lean idle, rich wot, NO boost leak. any thoughts?

[Supra mn]

Nick M:
The idle seems too lean, which may not necessarily be a problem, but it definitely bugs me, when before the upgrades it idled in the 15s.
WOT is extremely rich and seems to be making me lose power. Seems to, I could be wrong.

You need to adjust the deadtime compensation to match the RC550s. Unfortunately, you can't adjust it since its burned into the firmware of the ECU. Have you measured Vf at idle. It will tell you what the ECU is doing to compensate.

Vf at idle was 3.5v, revved it up to 2500 and let off and it idled at 5v.
Also, they're rx7 denso 550s, purple top, part number x-2020. If that makes a difference.

And ATL88:
I removed the pcv pipe running above the valve covers and flushed all the gunk out. Replaced all those hoses, valve cover hoses have hose clamps on them. The only old hose is the one from the block to the valve cover. Injector O-rings are new. But it could be intake mani gaskets.

 

[Nick M]

Nick M:
The idle seems too lean,

It is. Turn the screw back in some. Full load can be 9.7-10.5, depending on conditions. That is the goal because the car does not have forged pistons and to control detonation. And yes, 9.8:1 will make less power than 10.5:1.

 

[3p141592654]

This might seem counter-intuitive, but the lean idle and rich open-loop are related.

Ironically, your attempt to lean things out by backing out the screw actually makes things worse at full throttle. The ECU is getting a very low air flow signal at idle, which means the basic fuel model will compute a very low injector duty cycle. The ecu compensates by maxing out the trim variable so as to add much fuel as it can (Vf=5V). This maxed out trim variable then gets used open loop as well, and you have way too much fuel at the high end where the air-bypass screw has very little affect. You need to get back to a stoichiometric mixture (14.7) at idle, it needs to be cross-counting on the O2 sensor. So that screw needs to be closed down, and adjust the fuel pressure to compensate. The AFM is the primary variable in the fuel calculation, so a screwed up air flow signal along with built-in errors in the deadtime compensation for your non-stock injectors, means you have horribly messed up the input to your fuel calcs. Anytime the idle is not stoich, nothing is going to work right.

 

[Supra mn]

This might seem counter-intuitive, but the lean idle and rich open-loop are related.

Ironically, your attempt to lean things out by backing out the screw actually makes things worse at full throttle. The ECU is getting a very low air flow signal at idle, which means the basic fuel model will compute a very low injector duty cycle. The ecu compensates by maxing out the trim variable so as to add much fuel as it can (Vf=5V). This maxed out trim variable then gets used open loop as well, and you have way too much fuel at the high end where the air-bypass screw has very little affect. You need to get back to a stoichiometric mixture (14.7) at idle, it needs to be cross-counting on the O2 sensor. So that screw needs to be closed down, and adjust the fuel pressure to compensate. The AFM is the primary variable in the fuel calculation, so a screwed up air flow signal along with built-in errors in the deadtime compensation for your non-stock injectors, means you have horribly messed up the input to your fuel calcs. Anytime the idle is not stoich, nothing is going to work right.

The screw on the AFM is completely closed. Adjusting fuel pressure makes almost no difference in idle. I had it at 23 psi, 30, 33, and even tried 40 to see if it would richen it up at all. The richest it idled at was high 16s to low 17s.
Could the air filter adapter affect how much air is measured, since it reads by the way the air flows through. So if the adapter is blocking any of the honeycomb, it would read incorrectly, messing up how the ECU interprets it? I have a feeling i'm completely wrong there though.
I didn't have any problems with the stock afm, but the honeycomb on the lex afm is larger.
Also, I removed the negative on the battery to reset the ECU and all stored codes, calculations, etc. This didn't change idle either.

 

[3p141592654]

Measure the frequency of the Lex AFM KS signal at warm idle (650 rpm). Should be 15-20 Hz if its working normally. When you say the screw is closed, do you have a special one that completely seals off the bypass slot, or it is just in as far as it will go.

Turbulence on the input of the AFM will cause havoc, but usually the idle will be very erratic. Doesn't fit your symptoms. Injector flow goes as the square root of pressure, so you have to crank it up a lot to make significant changes.

 

[3p141592654]

What are the chances your injectors are high impedance? That would explain everything if they are.

 

[Nick M]

Also, I removed the negative on the battery to reset the ECU and all stored codes, calculations, etc. This didn't change idle either.

This is another "thing" that needs to go away.

 

[Supra mn]

Measure the frequency of the Lex AFM KS signal at warm idle (650 rpm). Should be 15-20 Hz if its working normally. When you say the screw is closed, do you have a special one that completely seals off the bypass slot, or it is just in as far as it will go.

Turbulence on the input of the AFM will cause havoc, but usually the idle will be very erratic. Doesn't fit your symptoms. Injector flow goes as the square root of pressure, so you have to crank it up a lot to make significant changes.

I checked the afm per the tsrm about a week ago. Don't remember the numbers, but it checked out fine for the temperature.
The injectors are high impedance, but there have been quite a few people running them that say they work with no problem:
http://www.supramania.com/forums/showthread.php?116683-RX-7-550cc-Injector-Clarification/page3

 

[3p141592654]

Let me guess, you also have the injector resistor pack installed as well.

By the way, testing per the TSRM doesn't tell you anything useful about the air flow signal. Its equivalent to measuring the oil pressure sensor and then saying the temp sensor is working.

 

[Nick M]

The injectors are high impedance, but there have been quite a few people running them that say they work with no problem:

They do not fire nearly as fast with the low current. Quite a few say nothing is wrong on SM, when it is. You can not have the screw all the way out. Well, you can, but you will have some "backrap" and rumble from the idle.

 

[7M4EVR]

This is another "thing" that needs to go away.

Can u explain? Doesn't resetting the ecu clear the learned fuel map adjustments? I have heard people claim their car runs worse after a reset until the ecu can learn and make adjustments.

 

[jetjock]

It also resets the learned idle value. How different the engine runs thereafter depends on how screwed up it is. I think what Nick's saying is, generally speaking, resetting the ECU will not fix problems and if the engine in question does run differently it was a POS to begin with. Remember, the cars didn't have to be "trained" when new...

 

[3p141592654]

16 bytes of learned data can only do so much....

 

[Nick M]

resetting the ECU will not fix problems ..

Exactly. It isn't a Microsoft program that needs a "reboot". I realize pulling the fuse is only 30 seconds of their life, but it is not a diagnostic tool.

I reset the computer but it didn't do anything

Actually it did, but it didn't fix the intake leak.

 

[Supra mn]

I reset the ECU to see if it would change anything. People give the advice to do it, I figure it doesn't take long and doesn't cost anything, why not give it a try. It didn't work, as you already know. Now I know, and now I'm reassured.

Clarification, since I wasn't specific enough, and all of the advice I keep getting is that I can't have the lex afm screw all the way out (and by all the way out, I mean backed out to its furthest point, not actually taken out of the afm):
The screw is the original screw, not some reg reimer upgrade screw or anything. It is now (and has been since the original post) all the way turned in. (and by this, I mean screwed as far in as the original screw will go)
I will try and back it out a few turns to see if it changes idle, but I had a live feed camera set up showing my gauge while I was turning the screw in originally, and idle did not dramatically change at any point, and only an overall change of .8 afr over the span of all the way backed out to all the way screwed in at one set fuel pressure level.

As for the injectors, more clarification: The injectors are high impedance, purple top, rx7 injectors, part number 195500-2020. They fit into the stock harness, plugs matched. They were flow tested and cleaned prior, all passing and within limits of eachother. The resistor pack is still installed.
Many people on the forum have used them "without problems." As in, many people have used them without extremely lean idle and extremely rich WOT, or any other majorly noticeable problems that would lead them to believe the injectors would not work.

3p: You said I need to get the idle to stoich, and it needs to be cross-counting on the O2 sensor. Could a bad turbo elbow gasket leak in air, making the idle seem lean to the O2/wideband sensor? Not sure if my gasket is leaking or not, but its a possibility.

 

[Dirgle]

What are the chances your injectors are high impedance? That would explain everything if they are.

Let me guess, you also have the injector resistor pack installed as well.

.....As for the injectors, more clarification: The injectors are high impedance, purple top, rx7 injectors, part number 195500-2020. They fit into the stock harness, plugs matched. They were flow tested and cleaned prior, all passing and within limits of eachother. The resistor pack is still installed.

You do not want the resistor pack in line as the injectors are already high impedance. You only need the resistor pack for low impedance injectors.

 

[Nick M]

I don't know if Suprasport is still making the Lex replica housing. If they are, pick one up.

 

[ATL88Supra]

if i was you i would take off all of your mods get it back to stock then repeat one mod at a time

 

[3p141592654]

He just needs to fix one obvious problem, and a second potential problem. He's running his high impedance injectors at about 8V because he still has the resistor pack installed. Its insane and shows a total lack of understanding. The deadtime is about 280% more at 8V than 14V. The high impedance injectors are already inferior for deadtime, and this just makes it so much worse.

Second, he may have an intake leak adding to his woes. That needs to be checked but first fix the dumb stuff.

 

[Supra mn]

You guys are geniuses.
Found a plug at a scrapyard incase this didn't work and needed to still use my resistor pack, soldered the 4 wires together, plugged it into the stock harness.

Idles perfect. Stronger vacuum. No more "misfire" sound. Not sure about normal driving or WOT since the first snow just hit and the roads are salty as all hell.
Fuel pressure set to 28 psi with the vacuum hose on, lex afm screw as far in as it'll go. So I'm not sure where fuel cut will be, but I'm sure it'll be high enough.

I'm not sure how people run these same injectors without any modifications, maybe they have an safc to tune it out.

Either way, this seemed to have fixed my problem. I'm just learning as I go. Thanks for all the info and help guys!