Runs great cold until you turn it off. If you start it back up constant 11.8ish AFR

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#1
Hey guys I've got a mk3 with a semi-fresh 7m

Here is the setup:
MAFT-PRO piggyback speed density conversion
89 turbo ecu
Rc 550s
Walboro 255 pump
Cx racing 67mm t4 turbo
Innovate LC-1 wideband and simulated narrowband for ecu
Stock ignition system

Alright so this ordeal is really strange. On the cars first start of the day it will run fantastic. Any change I make to the tune on the maft will immediately go into effect and all is well in the world. If I drive the car around and get it completely warm nothing changes.

If I turn the car off and try to start it back up it runs like absolute ass. It is trying to drown itself in fuel. At idle and at every load condition it runs 11.4 - 12.4 afrs. If I come to a stop it really wants to die and usually will. Nothing I change in the tune has any effect. Sometimes if I rip like 10% out of the mainscale it will lean out for a second then go straight back to super rich. I've replaced coolant temp sensors and adjusted everything to within spec. All sensors give good reasonable readings on the maft. Nothing is out of the ordinary except the afrs. I can tell they are accurate because of how terrible it runs and the suit gathering around my exaust. It's got to be some wacky electrical issue but I dont know what to check.

All of this is 100% repeatable by the way. It is not a sparratic issue

I'd appreciate any sort of input on this. Thanks guys!
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#3
Some ideas / questions:

Are you powering up the maft- pro first via a delay before starting?
It's much slower to boot than the stock ECU.

Are you getting any codes?

Have you been through this thread at all?
http://www.supramania.com/forum/threads/afr-problems.90541/

Yes the maft boots before the ecu when I turn the key to ignition

I'm currently getting no codes. I had unrelated codes recently so I know the celis functional.

I have been through that forum several times but it doesnt quite go through what I'm experiencing. Whenever it goes into this condition the afrs are always around or a little below 12 under any load (except boost it seems as though it correctly follows my tune under boost). During throttle transition it will lean out but only to like 13ish max
 

Piratetip

Far From Maddening Crowds
Staff member
Super Moderator
Authorized Seller
Joined
Dec 30, 2005
Messages
1,620
Likes
142
Location
MKE, WI
#4
Hmm..ok.

The only ideas I have are:

It's input coolant temperature to ECU is much colder than it should be.
Same for IAT temp.

Or

The O2 sensor input to the ECU is far too lean than it should be.

Have you monitored these signals to clarify they are in spec when the engine is hot after restart?
Compare them against the cold start / warm running engine.

Could be other sensors also.

What are your VF readings? Learned vs. O2 feedback modes.
Compare the readings cold vs hot start.
 

Piratetip

Far From Maddening Crowds
Staff member
Super Moderator
Authorized Seller
Joined
Dec 30, 2005
Messages
1,620
Likes
142
Location
MKE, WI
#5
Sounds like your open loop tuning is working.
But mid load tuning is not, this sounds pretty common with maft-pro though fighting against you.
Closed loop tuning with O2 feedback is probably giving you issues.
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#6
Sounds like your open loop tuning is working.
But mid load tuning is not, this sounds pretty common with maft-pro though fighting against you.
Closed loop tuning with O2 feedback is probably giving you issues.
Yeah that is most likely the case. I've seen others struggle with similar issues

The confusing thing to me is how perfectly it runs from a cold start. It will run fantastic all day as long as you dont turn it off.
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#7
Just drove the car again with the same results. I tried hooking up another coolant temp sensor in free air to see if it was based on the ecus interpretation of the coolant temp. It had. No effect on my symptoms
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#8
Here is another update on the subject of closed loop tuning. My vf is constantly reading between 1.7 and 1.9 when in the extremely rich condition. I've yet to measure when it is running good
 

Piratetip

Far From Maddening Crowds
Staff member
Super Moderator
Authorized Seller
Joined
Dec 30, 2005
Messages
1,620
Likes
142
Location
MKE, WI
#9
Which VF?
Learned or O2 feedback?

You are going to want to measure both.

Do you know how to access both?
 

Piratetip

Far From Maddening Crowds
Staff member
Super Moderator
Authorized Seller
Joined
Dec 30, 2005
Messages
1,620
Likes
142
Location
MKE, WI
#11
So that's the learned then. To see the O2 feedback you have to connect pins E1 and T.
It's backwards from learned values. 5v rich or 0v lean. Listed as 2500rpn.

The learned value is what you will want to use anyway.
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#12
So that's the learned then. To see the O2 feedback you have to connect pins E1 and T.
It's backwards from learned values. 5v rich or 0v lean. Listed as 2500rpn.

The learned value is what you will want to use anyway.
Alright I did a few things and the car is being steady for now. I recalibrated the LC-1 and reset the ecu just for giggles. After that the issue still came up. I decided to mess around in the maft just to see if anything would get better. I switched to ve table 1 and adjusted the user tune values to compensate. I'm not sure why but for now this seems to have resolved my issue. The VE reading never changed, it still only ever measures 0v or 1.5-1.9. With the car completely warm I shut it off and turned it back on. Unlike last time, nothing changed and the car was running good just as before. I'll update this with more info tommorow after the car is completely cooled down
 

Piratetip

Far From Maddening Crowds
Staff member
Super Moderator
Authorized Seller
Joined
Dec 30, 2005
Messages
1,620
Likes
142
Location
MKE, WI
#13
Sounds like you might be getting somewhere.
Though you probably just reset the past learned values from the ECU.
It might revert back after some driving, but you will see.

Just for reference but you probably know this:

VF values.
0v - ECU is decreasing fuel by 11-20%
1.25v - ECU is decreasing fuel by 4-10%
2.5v - ECU is Normal +/- 3%
3.75v - ECU is increasing fuel by 4-10%
5v - ECU is increasing fuel by 11-20%

So based on your current VF readings the ECU is receiving an input telling it that it's still lean.

Are you able to measure and adjust the simulated narrowband output to the ECU?
Is it operating normally?
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#14
Sounds like you might be getting somewhere.
Though you probably just reset the past learned values from the ECU.
It might revert back after some driving, but you will see.

Just for reference but you probably know this:

VF values.
0v - ECU is decreasing fuel by 11-20%
1.25v - ECU is decreasing fuel by 4-10%
2.5v - ECU is Normal +/- 3%
3.75v - ECU is increasing fuel by 4-10%
5v - ECU is increasing fuel by 11-20%

So based on your current VF readings the ECU is receiving an input telling it that it's still lean.

Are you able to measure and adjust the simulated narrowband output to the ECU?
Is it operating normally?
I'm not currently aware of how to measure the narrowband output but I'm sure there is a way. I'll look into it more tommorow after I drive the car and see how it acts
 

Piratetip

Far From Maddening Crowds
Staff member
Super Moderator
Authorized Seller
Joined
Dec 30, 2005
Messages
1,620
Likes
142
Location
MKE, WI
#15
I would be interested to find out if the narrowband also is able to simulate the O2 cross counts the stock ECU looks for during closed loop operation.

You could just backprobe the narrowband output signal from the wideband controller with a good Fluke DVOM.
That will tell you.
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#16
I would be interested to find out if the narrowband also is able to simulate the O2 cross counts the stock ECU looks for during closed loop operation.

You could just backprobe the narrowband output signal from the wideband controller with a good Fluke DVOM.
That will tell you.
That's starting to get a little above my pay grade. I dont have that sort of equipment at the moment either unfortunately. Im limited to educated guess diagnosis at the moment (emphasis on limited). A new development is that it seems to be happening more randomly now. It still almost always happens after a hot start but it has also started doing it on extended drives (anything over 10 minutes).

It is extremely strange and frustrating. I know that others have used the simulated narrowband with no issue and some ever run with only a wideband and no stock o2. I've tried this config aswell with no success. The exact problem still persists
 

Bmettie

Member
Authorized Seller
Joined
Apr 27, 2010
Messages
200
Likes
2
Location
Florida
#17
What resistor are you using for the IAT signal to the stock ECU?
Did you do the HAC sensor mod?
Both will change tables in the stock ECU

I ran a MAF-T Pro for years
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#18
What resistor are you using for the IAT signal to the stock ECU?
Did you do the HAC sensor mod?
Both will change tables in the stock ECU

I ran a MAF-T Pro for years
I bought the car with the maft installed. I'll have to investigate a bit to give solid answers. Even if these were out of wack, would it cause the symptoms I described? Runs great then it turns on me and the tccs wont let go of fuel control
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#19
What resistor are you using for the IAT signal to the stock ECU?
Did you do the HAC sensor mod?
Both will change tables in the stock ECU

I ran a MAF-T Pro for years
Also can you specify what IAT resistor you are talking about? I've not read anything that suggests the need for one
 

Bmettie

Member
Authorized Seller
Joined
Apr 27, 2010
Messages
200
Likes
2
Location
Florida
#20
I bought the car with the maft installed. I'll have to investigate a bit to give solid answers. Even if these were out of wack, would it cause the symptoms I described? Runs great then it turns on me and the tccs wont let go of fuel control
I assumed you did the install.
Has it always done this for you? If i were you i'd check all wiring, MAFT Pro is wiring specific. Check the simulated narrowband output settings and wiring.

Also can you specify what IAT resistor you are talking about? I've not read anything that suggests the need for one

Lucky i still have these bookmarked...I'm sure you've been to fullthrottletech.com already. They are old threads at fullthrottle but might help you out (set the "from the" display options to "beginning" on every page you go to)

Tons of info buried in here http://www.turboninjas.com/supra/maft/

resistor info
http://www.turboninjas.com/supra/maft/8 - Vehicle Specific Info/MK3 Supra/MAFT Pro and Mk3 Supra Error Codes (CRE).htm


Use the VF signal as a "window" into what the stock ECU is doing. Your ECU is pulling fuel based on closed loop feedback from the narrowband input at the ECU (in your case this is coming from the simulated narrowband output from the maft). The ECU will ALWAYS try to bring fueling back to 14.7 when in closed loop.
 
Last edited:
Likes: debrucer

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#21
I assumed you did the install.
Has it always done this for you? If i were you i'd check all wiring, MAFT Pro is wiring specific. Check the simulated narrowband output settings and wiring.




Lucky i still have these bookmarked...I'm sure you've been to fullthrottletech.com already. They are old threads at fullthrottle but might help you out (set the "from the" display options to "beginning" on every page you go to)

Tons of info buried in here http://www.turboninjas.com/supra/maft/

resistor info
http://www.turboninjas.com/supra/maft/8 - Vehicle Specific Info/MK3 Supra/MAFT Pro and Mk3 Supra Error Codes (CRE).htm


Use the VF signal as a "window" into what the stock ECU is doing. Your ECU is pulling fuel based on closed loop feedback from the narrowband input at the ECU (in your case this is coming from the simulated narrowband output from the maft). The ECU will ALWAYS try to bring fueling back to 14.7 when in closed loop.
Thanks that was all very useful I've been studying up on all the info.

My maft is currently setup to use the integrated voltage outputs instead of resistor networks for the air temp and hac signals. I dont need the outputs for anything else so I see no reason to change them at the moment. I dont mind the cel as long as it isnt causing any symptoms. I'll be messing with the car today and verifying the wiring situation again. It seems to me that wiring is has to be the culprit
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#22
Alrighty so for the sake of experimentation I unplugged the o2 sim wire and reset the ecu this seems to have completely eliminated the main issue I was having. The symptom of afrs diving is no longer present.

I have new, more typical maft issues now

After the reset I warmed up the car and tuned idle and some tip in while sitting in my driveway. The car idled smoothly at 800rpm and consistantly 14.7-15afr. VF was reading a complete 0.00



After I shut the car down I messed around in the garage then came back to restart the car. Upon restarting the car wanted to run rich 12-13 at idle. VF was reading between 1.90 and 2.00 shifting very rapidly. I could tune around it by pulling out a bunch of fuel but that causes a mess whenever you restart the car or tip in throttle because it wants to go really lean.

The car at this point is sending absolutely no O2 signal to the ecu. The wire ends right at the pin so I no it's not receiving any interference.

The car ran beautifully with VF at 0.00 (assuming it was in open loop bc of this reading) but on restarts it goes to around 2.00 and the car does not want to behave.

Any suggestions as to what is causing my Vf to sit at this reading when the ecu has no inputs?
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#23
**BIG DEVELOPMENT**

I was trying to narrow down the cause of my car entering closed loop and it seems as though it is coolant temp related.

I reset everything and ran the car until it warmed up. It ran great as usual when displaying a 0.00 VF on aux1. When I keyed the car off then turned it back on it started and ran good for a few seconds before VF shot straight to 2.0. Once I got that reading the car ran like poorly and idle was much less stable, a result of quite rich AFRs.

With the car still warm I unplugged the coolant temp sensor and plugged in a spare unit, leaving it outside of the engine to produce ambient air temp readings. I started the car in this config and it ran like garbage, but VF stayed at 0.00

This is where I am stuck at the moment. I'm unaware of a way to modify the coolant temp circut in a way that would force open loop without giving false temp readings that make the car run poorly. Any suggestions would be greatly appreciated
 

MK3Supra17

Member
Authorized Seller
Joined
Jun 8, 2011
Messages
37
Likes
1
Location
Toronto
#24
Reconnect the IAT, save the V-Out for something useful like ETPS.

Why do you want the car to run in open loop all the time? Leave the stock O2 sensor in place, and use your wideband only for monitoring. In all honesty ignore the wideband for anything other than WOT. VF is the guy you are looking to make happy here. You really cant have a 0% (2v to 2.5v) correction in VF(fuel trims) and your afrs not be around 14.7, if that's the case your wideband is junk or theres something seriously going on with the stock ecu.

Adjust your mainscale to satisfy VF at idle to get 2 to 2.5v and then tune the rest of it. VF should be displayed on either AUX 1 or AUX 2 depending on how it was installed. If VF cannot be satisfied within a -20% to -25% correction(for 550s), youre looking for mechanical faults.



Despite what a lot of posts will say about re keying the car, i have never seen a change in closed loop operation or cold vs hot start strategies. If its setup right once in closed loop regardless of temperature you should have your nice 2v to 2.5v VF reading.

The maft pro or any afc for that matter has way less of an impact on the ecu than people think. All it does is scale your airflow signal, so id say dont over think too much into the maft.
 
Likes: debrucer

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#25
Reconnect the IAT, save the V-Out for something useful like ETPS.

Why do you want the car to run in open loop all the time? Leave the stock O2 sensor in place, and use your wideband only for monitoring. In all honesty ignore the wideband for anything other than WOT. VF is the guy you are looking to make happy here. You really cant have a 0% (2v to 2.5v) correction in VF(fuel trims) and your afrs not be around 14.7, if that's the case your wideband is junk or theres something seriously going on with the stock ecu.

Adjust your mainscale to satisfy VF at idle to get 2 to 2.5v and then tune the rest of it. VF should be displayed on either AUX 1 or AUX 2 depending on how it was installed. If VF cannot be satisfied within a -20% to -25% correction(for 550s), youre looking for mechanical faults.



Despite what a lot of posts will say about re keying the car, i have never seen a change in closed loop operation or cold vs hot start strategies. If its setup right once in closed loop regardless of temperature you should have your nice 2v to 2.5v VF reading.

The maft pro or any afc for that matter has way less of an impact on the ecu than people think. All it does is scale your airflow signal, so id say dont over think too much into the maft.
I really appreciate the input! I will most likely end up reconnecting the stock o2 i was only interested with experimenting in part throttle tuning and other capabilities of the maft pro.

Here is where I'd like to further clarify the issue I'm having: the 1.90 to 2.0 VF is entirely unaffected by mainscale or any other adjustment. No matter how much fuel I add or subtract, the VF will remain the same. I have never had any reading other than 1.90-2.00. The tuning adjustments DO affect the afr and I can tell because of how the car sounds and the smoothness of the idle when I get between 14.0 and 15. I have no clue what would cause this reading and set of symptoms. I'm determined to find out out of spite at this point lol. It could all of course be an ecu issue but I'll reconnect the stock o2 before I worry about that
 

MK3Supra17

Member
Authorized Seller
Joined
Jun 8, 2011
Messages
37
Likes
1
Location
Toronto
#26
When it comes to ECUs engineers will add in backup strategies known as substituted values just to get the car to run. So im going to stress leaving everything hooked up the way Toyota designed it.

Put everything back besides your airflow signals and what not that go to and from the maft play around with your fuel pressure and main scale to get a 2v VF. Should be around 40 to 43psi(IIRC) and -20-25%,respectfully. With those settings the car should be running like it did with the stock AFM and 440s. What are you running for airflow signal? speed density, afm or maf sensor?
 

Jayybrue

New Member
Joined
Sep 1, 2019
Messages
17
Likes
1
#27
When it comes to ECUs engineers will add in backup strategies known as substituted values just to get the car to run. So im going to stress leaving everything hooked up the way Toyota designed it.

Put everything back besides your airflow signals and what not that go to and from the maft play around with your fuel pressure and main scale to get a 2v VF. Should be around 40 to 43psi(IIRC) and -20-25%,respectfully. With those settings the car should be running like it did with the stock AFM and 440s. What are you running for airflow signal? speed density, afm or maf sensor?
I'm running speed density

I'll reconfigure everything to a reasonable setup like you mentioned and report back
 
Joined
Sep 1, 2019
Messages
17
Likes
1
#28
Stock o2 has been plugged in and it runs as good as youd expect. Its obviously still learning but afrs at low and mid load are on the money (mids are actually a lil lean 14-14.6). Now it seems as though the maft isnt doing its job up top at high load. Boost leans the car out. It's possible I just need to add more fuel in the user tunes or ve table. I'll be doing more tuning later today to verify
 
Likes: debrucer