I am going to steal this write-up from miataturbo.net so all rights are to them, but this should help out..
The topic of hooking up manual boost controllers may prove to be confusing to some, or maybe somebody has bad boost sag on the high end of their kpa logs? I figured this could be useful to all. Specifically, maybe some folks seem to be having boost curves that look like this, and perceive it as justification for the fact that their manual boost controller just isn't cutting it, and they need to go out and spend money on a fancy electronic boost controller.
So, does that look familiar? If so, save your money. Fixing this problem is going to cost you about $3.
First, let's take a look at why this is happening. Here's a simplified diagram of your turbo system, where we have the turbocharger itself, then the intercooler, and then the throttle body.
Point "A" in this system is where a lot of folks have their boost controller connected. It's that nipple that came from the factory on the side of your compressor housing, probably with a hose already attached between it and the wastegate actuator.
Well, that's just stupid.
What's happening here is that your boost controller is in fact maintaining a constant level of boost, however it's doing it in the wrong place. Specifically, it's maintaining a constant level of boost at the compressor, but that's not what your engine is actually seeing.
Confused?
Yes, behold the simple drinking straw. Solver of great mysteries.
Here's a quick experiment. Stick a drinking straw into your pie-hole, and blow through it. Not too hard, very gently in fact. Very, very gently. This ain't Hustler's mom we're dealing with.
Feel the resistance that the straw is offering? No? Of course not. At the rate at which you are blowing into it, the straw is not much of a restriction at all.
Now, blow harder. And now you start to feel the straw fighting you.
A funny thing happens when we try to flow a gas through a restrictive orifice. The more we try to flow through the restriction, the more restrictive it becomes. In practical terms, at low rates of flow, we get very little pressure loss across the restriction. As flow increases, so does the pressure loss. And it's not linear, either. The magnitude of pressure drop increases almost exponentially with flow rate.
Now, it may not look like one, but your intercooler is a drinking straw. A large, heavy, aluminum, multi-faceted drinking straw. Or, at least, it exhibits a lot of the same characteristics as one. So, back to the diagram:
Say that we have our boost controller set such that we see a peak of 12 PSI in the intake manifold. At 4,000 RPM, we reach that point. We are, incidentally, flowing about 130 CFM through the intercooler (our engine has a 100% VE at all speeds) and we're loosing about 1 PSI across the intercooler. So while the pressure at point B is 12 PSI, the pressure at point A (which is what the boost controller is seeing) is actually 13 PSI.
Now, we increase the speed to 7,000 RPM. At this point, we're moving about 220 CFM, and yet, what's this? The pressure drop across the intercooler has increased to 3 PSI! We didn't even double the flow, and yet we tripled the drop. (Well, I'm ignoring the fact that these are relative, rather than absolute pressure values, but you get the idea.) So now, even though the MBC is faithfully holding 13 PSI at the compressor, we're only seeing 10 PSI at the manifold.
The solution here should be pretty obvious by now. Move the boost controller from point A to point B.
By doing this, we are now telling the boost controller, in essence, "Hey, I want you to do whatever it takes to maintain a constant pressure at point B in the system, and to hell with what's going on over at the compressor." And it will comply. (Boost controllers are pretty simple-minded like that. They don't question orders.) Specifically, it now does not matter what the drop across the intercooler is, at least insofar as your actual manifold pressure is concerned. As drop across the IC increases, the boost controller will cause the compressor pressure to increase accordingly. So by the time you get to 7,000 RPM and are experiencing 3 PSI of drop across the IC, the pressure at point A will be up to 15 PSI, and you'll still be getting your 12 PSI at the manifold.
And here's everything you need to make it happen:
Yup. One 1/8" NPT hose-barb fitting. About $3 at your local ACE Hardware store. Drill ye' olde hole into the pipe which leads up into your throttle body, install this fitting into it, and plumb a hose from there to your MBC. Using all-silicone tube? (fag.) Well, just drill a hole in the colid-side end tank of the IC itself. Anywhere is fine so long as it's after the IC core, and before the throttle body.
You'll probably have to turn the MBC down just a tad in order to achieve the same peak boost you had before, as it's no longer having to factor in even the smallest IC drop.
