake Pressure FPS
1 28 450 271.6
2 28 472 275.3
3 28 480 275.8
4 28 482 271.7

Flow test involving a mildly ported testhead
As an experiment we have also taken a mildly ported large port head with standard port mouth sizes to mate up to the SEM manifold. (Again flowed at a 10" test pressure). As you can see although there is a drop it is considered quite minor in the grant scheme of things averaging between 2.5-3% drop across the complete lift range.

The most surprising thing here though is that although runner 1 flows physically worse than runner 3 it only exhibits a 2.5% drop across the range against the 2.9% drop of runner 3 which will be down to the slower runner velocities that were recorded. Which goes to show that runner flow without the equation of the head can be quite deceptive and should only be taken as a guideline as to how well they are matched to each other not of their final ability. I also shows that port/runner velocity plays an important part in final combinations.

Common to both runners tested on the head; when measuring port/runner velocity at the port mouth of the head with the intake bolted on the port velocity became approximatelly the average of the two items when tested individually.
For example runner one measured 269fps on it's own and port 1 measured 227fps on it's own. Once the intake was bolted onto the head the port/runner velocity averaged out at 245fps. (All three velocity measurements taken as central to the runner/port as possible).


Final Overview

The SEM manifold for use on a standard head application will potentially see small gains in tq and hp in the mid to top range over a standard manifold due to the high runner velocities, providing it is correctly matched with the right throttle body (not too large) there will be improved throttle response over a standard manifold. There may be a small sacrifice of power in the low rev range as it will take extra time to fill the larger plenumb chamber however this would have to be tested on a dyno to verify.
For big power/high rpm application it would be considered a highly desirable manifold to have specifically because of the large plenumb chamber giving a large storage capacity of boost pressure for the cylinders to draw from. This means less of a pressure drop during gear changes and would be ideally suited to track orientated vehicles whether drag or circuit. For drag race/big turbo - high boost application it would be advised to open the ports up to slow the runner velocities down accordingly, whereas circuit cars with smaller turbo's and lower boost settings will benefit from the higher velocities for out of the corner throttle response on the standard port/runner size. Comparing the figures of the SEM manifold against existing test results floating around the net not only does it outflow the best manifolds out there at the moment but it is also the most consistant across all four runners even outdoing my previously favoured RMR manifold which has a
2.8% variation across the runners.

Courtesy of http://www.JNLRacing.com

Feel free to recirculate as desired, I've also attached a few dodgy pics of the phone of the setup. Unfortunatelly my camera battery went flat and didn't have a charger on me. At one point I'll do the AGU intake aswell as the AEB to compare it to. However I've just hit a massive back log as I'm taking delivery of another 4 jobs tomorrow so probably wont get onto it for at least another 2-3 weeks. Will get the manifold back in the post tomorrow by Special delivery as the couriers wont insure it. If you can also get me a price on a big port one for Karl as he's going to be needing one. I'll have to confirm throttle body choice before order though as we dont want it too big. For some reason the VAG turbo boys in America haven't realised yet that a forced induction engine favours a smaller throttle body than naturally aspirated. Same reason if you look at a small block chevy with a supercharger and a blow through carb setup they make the carb for
supercharger use out of a 750cfm body whereas a naturally aspirated equivalent engine would run either an 850 or 950cfm carb depending on cam and head spec.

--
Kind regards

Jean-Paul

http://www.JNLRacing.com