seriously- ideas to increase idle speed.
 

I know I have talked about this before. But really no answers were available at the time. Maybe there are now? The ideas i have heard are
1- drill small hole into the throttle body disc--how small? anyone done this?
2- install a mechanical hold spot on the accelerator pedal?
3- modify the neutal switch on the transmission?---How?

I REALLY would like to have an idle speed of 900-1K. engine is much smoother in that range. My current idle is ok at 800 or so but it does do a little search when you are stopping(drops to 700 or so for 1-2 secs)
answers/ideas anyone
olddragger

I completely agree with you. Car idles perfectly smoothly at 900 rpms, but once fully warmed up, it slows down and vibratres.

Drilling a hole in the throttle plate would almost certainly work, but could result in turbulent flow at partial throttle. At best it could sound terrible, and at worst, it could completely throw off the ECU's fuel calculations.

I don't think #2 is an option, at least not for me. I have tried to manually (pedally?) hold 900 rpms and it seems like the electronic throttle is either OFF, with an idle of 800, or when activated, opens enough to raise my rpm's to about 1200. I can't maintain anything less than that. Can you?

Altering the neutral switch would be great. My engine speed always stays right at 1000 rpms until the car comes to a complete stop, even when the clutch is in and I am coasting. There has to be a way to fool this into thinking that the car is still rolling.

Do those with standalone ECU's alter their idle speed?

A small hole in the throttle plate does nothing noticable from a turbulent flow standpoint. It's like a small vacuum leak. My RX-7 has such a hole in it's throttleplate.

You need to take a valve and install it on a vacuum line that runs from before the throttle plate to after it. Adjust this valve's airflow. You can make a quick one fairly cheaply from hardware store plumbing parts to test how well it works and what the side effects are.

This has been my biggest bitcha bout the car, although its been better as of late with the new engine mount braces.

I'm willing to even front the R&D costs to come up with a solution, as running a standalone seems to make the idle worse in general unless you go stupid rich with it.

The hole in the throttle plate is one idea, I tried briefly routing air around the Throttle plate per RG suggestion but the car didn't seem to like it. I may play with that somemore.

I like the neutral switch idea. I just don't have the electrical background to fashion up a switch or relay.

One other idea would be to fool the temp sensor at idle to run the car a little richer and high in rpms. maybe a rpm switch combined with a relay?

This whole idle thing has me intrigued . I think the 8's idle is very smooth & see nothing to improve on ...... what is the big deal with it ?
OD saying his idles at 800 does make me wonder if US spec cars are set a tad lower than JDM cars like mine at 830 - 840 (scanalyser reading) .

For the people with an aftermarket intake there is the unused VFAD nipple after the throttle plate.

A hole could be drilled in the resonator attached to the accordian tube to get the air before the throttle plate.

Ok--- now I have something else to tinker with:). Thanks for the idea RG. I have the ports available--I will have to find the type of valve to use--shouldnt be hard.
olddragger

trick the ecu into thinking the car has not come to a complete stop... or wait for a flash tuner. i am suprised rb did not cover this with there flash..

beers :beer:

Just about anything you do will be counteracted by the PCM eventually.
The target RPM is preset and it will do what ever it thinks is needed to keep the RPMs in the set range.
It will pull fuel, close the TB and even retard ignition timing.

I've tried a bypass valve (basically a controlled and metered vacuum leak), a clutch switch bypass, screwing with coolant temp sensor values, forcing airflow meter values up and down and air temp sensor modification.
Nothing creates a permanent change, except the coolant temp adjustment, but that interferes with the fan turn on and makes the idle mix a bit rich. It also ruins the throttle feel at low load and RPM.

Changing the target values in the EEPROM is the only viable option, other than going to a mechanical TB.

BTW - keeping the idle above 900 or so will dramatically raise the CAT temps.

mine doesn't need an idle speed increase, why does yours? :dunno:

but fwiw, the ability to change that preset is not too far away, it's just a piece of software code like all the rest of it ...

Many 8's have a droopy idle. None of the flashes have addressed it.

Mine isn't particularly droopy, but it is an interesting problem to me.

The Cobb AccessPort is still a few months away, but for many people it will not be a viable option - especially those that aren't FI...

MM, since you mentioned the AccessPort, do you think that would be a mod that would go well with my desire to control my ignition timing for higher shots of nitrous? If so, I'll kepp an eye out for it and get it when it comes out. Thanks.

fwiw, it's probably more than a few months ... DNKV's 04' RX-8 idle seems fine to me too :dunno:

I don't know that the AccessPort would help with the nitrous thing since there is no sensor in place on the Renesis to indicate the onset of the spray.
I think you would still want something external.

The AccessPort is physically done. It has been since November. The release is set for April.

MM, I hate to bother you with questions about Cobb stuff but their website is nil on the RX-8 AccessPort. I guess there are no external sensors on the AccessPort to allow trigger voltage(?). BTW, did you get my p.m. asking about help on a plug-and-play ignition controller module for precisely this purpose?

EVERY RX-8 I've ever been in has a MUCH smoother idle than my RX-7 has and I've never wanted to change it.

what about a temp fooler that is rpm dependant? Couldn't you build a device with a rpm switch that fools the temp sensor only below 1000 rpm? all you would need to do is get the temp sensor to register about 155 degrees which implies a idle value of about 850-900. Scotts cooling mod should still kick the fans on when needed, so you wouldn't overhead at idle.

just a a thought, but seems to cover the issues you mentioned.

^^ Yep. That would do it.
Break out the soldering iron.

cool. Do you have an RPM switch, I'm assuming you would know how to wire the switch to trigger the temp fake.

i would build a curcuit to manipilate the tps voltage.. what ever the stock tps voltage is, you need to raise just a little bit at idle. some sort of resistor or potentiameter. ecu sees a higher voltage and will control the throttle body to open. but i would be carefull, you dont want the voltage to get stuck at wot. but the question is does the stock ecu know what to do if it see a tps voltage that is too high. will it just circumvent the tps with a default value and render it useless, and then look at map sensor instead. then throws a code (tps voltage to high)

If you guys can better explain how all the parts work and fit together ie tps temp stuff etc. I can give you a basic mockup of a PCB that will give you the electric output you need where ever it is you need it. Should probably be able to give you part numbers from a cheap and reliable source for resistors, transformers, switches or whatever else you would need to solder something up.

then on second thought, most of you guys can prolly do this quicker and more efficiently than i could

edit: turbosa22c just edited his post with some interesting thoughts... altering voltage would be useless if it just circumvented and throws a CEL as the 'brain' learns

well, the TPS is a pulse width modulation system, so I don't think a simple resistor will do the job, I had that thought, but MM pointed out that wouldn't work.

There is some margin for error with the stock PCM, as I read in a abreivated workshop manual I saw at 7stock. I think the idle rpm can be +/- 100 and still be ok, which means that if you could build a PWM controler it could work.

my idea about triping the temp sensor below a certain idle rpm seems like it would have the fewest issues since the car just thinks the engine needs to warm up more.

Modifying the TPS output will only tell the PCM that you have pressed the gas or that the throttle is more open that it is and it will set up a feedback loop that will yield a new calibration for that sensor until it goes out of range.

What you want to do is have a circuit that reads the TPS output and set a binary "ON" when it sees idle (less than 1v on PCM terminal 1J).
Then, when the switch is set to "ON", and electronic switch across the CTS will increase its resistance to simulate something lower than 176° or so (which is a total resistance of about 300 ohms, which means adding about 60 ohms).

if you could just go to the source and modify the code itself. everyone is just waiting for the cobb access port. the question is will it allow us to do what we want or will it just be another tuning device that will keep us wanting more.

making the computer think its colder is going to make it run richer. it will think its still in warmup and closed loop may not come on.

i dont know, everycar is different, and the 8 is as different as it comes.

We have already figured out what it will do at the desired temp reading.

Yes. It allows you to completely change every single aspect of the OEM system.
I know it for a fact. I played with it myself in November.