nice post

pimpsmurf wrote:

Using an intake cam in place of the exhaust cam has been a fascination on the board recently. I happen to have an extra intake cam here (BIG THANKS TO Sorority Demon for that!!!) and decided to use it while building my engine in the next 2 weeks. Given the little information released (basically teasers at this point, IMO) I have decided to do the math and provide information that might help someone, myself included.

Here is a picture from the FSM of the stock Cam profiles:

http://myspecv.com/fs/pimpsmurf/Spec-V% … ofiles.jpg

Wow those are pretty lines! WTF do they mean?
a: 224 = exhaust cam duration
b: 244 = intake cam duration
c: 0 = TDC - the intake opens at TDC and they are just showing that by having a gap labeled 0 degrees
d: 64 = degrees after the cylinder hits BDC before the intake closes all the way.
e: 3 = degrees after TDC when the exhaust closes as the intake charge is begining to fill the cylinder. This is also the overlap of the cams.
f: 41 = degrees before BDC where the exhaust cam opens and begins flying out of the exhaust.

TDC = top dead center
BDC = bottom dead center
overlap = degrees of engine rotation when the intake and exhaust cams are both open.

Another factor that has been struggled with on the board is the fact that the alignment down on the cam, which fits into the cam gears is not in the same place on the two cams. This is unfortunate as it's actually off by a partial gear "tooth". This means recentering the 2nd intake cam in the exhaust port isn't possible with a stock gear. Tonight we will discover if it's possible with the one piece of technology WRP made which is actually worth anything at all will help us accomplish this feat.

As many of you know, the exhaust cam sprocket Will had made was rekeyed to offer +2 +4, -2, and -4 degree exhaust cam timing changes. One tooth is 8 degrees (roughly?) so this allows you to step in 2 degree increments instead of 8 degrees (by moving the alignment 1 tooth off.)

moving the cam 1 tooth = 16* crank.  since cams spin twice for every one time the crank spins

8 degrees in either direction is silly with stock cams, and can be dangerous with two intake cams as has been demonstrated. (12 degrees or so off actually caused the header to glow red hot and overheated the narrow band o2 sensor! haha)

Now to the fun part: geometry FTW!

The center of the exhaust cam is 248 degrees I believe, and the duration is 224
The center of the intake cam is 122 degrees, and the duration is 244 degrees

Now comes the more tricky part.

Overlap is a delicate thing, which is why companies like JWT and BC are in business. It is somewhat skill and knowledge, and to another extent, partially trial and error. In an engine with a high volume turbo such as a SC61 or GT35R, overlap isn't going to hurt you nearly as much as you are able to move more air with a lower pressure, so the intake pressure doesn't vastly overpower the exhaust pressure (typically well under 15psi, best I can tell) and cause your unburnt fuel and air to spray out of the exhaust. At higher RPMS, this isn't as big of a deal because 3 degrees of overlap is only affective .41% of the time, and at 7000rpms, the overlap is roughly .000035 seconds long. hahaha

I suppose this is where the aftermarket companies make their money! The problem with caming our car is a slight loss in low end torque. The reason for this is because at low RPMs, the overlap lasts for long enough to actually allow too much of the hot exhaust gasses to be pulled back into the cylinder causing less air to be drawn through the intake. The gain from the cams is both in duration, and base angle, as well as the overlap. The higher overlap of JWT cams gives just enough time at the higher RPMS (around 4000+?) for the overlap to actually do something. It is long enough for the vacuum of the exhaust gases rushing out to suck some intake air with it. Tuning this is a long task, which is why we are seeing BC cams take so long to come out (with such great dyno numbers) and they are still working on stage 3 cams which should be a great race-level cam for our car.

In order to add to the complexity of this issue, our beloved QR25DE has a system called CVVT. This, if I am correct, is a solenoid hooked to the ECU, and when the ECU tells the solenoid to open, oil pressure is allowed into a clutch-type device on the intake cam sprocket (which is heavy as fuck BTW) and causes the intake cam to change it's base angle (but not the lift, like vtek.) I'm not sure, but I would think that it simply allows you to add 15-30 degrees of movement on the intake side, adding more overlap. I think the purpose of the electrical component is to turn the system on and off, and the oil pressure (which rises with rpm) is what increases the amount of overlap. Important shit. I wish someone would point me to the right place in the FSM for more definitive information because I am still searching and can't find it.

the CVVT is not that simple.  The electronic component is a PWM solenoid that controls how much oil it actually sees.  Theres a target table in the ecu somewhere for it.

Back! Back I say! No more uneducated guesses at internal combustion theory for you! Back to the math!

We run into a problem with the math at this point. I do not know exactly where the center of the cam lobes are relative to the teeth on the sprockets. I am going to eye ball this as best I can, with my very good eyeballs. (This is where the technical becomes voodoo! When someone starts guessing angles!)

A couple pieces of terminology: advancing (+2-4) the exhaust cam timing causes the exhaust valves to close sooner, which DECREASES OVERLAP. Retarding the timing increases overlap. Many get this confused with ignition timing, where retarded timing is "better for boost." This is "base" overlap because the CVVT, unless modified, if I am correct, adds more overlap as oil pressure (read: rpm) increases to get some of that "free" power. highly modified engines vary in setup and some will want more, or less overlap than others. My individual setup (a t3t04E-BB-57-60 .63ar turbonetics turbo) requires much more PSI to move the same amount of air as a bigger or more efficient turbo would. Many people have read (over and over) that you do not want *any* overlap on a boosted engine. This isn't true, but with my setup, no overlap at all is better than too much as I will be throwing off my AFRs sending extra air/fuel mixture out of the exhaust and hurting power. Moving timing around is a two edged sword however. When you advance the exhaust cam timing to decrease overlap, you also cause the exhaust to open sooner on the power stroke. This is how you can fudge up and send still-burning air/fuel out your exhaust and maybe even break something expensive. Moving a stock cam around with a modified sprocket is child's play. You can simply go to +2 or +4 for boost, or -2 or -4 for N/A for more overlap. Moving around an aftermarket cam with a longer duration is more difficult.

Now I think it's time for more math and MORE PICTURES!
Counting from the marker on the exhaust cam sprocket, counterclockwise against engine rotation to the centerline of the cam lobe is:
6.5 teeth to center on intake cam
1 tooth to center on exhaust cam

There are 44 teeth on the cam sprockets, so 360/44=8.181818 degrees per tooth. We will just say 8 since we can't always be sure how CVVT is going to munge up our math!

So, we would need to count 5.5 teeth over to get the centerlines back together. The problem is this. The cam has a longer duration (by 20 freaking degrees!) so if the centerlines were lined up (ie, 5 teeth and a -4 setting on the sprocket) then you are still adding 10 more degrees of overlap to the 3 you had to begin with! FUCK that is a lot. 6 teeth is even worse making a total of 17 degrees of overlap, not to mention the fact that the exhaust valves are opening earlier too! roughly 4 degrees earlier on the 6th tooth over.

So, lets assume that in my high psi setup, I am going to want to reduce overlap (ideally to the point that there is just enough time for the intake charge to PUSH the exhaust out of the cylinder without loosing any boost) and lets assume that we want the exhaust port to open as late as possible to allow most of our boost-and-meth-fortified intake charge to burn. (NOTE: Even though most of the power is created earlier in the power stroke, only with very high octane fuels could we delay the process enough to make a difference, which is why many high end drag cars run 100% methanol as a fuel source.)

We could go so far as to eliminate all overlap by using the 4th tooth. This brings us to 1 degree of base overlap and DAMN! 19 degrees later than normal exhaust valve opening. This is interesting, and may be the best route for me. Having already purchased a WRP cam sprocket on a whim, I'm going to experiment with different settings. You see, the double edged sword of the increased exhaust duration is that you need to open soon enough to allow the engine to spool the turbo quickly, but also not loose too much boost from overlap. This is tuning, and this is why the duel-intake-cam idea isn't perfect. It's no BC cam, and that is for certain.

youll have 19 degrees earlier than normal valve opening.  right now that cam your describing on the 4th tooth has only 203* of duration

Now for more pictures!
http://big_j.myspecv.com/sprocket.jpg
Timing to 6 degrees is NOT OK. I explained above how the cam timing moves around when you change to different teeth.

http://myspecv.com/fs/pimpsmurf/sprocket-timing.jpg
Here is a WRP modified exhaust cam sprocket. The black numbers are the 4th and 5th tooth I have been refering to, and the red labels are the degrees of exhaust cam timing modification.

Now to wrap things up, in my mind anyway...

I believe that I am going to start with the 4th tooth over with +2 on the WRP sprocket. This is going to eliminate the base overlap and add a 1 degree buffer of resistance to the CVVT's effects, hopefully preventing boost leak in the higher rpms. This *WILL* have the negative side effect of loosing low end torque, probably a lot of low end torque. With my butterflies removed, I have no torque until 3krpms anyway, but I believe I will reinstall the butterflies and configure an output on megasquirt to enable the butterflies at 5kpa or so. Maybe this will return some of my low end power for street driving and perhaps give me that intake scream I used to love so much. I may end up eliminating the WRP sprocket and going directly to 4 teeth on stock cam sprocket timing to get 1 degree of overlap back at low load/rpms.

open the butterflys at 100kpa, no point in having them on before you go into boost

There is a whole lot of assumption in this post, and I welcome anyone to post information or questions. Hopefully we can all get to the bottom of this together.

-JNY

Last edited by darsondriss (2008-09-30 10:00:51)

i know you were talking about the intake cam in the exhaust side but what you described is wrong.  you cant have a the valve opening later and closing earlier and pick up duration.  the setup on 4 teeth would have the valve opening 17* earlier.

no not the whole cam in the exhaust sprocket just leave that alone...

say you have a cam that closes 3*  and opens at 200*
this is 163* of duration
then say you swap in the cam that closes 2* earlier and opens 17* later than the old cam.  that gives it closing at 1* and opening at 217*
this is 143* of duration

what you the cam is going to do is close 2* earlier and open 17* earlier.

why is each tooth 12*?  and idk if cvvt learns or not...  but everything seems to with the ecu so idk

i did 16* when i switch to crank angle.  but the other 2 are relative to the old cam not the crank.

dyno time