CEL and TCS light

Joined
3 March 2023
Messages
7
Hi guys,

I purchased an 1993 NSX last week in Melbourne, Australia and drove it across the country to Perth - about 3400KMs.

After the first 12 hour day of driving I had a TCS light come on, shortly followed by a CEL.

I've done a bunch of reading here about these lights and issues, but when I decoded the CEL it gave me 1 long and 3 short - I understand this is a BARO sensor issue but I can't find much information or history of this issue. The code list says it is an ECM issue but not super keen on replacing the ECM straight away.

The TCS light doesn't concern me too much but does anybody have any advice to fix this? Should I replace the main relay first?

Thanks a lot, happy to be part of the club now.

CheersIMG_7697.jpg
 
The barometric pressure sensor is inside the ECU, so if it's gone bad there's really nothing you can do about it besides find a cheap donor ECU (most 90s Honda ECUs use the same PCB-mounted baro sensor) and hope swapping the sensor over fixes the problem.

The fact that both lights came on at roughly the same time could definitely indicate some electrical fault somewhere rather than actually failed TCS or ECU. Do the lights stay on after unplugging the battery for a bit to clear the error memory?
 
The barometric pressure sensor is inside the ECU, so if it's gone bad there's really nothing you can do about it besides find a cheap donor ECU (most 90s Honda ECUs use the same PCB-mounted baro sensor) and hope swapping the sensor over fixes the problem.

The fact that both lights came on at roughly the same time could definitely indicate some electrical fault somewhere rather than actually failed TCS or ECU. Do the lights stay on after unplugging the battery for a bit to clear the error memory?

The lights come on after I do the clock reset but not immediately. It only happens after a few hours of driving, ie when the car is hot and has been running reasonably hard. When I start the next day it doesn’t turn on for a few hours at highway speed.

Just a little unusual because it doesn’t seem to hamper the driving or performance at all, as in there is literally nothing different about how the car runs when the lights come on
 
As noted by MotorMouth93, the barometric pressure sensor is inside the ECU. It is functionally similar to the MAP sensor; but, its purpose is to correct the fuel pulse width calculation for changes in atmospheric pressure. Atmospheric pressure changes the volumetric efficiency of the engine so they add a correction to the fuel pulse width calculation to do this (different ECUs may do this slightly differently).

I will speculate that the base fuel maps in the NSX are likely based upon operation at sea level atmospheric pressure. Since you are in Perth which is nominally at sea level, non operation of the sensor may have no material impact on the operation of the engine as long as the sensor failure mode does not migrate to shorting something out or you don't go driving at higher altitudes. Also, the atmospheric correction is most important during start up and high engine loads when the engine is in open loop. During closed loop engine operation O2 feedback may deal with the required fuel correction just fine; but, it may cause higher fuel trims which then may trigger a CEL for that at some future point. Again, you being at sea level may mean that this is all a non-factor.

As noted, if the barro sensor has gone bad you will have to open the ECU cover and get eyes on it. If you can get a part number off of the sensor it may still be current or have a modern equivalent and by doing a Google search on the part number you may be able to find a replacement form someone like Digikey, Mouser or Allied Electronics. Otherwise, it's find a suitable donor from some salvaged ECU.

Now this is where things get weird. I highlighted if in the previous paragraph because the problem may not be the barro sensor. If you look at the trouble shooting procedure for the barro sensor in the service manual it talks about the connection to the TCS control unit and that one of the fixes for the 13 error code is replacement of the TCS control unit, the others are replacement of the ECU or repair of a wiring short. Given that you also have a DTC for the TCS unit you may have a common problem that is not the barro sensor itself. I have not sat down with the wiring schematics to figure out what this odd commonality is between the two. I do know that there is a serial data line from the TCS to the ECU; but, this does not appear to involve that interconnection. I recommend that you get a copy of the service manual and go through the test procedure for the barro correction (Honda calls it atmospheric pressure ) to see if you can figure out where the problem. You might be incredibly lucky and it is just an external wiring short on terminal D5.

Since barro correction for you is probably a non issue, you can probably ignore the problem in terms of engine operation as long as you stick around sea level. However, this involves disconnecting the CEL if you don't want it on all the time. The problem with that is that when a future problem does emerge you won't get a CEL warning.

Always been a fan of the original red with black roof. Looks very shiney!
 
Thanks so much for your super detailed comment, that helps a lot. Even if I wanted to there’s pretty much nowhere at high elevation that I could drive to, Perth being as flat as anything just about everywhere. I actually missed the section of the manual about troubleshooting the baro sensor so as soon as I get a chance I’ll have a go at that.

The paint on this car is literally immaculate, looks like the day it rolled out of the factory. I can’t say red is my first choice but it definitely turns heads and with the red roof it really does look great
 
As noted by MotorMouth93, the barometric pressure sensor is inside the ECU. It is functionally similar to the MAP sensor; but, its purpose is to correct the fuel pulse width calculation for changes in atmospheric pressure. Atmospheric pressure changes the volumetric efficiency of the engine so they add a correction to the fuel pulse width calculation to do this (different ECUs may do this slightly differently).

I will speculate that the base fuel maps in the NSX are likely based upon operation at sea level atmospheric pressure. Since you are in Perth which is nominally at sea level, non operation of the sensor may have no material impact on the operation of the engine as long as the sensor failure mode does not migrate to shorting something out or you don't go driving at higher altitudes. Also, the atmospheric correction is most important during start up and high engine loads when the engine is in open loop. During closed loop engine operation O2 feedback may deal with the required fuel correction just fine; but, it may cause higher fuel trims which then may trigger a CEL for that at some future point. Again, you being at sea level may mean that this is all a non-factor.

As noted, if the barro sensor has gone bad you will have to open the ECU cover and get eyes on it. If you can get a part number off of the sensor it may still be current or have a modern equivalent and by doing a Google search on the part number you may be able to find a replacement form someone like Digikey, Mouser or Allied Electronics. Otherwise, it's find a suitable donor from some salvaged ECU.

Now this is where things get weird. I highlighted if in the previous paragraph because the problem may not be the barro sensor. If you look at the trouble shooting procedure for the barro sensor in the service manual it talks about the connection to the TCS control unit and that one of the fixes for the 13 error code is replacement of the TCS control unit, the others are replacement of the ECU or repair of a wiring short. Given that you also have a DTC for the TCS unit you may have a common problem that is not the barro sensor itself. I have not sat down with the wiring schematics to figure out what this odd commonality is between the two. I do know that there is a serial data line from the TCS to the ECU; but, this does not appear to involve that interconnection. I recommend that you get a copy of the service manual and go through the test procedure for the barro correction (Honda calls it atmospheric pressure ) to see if you can figure out where the problem. You might be incredibly lucky and it is just an external wiring short on terminal D5.

Since barro correction for you is probably a non issue, you can probably ignore the problem in terms of engine operation as long as you stick around sea level. However, this involves disconnecting the CEL if you don't want it on all the time. The problem with that is that when a future problem does emerge you won't get a CEL warning.

Always been a fan of the original red with black roof. Looks very shiney!
Is there a resource regarding disconnecting the CEL? I've had a search but can't find a good procedure for it. Does it require taking the cluster/dash out/apart?
 
The section of the early service manual covering the barro sensor test is 11-54.

I think the common procedure for disabling the CEL without chopping wires is to remove the lamp from its socket which requires removal of the instrument cluster. Since your car is a 1994, you might want to plan for a complete lamp replacement since cluster removal is a non trivial exercise. Removal might have a side benefit because with the instrument cluster out you can check the capacitors on the instrument cluster circuit boards for leakage which results in damage and erratic operation.
 
WHRDNSX is on the Gold Coast side of OZ, he is a good source.

I am in NZ, live near a cheap junkyard full of JDM cars, and can send parts if you go that route.

I also recommend rebuilding the cluster. And converting all the small wedge bulbs to LED.

I don't recommend disconnecting the MIL (aka CEL), you can depin the individual terminal from the green JAE IL-AG5 30P connectors that feed the cluster without removing it.
 
WHRDNSX is on the Gold Coast side of OZ, he is a good source.

I am in NZ, live near a cheap junkyard full of JDM cars, and can send parts if you go that route.

I also recommend rebuilding the cluster. And converting all the small wedge bulbs to LED.

I don't recommend disconnecting the MIL (aka CEL), you can depin the individual terminal from the green JAE IL-AG5 30P connectors that feed the cluster without removing it.
Don’t have any manual transmissions laying around over there in NZ do you? 😉
 
Back
Top