Engine Management Computer

This is meant to be a VERY brief and simplified explanation of how the computer works, not an exhaustive dissertation on fuel injection. There are some very good ones out there. This discussion will focus solely on the engine management aspect of the computer also, i.e.- no cruise control info. Also, this applies to the stock ECU, not sure if the aftermarket computers operate the same way.

The computer has a pre-programmed "map" of what it should be doing in terms of fuel injection for given levels of RPM, load, etc. If the computer is supplying fuel to the engine based on these maps, it is operating in what is know as open loop mode. This is the simplest mode of operation. It is also not the most efficient.

One common question is "How many mods can I do before I need to upgrade the computer?". The question has to do with what you can do to the car before it starts to run lean and experiences detonation, which will kill a rotary. Mods that improve airflow start to become dangerous if they let it flow too freely, and the stock computer does not provide enough fuel to go along with the extra air. Since the stock computer uses static maps, it either needs to be reprogrammed (see M2 Performance, Pettit, XS). See below for more info --Editor

Date: Mon, 14 Aug 2000 13:21:36 -0700 (PDT) From: e s (rx712psi@yahoo.com)
Subject: Re:(rx7) [3] Intake, DP, CB and no ECU upgrade? LONG!

I'm going to throw some numbers out on to this thread. I know its been cover MANY times, but I have never seen numbers other than my own readings so here goes.

My mods probably wont directly relate to anyone else since everyone does the catback before a midpipe, but I figure the midpipe will make a bigger deference than the catback, & considering I was only running about 8-8.5PSI on the secondary turbo the "intake, DP, CB" configuration at 10 PSI will probably be equal if not worse than my config.

Disclaimer: These are rough measurements taken with a very high quality Fluke 87 volt meter on a very low quality Bosch O2 sensor, & an Autometer boost gauge.

OK, Once I had the intake + DP config I came real close to overboosting one cold morning so I installed a needle valve into the wastegate actuator pressure line. This left me with 10PSI pri, & 8-8.5PSI sec. (don't ask me why, but the Profec B seems to have fixed it)

My maincat clogged so I installed a midpipe, & figure that with only 8PSI on the second turbo that I'd be pretty safe,

Well I did runs with the intake + DP, & measured around .97x, but as the RPM rised above 6K the O2 voltage would drop down to about .927ish I figured that to be plenty safe.

After installing the midpipe, I was measureing the .97v. (around 3K on the tach) but at 5K with 8PSI the voltage would drop down under .9v. With it going as low as .81-.80v. by 6.5K.

That is when I decided that I wouldn't run over 4-4.5K until my Power FC arrives, in fact I rarely drive the car right now.

So that is proof (kinda) that 3 mods with no fuel mods other then a K&N fuel filter, & no stock filter make the FD R-REX run lean.

_________________

To summarize, there seem to be two approaches to engine management computers. One is to have separate components (computers) for things like boost control, fuel control, etc. HKS and Greddy take this approach.

The Electromotive and other high-end computers take the other approach by combining all of these functions in one unit. This approach is preferred by the guys who are doing heavy mods to their cars.

Also, the computers can be split according to whether they are stand-alone or piggyback. (Piggyback meaning that the stock computer remains in place, and the other computers that are piggybacked on modulate the signals going into/coming out of the stock computer.) The PFS is an all-in-one unit that piggybacks on the stock computer. The rest of the high-end computers are stand-alone, and eliminate the need for the stock computer, at least for engine management purposes. The stock computer still needs to be retained for such things as air conditioning, etc. --Steve

______________

Regarding engine control and management, it seems like most people will do something like the following (and I am REALLY guessing at some of these numbers):

	ECU reprogram		$700 (or PMC $1500)
	J&S knock sensor	$500
	Boost controller	$400
	Crane HI-6		$300
	=================
	TOTAL			$1900 - 2800

I think you can get an Electromotive for about $2500 that does all of these functions (not sure if there are additional options required but not included in this price, or if this was for a unit that would even work in our cars).

Has anyone done a cost comparison like this? I will probably be doing this over the winter or next spring and thought I would see if anyone has looked at pricing yet for a complete engine management system.

The individual component route seems like it might be a little easier to install, but you would get a lot better total solution from the Electromotive.

Note: since posting this, someone said that the PMC has boost control built in. I changed the cost numbers to reflect this. --Steve

______________

Shiv writes about buying engine computers from people who are running cut-rate parts businesses from their dorm rooms. I changed the name to protect the innocent, and filled in the names of the "real" tuners since people may not have known who Ray (SR) and Brian (M2)are. --Steve

__

Date: Tue, 28 Mar 2000 10:54:02 -0800
From: "Shiv Pathak" (shivp@worldnet.att.net)

No offense, but where did [vendor] learn how to create baseline ignition maps? Through trial-and-error like SR Motorsports, Mostly Mazda, etc,. Or did he copy the maps from someone them? To me, A product such as this is only as important as its after-sale support. Not only must the support be generous, but it must also be knowledgeable. Especially when you're trying to tune an engine as fragile as ours.

Come on guys... this market is small. Let's do our best to support those who have done their part to advance the state of the art. This means buying from the real tuners. You know, the guys who use the money they make in the aftermarket to race and then, in return, develop new products. It's really sad when these hard-working guys chose not to develop and market an otherwise brilliant product just because they fear it will get copied a couple of months later and sold out of someone's basement. I've seen it happen too many times. Sorry for the rant...

______________

Date: Thu, 20 Apr 2000 01:51:52 -0400
From: Wael El-Dasher (wael.el-dasher@efini.net)
Subject: Re: (rx7) [3] 850cc installed in primary rail and other questions

M2 educated me about the fuel system a few days ago when we were discussing a setup for my car. (snip)

I have a fuel pressure gauge in my car so I am able to monitor it (with peak hold) while driving and set it up accordingly. I was considering using larger injectors, but the EFI PFS/RUN DMC/PMS PMC can't varry the ratio of primary and secondary injectors like the MoTeC can, so the stock ratio must be maintained.

With M2's current MoTeC with harness & sequential turbo operation/oil metering for less than $4k, I would consider it. But the reality is I do not need it for my use of the car (street use/occasional track days). Unless one moves to a MoTeC that gives you that level of control, sticking to the stock injectors is the only option as several members had terrible luck with boring out injectors. I remember Westbrook, Chuck wrote at length on this subject a while back and methodically explained the steps to maintaining the ratio when using larger injectors. I can not remember if the injector sizes he recommended were bored out or not. Can the RP-Online Haltech E6K or the SR Motorsports Apex-i Power FC with Commander (I hate long names, so I mock them whenever I can) control the ratio like the M2 Performance MoTeC M4? Just curious...

______________

From: Tuck
Date: Mon, 23 Dec 1996 23:59:20 -0500

At 12:20 AM 12/23/96 -0800, NetBlazer wrote:

>>That is one of the reasons why the PFS computer is
>>capable of some self-programming.  The computer can read from the MAP sensor
>>and see that the desired amount of boost is being exceeded or not being
>>reached (no cats or partially plugged cats).  It will then experimentally
>>tweak the wastegate control until things are running properly.  That
>>variation between otherwise identical cars can explain why a computer like
>>the PFS one might be dialed in perfectly for one and not for another, 
>
>Exactly, but what happens when you finially decide you don't like sequential
>twin turbos, and don't like troubleshooting all those hoses which periodically

Personally, I'd like to stick with the sequential setup. No reason I can think of why you couldn't teach the PFS rig (either of them) to control boost on the non-sequential setup.

>crack and or pop off (I know replace them all with silicon hoses).

Gotta love that bright yellow silicon hose!

>go non-sequential you will be glad to have an aftermarket boost controller.
>Most of the ones made today have a learning mode similar to whats described 
>above.

That's true, but a simple price comparison makes it obvious why the PFS is even a price/performance leader. Sure you can just stick a damn stopcock in to bleed off pressure and set a peak boost level, but if you want real control you need something like the HKS EVC. The price of the PFS computer, which controls boost, ignition and fuel is less than an F-Con and an EVC, and the F-con and EVC aren't nearly as programmable anyway (The F-con just has that little 5 position global adjust as I recall). I suppose if you really wanted to run oodles and oodles of boost then the EVC would be a worthwhile companion to the PFS computer. The reason the EVC is so good at what it does is that it is a fairly expensive stepper motor rather than a simple duty valve like the stock unit.

>>First I'll have to define a few terms.  It was recently argued by my best
>>friend Kevin that the third generation cars only have one wastegate to
>>control boost.  This is not correct.  The factory service manual does not
>>describe it as such (instead calling it a control valve or some such crap)
>>but aside from the wastegate on the second turbo, the gate that dumps
>>exhaust into the second turbo to spool it up also acts as a wastegate for
>>the primary turbo.  It has it's own duty valve to control the gate actuator,
>>which determines how much boost is generated on the primary turbo just like
>>the wastegate on the secondary turbo.  The way the wastegate is controlled
>>is via a sort of arm, and that arm is in turn manipulated by a normal
>>wastegate actuator.  
>
>A waste-gate typically dumps exhaust to the other side of the turbine(s) 
>and out the tailpipe. For it to be a waste gate it must be allowing gases 

Typically, yes.

>to leave the system. The Turbo Control and Turbo Pre-control gates do not 
>allow any exhaust to bypass the turbines. True changes in these two gates 

While that is true, the thing DOES act as a wastegate in that it is used to control boost on the primary turbo. In that respect it can be treated as a wastegate. I think the appelation is accurate.

>changes how much exhaust from the rear rotor gets to the front turbo, but 
>it doesn't leave the system, it just redirects it to spin the secondary 
>turbo which is part of this system and is creating boost at the rpm where 
>the turbo control gate opens.
>
>On a side note the waste gate on my car (primary turbo turbine bypass gate 
>which is the only true wastegate on the twin turbo setup) Does an excellent 
>job preventing any boost spiking with an aftermarket EVC with its own large 
>solenoid. As for the other gates my car was setup several different ways: 

Of course, something like the EVC can dump vastly higher quantities of air than the stock duty valve, so it won't need a smaller restrictor in the pressure line, and as a result the reaction time will stay high enough that you won't get boost spiking. BUT an EVC is how much? 900 bucks as I recall (and no fuel or ignition control of course), making it damned expensive unless you really want to run uber-boost (and keep in mind that the newer PFS computer is perfectly capable of handling significantly higher boost levels because it can control both duty valves...). It is a tradeoff really. With something like the PFS computer you just stick in a few plugs and you never need to get under the car and screw around with anything. To install something like the EVC you have to get under the hood and screw around with the vacuum lines.

>First it came to me with the EVC controlling the turbo pre-control gate and 
>primary turbo wastegate as if they were both waste-gates. This resulted in 
>exagerated lag when the secondary turbo was brought online, but never ANY 
>boost spikes that showed up on a boost gauge.

Like I said, with the stock restrictor pill thing the reaction time of the wastegate stays high enough that boost spikes cannot occur.

>Second I set it up non-sequential with the turbo control and turbo pre-control 
>gates completly removed, and the waste-gate actuator hooked up to the EVC. 
>Still no spikes, stayed rock solid at preset levels to redline (usually 1 bar)
>This however did not prevent my stock computer from exercising a fuel cut which 
>cost me the rear rotor. (HKS FCD is only rated for 14PSI and I don't reccomend 
>exceeding this for extended periods of time, or high load (5th gear at 150MPH 
>in my case))

Interesting thing is that the stock computer has fuel maps all the way up to 13psi and above. The reason you can't go higher of course is that the fuel cut to the rear rotor kicks in as soon as it sees more than 11 psi for more than one second from the MAP sensor. You put a voltage clamp on the map sensor so the computer never sees over 10 psi and you're set, except of course that you still can't get at the maps for higher boost levels. The way the Tectom rig works is by sort of bubble gumming around the main processor. You remove the stock processor with the fuel maps and the boost cut on it and solder it into a daughterboard. That daughter board has eprom sockets that you stick your custom-burned maps into, and basically what they did was they sort of cracked the proc with the fuel maps and boost cut on it. They trick it into looking at those eproms instead of the internal stuff. That's why it keeps the original chip. Neat, but too complex for my taste.

>The above two examples proves (to me at least) that boost spikes can be 
>completly controlled using an aftermarket EVC, and only the single wastegate 
>on the turbos. This indicates that the stock wastegate/actuator can do a very 

That's true, but aside from something like a motec or electromotive or a similar such unit, you're still going to have to deal with fuel and ignition maps with the PFS rig. Might as well just get the newer PFS computer and use that, you'll still get your really high boost levels and you get it all in one simple package. The other advantage is that if you DON'T want to screw with it, it can reprogram itself to your machine, and if you want to fine tune it you can get the keypad and do it. From a marketing perspective it's a really trick box.

>good job of controlling boost spikes but the stock solenoid/restrictors as 
>described below by Tuck (if his info is correct) is not able to do its job 

Again, with the purple computer you do not need to change the restrictors, and you only had to do it with the gold one if you wanted to exceed the design specs on the unit. Keep in mind that the gold computer was designed to be a part of a STREET LEGAL (50 state) performance package, so it was assumed that the cats would stay on the car. Within that specification, the unit worked flawlessly because the limits of that one duty valve and the restriction of the cats prevented you from going significantly higher.

>properly and needs help by tweaking with the turbo control and pre-control 
>gates which should not be messed with IMO since they are doing a completly 
>different job which has its own timing issues, and were not designed for 
>the purpose of controlling boost spikes.

This is not correct. The valves that you refuse to call wastegates are used by the stock ecu to control boost on the primary turbo. That is part of what they are designed for. Of course you are correct that that same system is used to prespool the secondary turbo in the stock sequential setup.

>I still say lose the stock mess and go with a Purpose designed unit that 
>can spend 100% of its time doing its job. And if you rip out the cats 

Well that's your pervue, but in terms of technical sophistication and programmability the PFS purple box beats the EVC hands down. While the EVC will allow you to program a boost curve of sorts, it doesn't give you 1/10th of the resolution that teh PFS computer(s) do. The real thing you are paying for with the EVC is the expensive stepper motor/valve. You could just as easily replace the stock duty valves with something that can flow more, say the duty valves from a grand national (about 50 bucks or so each), and use the pfs computer to control that. The purple box of course is capable of controlling boost beyond the capabilities of the stock fuel system, so I'm not sure why you'd want to.

>non-sequential operation is the next logical step since the only reason 
>for sequential was the exhaust restrictions imposed by the cats / stock 
>exhaust.

The sequential oepration is also designed to provide boost as low in the rev band as possible. Whether or not anyone wants to admit it, going to nonsequential might free things up a bit at the top end, but it simply must increase the time it takes the turbos to spool up initially. Since you can easily get over 15psi at the top of the rev band without switching to non-sequential, I can't think of any reason to reduce drivability (however slightly) for what is probably a minor gain at best.

_________________

A knock sensor is an almost indispensible item for a rotary if you are going to modify it. One note - if you buy a high-end computer, it will probably include a knock sensor, so you will not need to buy a separate one (I am pretty certain the Electromotive does, and am guessing the Motec does too). --Steve

________________

Date: Fri, 17 Sep 1999 17:58:58 +0000
From: "David Lane" (dlane@peabody.jhu.edu)

I believe one of the keys to the puzzle is the variability of where our cars are, and how they are used.

For instance:

Some of us live where we can get 94 octane pump gas. Others are limited to 92 octane, and some formulations are better than others.

Some of our cars have to cope with sub-freezing temperatures. Others stay in the balmy zone all year around.

Some of our cars have early ECUs which, if I read the list correctly, behave differently than the later ones.

Under given circumstances, some of our cars will show boost creep. Others won't.

Some of our boost gauges are not as accurate as others, and may be reading a PSI up or down.

Some of our cars are modified for the occasional street romp. Others are set up for the strip--some for road race tracks, others for autocrossing. Then there are the cars that are hardly revved above 5k, owned by people who just love them because they feel and look great.

So, a given set of modifications which will cause no trouble on a 94 octane car, located in the South, and never used for competition will be disastrous on a machine run hard in freezing weather at the track with 92 octane fuel.

While you may want to argue the technicalities of it all, the bottom line is that the only way anyone can assure someone else that suggested modifications will be safe is to stay very conservative. In my report on Mostly Mazda last year, I mentioned to Brian that one of his customers was using an M2 chip at higher boost than recommended--and was very happy with it. Brian said his chips were all set up for 92 octane California fuel, and that they had a lot of richness built into the maps. We can just call it "head room." Advocates of user programmable ECU upgrades like the PFS unit don't like the added richness (slows them down), so they prefer to tune their own maps, usually with less headroom. Those who install fully programmable units like the Electomotive say that for cars in my area (100 plus temps in the Summer with rare plunges to 0 F in the Winter, picky owners develop four different map sets; one for each season.

So, what's "safe?" Maybe a better question is, "What is your tolerance for risk?" Mazda delivers the car to you with enough headroom to hopefully be able to cope with all of the variables listed above at 10 psi max boost. If you are modifying the car to produce significantly more power, you are not really safe unless you duplicate the headroom Mazda built in--albeit at a higher boost level. That's why Brian wants you to get an ECU upgrade with just an intake and catback. The other reason is that he says he has figured a way to work around the stock boost control limitations, so that when you start working your way up the modification/power ladder, you won't deal with boost creep. Note that I mention the M2 upgrade sequence simply because I have not talked to Cam about his ECU, and of course Peter uses a programmable unit. Nevertheless, it may be (just guessing here) that cars with the M2 chips just happen to not be the ones with the boost creep problems.

In the end, if you have a high tolerance for risk, and use your car only within certain boundaries, you may be able to get away with less headroom in the system. However, if you don't have such a high tolerance for risk, you absolutely must maintain as much headroom as you can--just in case the day after you filled the tank with 92 octane the temperature hits a record low and some guy in a hopped up Mustang doesn't want to give up until you are going over 140 mph.

_______________

Date: Thu, 10 Feb 2000 23:35:17 -0800
From: Max Cooper (max@maxcooper.com)

> Are the stock fuel maps good up to 15 psi if you remove fuel cut?

No. You can't reasonably remove the fuel cut without limiting the boost the computer sees, so you are asking if the computer's 12 psi maps are good for 15 psi. No, they aren't. And if you have increased the flow capacity of the intake and exhaust, they might not even be adequate at 12 psi (my car detonated at this pressure with the stock main cat and computer, using an upgraded intake, ic, dp, and cat-back above 4500 RPM).

The stock ECU is tuned for the flow characteristics of the stock intake and exhaust. Unlike the more common mass-air control systems, the [3]'s speed density system CANNOT adjust for a better flowing intake, intercooler, or exhaust. Mass-air systems have a flapper door or other means of measuring the amount of air the engine ingests and provides fuel based on that measurement. On the speed-density system of the [3], no such measurement is made. Instead, the computer reads the engine speed (RPMs) and air density (its temp and pressure) and delivers fuel according to a map that was programmed based on the stock intake and exhaust systems. Those maps are no longer valid when you change the flow characteristics of the intake and exhaust systems. Some people have argued on this list that the air that flows through the engine is simply a function of boost pressure; that 10 psi before mods makes the engine flow the same as 10 psi after the mods. This is simply not true. For simple evidence, consider that the reduced back pressure of the exhaust system will leave less exhaust gas in the combustion chamber and that leaves more room for fresh air and thus a need for more fuel. Got it? The modded engine flows more air and thus needs more fuel. The computer knows only about engine speed and air density - nothing about flow so it can't compensate for these mods. The fact that you can run some mods is only because they don't take the engine out of the rich safety zone built into the stock fuel maps. Of course, it does put you closer to the edge, and eventually over the edge if you add enough mods, or the air is cool enough, or you get some bad gas, ...

It is foolish to run a lot of mods with the stock fuel maps. You won't get the power you are looking for and will spend more on an engine rebuild than you would spend if you did the right thing and got a computer to match your mods in the first place. A lot more -- don't do it.

______________

Date: Wed, 12 Apr 2000 17:26:23 -0400
From: "\(Mr\) Sandy Linthicum" (sandy-linthicum@nc.rr.com)

Doesn't matter what CPU you use, 1/2 second distance from real time control in closed loop will cause you to go lean and detonation in race/track conditions. Its true for the Motec as well as any other engine computer.

You may be implying using O2 readings to adjust fixed maps - which is fine. Running real closed loop means you ajust fuel, etc. based on the O2 sensor data (this is the major controlling factor). In WOT conditions you cannot afford to be 1/2 second behind.

The Motec allows you to configure the closed loop operation by rpm & load (map sensor) and typically you exit closed loop if Map or RPM show hard use.

Date: Fri, 14 Jan 2000 10:51:50 -0700
From: "Owens, Shaun" (ShaunOwens@aec.ca)

Date: Sat, 18 Sep 1999 16:07:57 -1000 (HST)
From: F8LDZZ (f8ldzz@lava.net)

> Any idea why it doesn't have a mass flow (AFM) sensor? Cost? Non-linearity?
> Operating conditions range too widely?

• cost
• "low tech"
• too restrictive

MAP-based systems are linear.

> Seems I remember someone posting the downside to a mass sensor vs the upside
> to a pressure system and that given the operating conditions, it was a good
> trade-off. Anybody know the facts/arguments?

AFM - generally more "accurate" in terms of air flow measurement
    - replacement cost "high" - over $1000 new
    - "restrictive" versus a MAP system
    - "low tech" due to moving parts (with exception to Hall Effect)

MAP - cheaper to replace (usually <$500 for pressure sensor)
    - "high tech" due to solid-state electronics
    - "less restrictive" - MAP readings can be "calibrated" to (any)
      intake opening size

_______________

Date: Sat, 18 Sep 1999 12:47:54 -0700
From: "Michael Card" (cul8r@my-deja.com)

I've been trying to find out what the EXACT fuel cut off points are on the 93s. I have an early run edition, one of the first batches sent over from Japan, and was wondering what fuel cut points were.

According to Rob:

The ECU will cut the fuel to the rear rotor when this boost level is reached at this rpm:

1000 rpm   11.8 psi
2000       11.8
3000       12.4
4000       13.3
5000       11.3
6000       10.7
7000       10.7

The shops all told me 12 lbs was the cut off...so kinda wondering. Anyone out there know exactly how the stock chips funtion?

_______________

From: Jay (styk33@yahoo.com)
date: April14, 2000

>What happens when the computer makes the 20K mileage switch, 
>i.e.- what changes at 20K miles?

I know you inquired last week about the milage switch. If you want to see what happens to your car with it 'deactivated', try this.

Disconnect wire 1N on the ECU. Connect the ECU side of that wire to a 1.2DCV+. I am unsure of what resistance is needed on the other side. You might try getting a 12VDC zener diode from radio shack and put it inline.

If you are still curious and don't feel like playing with the ECU wires(don't blame you if you don't), call Brian at M2. He might have more info if you need it.

The PFS computer is a piggyback unit that works w/ the stock ECU. It modifies the signals coming from the stock ECU and reports back to it that everything is still alright.

It is very programmable, which is a benefit and a drawback. You have to know what you are doing. You also have to buy extra software for a PC interface, or buy a keypad to program it. Unless you have PFS install and tune it on your car.

Some people have posted that they have had success with the PFS computer. Brooks' note below is a negative one, echoed by others who have had problems. No concensus on this yet.

Date: Mon, 29 Dec 97 22:54:33 -0500
From: Brooks Weisblat

Yes, I'm saying that my pre-programmed chip (Pettit) works MUCH BETTER than the PFS box......hands down...no question about it.....im never playing with the stupid #'s all the time.....fouling out plugs....detonating because Peter said to hold the throttle down until the computer brings down the boost and learns the curve....

You can't say that my car was at fault either.....without the PFS box my car ran fine...after returning the PFS box and installing a Pettit computer my car was hitting 11 second timeslips with 14-15 pounds of boost.

___________________

Date: Mon, 29 Dec 1997 23:47:00 -0500
From: Tom Gandey Not to say that the PFS box is a P.O.S., but Brooks has proved that Cam's (Pettit) maps DO WORK, and can better anything you have produced.

I believe Mostly Mazda will reprogram your stock ECU from the sounds of Jim's note. --Steve

Date: Fri, 21 Nov 1997 10:30:41 -0800
From: "Jim LaBreck (ECA)"

Brian (at Mostly Mazda) said that this was easily the most drastic ECU he'd done, but it will work well for most intermediate levels of modification up to full out gutting of the intake and exhaust and non-sequential modifications, among others.

I was told that it would control boost in sequential systems without the need for a separate boost computer, but non-sequential systems will definitely need a controller, as the ECU no longer has the means to regulate the activity of the second (rear) turbo in that setup. It has fuel maps up to 22 lbs. of boost, but with systems with stock injectors, they'll reach their maximum duty cycle at about 16 p.s.i. I would not recommend going much higher without high octane fuel, an upgraded fuel system, and/or larger injectors.

The fuel maps are beautiful, by the way... something that it would take hours to achieve with a gold or purple box, plus you don't have to find a place for that extra box to live in your car, which is one reason why I chose to have my ECU reprogrammed.

HKS takes the component approach to computers. They have several:

• FCON: Fuel CONtroller - controls the injectors. If you increase boost, you will need to increase the amount of fuel fed to the engine.
  
  
• Fuel Cut Defenser: the stock ECU will cut the fuel to the engine if the boost excedes a certain maximum. The Fuel Cut Defenser solves the problem of what to do if you do want to increase boost (e.g.- with the EVC).
  
  
• EVC: Electronic Valve Control - controls boost.
  
  

___________________

Date: Fri, 30 Apr 1999 10:50:51 +0000
From: "David Lane"

AIC: Additional Injector Controller

I have an older model HKS AIC. In addition to setting the boost and rpm levels at which the additional injectors begin to fire, it has two other controls to specify the rate of increase for fuel flow as a function of boost and/or rpm increase. These "rate of increase" controls are linear. That is, the "curves" are straight lines. Both curves combine as both parameters increase. Obviously, the RPM trigger does not activate if the minimum boost has not been reached.

My old unit has an LED bar graph to show the percentage of available fuel being used (duty cycle). It also has lights to show when each function (RPM or BOOST) has reached the threshold. You can use this information in conjunction with your tach and boost gauge to confirm that things are happening when you want. If I read the lit correctly, the newer units allow you to program in these parameters directly, so it should be easier to use.

Do not, under any circumstances, rely on one of these things without having some way of monitoring the results. A decent A/F meter is indespensable for tuning. Without one you are flying blind. AICs do not offer the finely tuned options of an aftermarket engine management system, so even an inexpensive A/F meter can get you in the ball park. After months of messing with mine, trying to tune it with an EGT gauge, I installed an A/F meter. In 10 minutes, I had the system working properly. Once set up, you can basically leave it alone.

One other warning: These units do not have access to data on coolant or ambient temperatures, so if you live in a place with wide temperature variations, you may want to tweek the boost gain up or down a notch when the seasons change . In other words, locate the control unit where you can get to it.

Within the above restrictions, I am very happy with my HKS unit. Once set up, it has functioned reliably and consistently. Also, it has an interesting "test" function which fires the added injectors full blast to verify that they are working. This extra shot of fuel can be used to load up the exhaust with the kind of stuff which makes for spectacular flames out the back. My exhaust system is old, but when I replace it, I will certainly experiment a bit with the "test" button to see if I can't figure how to create a subtile warning to those who follow too closely.

___________________

Date: Fri, 17 Mar 2000 17:13:47 +0000
From: "David Lane" (dlane@gigue.peabody.jhu.edu)

> I will (of course) need to upgrade my fuel.  I am thinking about the 
> HKS AIC with 2 injectors in the Greddy Elbow.  I will be running 15-20 
> psi of boost on the stock turbos. I'm also upgrading the fuel pump and
> adding an aftermarket fuel pressure regulator. This will all be
> dyno tuned.   Ive heard that the AIC is kinda hard to tune, any
> opinions out there???

The HKS AIC is actually easy to tune. The problem is that the unit itself gives you no frame of reference for your air/fuel ratio--which is what you need if you are going to tune it properly. You say you are going to run 15-20 psi on stock turbos. Even the lower side of that will put you into very dangerous territory unless you have a wide-range A/F rig hooked up for the tuning. Frankly, I agree with Max that 20 psi is beyond the design parameters of the stock turbo system. What I have to say would be best applied for, say 14-15 psi. I defer to 3rd gen owners for better figures.

The system you propose (boost dependent FPR + AIC) is the one I use on my turbo GSL-SE. There are some limitations to this kind of rig.

1. Neither device "knows" about the other, and your ECU hasn't a clue that it is getting "help," so you have to figure out the best way to integrate them yourself.
   
   
2. The AIC has no feed for engine temperature or ambient temperature--just boost and RPM. Thus, if you tune it on a hot day, it may be a bit off when the temperature drops. This would not be much of a problem in Southern California or Florida, but it is a concern in locations where there is wide variation in temps over the year.
   
   
3. The boost dependent FPR works through the stock injectors, and is thus affected by all the parameters measured by the stock ECU (as translated into the injector duty cycle). The boost dependent FPR also works through the additional injectors, but the AIC only "sees" boost and RPM. You can see how this can become confusing when you are trying to make fine adjustments.
   
   
4. Since the ECU, the boost dependent FPR, and the Additional Injector Controller do not "talk" to each other, and do not operate with the same input, you cannot control your fuel mix as consistently as you can with other fuel enrichment options (aftermarket chip, piggyback controller, Motec, Haltech, etc.). This is not much of a problem with a car like mine, running relatively low boost with a lot of headroom, but I am skeptical about the wisdom of using it for the kind of boost you specified. Your engine would have little tolerance for small variations in A/F mix. Besides, at those kinds of boost levels you would probably need some kind of ignition timing control to keep the engine happy. You get the benefit of integration and timing control in the more comprehensive aftermarket systems--even a relatively inexpensive chip upgrade.
   
   
5. Since the system you propose is relatively imprecise, and can change performance (literally) as the weather changes, you will need some way to monitor it at all times. I am a big proponent of inexpensive A/F meters for cars like mine, but such a meter does not have the resolution to tune the car properly for the kind of boost you want to run. Still, I recommend one as a gross indication of the health of your system. At the same time, a J&S knock sensor is mandatory to keep you safe when small changes bring you too near the edge of detonation. You said your car would be dyno tuned, but that tuning may not be accurate over a period of months and under changing conditions. Boost dependent FPRs are not particularly noted for being consistent over the years. As a final caution, I understand that dyno tuning by itself is not a guarantee of good results. I don't have the details, but one of the true gurus out there told me it was tricky--that what looks good on a dyno is not always the best setting for real life.
   
   

AICs are not cheap, and you could probably get a suitable chip for your ECU that would give you better results with more control for the same money. If you really MUST get to 20 psi, I should think you would need an entire aftermarket engine management system, and you would have to ditch the stock turbos. Max Cooper covered that topic much better than I ever could.

On the other hand, if you are looking for a relatively low-tech approach for fuel supplementation with an aftermarket turbo, or for safety if you are going to push stock boost pressures by a couple of pounds, a boost dependent FPR coupled with an AIC is a reasonable approach. Just don't forget to add a J&S knock sensor to your budget. Sooner or later you WILL experience a lean running condition--either as you initially try to tune the system or later in its life when something goes on the fritz, clogs up, or comes loose. Maybe you will just get a little frisky with boost pressures to see if you can't just get a little more power out of it.

Better to see the lights on the J&S flicker than to hear the popcorn sound and see the smoke.

___________________

Fuel Cut Defenser:

Mon, 10 Nov 97 18:33:36 -0500
From: "Kyle Krutilek"

I may be off with the following reply, however, this is how its been explained to me, here goes.

The FCD lies to your computer regarding boost levels. It prevents the ECU from cutting fuel in the event of excessive boost at high rpm. This has been referred to as sounding like a "dog barking", however, they are referring not to the car but to the sound that the rx7 owner makes when he/she thinks that the engine just took a dump!

To paraphrase ... if your not running enough boost to hit the fuel cut, you don't need the FCD.

The FCON will allow you to add the additional but necessary fuel when running much higher than stock boost. This is necessary because the FCD will be lying to your ECU which will of course have no idea that its owner is feeding the engine such high boost and therefore the ECU will probably only feed the engine enough fuel to run its normal fuel delivery for an estimated 8 psi or so boost. Probably not enough fuel for the gutsy owners that push it into the teens. May want to invest in an EGT to meter fuel dosage with the FCON, some use air/fuel meters or O2 sensor, matter of preference or faith. There are higher forms of intelligent metering and thought for that matter available to tune the engine.

I have heard of rx7s running open intake/exhaust with a FCD/FCON with performance described as "runs like a motorcycle". Stock engine in this case was short lived.

_________________

Date: Mon, 10 Nov 1997 22:35:11 -0500
From: Rob Robinette

I just finished installing an HKS FCON and Fuel Cut Defenser. Before I installed them I did some 3rd gear boost tests. I was getting 10 PSI, 8, spike to 15, and then steady 7 to redline. I believe the fuel cut was occurring and the ECU was resorting to the safe 7 PSI level.

After I installed the FCON and FCD I got 10, 8, spike to 15, and then steady 12 PSI.

This is the important thing: The only mods I have right now are RB intake (with stock airbox), K&N filter, 3" straight pipe cat back, and no air pump and yet I really believe I was hitting the high boost fuel cut when the second turbo came on line and spiked. I'm really surprised that the FCON and FCD solved this 7 PSI boost problem. Just wanted to let you know that even relatively mild mods can cause problems in the 3rd gen. I can't wait to get the down pipe on.

___________________

Date: Tue, 11 Nov 1997 07:59:38 -0500
From: Brad Franklin

Just a word of caution, you guys should be careful about hitting fuel cut. It is there to prevent a catastrophe, and is not to healthy itself. Think about your turbo forcing so much air into your engine, then the computer says "too much boost" and cuts fuel. It does the trick, but for a short duration, your engine runs very lean, which can cause severe detonation for a short period. And we all know what detonation can do to apex seals. Be careful!

___________________

Date: Thu, 18 Mar 1999 10:03:01 -0500
From: Tom Walsh

I know someone has already replied to you telling you not to use the FCD without a computer, but let me explain why you should not use the fuel cut defenser.

The stock ECU has a programmed fuel cut that varies depending on RPM. During the switch over (4000-5000 RPMs) is as high as 13lbs, but comes back down to 10lbs as you approach the redline.

The Fuel Cut Defenser plugs into the MAP (Manifold Absolute Pressure?) sensor, located on the firewall in the area of the throttle body. The FCD then supplies the ECU with a modified reading of the MAP sensor. This modified reading is basically a lie to the ECU to prevent it from hitting the fuel cut (normally 10lbs of boost).

Now lets think about this. The MAP sensor's job is to supply the ECU with a current pressure reading so the ECU can add the correct amount of fuel to the mixture and prevent a lean condition. Now, if you are going to run the car at 14lbs of boost (which I do not recommend, but that is a personal recommendation), and the FCD is telling the ECU you are only running 10lbs (to avoid fuel cut) there is a difference of 4lbs of boost that the ECU has not added fuel for, thus you are at a very lean condition and this can lead to detonation.

So in short, if you are planning on running your car above stock fuel cut, get a computer that will allow you to do so (Pettit, PFS, Wolf 3D, G-force, etc...) by removing the fuel cut from the ECUs program.

Who would use the fuel cut defencer then? The only people, I feel, that have a need for this type of device (FCD) would be those of you that live in a cold environment, but have the car stock. Often times, with the cold air, your car will run a higher amount of boost and you might occasionally hit fuel cut.

So why fuel cut instead of ignition cut? "Fuel cut" cuts out the fuel being injected into the motor during the intake cycle of the motor. This causes a momentary lean condition inside the motor. Is that bad? Yes. So why did they use an ignition cut instead. This (fuel cut) is actually a standard practice across the industry. It was determined, some time ago, that ignition cut was not the best solution to the problem due to the fact that the unburned fuel remains inside the combustion chamber and can "wash" away the lubrication that lines the combustion chamber walls. There is also the possibility that emissions also had a determining factor into this but I am unsure as to how much.

The Simple Skinny: If your car is stock and you hit fuel cut every once in a while, while driving around town, then you should get the FCD because the fuel cut is doing damage, however slight, to the motor.

If you want to raise the boost levels over the stock fuel cut level (perhaps using a Profec B, which I use) get a modified ECU that has had the fuel cut removed.

___________________

EVC (Electronic Valve Control):

Controls boost pressure.

Date: Thu, 17 Aug 2000 10:41:12 -0400
From: "David Lane" (dlane@gigue.peabody.jhu.edu)

> Now that carparts.com has HKS products I might buy a boost
> controller. But they have the HKS EVC IV and the original EVC.
> Which should I get for my car?  I've always heard that fuzzy logic
> is not good for third gens. Does anyone know if the original EVC
> has fuzzy logic and if it is dual solenoid?

I recently switched from the EVC I to the EVC III (similar, I think, to the IV--at least I am going to assume so for the purpose of this message).

Unfortunatley, the information below may not be applicable to your case, because I doubt that the "original" EVC (EVC I) is still available. Rather, they are probably selling the current fuzzy logic EVC IV, and a more basic unit--probably designed to compete with the Gready Profec B at a lower price point, and without some of the fancy electronics. The HKS website shows something called the "EVC EZ" in that category. I suspect these are the two units priced at Carparts.com under the titles of "EVC" and "EVC IV." Unfortunately, I wrote the following comparison and contrast between the original EVC I and the EVC III (which I assume to be similar to the EVC IV) before checking out the Carparts.com web site. You might be interested in reading it anyway--just so you will have a better idea of what questions to ask if you call Carparts.com about the less costly unit. Also, it is pretty complete about how to deal with the fuzzy logic unit. Finally, I am too lazy to re-write it.

The EVC I came in two versions--one for integregal wastegate, and one for remote wastegate. The EVC III has a switch on the back. I believe the "single" and "remote" designation is about one or two ports on the side of the wastegate.

The EVC III is much easier to use for a variety of reasons. The fuzzy logic feature can be bypassed. The fuzzy logic itself is pretty neat, but it requires you to do three identical runs in "learn mode." The instructions for doing this with sequential turbo systems require you to begin the runs when the 2nd turbo comes on line. Obviously this is difficult to nail exactly the same way three times, and they say as much in the instructions. In fact, it is at that point they tell you how to bypass the fuzzy logic feature.

For what it's worth, I ran mine for a day with fuzzy logic bypassed--couldn't get the learn feature to "take." The next day, I was successful. My impression is that the fuzzy logic is an improvement.

The EVC I has the same features as the EVC III running without the fuzzy logic. However, there are the following operational differences:

1. Scramble boost control is an add-on box with the EVC-I. It is integrated into the EVC III.
   
   
2. Adjustments for Low and High boost on the EVC I are trim pots in the back of the unit. Adjusting this is awkward in that you have to dismount the unit and turn the screws--which is hard to do by yourself under boost. Thus, you have to turn a little, test, and repeat until you get the results you want. In contrast, the EVC III has a digital interface, so you just dial up your target boost levels directly, and make a small modification or two if you don't get the anticipated results on your boost gauge.
   
   
3. The EVC III has an overboost protection mode that you set where you what it. If the overboost threshold is exceeded, boost drops to wastegate level. The EVC I does not have that feature.
   
   
4. Both EVCs have a manual control so you can mess around with boost without tampering with your Low or High presets. The EVC I is a knob on the face of the unit. You watch your boost gauge and turn the knob until you get what you want. The EVC III, is set by punching up the boost you want on the digital display (same as with any other setting).
   
   
5. The EVC III has an "offset" feature to calibrate the unit up or down a little to compensate for installation variations.
   
   

___________________

Date: Fri, 03 Apr 1998 01:29:19 -0800
From: Andrew Chiao

The electronic boost controllers such as the HKS EVC IV and the other ones out there "trick" the wastegate to stay close when it would normally open. With the use of a stepping motor attached to the elec. boost controller, a lower pressure reading is sent to the wastegate than there actually is. For example if your wastegate is suppose to open at 10psi, but you want to turn up the boost to 12 for instance, then how are to supposed to reach 12 when it closes at 10? This is done, for example, by telling the wastegate it's at only 8psi when it's actually at 10, and telling it its at 10 when it's actually 12. Therefore you could reach your desired 12psi. This is my understanding of how it works, but I'm pretty sure that it's right.

__________________

Date: Fri, 03 Apr 98 08:21:50 -0500
From: "Linthicum, Sandy"

The controller bleeds off air from the line controlling the wastegate so the flow/pressure in the line that would normally have the wastegate fully open at 10psi (ie.pressure on diaphram overcomes the spring pressure holding the wastegate closed) does not occur till an actual 12psi. Only a limited amount of air can be bleed off by the soloniod, which is the reason for a restrictor in the line which limits the amount of air that can pass.

The computer provides fuel based on TPS, MAP, and other sensors. As boost rises, it will provide more fuel up to about 12psi which is where the stock ECU does a fuel cut (total) to protect the engine. This is one of many reasons you need a fuel computer if you are going to run boost over 12psi.

__________________

Date: Fri, 3 Apr 1998 11:34:55 +0000
From: "David Lane"

I can't speak for 3rd gens, but the HKS EVC on my car does not work the way a previous poster implied. What follows is for a typical wastegate on a single turbo system.

First of all, the wastegate OPENS to create a path around the turbo for the exhaust gasses to flow. Most wastegates have a valve that is controlled by a diaphragm. Boost pressure is brought to one side of the diaphragm, tending to open the wastegate. A spring acts on the other side to keep the wastegate closed. As long as the spring on one side of the diaphragm is stronger than the boost pressure on the other side, the wastegate will remain closed. However, there will come a time when the boost pressure overpowers the spring and allows the wastegate to open.

Now, the problem is that the wastegate tends not to "pop" open all at once. If you are set for 10 psi, the typical wastegate will start to open a little before that, in effect cracking the valve open before necessary which makes the engine take a small amount of extra time to arrive at max boost. The theory is that you can increase power if somehow you can positively hold the wastegate closed until exactly 10 psi.

The aftermarket electronic boost controllers feed boost pressure to the back side of the diaphragm (the same side as the spring) to assist in holding the valve closed untill exactly the right moment. This results in slightly enhanced performance. However, the reason many people turn to these devices is not so much for additional power as for better control of boost levels. However, it must also be said that if you have better control of your boost levels, you can set it a bit higher without worrying about it changing by itself. In that sense, the car does end up putting out more power in daily driving.

The spring and the air pressure used to control most wastgates vary in value depending on ambient temperature, humidity, etc. In some cases (depending on your local climate), your boost pressure will be plus or minus a psi between the morning and the afternoon. This is not a problem for a stock turbomotor with huge amounts of head room built into the system, but for those of us who want to run nearer the edge, it is critical.

Secondly, since the EVCs are electronically controlled, you can change the boost setting remotely from the head unit within easy reach. The HKS, for instance, offers pushbutton access to a low setting, a high setting, and a variable setting--all of which you can set to your taste. I believe the Greddy unit is similar.

Finally, the more recent units offer a "temporary" setting which allows you to trigger additional boost of a pre-set amount for X time--after which the standard boost setting is re-established. This allows you to run full throttle at a conservative boost setting for daily driving, but to have instant access to your maximum setting if you need it in anger. A secondary benefit of this feature is to avoid wheel spin in your lower gears, while increasing boost for the upper gears.

While this post is accurate for single turbo, single wastegate sytems, I don't have the background to apply it directly to the more complex sequential turbo set-ups on 3rd gens. Earlier posts indicate that electronic boost controllers are in use by some list members, so I can't be too far off.

Date: Sun, 15 Mar 1998 21:52:49 +0000
From: "Steven M. Robertson"

If I were to recommend any one thing of Knight Sports, it would have to be the boost controller EBS III. It is made specifically for our cars and also controls the sequential turbo switch. It is supposedly quite easy to install, about the same as a boost guage and turbo timer with a good harness together. Those of you who have gone non-sequential can possibly get a version without the SCA (sequential control adapter). Remember never to set ANY boost controller over 0.75 kg/cm2 (~10.8) without a fuel cut defender and never above 0.80 without extra fuel management. Being a conservative company, Knight Sports never recommends setting boost above 0.90K (~13.0) even with their chip, though the chip has fuel maps for up to 1.0K (~14.5 depending on who you talk to).

Date: Wed, 24 Dec 1997 23:49:48 -1000 (HST)
From: F8LDZZ

(It) bumps up the fuel maps a bit but not any significant change. SFC stands for "Simple Fuel Computer". What were you expecting? All it does is intercept the signal coming from the MAF sensor and modifies that signal going back into the ECU. From there, injector pulsewidths can be changed.

Yup, APEX'i only makes the SFC in a single, universal version. This unit will work on almost all turbocharged vehicles coming out of Japan. You can change the configuration by the single knob and DIP switch bank in the rear of the unit.

APEX'i also makes a BFC (I think) that's just a plug in unit with a single rotary knob to adjust rich/lean over the entire RPM band. These are vehicle model specific, and I don't think there is one that applied to any of the RX-7's. This is basically what the HKS PFC F-CON does.

For $400, the only thing that comes close is a Field Hyper-SFC unit. A brand new HKS PFC F-CON costs closer to a US$1,000! Add a GCC II for another US$400, and you got one helluva electronics package with almost no money in your wallet. The APEX'i SFC does what the F-CON and GCC II do combined at a whomping 1/3rd of the price!

APEX'i was started by a bunch of HKS tech engineers that were tired with the high cost of the HKS stuff. If you noticed, a lot of the APEX'i components can be easily compared to a lot of the HKS stuff, especially with electronic units. These guys are as good, if not better due to lower MSRPs, than HKS. Although most US consumers only see the APEX'i Skyline GT-R in ads, they have far more knowledge than with the R32's and R33's...

Personally, I never really liked *ANY* of the US reps for these Japan companies. This includes HKS and Trust/GReddy. I think all of them have no idea what the hell they are selling and info about their own products.

> I would avoid any boost or fuel products from them. The boost
> controller really has a large spike when the turbo switches.

This is too much of a blanket statement. I've seen lots of problems with boost spikes with almost any boost controller system out there for the FD, including the PFS Purple/Gold boxes; do you mean these are a waste of time? I've seen an AVC-R installed in a black FD locally that had rock solid boost levels when under WOT. The good thing about the AVC-R is that it has its own atmospheric sensor, so basically it does what an HKS III ATM does. With the ATM sensor, the EBC has a much more accurate control of the boost levels. I don't know what your experiences with the AVC-R is, but it could be something else that was causing the boost spike. Another side note, the AVC-R has an injection pulsewidth monitor built into it. This might or might not be of use to you.

What makes it more attractive is that you can change almost any part of your fuel system, and you can always go back and retune the SFC, unlike other systems. Changing fuel injectors? Changing fuel pump? Adding a boost inline pump? Adding a boost dependent fuel pressure regulator? No problem, just retune the SFC...

The SFC runs 5 rev levels from 800RPM to 7200RPM in 5 discrete increments. You can dial in +-30% on the fuel injection pulsewidths at each level. Be VERY CAREFUL about dialing in too much at the top end as you might be saturating your fuel injectors.

Installation is very easy, as there are only 5 leads. +12VDC constant, +12VDC ignition, ground, input MAF, output MAF. That's it.

We can dial in the SFC to run real rich (solid 1 red bar on a Cyberdyne AFR gauge) at over 0.9VDC on the O2 sensor if you want. This doesn't really mean much though.

Some of the downsides are that the unit might be (not confirmed) modifying pulsewidths full-time. This means under part throttle conditions, this might not be a good thing. Also, like I stated before, it's very possible to dial over 100% on the top end, since the SFC is designed to go +30% on all RPM levels; you do *NOT* want to saturate your fuel injectors! You might want to keep the unit in a semi-secure are, since it's very easy to change the settings from what you've safely set them for. A friend took his FC to the Mazda dealership, and the idiots there set all his SFC settings to -30%!?!?! Lucky thing he didn't hammer the throttle when he got the car, but the car did bog on him trying to drive out of the dealership lot!

________________

Date: Thu, 25 Dec 1997 03:03:55 -0800
From: "Hung-Jen Hung"

I just double checked the CAR BOY Jan 1995 issue. AFC works only for RX7 FC3S 13BT. You need an AFC II for FD3S 13B-REW.

________________

Date: Wed, 21 Apr 1999 18:01:15 -1000 (HST)
From: F8LDZZ

A'PEXi Super AVC-R

Recommend.  
Advantages:  *closed-loop operation - dedicated boost sensor
             *"self learning" 
             *two boost levels
             *fuel injector pulsewidth monitor + warning
             *1/4-DIN size
             *smallest electronic valve in the industry - easier
              to install versus HKS EVC or GReddy PRofec untis
             *Probably the best EBC out there for the money - street 
              priced under $450 if you look around
Disadvantages:  *pulsewidth LED's not too sharp - kinda floods the entire
                  face
                *takes a few seconds to toggle between the two boost
                  levels
                *not always easy to set the self-learning right
                *boost/vacuum in bar/mmHg

Jim, I think you might want to wait for the new A'PEXi Super AVC-R. It just came out in Japan, and it blows away anything that's out in the US now. The face glows Indiglo blue...it might match your car. :)

________________

From: Azeem (araja@kestelsolutions.com)
Date: February 23, 2000

I've been using the new AVC-R in my FD for a few weeks. Here is what I think about it, plus some notes on its operation.

Summary: The unit works fairly well in my FD. It keeps boost stable, controls the transition spike, controls initial boost spiking, is temperature stable, and is extremely flexible/programmable.

Positives:

• LOTS of features to achieve desired boost levels and boost pattern (see below)
  
  
• Installation is straight forward (but you will need to splice into ECU wires!) FD specific diagrams are included. The instructions to 'T' the precontrol & WG lines to the AVC solenoid work well, and the stock solenoids must be disconnected.
  
  
• Can turn the unit off (feature of any EBC)... I really like 7psi max boost for 'mellow' driving, the lunging forward of the car goes away, great in the rain, extra-safe for motor, etc.
  
  
• Boost pressure on AVC matches my autometer boost gauge EXACTLY
  
  
• Digital display mode and peak-hold feature are very useful.
  
  
• Digital peak-hold readings for boost show true spike values - these appear to be a little higher than what I can see on my boost gauge (needle may not react as fast as the electronics in the unit and no 'hold' feature on gauge)
  
  
• Smooth transition... I have heard that some boost controllers cause a jerky/violent transitions when not at full throttle. The AVC does not do this, it provides smooth boost thru the seq. transition regardless of the throttle position.
  
  
• Boost does not vary as temp varies (I have not been able to test with very hot or very cold temp's yet, but boost is constant 80deg to 40deg.
  
  
• Very cool 'hi-res' display
  
  
• Amber lighting in AVC buttons matches FD dash lighting exactly.

Negatives:

• While the unit works fairly well on the FD, it is not perfect... I will sometimes get 0.5-0.8psi spikes at transition; and once in a while the 2nd turbo wont build full boost - 10psi instead of 12psi (but this may be an issue with my car, not the AVC.)
  
  
• Tuning is very time-consuming; basic tuning takes a few hours, but getting everything dialed in perfectly will take DAYS... once the unit is set, however, it does not require constant tweaking.
  
  
• Self-learn mode does not work well on my FD (see below)
  
  
• Documentation needs improvement... the installation & basic tuning sections are ok, but the manual does not even explain what the 'advanced' features do, much less how to set them up... I spent a few hours on the phone with Apex tech. support to really figure out how everything works...
  
  
• No way to vary boost in different gears -- the initial spike reduction feature is setup per gear, so can be confused as a 'sustained boost' setting per gear, but it is not.
  
  
• No good place to mount 'head-unit', especially on touring model. (Editor's Note: Rick S. of Houston mounted his on ne of those cell phone flexible stalks, on the passsenger side - looks like it works great.)
  
  
• Graphing displays are not very useful, they are usable for seeing general fluctuations and spikes, but you cannot determine exactly how big the fluctuation is (there aren't any numbers on x-axis)...
  
  
• Only the graphing modes can be 'recorded', not digital values.
  
  
• Clicking of solenoid if LOUD... really only a problem in low-speed, around-town driving, and I am getting used to it.
  
  
• Display is in metric.

• The unit allows setting desired boost levels in increments of 500RPMs As a result, it is possible to run one or more boost levels on the first turbo, different boost level(s) on the second turbo, and to dial-out the transition spike.
  
  
• The unit also has the ability to dial out initial spiking, i.e. quick no-throttle to full-throttle at 3K. This feature reduces the max boost a little, when going from negative pressure to boost... so it has the desired affect of reducing the initial spike, but not impacting sustained boost. The amount of initial spike reduction can be set differently per gear.
  
  
• The unit has its own pressure sensor (independent of the control solenoid) reading manifold pressure. Using this pressure sensor, the unit provides two feedback routines which keep the boost under control. One is a 'generalized' routine, working all the time, making small real-time corrections to keep actual boost at desired boost (basic closed loop operation). It can not correct large mismatches between actual and desired boost, but will keep boost stable once the unit is tuned properly. This routine works well in my FD.

  The other routine is a 'learn-mode' which helps in programming the unit. It updates your programmed values (duty cycle numbers), at setup time or on an on-going basis, again trying to keep actual boost at desired boost. This routine does NOT work well with my FD. It kept trying to dial out my transition spike corrections, thus dialing the spike back in. The learn mode can be turned on/off per gear... I have it turned off in all gears - so I had to setup the duty-cycle numbers for all of the 500rpm values myself.

Finally, for more info on the unit: