2JZ-GTE Single Turbo AEM version 2 support | AEM
11 posts / 0 new
Last post
kingtal0n
Offline
Last seen: 3 years 8 months ago
Joined: 01/31/2015 - 18:00
2JZ-GTE Single Turbo AEM version 2 support

Hello, 

I set my idle per coolant temperature so 197*F is 850rpm. Idle target: 850rpm. I want my stock 2jz to idle around 850.
 

The engine turns many revolutions, as much as 12-15 cylinder events, or ~4 revolutions, before fuel is injected. Maybe the cam/crank angle sensor is weak? Is that a common issue with the JDM 2jz? The "crank inject all" feature is enabled. The pulse, when it finally comes, is instantly detectable and the engine proceeds to idle smoothly without any indication of there being any excess fuel, or any need for it (the combustion event is never weakly presented, every, upon any occasion once it is detectable).
 

For reference, my
Crank injector time at 0% TPS is 0.868ms, and my coolant temp initial crank pulse is 11.1ms. If I understand correctly, the engine is supposed to get 11ms to all injectors as soon as the AEM gets any kind of signal from the sensor. Followed by a 0.868ms shot to the next injector in the cycle, and so forth to each injector. The engine never floods with these settings. It has always just been a waiting for that initial crank 11ms pulse and the engine will run normally each time.

AEM_NS
Offline
Last seen: 58 min 37 sec ago
Joined: 06/06/2014 - 09:31
What firmware version are you

What firmware version are you using?  How soon after you start cranking does Stat Sync'd come ON?  How long after Stat Sync'd comes ON does the engine fire and start?  A crank injector time of .8ms sounds incredibly small - especially for stock injectors.  No fuel or spark is delivered until Stat Sync'd turns ON so, if it takes a few revs to get the EMS to recognize the cam/crank pattern, it'll take at least that amount of time before any fuel is delivered.  The typical ballpark rule of thumb is to have something like 4 times the amount of fuel needed to idle delivered during cranking.  So, if the engine idles happily at ~1ms at idle, you'd need to have about 4ms of fuel pulse during cranking and even more if the engine is cold. 

If it takes a long time to have Stat Sync'd turn ON during cranking, there are possibly some changes that can be made in the cal to improve this.  But first you need to establish if this is even part of the problem or if it's purely a cold start tuning issue.

kingtal0n
Offline
Last seen: 3 years 8 months ago
Joined: 01/31/2015 - 18:00
I am using AEM v2 the latest

I am using AEM v2 the latest firmware (or at least < 3 months old)

As SOON as stat sync'd comes on, the engine starts and purrs. The amount of fuel is perfect for it to fire under cold OR hot conditions. They are 770cc/min.

I need to help it stat sync'd sooner. It's almost as if the ECU is resseting when I crank. I tried a new battery and jumper cables to see if it was a voltage drop related.
I also tried logging the cranking, but as soon as I crank, the AEM stops logging. My next idea is to run a solid 12v from the battery directly to the ECU to rule out any relay related switching problems.

AEM_NS
Offline
Last seen: 58 min 37 sec ago
Joined: 06/06/2014 - 09:31
Let's clarify... it's almost

Let's clarify... it's almost if the ECU is resetting or the ECU is actually shutting off and then turning back on during cranking?  You should be able to see this while cranking if comm'd up to the EMS because comms will drop out then then reconnect.  If this is the case then your battery voltage is dropping down too low during cranking.  Re-reading your last post and you say that the EMS stops logging during cranking.  You definitely have a voltage drop issue.  Are you using a small "racing" type battery?  Is it mounted in the trunk?  These two things combined can create voltage drop issues like this.  

kingtal0n
Offline
Last seen: 3 years 8 months ago
Joined: 01/31/2015 - 18:00
Issue #1

Issue #1
the default setting for the tps->load to take over is too low, and should be required to "find on purpose" by the user. The initial setting should be null, or large enough to avoid use until the end user decides to set it up on purpose. Otherwise, a new owner might expect the full capability of the factory map to be available, and it is not. The oem software claims that it will fuel cut at 17.2psi and it does not because the default setting is too low for the end user to make full use of the OEM map sensor.

 

Issue #2: Voltage drop

I did a little more testing and the internal ECU log shows no delay whatsoever between the battery voltage drop that occurs when you start cranking, and the actual stat sync'd event, whereas in reality there is about a 5 second delay between them.

In other words, If I tell the computer to log data internally, and I start to crank the engine. The engine takes 5~ seconds of cranking before it starts. But when I pull the ECU log, i See the battery voltage drop and immediately right after the engine RPM comes up and stat sync'd is on. So its almost as if the ECU is "forgetting" 5 seconds of reality, which is when the communication drops off (the ECU disconnects from the laptop when you start cranking).

 

 

AEM_NS
Offline
Last seen: 58 min 37 sec ago
Joined: 06/06/2014 - 09:31
I already answered Issue #1

I already answered Issue #1 in your other thread.  I think unfamiliarity with how certain conditions are handled in the sw/fw lead to the problems you encountered.

Re Issue #2, the ECU isn't "forgetting" 5 seconds of data.  The ECU is most likely physically shutting down during cranking because your voltage is dropping thus making the log stop recording.  When the ECU turns back on because the voltage starts to increase after the initial current inrush to the starter motor, it starts logging again.  It didn't "forget" what happened during those 5 seconds.  It wasn't even recording what happened because it seems that the EMS is shutting off briefly while you crank.

kingtal0n
Offline
Last seen: 3 years 8 months ago
Joined: 01/31/2015 - 18:00
Okay, thanks for all your

Okay, thanks for all your suggestions, I am trying to get the owner to do something about it. Maybe we need a little battery pack to keep 12V at the ecu while the owner turns the key? Wouldn't that work, a small circuit with a capacitor? We tried jumper cables to an additional battery with the same result but we have not yet investigated further wiring/grounding issues. Well, I would have, but the owner insists he will be re-doing the entire chassis harness sometime this year, so I am supposed to wait. The AEM 3-bar map will go in when E85 comes in.

spoon_ek9
Offline
Last seen: 3 years 4 months ago
Joined: 11/23/2014 - 20:59
stop being stubborn, just

stop being stubborn, just because you "think" it is not making sense to you doesnt mean it is not how it works.  

you are using your equipment beyond it was designed. the tps to load is only for the map sensor is overload and keep the engine running, it is not for overload the sensor and run safe.

be nice to yourself and get the right sensor if you keep overload your sensor, 2 bar sensor is just not enough for your use.

remember, it is a fully standalone ecu, understand how it works and use it as how it is designed, then you will be fine.

for your voltage drop, try using a bigger cable for your starter and ground, it should help.

 

 

GK System - Efi and data acq specialist

kingtal0n
Offline
Last seen: 3 years 8 months ago
Joined: 01/31/2015 - 18:00
I think you are missing the

I think you are missing the point. The default setting for map to tps voltage in the ECU is something like 4.9v or 4900mv which is about 17psi in real life pressure on the stock sensor. yet the default tps->load value is only 8psi. That is what doesn't make sense, that is the an oversight causing the problem here. Why would the default overboost map failsafe be set to half the sensor's max value, once the max voltage is reached?

 

You dont need a map sensor at all to run an engine. Its just extra information that happens to be useful. I don't need a MAP sensor- or a wideband sensor, or an EGT sensor, I've been tuning engines before all this fancy electronics were available by playing with a carbureter and some spark plugs. If I want to ignore the map sensor after 4.8volts then its possible using this tps->load map, however, not using the default values. The reason this is a bug and not just a setting to adjust as a new ownership of the ECU, is because the factory default for boost fuel cut voltage also happens to be around 17.2psi ----- one step above the default voltage that trigger the tps->load map, so the advertised boost fuel cut never happens, which is a dangerous, stupid out-of-the-box setting. They should adjust it out of range- 5.5volts or 15.5 volts or 200000 volts, something out of range so the out-of-the-box setting of 17.2psi actually works and gives a fuel cut. Or, they could simply lower the default boost fuel cut to 15psi from 17.2psi. In other words, don't send out a plug and play ECU with a bad default value for no good reason and just assume the new owner will be testing every sensor and counting all the teeth to make sure every default value in the computer is accurate.

I tune more than 10 different computers, they all work from the same fundamentals. You come to expect that the default out-of-the-box values from companies that claim plug and play ECU will make sense, like the crankshaft trigger wheel teeth numbers and the coil dwell settings. That stuff is supposedly "plug and play" and it speeds up the process of getting a car tuned quickly. If a default value is erroneous and dangerous- such as one that causes a default setting for boost fuel cut to not work- that is an oversight by the manufacturer. The AEM ecu has always seemed very complete and easy to work with but there was absolutely no way I would have known to check the max voltage of the map sensor in advance to compare with the default setting for the tps to load (out of range of map) nor would I have guessed that a 2-bar map sensors who maximum boost is at least 15psi would be set to only 8psi as a default load value for the conversion! What kind of trickery is that?

 

More info that might be useful for others: 
I see now by looking over my logs that the OEM 2-bar map is really only good to use until about 13psi then it starts to get "flimsy" and somewhat unreliable. So for anyone attempting to utilize an 18psi boost setting (or around that) on the OEM 2-bar 2jz-gte map sensor, they should use a 4,700mv (well before 4900mv) setting to trigger the tps->load map, and run it at 17psi flat (just below the default fuel cut). This will be your final line so you put all of your fuel here for whatever max boost you want. If you want 21psi then you need to put 21psi worth of fuel in the 17psi row where the AEM will lock you at your TPS setting. Once the boost passes your map trigger voltage it will jump to the 17psi line and fall rich enough for 21psi, which comes quickly when you run a large single turbo on a 2jz-gte so you only get a slight rich dip until it flattens out. The ONLY drawback to doing this is that you will no longer have a boost based fuel cut, so if boost pressure spikes the engine will go lean. I would never drive around the car like this- its just for that one pass on the dyno ex. You went to the dyno with the setup before you did anything to it, before the E85 and new fuel system, before the upgrades and larger map sensor, you wanted a baseline run to beat later. So you benchmark the engine- but you really wanted to see what it would do on pump gas at 20psi just once before all that upgrading...

spoon_ek9
Offline
Last seen: 3 years 4 months ago
Joined: 11/23/2014 - 20:59
it is funny to hear you say u

it is funny to hear you say u dont need map and lambda sensor to run your computer controlled engine, then why would u get one? cant u just make it carbureted turbo? lol

AEM already said it is just a start up file and it is not recommended to run it and think it is fool proof. this ecu is just a tool, it is only great when the user use it correctly.

i dont understand in what part you dont understand a 2 bar sensor mean.  running an engine over one bar of boost with a 2 bar absolute sensor and yet complaining resolution and accurancy and over voltage is just beyond me.    

the plug and play just mean you dont need extra wiring like a professional race grade ecu which require to design and build a harness for that specific application. it doesnt mean it is like an oem ecu.    

the tps to load is not for u to do a voltage clamp like a piggy back setup. it is just for case like the map sensor is totally failed and you can still limp your car back to a shop. using that function to "overcome" a sensor overload is just not wise.  tps and map is not really related in any function.  i think it is more dangerous for you to "think" of work around than making use the correct sensor and setup.  

 

 

GK System - Efi and data acq specialist

AEM_NS
Offline
Last seen: 58 min 37 sec ago
Joined: 06/06/2014 - 09:31
I think we've reached the end

I think we've reached the end of this discussion now guys.

Topic locked