I constructed my boost controller after the one Brooks Wesblat uses in his car. Fully detailed instructions on of how to create your own boost controller for about $ 10 are available at his site.
There are two types of boost controllers available; electronic or manual. Electronic controllers use a computer which learns the boost curve of your particular car and driving style. Once setup, they work very well and control most of the boost spiking which occurs during the transition from the first turbo to the second. The only downfalls to electronic controllers is that they are expensive (around $ 400) and don't always work well with sequential setups. I'm not comfortable with spending that much money on something that wasn't made to work with the turbo setup in my car. Manual controllers simply involves inserting an adjustable air valve to the wastegate. By adjusting the valve you are simply telling the wastegate how much to open. This setup is extremely simple, very inexpensive, and used by many. The only downfall I've seen in using the manual controller is the extensive boost spiking. Sometimes my boost will spike as much as 3 pounds, which can be very harmful if there isn't sufficient fuel to compensate.
One way to help eliminate spiking (which I pulled off of Rob Robinette's site) is to insert another adjustable valve in the prespool control line. When adjusted, you have control over when the wastegates open. By opening the valve, the wastegates open sooner which gives them less time to spool up. When the second turbo comes on, the isn't as much of a surge of air rushing to the engine. This in turn eliminates some of the spiking. If the wastegate opens too soon, however, then the turbos haven't spooled up enough and your boost response will be delayed. If carefully tuned, there is a nice medium which eliminates some spiking and doesn't substantially effect turbo response.
I recently purchased a boost controller from Darren Dawes for $35. This boost controller is a pressure valve that is supposed to control boost in the same way much more expensive electronic boost controllers do. The controller is installed the same way as the bleeder valve is installed. It simply replaces the stock boost line which goes between the turbo nipple and the wastegate. After the installation was completed, I took it for a test run and the following conclusions were made:
1) The boost came on stronger and felt smoother. It is hard to explain, it is more of something that you just feel when you get on the throttle.
2) Boost spiking was minimized to less than 1 PSI.
3) More boost is made in the higher gears (3,4,5). The first two gears make about 1 less PSI of boost. Therefore, always set the boost in the higher gears.
4) I didn't notice any boost creep. The boost held steady regardless of the gear I was in.
5) The bleeder valve that was previously installed to help balance out the two turbos was still needed. With the valve opened too much, there was more output from the second turbo. With the valve closed too much, there was slightly more output from the first turbo (about .5 PSI).
Note: All of the testing was performed with the stock catalytic converter installed.
Just when I got my boost operating properly with almost no spiking, I decided to replace my vacuum lines with blue silicon lines to match my engine compartment and at the same time put the new boost controller in the ash tray portion of the center console similar to the way I had my bleed valve mounted. One thing I did notice about the silicon hose but didn't think it would really matter was that the inside diameter is smaller than the hose I was previously using (fuel hose). Once I had everything mounted up and looking nice, I took the car for a spin. Wow is this thing noisy! It whistles very loudly, too loud to be mounted inside the car in my opinion. The next thing I noticed was my boost pattern. The first turbo came on strong at about 13 PSI, just where I wanted it. The second turbo also came on strong, but a little stronger than I preferred. It shot up to 17 PSI! I quickly let off the throttle and gave 3rd gear another run. Same thing, 13 on the first and 17 on the second. The boost worked great prior to changing the hose and relocating the boost controller from the engine compartment to the console. I then turned the boost down as far as it would go and made another 3rd gear run. I got about 10 PSI on the first turbo, a very large and lengthy dip during the transition, and about 14 on the second. I also received a rather large boost spike of nearly 2 PSI. The first thing I did was pull off all of the hoses and make sure there were no blockages. Next I hooked up a vacuum line directly from the wastegate to the turbo vacuum nipple (no boost pill). My pattern with this setup was 8 - 7- 8. This translates to 8 PSI on the first turbo, a small dip of 1 PSI during transition, and 8 PSI on the second turbo. This told me that there was nothing wrong with with my turbo system and the problem most likely lied in the boost controller. I put the boost controller back in the engine compartment and used my larger diameter fuel lines (6 mm) to hook it up. After a test drive, the boost gauge worked normal again. My conclusions from this were as follows:
The vacuum line to the console were too long, causing the turbos to receive a delayed vacuum signal from the boost controller. This was probably the reason for the large boost dip, lengthy turbo transition, and large boost spiking. The boost control problem was most likely caused by the smaller diameter hose (4 mm) combined with the lengthy amount used (about 15 ft). This caused enough restriction in the vacuum lines to act as a small boost pill which made it hard to keep boost below 13 PSI. My solution to the problem was to keep the boost controller in the engine compartment, use the least amount of vacuum line as possible, and use 6 mm vacuum hose. With this setup the boost operated flawlessly.
Approximate Performance Gain:
Boost controller - approx 15 RWHP for every 1 PSI increase