Setup of car before:

Efini Y-Pipe
HKS EVC IV (boost controller)
3" Cat-back Exhaust (Dynomax Ultra-flo muffler)
Stock Air Intake

With that setup I'd noticed my boost levels jumping up as high as .86 bar (~12psi) with the EVC set at .70 bar (~10psi). I would commonly see it bounce over right when I went from vacuum instantly to WOT by .03-.08 bar. With boost set down to .55 bar it bounced up as high as .72 bar.

Setup of car after:

Efini Y-Pipe
HKS EVC IV (boost controller)
3" Cat-back Exhaust (Dynomax Ultra-flo muffler)
Apexi Power Intake

I removed the stock air intake and replaced it with the Apexi. I don't see any big boost jumps now and the EVC is able to keep it within .02 of the setting (set either at .70 or .55).

Hypothesis:

The EVC had a hard time figuring out the airflow rate from the stock air intake and thus effecting its ability to predict when and how it would need to control boost pressure.

Since the stock air intake is a box with restricted flow if you're off boost air fills the box and sits inside and when go from coasting to WOT you suck all the air out of the box creating a pressure drop inside the box. The airflow into the box is so poor it isn't able to maintain a consistant pressure level.

With the Apexi intake (open intake) the pressure doesn't change significantly and the EVC is able to better predict and control what the boost will be based off of a more consistant airflow.

Question?

Is the EVC causing the boost spike with the stock air intake by venting things later than normal because it thinks the airflow is lower?

Does the stock air intake itself cause boost spiking to a small amount but the factory AFR is so rich nobody notices?

Do other box style intakes have this issue (M2/Cheap Bastard/etc)?

 

I didn't have a problem with the stock intake and my Profec B at all. In fact, when I installed the PFS SMIC and intake, boost rose faster than with the stock intake (and was 2 psi higher), enough that I had to adjust the balance control on the Profec.

 

How does the Profec B regulate boost?

On the EVC you calibrate it by running 3 WOT throttle runs (in preferrably 3rd but 2nd works) from 2500 through 7000ish. It then learns the base boost pattern of your car and regulates/normalizes it if it can to your desired boost level (i.e. my car runs 10-10-10 instead of 10-8-10). With the open intake the spool of the boost is noticably faster than it was with the stock air intake (probably only 3 updates (not sure if they are a whole second or a bit less) on the EVC now to hit .7 boost where the stock air would get to .55 or so about the same then take a bit longer to get to .7 (probably 4-5 updates)).

 

I suspect that your problem is with the boost controller, rather than the change in intake. I have never heard of anyone removing a boost spike by increasing airflow.

 

Ah, so it uses a sort of "fuzzy logic" to control boost. That is the cause of your problem. The Profec B uses no "logic" whatsoever. Using trial-and-error, you tune both the wastegate response and the boost level. Simple and reliable -- and cheap!

 

So really what this means is HKS EVC IV + Stock Air Intake = Boost Spike

So adding a boost controller (depending on what controller) can actually cause more problems than it can produce reliability.

Caveat Emptor and good to know this is a specific situation and not a widespread issue.

 

perhaps while installing the apex airfilter, you knocked something back in to place so it started working again.

 

Using fuzzy logic boost controllers is not recommended on 3rd gen RX-7s.

 

Care to elaborate with your reasoning? From reading I can't find anything that says you shouldn't use one on a RX-7 (Turbo II or FD).

From searching around the web it looks like the HKS EVC IV, Apexi AVC-R and the Greddy Profec A are all "fuzzy logic" controllers so this would apply to any of those popular models and perhaps others as well.

 

The fuzzy logic controllers sometimes get confused by the sequential operation of our turbos. In fact it has long been known that the Profec A works poorly with the sequential twins. The fuzzy logic works great with non-seq or single turbo setups.

That said, there are many people that run the HKS EVC and the Apexi AVC-R without issues.