Hybrid energy flow arrows

[NickySB]

Hi everyone

Since my engine change in late 2021/early 2022 I’ve has an odd issue of the energy flow arrows for the hybrid not showing. Everything else on it replicates - charging Battery, spinning wheels but the arrows have gone awol.

it’s been a bit of a ride since the engine was replaced with the 3rd or 4th hybrid Battery rebuild then the local garage blowing up the inverter some 6 months or so later during welding for the MOT, however not a Toyota/Lexus hybrid specialist nor the dealership themselves are familiar with this scenario or have been able to find the cause.
 

I wanted to document my attempts to resolve so far in case it helps someone else out when I (hopefully) get a fix. 
 

1) no apparent hybrid Battery wiring harness damage- there doesn’t appear to be any broken/pinched/visibly degraded wires and all fuses are good.

2) a lot of Prius owners loose their screen when the 12v Battery thermistor fails. I’ve tried changing mine, no change.

3) using TechStream I had codes B1498 - Communication Malfunction A/C inverter local & P3108 Lost Communication with AC system control module. After reading up I found during engine changes the AC compressor connector can get coolant in it and corrode. Checked the socket and it was corroded.  I had also recently had the system regassed at Lexus (after the engine change I’d left it empty ) and it held pressure but wouldn’t blow cold. I tried to clean the socket on the compressor up but when the plug went back on it went back on out of line and bent the pins, so I fitted a replacement compressor, B1498 is now historical but p3108 remains. I’ve not had it regassed to try the AC function yet. 

4) there’s nothing on TechStream when I go into the hybrid summary which shows any relevant degradation. 
 

5) I’ve checked the voltage and resistance on pin receptors 5 & 6 on the black plug which at attaches to the compressor and it’s within spec of the Lexus manual 

 

6) CAN BUS check on TechStream shows all units are white which means Currentlu Communicating on the CAN BUS.

the saga continues 

[hazelnut]

Hi Nicky,

I have a 2009 GS450h (working) and also a 2006 GS450h (not working).  I have managed to successfully clone a GS450h Hybrid ECU for my 2009, so I know enough about the CAN system to be dangerous, but maybe not helpful. 

ISTR that the Hybrid ECU has a dedicated CAN connection to the instrument cluster and infotainment system. I have sometimes seen the energy monitor 'glitch' like this while driving, and my assumption is that the CAN messages for this are low priority. Perhaps only sent out every second. My other guess is that other higher priority CAN messages are being sent instead. To figure this out conclusively you would need to attach a CAN BUS sniffer and then do some detective work to figure out which messages do what.  It may be that the hybrid ECU is constantly trying to get a response from something on the BUS that isn't responding (such as the AC?) 

-Hazel

PS I think the same BUS is used to the instrument cluster to show the kW display, so if that is not working that Bus may be broken.

PPS techstream talks on one of about 6 busses on the Hybrid ECU, so being able to talk to the hybrid ECU does not mean all its BUS's are good.

[NickySB]

Thanks that’s useful background. Yes the dash display for the Battery charging is also absent of arrows but does show the wheel & battery 

[NickySB]

A few more hours fault finding today using the guidance in the manual via Lex Tech. I checked the connector to the compressor again just to make sure I got the same. I can’t have checked properly before as I got different readings. The resistance was within spec ( under 1 ohm) but voltage was not, it should be under 1 v in the specified condition (power on, pin 5 to pin 6) but it reads 11.74v same as the Battery, so it looks like the cause is either the connector or the wiring harness. 

[NickySB]

Gone through the wiring diagram and looking at the circuit start to finish there doesn’t appear to be anything that drops the voltage so I’m not entirely clear why it should be testing below 1v, more investigation to follow 

[NickySB]

I’ve gone through the fault finding from start to finish , regardless of the flow chart, and other than having excess voltage as mentioned above between pin 5&6 on the compressor connector, the only other finding was poor ( over loaded) resistance reading between DOUT ( pin 3) on the connector & body ground.  
 

The flow chart points at the connector/harness for the excess voltage and the ECU for the resistance issue.

Initial thoughts are I’m still struggling to see how the voltage gets dropped from the circuitry so I’m inclined to focus on the ECU 

[NickySB]

On 9/1/2023 at 11:04 AM, NickySB said:

Thanks that’s useful background. Yes the dash display for the battery charging is also absent of arrows but does show the wheel & battery 

I should have clarified though that the kw gauge still works 

[hazelnut]

Hi Nicky,

Is the A/C control module not built into the A/C compressor?  I see you have tried replacing the A/C compressor, but I wonder if the harness is also corroded? I would be tempted to make a test lead that would allow you to easily plug in another known good compressor, from a friends car maybe? 

[NickySB]

Hi, yes you offer a good perspective and I think you’re probably right. I believe it is part of it ( but I’ll happily be corrected by those more in the know). I will try to conduct a test with another harness/compressor. Thanks 

[NickySB]

I realise I'm well overdue an update on this thread. Not been able to borrow a compressor as I don't know any other GS owners.  I've gone through the Lexus testing relating to this specific code (P3108-535). The guidance says DTC P3108-535 may be stored due to an open circuit in the SMRB circuit or SMRG circuit Voltage& resistance checks, but I checked the various voltages/resistances on 2 separate occasions and observed the following (under various power on or off scenarios as dictated by guidance):

• CLK port on motor connector to ground is good for voltage
• CLK port on motor connector to CLK on ECU connector is good for resistance
• CLK on ECU connector to ground is good for resistance

 

• ITE on ECU connector to ground is good for voltage
• ITE on ECU connector to DIM on motor connector is good for resistance
• TE on ECU connector to ground is good for resistance

 

• ETI on ECU connector to ground is good for voltage
• ETI on ECU connector to DOUT on motor connector is good for resistance
• ETI on ECU connector to ground is good for resistance

 

• DOUT on motor connector to ground is NOT GOOD for resistance
• DOUT on motor connector to ground has good voltage

 

• DIM on motor connector to ground has good voltage

The obvious suspect in that is the DOUT (3rd port from left) on motor connector, which would imply its a problem somewhere in the harness) however the guidance continues to point at hybrid ECU or hybrid ecu wiring harness ( i think that is P/N 82121-30D20). When the DOUT is checked, the ECU connector is in place and the power is off. As the ECU is connected at that point I can't rule it out. When ETI is measured from the hybrid ECU connector to DOUT for resistance, it is within spec.

[NickySB]

Finally managed to get the terminal pins out today. I was expecting to find corrosion or a snapped pin within but sadly nothing found - all connections were good and clean.

[NickySB]

Installed a fly lead from the connector back to the ecu pin, checked again but code still remaining. That would rule out the harness. 

[NickySB]

After ruling out the harness I’ve gotten hold of another hybrid ECU. I’m aware these are coded to the car so wasn’t expecting a miracle with a swap over, but just enough to see if the code (P3108) still remained or not. The only code that came up was for the immobiliser (as the ecu isn’t coded to the car - B2799) thus giving me a good degree of confidence that my ECU is somewhat unhappy.  I also noticed when I connect the replacement ECU , TechStream picks up the VIN, however on my original it doesn’t. I’ve tried to rewrite my VIN to the original ECU but I get an error saying function has failed, so another marker to confirm ECU has had a bit of a wobble. 

[NickySB]

I managed to code & initialise the replacement ECU and start the car - irritatingly code P3108 has returned, and there was no change to the screen etc, so I'm back on the original ECU for now whilst I have a think about where to go with this next😒

[NickySB]

I tried to go back to my original ECU and the 12v Battery didn't last long enough for the 30 minute initialisation, so couldn't revert. Since running on the replacement I'm getting good old P0A80 (replace hybrid Battery pack as current, pending & history) and the standard 'check hybrid/vsc/ecb' messages. Techstream shows the voltage blocks looking pretty good (all around 16.5/16.6v except for block 4 which is at 15.36, so slightly lower than the rest. Could be that the Battery needs another coincidental rebuild (as it tends to need a rebuild every yr or so with a cell ) or it could be that now I have a completely functioning ECU its spotting issues the original (Degraded) one missed.

[NickySB]

I've managed to obtain another complete harness (even though I've done the fly lead test) I wanted to see whether tech stream still received P3108 - and low and behold it does. I should clarify I've only connected the compressor black connector (6 wires) to the smaller hybrid ECU connector as this was the same one used for the electrical testing featured above.

The lexus guidance says that there are 3 potential causes for this issue - harness, compressor or ECU, having now tried all 3, it's clearly something else! 

[NickySB]

Some more testing and things are still no clearer. I'm considering that the lack of hybrid energy flow and the P3108 are separate issues, but can't say either way atm as both are proving to be head scratchers. The freeze frame data given from P3108-535 is not indicating any clues.

With the black air con compressor connector disconnected, I still have code P3108-535 & no additional codes (other than the expected code of B1423 for the lack of gas in the AC system).

With the black connector and the orange hybrid plug disconnected (ie the compressor completely isolated),  I still have code P3108-535 & no additional codes, but I then can't talk to the air con in the Air Con module of tech stream, understandably so.

With the black air con compressor connector reconnected, and the orange left disconnected, still no change of codes, I still have code P3108-535 & no additional codes.

I've considered buying a 3rd compressor and testing to that but given P3108 has been present in the original and the replacement I'm not convinced anything will change.

I've attached a visual of the wiring testing.

There are a few things I'm considering atm:

1. There's a bad earth/ground somewhere- that's based on the lack of resistance readings given above
2. The hybrid cable to the compressor is corroded - but there's not a lot to draw me to that atm given there is not much change in state when it is disconnected.
3. There could be an issue with the relays in the hybrid Battery junction box, where the SMRB or SMRG relays are housed, however again there's not much to point me to those atm, as I've no other indications from the system that these are stuck/failing.
4. Potentially the AC Amplifier, however as no codes relate to that, and this circuit appears to be outwit the Amplifier, then I've no reason to suspect it.

 

 

[NickySB]

The search for P3108 continues. As above, I was considering the AC Amplifier although really without cause except its prominence in the circuit (it's hidden behind the glove box). I managed to obtain another of the same part number. Before fitting I tried to read codes with each AC Amplifier socket removed one at a time but Tech Stream wouldn’t talk to it regardless. On receipt of the new one, I swapped them over and tried Tech Stream again. Initially, there was no P3108, but it soon returned (I suspect as I had a 12v Battery drain in the middle of this work the car forgot it for a short time).  Tech Stream can’t talk to the replacement AC Amplifier which is odd as I don't believe they are required to be re-coded to the car.

I’ve also obtained another hybrid Battery donor. The donor had the usual copper corrosion, so I’ve cleaned up the bus bars again, rebuilt it and reinstalled. There is still no change. One of the suggestions from the hybrid specialist when I lost the energy flow arrows was that it could be the wiring harness in the Battery itself, but given the recent donor swap and no change, it rules that out too.

[NickySB]

Spent the last few days attacking the fault finding again. I've checked the CAN BUS whilst asleep between port 6 & 16 of the OBD reader. It came in at 62.2 ohms. This seems to be a generally accepted reading. This doesn't consider testing on the gateway modules or items with no terminal resistor.

I followed fault finding for the newer GS which seems to be mostly the same as this model. I seemed to get a different set of results this time though, whether that's following on from fixing other stuff within the broader circuitry (had some ABS sensor issues) or the replacement hybrid ECU I'm not sure.

I've checked all the grounds within the harness and they're all good (ie clean and fitted well).

Here's the differential updates in comparison to first and second time around.

• Ground on AC motor connector to body ground good (measures 0.04 ohms)
• Ig on AC motor connector to ground with ignition on 11.23v (lower than I would like but just within spec of 11-14v)
• Resistance checks between AC motor connector to corresponding port on hybrid ECU connector for CLK-CLK/DIN-ITW/DOUT-ETI/STBI-STB all come in at 0 ohms.

• CLK port on motor connector to ground is good for voltage - this time came in at 8.43v - too low (should be 11-14v)
• CLK port on motor connector to CLK on ECU connector is good for resistance- still is
• CLK on ECU connector to ground is good for resistance - still is

 

• ITE on ECU connector to ground is good for voltage- still is
• ITE on ECU connector to DIM on motor connector is good for resistance - still is
• ITE on ECU connector to ground is good for resistance - still is
• DIN on motor connector to ground has good voltage - this time came in at 8.43v - too low (should be 11-14v)

 

• ETI on ECU connector to ground is good for voltage - still is
• ETI on ECU connector to DOUT on motor connector is good for resistance - still is
• ETI on ECU connector to ground is good for resistance - still is

 

• DOUT on motor connector to ground is NOT GOOD for resistance - now tests good
• DOUT on motor connector to ground has good voltage - still is (under 1 v)

On the newer GS the suggested fix is to replace the power management ECU, however I'm not sure the same can apply on this variant and it doesn't seem to have the same.

I think my 12v Battery is getting weak as it doesn't last long when I need to use Tech Stream and drops voltage quite quickly (it is 6 yrs old) so I'll replace that too before testing more.

[NickySB]

Cross checking the voltage changes against the correct model of GS fault finding, the check between the IGN & GND on the voltage is coming out too high, as according to the guidance it should be under 1V.

That would imply its still the harness, so after I chance a new Battery the next step will be doing a proper test with the replacement harness. Its a major hardness so swapping it out, when it goes from under the fuse box, to the ECU's, and all round the back and bottom of the engine, and out again, it's a massive piece of work.

[NickySB]

I remembered overnight having the discussion before with an electrician colleague who reviewed the circuit diagram and highlighted there was nothing in the circuit to make it produce under 1V when powered on between IG & GND. I've picked another hybrid model (HS250H), and although a different engine, wanted to compare the fault finding. Also for that car it highlights 11-14v for this condition. I tried checking the R400H but that isn't in the fault finding.

 

[NickySB]

a newer version of the RX with the same engine confirms the same, so suspect this is a typo in the Lexus manual for my variant

 

[NickySB]

Back to fault finding today. After trying the AC Amp (the replacement unit decided to play ball so it just plug & play) and then also trying a power management ECU, neither yeilded any results. I decided to start dismantling my spare complete harness to understand how its connected. I've already ascertained that pins 1-4 of the AC Compressor connector (CLK/DIM/DOUT/STBI) all go to the ECU (as fault finding and wiring diagram makes the link), but I wasn't entirely sure what happened to the pin 5 & 6 (red wire & black/white wire). It was fairly clear from the diagram I've already posted (see 9th June) that they were ignition & ground but I wasn't sure where they went exactly. They both end up in the fuse box. The section of it which slides out has around 4 connectors, and they terminate in the largest one (see photo).

I tried connecting my complete harness into the fuse box, into the ECU & to the AC Compressor. Sadly it didn't work & TechStream couldn't talk to the hybrid ECU. It might have been due to the other 38 connectors & 3 earths in that harness being disconnected.

It's a massive job to change the harness (which was becoming my only remaining option), which I didn't want to undertake lightly, so went back to the diagnosis checks for a 3rd time, as it seems every time I test, the results change. Sure enough, it altered again - 

Hybrid plug disconnected doesn't give either codes (P0ADC-226 or P0AE0-228)

• CLK port on motor connector to ground is good for voltage
• CLK port on motor connector to CLK on ECU connector is good for resistance
• CLK on ECU connector to ground is good for resistance

 

• ITE on ECU connector to ground is good for voltage
• ITE on ECU connector to DIM on motor connector is good for resistance
• TE on ECU connector to ground is good for resistance

 

• ETI on ECU connector to ground is now not good for voltage - when I first turn ignition on it gets about 6v, and then it drops off second by second until it ends up at 0v.
• ETI on ECU connector to DOUT on motor connector is good for resistance
• ETI on ECU connector to ground is good for resistance

 

• DOUT on motor connector to ground is now good for resistance
• DOUT on motor connector to ground has good voltage
• DIM on motor connector to ground has good voltage

Clearly the ETI readings (which I've now checked 3 times today alone) are showing a degradation. The manual points towards the compressor, and although I've already changed that once, I'm going to get another just to be able to plug in without fitting to see what happens to the codes. Interestingly, I've also seen the arrows on the hybrid energy display reappear for around 2-3 seconds on start up - I am surmising that is due to the voltage being high enough at start up then decreasing as the car runs up.

The attached fault finding table might help someone else in the future (it's just an abbreviated version of the manual) for fault finding on code P3108-535/536/537, as well as my very crude wiring diagram for the air con compressor wires.

Lexus Hybrid fault finding table.docx GS450h 191 air con compressor wiring .pdf

[NickySB]

The fun continues! After putting the car back together , I took it out for a drive. About 15 mins it in generated a 'Check Hybrid System' message (no other messages) plus warning triangle and EML. A TechStream scan once I got home shows the usual P3108 & now P2799 which is described as 'Auxiliary Transmission Oil Pump Control Circuit High Voltage'.  I've not had any codes relating to the transmission oil pump to my knowledge in the past. The car had literally done nothing in between generating this fault and not having it, so I'm back to suspecting the harness again. The Aux Trans Oil Pump is on the same harness from what I can gather, as although I can't match the diagram of the connector from the manual exactly, there's one very similar on said harness (as I inspected the spare/donor harness) and the fault finding from the manual also highlights the harness or the hybrid ECU as a potential problem. Incidentally the few seconds of energy flow on start up have also disappeared again

Given that the ETI pin on the ECU connector is showing a steady voltage drop, and now the oil pump circuit apparently has a high voltage, I continue to suspect a compromised harness as the ECU has already been replaced once.  The fault finding continues......

[NickySB]

Reviewing my options I decided to disconnect the AC compressor connector (6 wires) and go for another drive, this time no hybrid issues (30 mins driving, came on yesterday after 15 mins) so perhaps it is JUST the compressor again - but lets see when another 2nd hand unit arrives.