3P's TCCS Disassembly/Analysis
- Thread starter 3p141592654
- Start date
Yes they do. The JDM ECU has O2 sensor heater control logic and fuel trim versus O2 so I am assuming it also has a cat. Is that not true?
Random question for you. You talked about warm up enrichment earlier, what exactly is done for warm up enrichment and what parameters have to be met to be in warm up enrichment? I have noticed that even on very cold starts (well cold for me, low 40's high 30's) the wideband will show AF ratios of ~14.7 almost immediately after engine speed is increased. At idle, it seems to be a bit rich, but if I drive off, as soon as rpm's come up the AF ratios are right around 14.7. I also have an issue with the car bucking and hesitating a little bit for the first couple minutes of driving, perhaps this is related? Also, during this time when it is cold and 14.7 AF, VF is right around 2.5 to 3.5 volts. I am using a MAFT pro piggyback which has a speed density conversion, if that makes any difference. AFM has been eliminated as well as the stock internal air temp sensor.
Awesome thread guys!
I do quite a lot of Toyota conversions, do you need detailed photos of newer ECUs? I usually have some of them handy. 1JZ-GE 4a/t and 5a/t , 1UZ-FE 1998-2003, 5VZ-FE 1995-2000, 3RZ-FE, JDM only
And I'd like to ask if you plan to make some sort of wiki on basic Toyota ECU algorithms outline. This would be helpful when trying to address some odd problem not described anywhere. For example I recently had an 1JZ-GE that would accelerate a bit slower past 4500 rpm exactly, the fix was to replace ECU, still don't know what's stuffed there. Also many older ECUs don't have many useful parameters available after ~2003 such as misfire counter, knock amount, target/actual VVTI position, many others...in most cases you're only left to guess.
I do quite a lot of Toyota conversions, do you need detailed photos of newer ECUs? I usually have some of them handy. 1JZ-GE 4a/t and 5a/t , 1UZ-FE 1998-2003, 5VZ-FE 1995-2000, 3RZ-FE, JDM only
And I'd like to ask if you plan to make some sort of wiki on basic Toyota ECU algorithms outline. This would be helpful when trying to address some odd problem not described anywhere. For example I recently had an 1JZ-GE that would accelerate a bit slower past 4500 rpm exactly, the fix was to replace ECU, still don't know what's stuffed there. Also many older ECUs don't have many useful parameters available after ~2003 such as misfire counter, knock amount, target/actual VVTI position, many others...in most cases you're only left to guess.
Its hard to discuss the fuel calcs without the overall picture of the ECU engine model, and I will write up a more detailed explanation of the fuel calcs soon.
After a cold start the ox sensor reads lean (0V) and won't do anything until it heats up to about 300C. During this time the ECU uses 4 cold start trim counters to add fuel to the basic model. These counters are set to specific values (from internal maps) at startup depending on coolant temp, and rapidly count down with time and temperature until they reach 0, at which point cold start enrichment is over. The longest one lasts until a coolant temp of 60C is reached, the others are faster. They also drop with time.
Now, the switch from open loop to closed loop happens when the ECU sees the O2 sensor switch from lean to rich. For this morning, which was 7C here in So Cal, it took 2 minutes before the ECU switched to closed loop. At that point the mixture will be forced to stoichiometric unless the ECU is forced open loop for other reasons (like high load). So, after cold startup, you will be running open loop, but the ECU is calibrated to get the fuel pretty close on startup to help with lightoff emissions.
Vf will read 0V when open loop. You should not see 2.5V right after a cold start.
You can see a datalog of my cold start this morning. Warning: this will waste 2 min of your life. Note that the purple line (short term fuel trim) stays at 100% for most of the datalog because fuel control is open loop after a cold start. At about 1:56 it starts to move indicating closed loop fuel control has started. Even though it in closed loop, the fuel enrichment continues to help the model stay close to the actual required values.
[video=youtube;cu6TJqXtBV4]http://www.youtube.com/watch?v=cu6TJqXtBV4&feature=youtu.be[/video]
After a cold start the ox sensor reads lean (0V) and won't do anything until it heats up to about 300C. During this time the ECU uses 4 cold start trim counters to add fuel to the basic model. These counters are set to specific values (from internal maps) at startup depending on coolant temp, and rapidly count down with time and temperature until they reach 0, at which point cold start enrichment is over. The longest one lasts until a coolant temp of 60C is reached, the others are faster. They also drop with time.
Now, the switch from open loop to closed loop happens when the ECU sees the O2 sensor switch from lean to rich. For this morning, which was 7C here in So Cal, it took 2 minutes before the ECU switched to closed loop. At that point the mixture will be forced to stoichiometric unless the ECU is forced open loop for other reasons (like high load). So, after cold startup, you will be running open loop, but the ECU is calibrated to get the fuel pretty close on startup to help with lightoff emissions.
Vf will read 0V when open loop. You should not see 2.5V right after a cold start.
You can see a datalog of my cold start this morning. Warning: this will waste 2 min of your life. Note that the purple line (short term fuel trim) stays at 100% for most of the datalog because fuel control is open loop after a cold start. At about 1:56 it starts to move indicating closed loop fuel control has started. Even though it in closed loop, the fuel enrichment continues to help the model stay close to the actual required values.
[video=youtube;cu6TJqXtBV4]http://www.youtube.com/watch?v=cu6TJqXtBV4&feature=youtu.be[/video]
Last edited:
what impact does the oil pressure sensor have on the ecu? When I rebuilt my harness i noticed it was split between the gauge and ecu.
What would happen if the ecu gets a 0 reading? what about 100psi reading?
What would happen if the ecu gets a 0 reading? what about 100psi reading?
grimreaper;1901639 said:
I'm curious about this as well, I'm using the harness to carry my aftermarket sender's signal up to the dash, so this means the TCCS is getting the aftermarket signal also. Hmmm wonder what its doing to the TCCS, I recall awhile back that the connector fell off the sender but nothing happened except a zero on the gauge.
CyFi6;1899372 said:
In plain terms...
www.autoshop101.com/forms/h25.pdf
It gets to your point around 11-13.
up to my experience, 1jz-gte (jzs171) and 4gr-fse (grx120) won't do anything to the fuel/ignition/auto trans system based on the oil level/pressure signals. They are there just for convenience. What ECU does is only forward these signals on the MPX bus to supply instrument cluster with the data it needs. I think that presence of oil signal pin in the ECU and lack of digital signal line to gauges may be strong evidence that the signal does matter and can't be ignored.Bmettie;1901902 said:
btw, decoding mpx bus format might appear to be another important outcome of disassembling denso firmware.
The OILP signal is funky with the 7MGTE, it is a pulsed signal. JJ has described the senders operation before on this site.
The ECU uses that signal and integrates it to form a simple yes/no flag at idle. If oil pressure is deemed to be too low at idle, the rpm is increased by 100 to 750rpm. The flag will not be updated until rpm exceeds 1000 rpm, then the flag is cleared if OILP is good. I believe the rational is that low OILP means a hot engine, and raising the idle speed improves the cooling ability.
There are 3 other flags that also set the idle to 750 rpm. They are high altitude (from HAC sensor), high average load over the current drive cycle, and defogger on. Of course, the AC also raises the idle but to 900 rpm. This is for the M/T ECU. The A/T ECU has a more complicated idle program.
Of course this is the late 1980's, so no MPX bus for this car.
The ECU uses that signal and integrates it to form a simple yes/no flag at idle. If oil pressure is deemed to be too low at idle, the rpm is increased by 100 to 750rpm. The flag will not be updated until rpm exceeds 1000 rpm, then the flag is cleared if OILP is good. I believe the rational is that low OILP means a hot engine, and raising the idle speed improves the cooling ability.
There are 3 other flags that also set the idle to 750 rpm. They are high altitude (from HAC sensor), high average load over the current drive cycle, and defogger on. Of course, the AC also raises the idle but to 900 rpm. This is for the M/T ECU. The A/T ECU has a more complicated idle program.
Of course this is the late 1980's, so no MPX bus for this car.
3p141592654;1317537 said:
was this ever confirmed for the GTE ecu's?
based on your JDM timing map, there are quite a few areas that the JDM map runs more advance. Add your previous comments about very active knock retard even during 'normal' driving on a USDM timing map, would you use the JDM timing map with out EGR?
The GTE code is too big so it cannot fit both the JDM and USDM maps, so there is no swapping between them, with a resistor, sadly.
I'm not sure I understand your second question. The JDM ECU runs those maps without EGR. I am not familiar with the minimum octane in Japan, but I would make sure to run a comparable or better grade here if I was running the JDM ECU. I currently have both the USDM and JDM codes loaded in my ECU, but haven't played around with running the two back to back to see how they compare.
I'm not sure I understand your second question. The JDM ECU runs those maps without EGR. I am not familiar with the minimum octane in Japan, but I would make sure to run a comparable or better grade here if I was running the JDM ECU. I currently have both the USDM and JDM codes loaded in my ECU, but haven't played around with running the two back to back to see how they compare.
Last edited:
3p141592654;1771388 said:Attached is an image of the timing map. Haven't figured out how to attach files to this post. I would not use this map without a very capable knock-sense algorithm monitoring things. The stock ECU knock sensor is fast and pulls timing a lot during normal operation.
3p141592654;1905810 said:
Might have stretched your words a bit in my first post.
The USDM timing map already has timing being pulled under normal driving conditions. The JDM map shows many areas with additional timing. In theory, youd risk MORE detonation/knock retard with the JDM map unless higher octane could be used.
Since the resistor in R606 wont work for the GTE tccs, this is all null and void.
Yes, the JDM map does run advanced in some areas compared to the USDM. I thought maybe regular Japanese gas was higher octane than ours, and Toyota felt comfortable running it more advanced, but I don't really know.
If I am remembering correctly, my comments about the knock sensors retarding timing were referring to during warmup. With cold coolant the ECU advances timing globally, and the knock sensors will hear knock under light throttle and retard it. Once warm, I don't see a lot of knock retard generally speaking, at least on premium fuel.
Yeah, the resistors don't do anything on a gte. There are two ADC channels that can be used to select different maps for minor things, but nothing spectacular, just little tweaks. I need to get my run of boards built so people can start running home brew code on our factory ECUs.
If I am remembering correctly, my comments about the knock sensors retarding timing were referring to during warmup. With cold coolant the ECU advances timing globally, and the knock sensors will hear knock under light throttle and retard it. Once warm, I don't see a lot of knock retard generally speaking, at least on premium fuel.
Yeah, the resistors don't do anything on a gte. There are two ADC channels that can be used to select different maps for minor things, but nothing spectacular, just little tweaks. I need to get my run of boards built so people can start running home brew code on our factory ECUs.
got ya, thanks for clarity on the knock retard portion. If you cant tell, I'm grabbing any tid bit I can in anticipation of this making it into my car..
Now that you've looked through the code, would you have any idea what would cause relatively severe power loss under normal engine conditions and yet worsens with heat? My suspicion has always been timing related. The power I feel during a warm engine (needle about 1/4 way up) and hot (needle up to normal for at least 10 minutes) is significant. A stock turbo can keep up easily with me where as I blow it away on a 'cool' engine; no exaggeration, i feel like im driving my old V6 camry on a very hot day. No codes. I have always suspected the knock sensors or engine temp sensor as having heavy influence,temp sensor tested ok, though the temp sensor was on the high end of the 'in spec' range. Is there anything else that can have a major impact on timing? This has been a LONG time gremlin and I have replaced oh so many parts with no results.
Edit: I've been following this thread a long time. My main hope was to purchase just to help diagnose this issue, I want to see what the computer is reading! OBD1 is too limited.
Edit: I've been following this thread a long time. My main hope was to purchase just to help diagnose this issue, I want to see what the computer is reading! OBD1 is too limited.
If you got hesitation you need to look at the primary inputs for fuel control. MAF, TPS, ECT, IAT, crank and cam position.