The following information has been reproduced from a post to a thread on idle instability issues on skylinesdownunder.co.nz forums by 'phantom'. All credit for this information goes to 'phantom' and thanks for his permission to reproduce here.

I'm having some technical difficulties with my digital camera, in particular focusing at close distances! As such my attempts to get a descent view of the AAC spool etc. wasn't to good. I have instead done an auto cad drawing of the R33 AAC valve. Bear in mind that this is a basic drawing for conceptual purposes only!! It will however give you a good understanding on how it works. Additionally I've also drawn a representation of my new spool.

	What does it look like? - AAC Valve

As states this drawing is a conceptual interpretation of the inner workings of the AAC and spool. The point of wear on these things occurs on the bronze bush insert on the right and the bronze orifice insert on the left. A loose spring will not pre load the spool at all and it will probably float around causing irregular idle conditions. When wear has occurred leakage through the bronze orifice ensues even before the taper section of the spool is exposed.
This results in an AAC without a significant airflow difference between open and closed and anywhere in between. Additionally, wear in the bronze bush on the right causes air to be sucked through the air supply side, into the cavity enclosed by the spool and the solenoid, and, subsequently through the pressure equalisation hole in the spool all the way through to the other side of the orifice and into the engine. This leakage bypasses the spool altogether. The spool has a hollow steel tube at the centre to support and guide everything.
The reason for it being hollow and having a hole on each side is to equalise the pressure on both sides of the spool such that pressure differences occurring as a result of general engine running don't cause the spool to move. Movement can only be caused by the solenoid plunger movement. This is typical of this type of valve and I have seen similar pressure equalisation techniques in industrial hydraulic and pneumatic proportional valves.

As the bush and orifice would be difficult to manufacture I did the next best thing and made a new spool instead. I placed the spool in a lathe and VERY GENTLY machined off the plastic leaving the hollow central tube. I then made the replacement in three sections, purely out of the difficulty in making such small components with the size of cutting tools I had available. The spring retainer is straightforward and can be very approximate in its sizing. The taper section was then turned down to a diameter measured with an inside micrometer on the bronze orifice hole. I made it slightly larger and polished it to size whilst in the lathe with some emery until it JUST slid into the orifice without binding. This took a few attempts to get right. The bush end does not need to have the taper as per the factory. This is probably just a manufacturing simplification. It was simply a cylindrical section, again referenced to the measurement of the existing bush diameter as measured with the inside micrometer.

I actually made it slightly longer than standard as far as the equalisation hole allowed on that side. The reason being that the additional length would assist in alignment once inserted into the bush, and stop the tendency for vibration to cause further wear. All the sections had a 4 mm diameter axially drilled hole on the lathe. Accuracy is highly important, as with close tolerances misalignment will lead to the spool binding in operation. I then slid the three sections over the hollow tube, carefully positioning the taper to the same position as the original. The cylindrical side placement was not as critical, and simply slid far enough up to the equalisation hole. I secured these onto the hollow tube with super glue. Don't laugh, but super glue is designed for very close fitting surfaces, and since the 4mm hole in the sections made a nice sliding fit over the tube, the super glue worked fantastically.

Placing the new spool back into the body and blocking the relevant holes then blowing through it revealed almost zero leakage when closed. Only once the spool was moved over exposing the taper did significant flow occur....just what the doctor ordered!!

	What does it look like? - Spool

Next a new spring was found and chopped and reshaped to the original length but significantly lighter. I wanted to maximise the movement of the spool with the solenoid force to amplify its effect on engine RPM. As stated in the past the pre load spring retainer is threaded into the body and its depth can be adjusted by turning the retainer. The glue applied by the factory over this was simply milled through on a milling machine until I just contacted the retainer. I then picked out the remaining glue in the retainer phillips head with a scriber. This then allowed me to get a wide flat blade screwdriver onto the retainer and screw it out. The indentation on the retainer is not exactly a philips head but really a cross with a circular centre. The factory obviously has an automated tool used for manufacturer adjustment prior to glue sealing.

Once back together and on the car, the engine was started and allowed to reach temperature. I set the RPM manually via the idle screw to approximately 650 rpm so as to not cause the solenoid to push out the spool trying to compensate. I then unscrewed the retainer until the engine RPM JUST started to increase, then turned it back in about a quarter turn. This ensured that with the given base duty cycle on the solenoid, the resulting force had the spool JUST about expose the taper. This would guarantee that any further increase in duty cycle by the ECU would immediately expose the taper and increase RPM rather than being wasted just sliding the spool whilst still on the straight portion in the orifice.

As I've stated, once this was done, the idle control returned with a vengeance...and has been operating beautifully now for about two weeks straight. I resealed the pre load adjuster with some silicone, as there is slight leakage through the thread. The silicone can be easily pried out in future should I need to readjust or disassemble.

Some may wonder why I bothered with this. Well, this cost me nothing...a new AAC costs $650. In addition I had the advantage of adjustability and therefore tunability to my specific engine. A brand new one would remain unadjustable.

Information derived from 'phantom' at skylinesdownunder.co.nz

Information on the page was last updated 10 July 2003