Right, back to how my 4A-GE controls its idle speed. Sorry, this will be a long post.
The fuel injection computer reads several sensors - but primarily the Manifold Absolute Pressure sensor. The more you open the throttle, the more the engine can breath and you get less vacuum in the inlet manifold. So the management computer can chuck fuel in based (initially) on MAP sensor readings.
This means that, as there is no Mass Airflow Sensor, it doesn’t care where the air is getting into the plenum. It can be through the throttle, through an air bleed, or through a failed gasket, the computer will fuel it appropriately.
With the throttle closed completely no air passes it. There is an air bleed adjusted with a screw in the top of the throttle body that sets the hot idle speed.
As an addition a solenoid valve opens when something with high electrical load is turned on. That adds a couple of hundred revs to stop it chugging when the head lights are on. This is the system I thought was broken on my car but it’s working perfectly. It’s just that I always have the cabin fan on and that’s one of the things that opens the solenoid.
That leaves the problem of a cold start. There are two aspects to this and they don’t talk to each other.
First you need additional fuel. There is a 5th injector in the plenum for that. A timer / temperature sensor in a coolant path on the end of the cylinder head controls it. Essentially if the engine is cold the 5th injector is turned on and it turns off when the engine warms up or after a certain amount of time. I expect it’s a bimetal strip with a heater round it. Now on my car (and I think many 4A-GEs) the car stank of fuel when it was cold. It must have been shockingly rich. So I’ve unplugged the plug on the 5th injector and it starts just fine without it. Maybe if the ambient temp got way below zero I might need it but, in general, in the south of England it doesn’t seem to do anything useful.
Secondly you need a fast idle. That’s done by the Auxiliary Air Valve that most 4A-GEs have problems with. It’s a wax pellet in a chamber that cooling water passes through. It expands as it gets hot. That pushes a valve closed in another air bleed round the throttle plate. Nothing sets the cold idle speed. It’s not adjustable as such. It’s just set by the amount of air that happens to pass through the AAV.
When the engine warms up, the valve closes and no air passes. Then your idle speed is set by that hot idle screw I mentioned earlier.
The problem with the AAV is that after 30 years it’s generally too far open when it’s cold and won’t close fully. Now being too far open when it’s cold just leads to a slightly high idle speed. Typically about 2200 revs. Part of that might be that cars with carbs and a choke button tended to idle fairly fast so that’s what they made this do. It’s only fairly recently, where the idle is monitored by the computer that cold idle speeds have fallen.
The real problem is that the AAV doesn’t close fully. You compensate for the air leakage by screwing the hot idle screw down a bit more. But then as the engine temp fluctuates, like after you’ve been spanking it, the idle speed also alters as the AAV shifts. It’s annoying. Usually to avoid it stalling when it’s really hot you wind up with a normal idle speed of around 1200 revs. I’ll say it again. It’s annoying. Really annoying. I guess when it gets really bad you wind up with the hot idle screw hard down and the hot idle it totally dependant on the AAV with no way to control it.
So what to do? Well the usual fix it to convert it to a ‘manual choke’. You need to defeat the AAV, normally with a tyre valve cap jammed in the feed hole in the throttle body. As I had it apart anyway I just made a new gasket with the hole blocked.
Next you need to create an air bleed that’s controllable from a switch inside the car. The best way is to use a Vacuum Switching Valve off the MK2. It has a screw in the top so you can adjust the amount of air it passes and, hence, the idle speed.
The other route is to use something like water solenoid that you’d find on the inlet of a washing machine. That’s what I’m doing for the moment. Of course you can’t adjust it. But so long as it flows too much I can always add a tap or just nip the hose to reduce the idle speed.
Either of these systems is either open or closed. There is no slowly reducing the idle speed. It’s fast or it isn’t.
In the traditional version you have a switch on the dash to operate the solenoid. Apparently you only need to run the fast idle for 30 seconds or so. You just turn it off when you get to the end of the road.
A variation of this is that you use a timer so you don’t have to remember to turn it off. I think this is how
jimi has his set up. By all accounts this ‘manual choke with timer’ is one of the best mods you can do to a 4A-GE.
So obviously I’m not going to do that… I want it to be fully automatic so I’ve bought a temperature sensor module. Hopefully I’ll strap the sensor to the the pipework feeding the waxstat and set an appropriate temp on the module and I’ll have created an electronic version of the existing set up.
Wish me luck!
Some final thoughts.
Why didn’t they do the cold start as a function of the engine management? Well I’m guessing they were running out of processing power in the ’80s microprocessor for one. But also there was probably no valve that could control the idle speed variably at the time.
Today you monitor the engine speed and control it with a variable air bleed. For cold starting you could richen the mixture by opening the injectors a little longer and increase the idle speed by opening the air bleed a little more.
Could I add computer controlled idle speed to an old 4A-GE?
It’d be an interesting project. I reckon an Arduino could do the computing. I’d need a feed of the tacho to know the engine speed. A feed of the the throttle ‘off normal’ switch to stop it trying to operate when you were driving, and a feed of the coolant temp sensor. I could dump the idle up valve because alternator load wouldn’t matter. It’d be compensated for automatically.
The problem is trying to find a suitable variable air bleed. I’ve looked and most of them are designed to screw into or onto the throttle body. I’d need a stand alone version and I can’t find one. That’s why I was curious about the Bosch one but it isn’t electronically controllable. The other (really neat) possibility would be to buy a valve and machine a housing to accept it that would screw to the bottom of the throttle body in place of the wax stat. But I can’t because, well, not @johnnybravo .
Hmmm. It’s not like I don’t have a massive pile of other things to play with anyway.
James
