I had been wondering about this for some time. Now that I have a datalogit I am finally able to get some hard data. It appears that severe cracking in the turbo housing around the turbo pre-control does have a significant effect on primary boost. In the following experiment I forced the pre-control door shut by blocking air going into the actuator with a home depot boost controller turned completely off. I determined this was safe since I was never reaching my target boost of 11.4psi bar even with the actuator 100% vented by the pre-control solenoid.

In the first picture you see the boost pattern in a 3rd gear run starting around 2500rpm. The primary never reaches its target of 11.4psi, it just maxes out at 10.95. Also, it doesn't even reach that level until 4000rpm. In the second picture you can see the severe cracks in the turbo housing that are causing the problem.

Since this problem would not have been identified in the normal debugging people do for boost problems, I wonder if it is more common than we realize. Perhaps many of the people who get tired of trying to fix their sequential systems have a problem like this which wouldn't be solved by any number of vacuum hose, solenoid and actuator replacements. As much as I hate to consider it, Since the primary is spooling so slowly, perhaps this is actually a good candidate for non-sequential.

BTW, I did numerous runs in this experiment. This was the best spoolup I got on the primary. The performance is much worse if the boost controller (ecu, powerfc, etc) is allowed to open the pre-control door at all.

 

That's the worse housing I ever seen.
I just PMed you.

 

I agree, that housing is nothing more than charcoal anymore.

Unfortunately, the cracks and the boost pattern alone are only circumstantial evidence. I personally would not make that conclusion until a set of turbos with clean housings solves the problem. But even then, it doesn't remove variables such as precontrol rod adjustment, actuator spring rate, etc.

But I think that a newer set of turbos with much cleaner housings (should run around $400) will help for many reasons.

Dave

 

I did do the pre-control rod adjustment, but the condition of the spring could still be a problem. It is really surprising that those cracks have that much of an impact. I expected that the pre-control door would always be somewhat open to regulate the boost. Therefore, the boost controller would be able to compensate for the cracks by keeping the door shut longer.

For the mostpart I am really impressed with the performance despite the slow primary. I can't imagine what 10psi at 3000rpm would be like

 

10psi at 3k is nice, but lame compared to 10psi at higher rpms. Boost down low really helps a rotary, but it doesn't make it into a V8. IMHO.

Dave

 

I get 14@2500 in 3rd+ but it's not even close to a V8's torque.

 

My slow primary boost problem has been solved! As I suspected the leaking precontrol door was the culprit. I bought new turbo housings from a fellow rx-7club member (thanks cewrx7r1). It made a huge difference. In the attached picture you can see a graph of before and after. In the before case, the target boost of 0.8 bar was never reached (even with a precontrol duty cycle of 100%). Lower boost targets were attainable, but not until well after 4K rpm. For the after case, 0.8 bar is reached by 3300rpm.

Note: disregard the boost difference in the region after transition since I had secondary boost set differently for the two cases. Also, powerfc (which logged this data) reports boost in units of BAR x 100. So, 0.8 BAR is depicted at 80.

Conclusion
If you have slow building boost on your primary turbo and all the control systems are functioning, you may have a very leaky precontrol door / manifold. Considering decent turbo manifolds are not too hard to find (read inexpensive) this is a great fix.

 

That was a great troubleshooting (a bit risky too, heh) step on your part to keep the precontrol wastegate door closed to test it. I'm glad you solved the issue! It's nice to see someone resolve their sequential system rather than turning to a non-sequential setup.

I recently buttoned up my sequential system as well. It's making a strong 13psi across the band, starting at ~3000rpm.