MR2 2GR-FE VVT related RPM spikes

[BikeMurgunder]

Engine - 2GR-FE 3.5L V6

Car - Toyota MR2 Turbo

ECU - G4+ Xtreme

So I am having this really strange issue. I recently realized my VVT oil valves werent wired correctly and i havent really had any control. So i fixed the wiring and everything is working now. Here is the weird part. When driving with a cold engine (less than 170 ECT), and VVT activated, I am getting weird rpm limits around 3000 rpm. Logs show that I am hitting 10000 rpm for split seconds. 

This never happened when VVT wasnt working (no power to control valves), and if I force a lockout condition for VVT, it goes away. Once engine coolant reaches 190 degrees, it acts completely normal with VVT activated. I have uploaded a log where I blipped throttle (3000rpm) after letting the car sit for 3-4 minutes after initial start. Environment temp is 90F.  I will also upload a log where everything is up to temp and im cruising around (for comparison).  To mitigate the problem, I keep VVT locked out until 176F ECT, but id like to know why this is happening.

Does anyone have any idea what I'm missing? Ive also attached trigger scope (at idle) for reference. 

 

And another issue I am seeing now. At idle, if i force VVT on at 700rpm (still calling for 0 degrees), I have two oil control valves showing the ECU is calling for 30-35% DC at 0 degrees.  ANy idea why the ecu would call for that? 

 

Thanks in advance for any help.

2gr-fe-mr2-6-28-21 -cold temp ait comp.pclr Trigger scope 7-3-21.llg Log 2021-06-27 8;14;32 pm 2.llg 2021-06-25 Cold Start RPM Limit.llg

[Adamw]

You have a lot of trigger errors in your logs.

Your cam hasing looks a bit different that our expected pattern, do you have the LH or RH intake cam connected to trig 2 (Looking from flywheel end towards front pulley)?

 

7 hours ago, BikeMurgunder said:

And another issue I am seeing now. At idle, if i force VVT on at 700rpm (still calling for 0 degrees), I have two oil control valves showing the ECU is calling for 30-35% DC at 0 degrees.  ANy idea why the ecu would call for that? 

That would be pretty normal.  With most VVT systems the base DC is around 40-50%, so cams wont move at that level.  For intake cams less than about 40% is retard (move towards home), more than 45 is advance.  The faster it needs to move the further the DC will move from the base DC.  

Note you shouldnt activate VVT at idle, it can drop oil pressure significantly on some engines when hot, generally they activate somewhere a bit above idle - 1200RPM or so.  

 

[BikeMurgunder]

4 hours ago, Adamw said:

You have a lot of trigger errors in your logs.

Your cam hasing looks a bit different that our expected pattern, do you have the LH or RH intake cam connected to trig 2 (Looking from flywheel end towards front pulley)?

I originally had RH intake (looking from flywheel) setup as trigger 2, but I changed it back to LH. It was an error in my wiring. I am pretty sure it should be correct in the Log 2021-06-27. 

 

4 hours ago, Adamw said:

That would be pretty normal.  With most VVT systems the base DC is around 40-50%, so cams wont move at that level.  For intake cams less than about 40% is retard (move towards home), more than 45 is advance.  The faster it needs to move the further the DC will move from the base DC.  

OK. I was not aware the ecu would force a base percentage to hold the vvt in place. Then my next question is why are only 2 being asked to hold at ~33% and not all 4? The documentation I have shows that toyota VVT in this motor has no intake "retard", just home position which is just 0. The factory VVT chart references 0 for most offload driving and only advances above 15% load (based on rpm). So the base 33% the link ecu is giving is just keeping the cams off the lock position?

 

4 hours ago, Adamw said:

Note you shouldnt activate VVT at idle, it can drop oil pressure significantly on some engines when hot, generally they activate somewhere a bit above idle - 1200RPM or so.  

Yes, I only had it active at that point for testing purposes. I wanted to see what was going on with only having 2 cams showing that base 33% percentage. 

 

I have attached another log where I was playing with the PID for the VVT. This definitely had the correct cam setup as Trig 2. Does the phasing still look off? There were errors at the start of this log, but as the log goes on, there were no new trigger errors.

Log 2021-06-28 6_17_26 pm.llg

[Adamw]

There are no RPM spikes or trigger errors showing in the most recent log.  Do you still have a problem?

 

11 hours ago, BikeMurgunder said:

Then my next question is why are only 2 being asked to hold at ~33% and not all 4? The documentation I have shows that toyota VVT in this motor has no intake "retard", just home position which is just 0.

It is a closed loop system, the DC will do whatever it needs to do for the cam to get close to target.  In your log the LH inlet cam is sitting at 0deg (matches target) with the DC at ~34%.  The RH cam is sitting at 1.8deg so the ecu has reduced the DC all the way to 0% trying to get the cam to 0deg but it appears it is already against the mechanical stop so it is not going to go any further.  You need to adjust the offset so that it reads zero when it is against the mechanical stop.

 

It looks like your PID still needs a lot of work as the cams arent controlling well at all.

[BikeMurgunder]

1 hour ago, Adamw said:

There are no RPM spikes or trigger errors showing in the most recent log.  Do you still have a problem?

Yes, my problem is that the errors all show up when the engine is cold, but goes away when warm. Those spikes happen constantly until the engine gets to about 180F. Is it common to leave a lockout of 176F for the VVT? It's the only thing that remedies the spikes. 

 

Also, you had mentioned that my cam phasing looked odd. Did it still look off in that more recent log (6-28-21)?

[Adamw]

I need the trigger scope to see cam phasing.  Please do a few more trigger scope captures around the same time/RPM when youi are seeing the RPM spikes.  

[BikeMurgunder]

Attached are a log of the rpm spikes and associated cam errors. Scopes called out too.

Log 2021-07-5 2;04;53 pm cam errors.llg

[BikeMurgunder]

I just realized i saved the scopes as JPG. I will redo them shortly with log files. 

[Adamw]

Where those scopes done at idle or higher RPM?  Can you also do one with the LH Intake VVT solenoid unplugged.

[BikeMurgunder]

I have attached trigger scopes with the following conditions:

Cold Start - Idle (1100rpm)

2200 rpm

3500rpm

4000 rpm

 

I also attached a cold(ish) start and did some revving to force cam errors. Again this all goes away when warm.

 

Do you want a trigger scope with the solenoid unplugged? or a data log?

Trigger Scope Log 7-5-21 4000 rpm.llg Trigger Scope Log 7-5-21 3500 rpm.llg Trigger Scope Log 7-25-21 2200 rpm.llg Trigger Scope Log 7-5-21 cold start.llg Log 2021-07-5 2;04;53 pm cam errors.llg

[Adamw]

Im still struggling to see an obvious reason for the trigger error.  One thought at the moment is possibly sometimes trigger 2 voltage doesnt fall low enough to count as a edge, the threshold it needs to drop below is about 1.0V and your scopes show 0.8V so there is not a whole lot of margin for error. 

Do you know what value pull-up resistor you have fitted on trigger 2? 

Can you set cam angle test to LH inlet RPM lockout to 500 and test pulse count to 3, then log it running for a little while like that with a few revs etc.  

 

[BikeMurgunder]

And a quick follow-up. Was this the waveform you were expecting to see from my motor? 

 

I have the book for this motor and I am not noticing mine seems to be shifted from this waveform.

[Adamw]

Your cam phasing is quite different to that pic from the manual but it is close enough to the pattern our ECU is expecting and our ecu seems to be happy with it when I test on the bench with a simulated pattern that matches your engine.  So I dont believe there is any problem there at this stage.

But just for your interest, if you look at the falling edge of the longest cam tooth in your scopes (marked in pink below) it is sitting right underneath the last tooth before the gap on the crank.  In the FSM that same falling edge is nowhere near the gap.

 

 

[BikeMurgunder]

Would it be beneficial to change over to rising edge? It would keep my edges further away from the gaps. Redoing the offsets isn't an issue. 

[Adamw]

No, as I said I dont see an issue with the location of your edges.  

 

19 hours ago, Adamw said:

Do you know what value pull-up resistor you have fitted on trigger 2? 

Can you set cam angle test to LH inlet RPM lockout to 500 and test pulse count to 3, then log it running for a little while like that with a few revs etc.  

 

 

[BikeMurgunder]

1 hour ago, Adamw said:

No, as I said I dont see an issue with the location of your edges.  

 

 

Sounds good. 

 

20 hours ago, Adamw said:

Im still struggling to see an obvious reason for the trigger error.  One thought at the moment is possibly sometimes trigger 2 voltage doesnt fall low enough to count as a edge, the threshold it needs to drop below is about 1.0V and your scopes show 0.8V so there is not a whole lot of margin for error. 

Do you know what value pull-up resistor you have fitted on trigger 2? 

Can you set cam angle test to LH inlet RPM lockout to 500 and test pulse count to 3, then log it running for a little while like that with a few revs etc.  

 

I have attached 2 logs. One is a cold engine (2-3 minute idle) and the other is a warm engine.

 

 

The LH cam sensor had a 2.2K ohm. I didnt switch them when i correct the wiring apparently

Log 2021-07-6 - LH Cam Test WARM ENGINE.llg Log 2021-07-6 - LH Cam Test.llg

[BikeMurgunder]

since the rpm issue happens with vvt active at anything less than full temp, i wonder if it could be oil pressure/flow related.  And is the vvt actuator not homing all the way due to varnish build up from the previous life of this motor.

[Adamw]

VVT movement shouldnt cause a trigger error, the trigger decoder is designed to cope with moving cam teeth.  

As I mentioned earlier I have tested your map on the simulator with a trigger pattern that matches your scopes.  I can move the cam about 25 deg further retarded than yours and more than 60deg further advanced without before causing a trigger error im not convinced the error is related to cam movement.  It sure is odd that it only seems to happen when cams are activated though, so I'm not writting this off as a non-posibility.  

But at this point Im more suspicious the error may be caused by your trig 2 cam voltage sometimes not falling below the threshold when cold.  It may just be you have slightly more ground offset or something when the solenoids are active when cold.  Depending on what value pull-up resistor you have on trig 2, we may need to try something higher resistance or possibly there is a ground offset between ecu and engine that needs fixing.  

[BikeMurgunder]

Ok so I have a few thoughts on this. My grounds are as such: I have the engine harness' two main grounds attached to their factory spots on each head. ECU earths are tied to those grounds through ecu plugs. My battery is up front and the negative terminal goes direct to chassis. Engine has 2 1-ga wires going from block to chassis. The only thing that stands out is I have the ecu mounted in the factory ecu case (hacked to fit). The factory case has a plastic barrier inside so the ecu body isn't directly mounted to car body. There is a buffer. I am using the ecu mounting shell and it is bolted to metal. 

Does the link g4+ series use the ecu case as a ground? Sensor grounds are internal, but where do they ground themselves in relation to engine grounds that an offset could form?

 

I am going to run a jumper cable from my battery negative to the block and see if that direct connection alleviates any offset through the chassis.

Also, my resistor was a 2.2K ohm on the LH cam sensor.

[Adamw]

ECU doesnot ground through the case so mounting doesnt matter.  

8 hours ago, BikeMurgunder said:

I am going to run a jumper cable from my battery negative to the block and see if that direct connection alleviates any offset through the chassis.

That will be well worth a shot.  If the chassis has some resistance then you could have some chassis currernt passing through sensor ground.  

 

8 hours ago, BikeMurgunder said:

Also, my resistor was a 2.2K ohm on the LH cam sensor.

If its not a big job, the next thing I would try is sniping that resistor and turn on the internal pull-up to see if we still get acceptable voltages.  The internal is 4.7K, so both the max voltage and min voltage on trig 2 will decrease. 

[BikeMurgunder]

Ok got an update. 

I attached a cable from the battery negative terminal to the engine block. It did not change anything for the trigger voltage to ground. 

As for resistors, I have attached 4 scopes with different resistors for trigger 2. The original scopes above were a 2200ohm. All of these ran EXCEPT for the internal pull-up. It would crank and start, then immediately die. 

Let me know what you think!

Trigger Scope Log 3200 ohm Trig 2.llg Trigger Scope Log 2700 ohm.llg Trigger Scope Log internal Trig 2.llg Trigger Scope Log 1K Ohm Trig 2.llg

[BikeMurgunder]

One additional update as well. I left VVT active (1500 rpm lockout) and unplugged the OCV's. Here is a log of a cold start and rev with them unplugged. No trigger errors and no rpm cutoffs.

 

I ran out of time today, but ill try to unplug one at a time.

 

 

 

Log 2021-07-11 OCV unplugged, cold start.llg

[Adamw]

Im still not seeing what is causing the trigger error.  BTW, the 3200ohm pull-up on trig 2 looks about the best comprimise in terms of voltages.  

I know it looks like the trigger errors are caused only when the cam moves but I dont think it is the actually the movement as I cant reproduce it on the bench.  Im suspecting it is something else related like possibly some interference on the cam signal when the solenoids become active or something.  

Can you try a cold start with all of the cam max DC clamps set to something like below.  This will mean the solenoids are active (so making electrical noise), but there wont be enough DC for the cams to move so that may help isolate if it is movement or noise that is causing the issue.

  

[BikeMurgunder]

Log here: https://drive.google.com/file/d/1OQxXDor54qhQw_6GjlT-XwkGNhJpJEuo/view?usp=sharing

 

Cold start with all 4 cams set with max clamp 20%. Errors are still there on all cams except RH inlet. I am going to open the loom and see what was different with that one. 

[BikeMurgunder]

My RH cam sensor had a 1.5k ohm (I didn't have a 1K for trigger at the time). So it seemes to react better than the other cams that had 2.2K? That make any sense?