supercharging and the induction control valve

[longassname]

As some of you know i've been researching supercharging the svx. Current status has me looking at using an ati procharger which does not require plumbing to the engine oil and therefore can be mounted on the lower side of the engine where there is more space and can run off of the front accessory belt without changing the crank pulley. Since the svx fuel injection pressure is maintained by a pressure regulator on a loop with a return line fuel tuning can be done with the simple addition of an fmu.

Now, a new idea has come to my mind and I thought some of our local subaru techs might have some input for me. The svx has an "induction control system" with an "induction solenoid valve between the left and right sides of the intake which the ecu opens and closes to control the "intake air flowrate." From the diagrams in the factory service manual it appears that for some reason which i don't know having this valve between the left and right collectors closed below 4200 rpms results in a higher pressure at the intake valve and higher volumetric efficiency and having it open above 4200 rpms results in higher pressure at the intake valve and higher volumetric efficiency. This is of course under naturally aspirated conditions. I'm hoping one of you subaru techs will know if turbo model subarus such as the wrx still incorporate this valve and if so how does it's function vary?

A second question also occurs to me at this point which i can't believe i've never pondered before. This valve is operated by intake manifold vacum controled by an electric solenoid valve. On forced induction vehicles where do you get vacume?

[oab_au]

Originally posted by longassname
As some of you know i've been researching supercharging the svx. Current status has me looking at using an ati procharger which does not require plumbing to the engine oil and therefore can be mounted on the lower side of the engine where there is more space and can run off of the front accessory belt without changing the crank pulley. Since the svx fuel injection pressure is maintained by a pressure regulator on a loop with a return line fuel tuning can be done with the simple addition of an fmu.

Now, a new idea has come to my mind and I thought some of our local subaru techs might have some input for me. The svx has an "induction control system" with an "induction solenoid valve between the left and right sides of the intake which the ecu opens and closes to control the "intake air flowrate." From the diagrams in the factory service manual it appears that for some reason which i don't know having this valve between the left and right collectors closed below 4200 rpms results in a higher pressure at the intake valve and higher volumetric efficiency and having it open above 4200 rpms results in higher pressure at the intake valve and higher volumetric efficiency. This is of course under naturally aspirated conditions. I'm hoping one of you subaru techs will know if turbo model subarus such as the wrx still incorporate this valve and if so how does it's function vary?

A second question also occurs to me at this point which i can't believe i've never pondered before. This valve is operated by intake manifold vacum controled by an electric solenoid valve. On forced induction vehicles where do you get vacume?

Hi Michael,

Good to see somebody is still looking at the possible. Is that blower a positive displacement or a centrifugal?

Fuel management unit? Depending on the pressure you are planing, a change of regulator to one that allows the manifold pressure to increase the fuel pressure in step with boost pressure, is the easiest way to maintain the fuel rise. If you are going for a boost over one atmosphere you may have to go to the piggyback management unit.

The principals of the IRIS are contained in the fact that the two halves of the engine are separated, so that the 3 cylinders on one side induct every 240 deg, thats why it is important to keep the duration of the intake opening down to 240deg. 240 deg X 3 cyl = 720 deg.

As the No.1 cylinder accelerates the airflow down the long track, the kinetic energy developed, wants to keep the flow going. When the first cylinder starts to close its inlet valve on this energy, the pressure at the valve end of the manifold will rise to above atmospheric, due to the kinetic energy. When this valve closes the next valve opens to a high pressure that starts flowing into that cylinder. This way the inlet airflow is not stopped at the end of one induction then started by the next it is a constant flow.

As the engine speed increases above 4000, the flow starts to lag behind. Thats when the butterfly valve between the two inlet tracks opens. The system then changes to a resonate system. The branched area of inlet manifold, with the butterfly valve open now forms a plenum chamber that terminates the resonate pressure wave, at that point. So that we now have a short, tuned length track, of about 14" that will resonate at about 5200.

The IRIS is not a new idea, though no one has used it so effectively as Subaru has. Chrysler used a form of it in the early 70s on the slant six and V8s. long inlet track, on the V8s the four-barrel carbs mounted over the opposite head. They called it "Ram charging". It only added positive pressure to the end of the inlet phase.

To my knowledge Subaru is the only one to have it add positive pressure to the start and end of the inlet phase. You could only do it to a 3, 6, and 12, cylinder engine. As the inlet has to be arranged to have one inlet valve closing as the next opens.

Subaru has designed this engine to operate as two 3-cylinder units. One cylinder inducting every 240deg. so with the inlet duration being 236deg, one valve is closed for 4deg before the next valve opens.
This short closing phase causes the rise in pressure in the inlet track, from the inertia of the air flow, to produce a positive pressure of about 4 to 5 PSI.
When the next inlet valve opens this pressurised air flows in like it was supercharged to the tune of 4 to 5 PSI. It's inlet phase accelerating the flow for the next cylinder.

The IRIS will operate the same blown or not. If a Turbo is used it will provide the bottom end torque that the turbo lacks. The manifolding won't cause restrictions or problems when on boost. It will still work the same. Don't take it off or it will die in the ass at less than 3000 rpm.

The four cylinder engines would have to have an inlet duration of 360 deg to arrange the same system. The WRX just uses a long inlet tract to provide some low speed torque to fill in the bottom end. They just rely on the boost to provide the top end without any resonance.

The source of vacuum is always present at the throttle butterfly, it doesn’t matter wether it is blown or not, When you lift the foot, vacuum is present in the manifold. If it’s a turbo there is vacuum at the inlet side of the impeller.

All the best, keep at it, it is possible.

Harvey.

[longassname]

Thanks Harvey,

Informative as usual. Ati's procharger is a centrifugal supercharger. A very high quality, high endurance, reliable unit at that. That's their basic claim to fame in the forced induction business. An fmu is just the type of thing you were describing, they plumb in after the stock pressure regulator and restrict the flow to the return line thus increasing fuel pressure based on manifold pressure (boost). Pretty much every supercharger company out there makes a line of them as they are by far the prefered method of fuel tuning and used on any car that has a return loop rather than computer regulated pressure. They come in varying base ratios depending on the pressure gain needed for your application and are somewhat variable--usually by about 4 pounds. The svx application will probably end up taking an 8 to 1 fmu which means 8 psi of fuel pressure is added for every one pound of boost at it's base rate. A screw on the fmu allows you to adjust down the ratio to as low as 4 to 1. You basically start with an fmu that will comfortably provide enough fuel to give the amount of horse power expected from your engine under the amount of boost you expect to generate, put the car on a dyno when you install the system with the fmu turned all the way up so you are running rich, and then turn the fmu down incrementally until you are running at the stoichiometric ratio.

I will only be running 5 lbs of boost but it will be intercooled so hp gain will be extreme on our 10 to 1 compression engine. The gain will be well over 100 hp but how much over is hard to say. At 10 to 1 compression and even a modest 5lbs boost running significantly lean will result in detonation so I'm kind of on the fence about starting with an 8 to 1 or 10 to 1 fmu. Chances are either one will work fine but no knowing for sure till on the dyno. The good thing about blowing a 10 to 1 compression engine is there will be pretty much no lag.

[michael]

I appologize if I sound like I'm talking out my ass, but I don't know much about this topic. A supercharger applies a boost of cold air to the air intake using a belt driven compressor correct? What type of piping is used to deliver this compressed air and what area is to be pressureized/before or after throttle? I was wondering if you could replace your ac compressor with a supercharger also how is the boost activated gradually or what, just curious?

[Subafreak]

Sound like a good idea. I keep going back and forth over supercharging or turbo charging. Where do you intend to mount it and witch pully are you gona run it off of? I was looking at mounting one underneath too, but it's just gona be tight anywhere under the hood.

[longassname]

The only dumb question is a question unasked.

A centrifugal supercharger which is what we are talking about here is the exact same as a turbo except it is driven by a belt instead of by exhaust pressure. It's a fan that blows air into the intake thus creating pressure in the manifold and forcing more air into the combustion chambers. This is called boost. Boost increases as rpm increases. Any type of forced induction actually increase the temperature of the air charge although that is not a desired effect...it just can't be avoided. According to boyles gas law when you compress air you create heat. The end result is that while the air charge is hotter because it is compressed there are still more oxygen atoms forced into the combustion chamber and therefore more fuel can be burned yielding more horsepower. To counter the temperature increase of the aircharge created by the supercharger a thing called an aftercooler is comonly used. An after cooler is basically a radiator that the air charge is blown through to cool it. Air to air aftercoolers are the most efficient. An air to air aftercooler is placed in the front grill of a car so that the fans pulling air accross the radiator also pull air across the aftercooler. The aircharge from the supercharger must be plumbed up to aftercoller (yes through a large 3 inch diameter tube) and out of the aftercooler to the intake. In the case of the svx space up there is very tight but it looks like it can be done. For the 3 inch tubing used in a custom job as it will be for the svx you commonly use mandrel bent 3 inch steel tube from a good muffler shot (only good muffler shops do mandrel bending by the way). Ready to install kits from big companies usually use fancy polished aluminum tubing so you don't feel as bad about dropping 5 grand. Connections between tubes and tubes and tubes and components are made using silicone tubing commonly available in both straight and elbow shapes and hose clamps. In most supercharger installations the tubing is run above the motor as that's where the supercharger is mounted. There's no space for this in the svx's case but there does apear to be space beneath the engine and there is a good sized space to come up in the back side of the passenger side of the engine compartment with conveniently enough leads right to the plastic intack hose coming from the mass air meter and going to the twin throttle bodies. It is my intent to cut out a section of that plastic hose so that i can plum from the mass air meter to the supercharger and then from the after cooler to the remaining section of the "hose" leading to the trottle bodies. This seems the sensible thing to do as that "hose" has numerous tubes plumbed into it for the egr system and is molded to fit around both throttle bodies.

[longassname]

Just to clarify, the induction control system Harvey and I were discussing is not a forced induction sytem and doesn't have anything to do with superchargin. It's a valve between the left and right sections of the upper intake on the svx. The physics behind it's function were unclear to me until Harvey explained them and i was asking because i thought they might have been effected or rendered null under high manifold pressure. To explain it in more simple terms than he was using for my benefit...it doesn't create manifold pressure, it momentarily creates pressure at the intake valve right at the time of opening to maintain better flow. It doesn't not pressurize air in the combustion chamber.

Are you seriously considering bolting a centrifugal unit on your SVX? Let me know if I can help, I helped build the only centrifugally superchaged 2.5RS in the US. We had excellent results, though it was loud and obnoxious in the extreme. One problem is the constant venting that you get from the blowoff valve when you come off the throttle... since centrifugal units never stop boosting, that air has to go somewhere. Care should be taken in the selection of the blowoff (dump) valve, make sure you go with an extremely reliable unit as it will be working harder than on a comparable turbo application.

[oab_au]

Originally posted by longassname
Thanks Harvey,

Informative as usual. Ati's procharger is a centrifugal supercharger. A very high quality, high endurance, reliable unit at that. That's their basic claim to fame in the forced induction business. An fmu is just the type of thing you were describing, they plumb in after the stock pressure regulator and restrict the flow to the return line thus increasing fuel pressure based on manifold pressure (boost). Pretty much every supercharger company out there makes a line of them as they are by far the prefered method of fuel tuning and used on any car that has a return loop rather than computer regulated pressure. They come in varying base ratios depending on the pressure gain needed for your application and are somewhat variable--usually by about 4 pounds. The svx application will probably end up taking an 8 to 1 fmu which means 8 psi of fuel pressure is added for every one pound of boost at it's base rate. A screw on the fmu allows you to adjust down the ratio to as low as 4 to 1. You basically start with an fmu that will comfortably provide enough fuel to give the amount of horse power expected from your engine under the amount of boost you expect to generate, put the car on a dyno when you install the system with the fmu turned all the way up so you are running rich, and then turn the fmu down incrementally until you are running at the stoichiometric ratio.

I will only be running 5 lbs of boost but it will be intercooled so hp gain will be extreme on our 10 to 1 compression engine. The gain will be well over 100 hp but how much over is hard to say. At 10 to 1 compression and even a modest 5lbs boost running significantly lean will result in detonation so I'm kind of on the fence about starting with an 8 to 1 or 10 to 1 fmu. Chances are either one will work fine but no knowing for sure till on the dyno. The good thing about blowing a 10 to 1 compression engine is there will be pretty much no lag.

OK, using an centrifugal, you won't have much boost at low RPMs like, 5psi at 1000 rpm will cause more detonation than 5 psi at 2000 rpm, time is the factor. So as the engine speed goes up the boost pressure can also go up, with less chance of detonation.

I think an intercooler is a bit over kill , at that low a boost the charge will not have much heat from the blower's compression and without a hot turbine driving it, the charge will not heat up very much at all. You also have to consider the length of time that the engine will be on full boost, 30 sec? then off the throttle and the system cools down again. Not quite the same as oval circuit racing where the boost is maintained for extended periods.

I find the description of the FMU a bit strange. You say it adds 8psi of fuel pressure for every 1psi of boost ?.

As you would know the regulator that is standard, alters the fuel pressure with manifold pressure. Taking it from about 22psi at idle to 36 psi at atmospheric pressure at full throttle. 8psi for each 1psi will force 8 times the amount of fuel out the injector?

May be I just don't understand the set up, I am sure you do, so no worries. 5 psi is a 34% increase in pressure will yield an extra 81 hp. in theory.

Harvey.

[longassname]

Originally posted by oab_au


OK, using an centrifugal, you won't have much boost at low RPMs like, 5psi at 1000 rpm will cause more detonation than 5 psi at 2000 rpm, time is the factor. So as the engine speed goes up the boost pressure can also go up, with less chance of detonation.

I think an intercooler is a bit over kill , at that low a boost the charge will not have much heat from the blower's compression and without a hot turbine driving it, the charge will not heat up very much at all. You also have to consider the length of time that the engine will be on full boost, 30 sec? then off the throttle and the system cools down again. Not quite the same as oval circuit racing where the boost is maintained for extended periods.

I find the description of the FMU a bit strange. You say it adds 8psi of fuel pressure for every 1psi of boost ?.

As you would know the regulator that is standard, alters the fuel pressure with manifold pressure. Taking it from about 22psi at idle to 36 psi at atmospheric pressure at full throttle. 8psi for each 1psi will force 8 times the amount of fuel out the injector?

May be I just don't understand the set up, I am sure you do, so no worries. 5 psi is a 34% increase in pressure will yield an extra 81 hp. in theory.

Harvey.

Well let's see. The complete combined gas law is P1V1/n1T1=P2V2/n2T2. Now since i don't know how much of the pressure increase is due to the increase in moles and how much is do to the increas in temperate since they are happening at the same time, I can't really figure out how much temperature increase there is due to gas laws. But assuming ati's #'s are correct there is a significant temperature increas at 8 psi...115 degrees, 71 from compression, 24 from inefficency of compressor, 20 from heat soak. Not worrying about being possibly grossly inaccurate if we say 44 for compressing to 5 psi then add the 24 and 20 that gives us an aircharge temp encrease of 88 degrees. That number could of course be way off not to mention it's based on accepting ati's numbers but if it's right....

I also noted that an aftercooler has the nice added effect of acting like a damper on high boost allowing the use of a more aggressive pulley which results in spooling up faster.

The aftercooler is admitedly a great added expense and pain to fit in. I'd love not to have to put it in but i'm thinking the results will be significantly effected. Hard work and spending my hard earned money aren't really my goals It also of course is necessary in order to leave the ability to creap up the boost until i get as much hp out of the motor as i dare. I can see myself now...some day down the road at the dyno with a bunch of pulleys and monitor looking for knock sensor readings.

It's my understanding that the factory fuel pressure regulator maintains fuel pressure at 36.3 psi, no? Damn service manual didn't say anything about 22 psi. Either way fmu will work the same. 8 psi added fuel pressure for each 1 psi of boost...adjustable down to 4 psi of added fuel pressure for each 1 psi of boost. The supercharger companies all have tables for different rate fuel injectors at different stock pressures that show how much fuel is added, supporting how much more horsepower for how much added fuel pressure. That's one thing you don't have to figure out on your own. If you know the horsepower gain you expect you know how much fuel pressure to ad for your particular injector set.

Now let's work this whole hp thing out. I haven't done that yet..doing now as i type. At sea level pressure of 14.7 psi your 34 % increase is dead on for n at 5psi boost assuming stock combustion chamber pressure at bdc is 14.7 and combustion chamber pressure with 5 lbs boost is 19.7 with no increase in aircharge temp. Of course there's the increase in aircharge temp I grossly calculated above so gain should be less...but then T in gas laws is in kelvin so 20 or 30 degrees farenheit after aftercooling isn't much of a performance detriment at all. Now this brings me to the real question in my mind. How off is the assumption that in the naturally aspirated motor a full atmosphere of pressure collects in the combustion chamber at bdc? If it only develops say 90 % of atmospheric pressure that's 13.23 psi which give us a much more promissing 49% gain of 113 hp. This is a serious question in my mind as I'm going to choose the boost i start with to get me in the 3 digit gained range. Next we should probably discuss combustion efficiency due to increased volatility under pressure and wether mechanical efficiency maintains a constant

[jsvxstyle]

just get a set of turbo cams, from one of my pal in l.a. with valve spring, as the intake valve on the eg33 can handle extreme heat. with no worryies, to an exstent. get the injecter slightly bigger like i have, and then youll feel the mid range and top end come in stronger with impressive accelaration, for the N.A guyz i discovered a way to advance ignition timing threw the use of the signal wire from the tps senor and the signal wire from the MAS flow sensor. after trying the device on the 2000 acura tl 3.2 with auto trans, it was impressive how quickly the car downshifted, and had a slightly harder pull from the start. its cost about 340 bucks, its a signal conditioner they have one for the svx, but never listed it at all, found that out threw the production department of venom and talking with the tech who built them.
i must get one for that price it aint bad, and see how it works on my svx

A centrifugal unit should have enough adiabatic efficiency to negate the benefit of an intercooler at anything under about 8psi.. I know that's the case with the Rotrex unit that we were using, your mileage may vary on another type. A simple top mount intercooler would give the best results in terms of pressure drop if you are creating enough heat to require one, one off an early Legacy Turbo may be sized appropriately to fit behind/above the throttle bodies. The new narrow intercoolers on the Forester Turbos may also fit back there, it will require some modification but I'm sure it's feasible. They're long and thin, it seems like they'd be perfect for this application.

Beware the beguiling force that is theory when discussing boost vs. RPM with a centrifugal. We found that in practice the boost builds up enough against the back of the throttle plate to give an almost instantaneous vertical torque "curve" when you come back on the throttle at any given RPM. The blowoff valve used, how much air it can flow continuously, and the type of valve that is used internally are all critical considerations when you think about how much boost will bottleneck as a result of its flow characteristics. Ours actually heated itself to the point where it began sticking closed, which is how the motor revved itself to 7000rpm with almost no throttle input, fuel cut, and blew #3 and #4. Care should be taken, and I'd recommend against the use of a lightweight flywheel. The more mass you have to slow down the boost "hammer" when the throttle plate opens, the safer you will be with regards to unexpected boost events such as the one we encountered. We were running a 9lb billet Fidanza unit and while amazing on an NA car, it was actually TOO responsive for anything but an absolutely pure race vehicle when used in a forced induction application.

Best of luck!

[longassname]

Originally posted by jsvxstyle
just get a set of turbo cams, from one of my pal in l.a. with valve spring, as the intake valve on the eg33 can handle extreme heat. with no worryies, to an exstent. get the injecter slightly bigger like i have, and then youll feel the mid range and top end come in stronger with impressive accelaration, for the N.A guyz i discovered a way to advance ignition timing threw the use of the signal wire from the tps senor and the signal wire from the MAS flow sensor. after trying the device on the 2000 acura tl 3.2 with auto trans, it was impressive how quickly the car downshifted, and had a slightly harder pull from the start. its cost about 340 bucks, its a signal conditioner they have one for the svx, but never listed it at all, found that out threw the production department of venom and talking with the tech who built them.
i must get one for that price it aint bad, and see how it works on my svx

I appreciate the interest in discussion but if we could i'd like to keep this one thread on supercharging limited to the technical. The treads on supercharging in the past have been hijacked to talk about every other idea in the world for hp and how much everyone wants it I'd like to keep this one the way it's going

[longassname]

Originally posted by Porter
A centrifugal unit should have enough adiabatic efficiency to negate the benefit of an intercooler at anything under about 8psi.. I know that's the case with the Rotrex unit that we were using, your mileage may vary on another type. A simple top mount intercooler would give the best results in terms of pressure drop if you are creating enough heat to require one, one off an early Legacy Turbo may be sized appropriately to fit behind/above the throttle bodies. The new narrow intercoolers on the Forester Turbos may also fit back there, it will require some modification but I'm sure it's feasible. They're long and thin, it seems like they'd be perfect for this application.

Beware the beguiling force that is theory when discussing boost vs. RPM with a centrifugal. We found that in practice the boost builds up enough against the back of the throttle plate to give an almost instantaneous vertical torque "curve" when you come back on the throttle at any given RPM. The blowoff valve used, how much air it can flow continuously, and the type of valve that is used internally are all critical considerations when you think about how much boost will bottleneck as a result of its flow characteristics. Ours actually heated itself to the point where it began sticking closed, which is how the motor revved itself to 7000rpm with almost no throttle input, fuel cut, and blew #3 and #4. Care should be taken, and I'd recommend against the use of a lightweight flywheel. The more mass you have to slow down the boost "hammer" when the throttle plate opens, the safer you will be with regards to unexpected boost events such as the one we encountered. We were running a 9lb billet Fidanza unit and while amazing on an NA car, it was actually TOO responsive for anything but an absolutely pure race vehicle when used in a forced induction application.

Best of luck!

If you'll note in the example given above abiatic efficiency is only responsible for 24 degrees f of the 115 f temperature increase of the aircharge at 8 psi. Sure it's nice to do every little thing you can to decrease temp increase like have a higher efficiency compressor...but I think it's effect is one of the smaller variables in the temp increase equation.

As far as bypass valves since i'll be custom fabricating the piping sure why not go overboard and use a vortech race bypass valve. Sounds like a good idea to me. It even comes in pretty colors.

No flywheel for me. Going to stick on the svx should actually mean less hp to the ground. I know the svx transmission gets ragged on a lot around here but i'm not one of those doing it. I keep my ragging on it to on the street. People seem to forget that when they go to manual they are going to 50/50 awd and will have less hp on the ground than with a 90/10 automatic that locks up............Now before everyone jumps on this idea, do it in another thread please

[oab_au]

You say that you are going to mount the blower in this way:
aircleaner, MAFsensor, blower, blow off valve, intercooler, throttle, engine.

Why this way, it is going to present a lot of problems. Can't the blower handle vacuum.?

It would be better mounted this way:
aircleaner, MAF sensor, throttle, blower, intercooler, engine.

This way you don't need the blowoff valve, which will just blow off air that has been measured by the MAF for injection and now has been lost.

Also when the throttle is closed the blower is under vaccum, so it does not pump air. The other way it will pump air, produce heat and use hp all the time.

If the blower does not have seals good enought to handle vacuum, I would modify it so it does or find another unit that will, as it will save a heap of problems.

Harvey.