S2000 g4x ect reading issues.

[Mikel58]

I'm having some high ect readings from my g4x in my s2000. I'm seeing 204f-210f while driving. No air pockets in my coolant or leaks. If the car is idling for more than a few seconds the temps start to climb. The highest I've noticed is 228. Once I get moving it drops to the 204-208 range. 

My fans are kicking on but they don't  seem to help. When I put my factory ecu back into the car I have no issues. I can leave it idling cook a meal and come back to the temperatures being normal. 

I tried changing the driven speed lockout value to a higher mph and a lower mph. No luck.

[Vaughan]

Is this based on the position of the coolant temperature gauge? Or do you have an aftermarket gauge or are you looking at obd2 data on the factory ECU?

[Mikel58]

I'm looking at obd2 data with the factory temps. And I have a guage art linkecu can guage that reads from link when I have my linkecu in. 

[Adamw]

I would probably start by checking the calibration of the ECT sensor.  You could do this either by removing it and dropping it in a jug of boiling water and confirm ecu shows near 100C.  Or you could run the engine with the stock ecu until it is hot, check OBD2 temp, switch off then unplug sensor and measure the resistance of the sensor.  Then put the Link ecu in run it until PC link shows the same temp as what you had before, then check the resistance of the sensor again.  If the calibration in Link is the same as the factory ecu I would expect resistance to be within a couple of ohms when the engine is at the same temp. 

The only other thing that would effect coolant temp while driving is tune related such as being too retarded or very lean, the fan control will have little effect when moving above about 10mph.  

[Mikel58]

I'll give this a try. I kept my stock ecu in the factory case. So it's easy to switch ecus.

[Adamw]

I looked at this a bit closer today, there is not a lot of solid info around for the F20C temp sensor, but im pretty sure from the few clues I found it should use the "Honda K20" calibration.  Our base map is set to "Std Bosch NTC" which I assume is what you are using.  The car that our S2000 basemap came from was pretty modified so it possibly didnt have the factory temp sensor in it.  The reported temp will be about 5degC too high if using the bosch calibration with the honda sensor so that may be what you are seeing.  

[mpetrillo]

I thought I was losing my mind until I saw this post. I was seeing high ECTs on my AP1 S2000 and finally decided to check the sensor calibration. 

I'm going to test this out today with my temp gun but according to other standalone ECUs, the AP1 ECT calibration should look like the following: 

C F Ohms
0 32 6205
10 50 3920
20 68 2454
30 86 1604
40 104 1072
50 122 738
60 140 535
70 158 387
80 176 284
90 194 210
100 212 159
110 230 121
120 248 95
130 266 75
140 284 60

[Ucf120]

Having the same issue Mikel is having with my AP1. We are both tuned through Evans Tuning around the same time.  Never had this issue with factory ecu.  
 

Car will be fine for a while but then suddenly while driving temps will start creeping up over 220.  Especially while the AC is on.  With AC off the temps tend to stay below 210 but with them on they go up above 220.  
 

I know Jeff had to tweak my IAT calibration because it was reading super high also..like 170 degrees at times when outside temps were 80. After he redid that calibration the IAT haven’t gotten more than 20 degrees above ambient.  Still having the ECT issue though and it’s making me nervous to drive the car without the computer plugged in all the time which is very annoying.

if anyone has a fix please let me know!

[Mikel58]

10 hours ago, mpetrillo said:

I thought I was losing my mind until I saw this post. I was seeing high ECTs on my AP1 S2000 and finally decided to check the sensor calibration. 

I'm going to test this out today with my temp gun but according to other standalone ECUs, the AP1 ECT calibration should look like the following: 

C F Ohms
0 32 6205
10 50 3920
20 68 2454
30 86 1604
40 104 1072
50 122 738
60 140 535
70 158 387
80 176 284
90 194 210
100 212 159
110 230 121
120 248 95
130 266 75
140 284 60

Let us know how that goes. I haven't  had much time to mess with my car recently. 

[Adamw]

11 hours ago, mpetrillo said:

the AP1 ECT calibration should look like the following: 

This calibration you quote matches our "Honda K20" built-in calibration, however It would be good to confirm the resistance of your sensor does match that rather than just assume.  The info that I have found seems to contradict that, there is definitely an F20C sensor with the same part number as the white K20Z - part number 37870-RTA-005 (white plastic) which uses that calibration you quote, but that is only meant to be for the DBW AP1, the cable throttle AP1 is meant to have part number 37870-PJ7-003 (black plastic) which is the same as the B18C and uses the Bosch calibration as far as I can ascertain.  

The difference is quite subtle, at 100°C (boiling water test), the Bosch resistance would be about 187ohm, Vs 159ohm for the K20.  

[Ucf120]

16 minutes ago, Adamw said:

This calibration you quote matches our "Honda K20" built-in calibration, however It would be good to confirm the resistance of your sensor does match that rather than just assume.  The info that I have found seems to contradict that, there is definitely an F20C sensor with the same part number as the white K20Z - part number 37870-RTA-005 (white plastic) which uses that calibration you quote, but that is only meant to be for the DBW AP1, the cable throttle AP1 is meant to have part number 37870-PJ7-003 (black plastic) which is the same as the B18C and uses the Bosch calibration as far as I can ascertain.  

The difference is quite subtle, at 100°C (boiling water test), the Bosch resistance would be about 187ohm, Vs 159ohm for the K20.  

As far as I’m aware, there is no factory DBW AP1. All AP1’s are cable throttle.  If I’m understanding what you are saying correctly, the calibration would be off which is why our cars are showing that they overheating correct?

[Adamw]

Sorry yeah it was the AP2 that you could get in both DBW and cable throttle right?  The 

So what Im saying is most Honda parts catalogues show cable throttle cars should have the black temp sensor which is apparently the same as the B18 which uses the Std bosch calibration, so that would mean the base map setting is right.   But your observed "increased temperature when driving" suggests it may be wrong.  

So it really needs one of you guys that has access to one to confirm, there is conflicting info and the two possibilities are close to a 50/50 split. - what colour is your sensor?  And what is the resistance at 100°C? 

[mpetrillo]

I confirmed yesterday on my AP1 that the settings above are much more accurate then the Bosch sensor that's configured in the S2000 base map. I think Link should update their base map to avoid confusion in the future. 

I have a GM IAT sensor and before the engine is running both IAT and ECT match now 

[Mikel58]

2 hours ago, mpetrillo said:

I confirmed yesterday on my AP1 that the settings above are much more accurate then the Bosch sensor that's configured in the S2000 base map. I think Link should update their base map to avoid confusion in the future. 

I have a GM IAT sensor and before the engine is running both IAT and ECT match now 

Did you change to the k20 ect calibration?

[Ucf120]

3 hours ago, mpetrillo said:

I confirmed yesterday on my AP1 that the settings above are much more accurate then the Bosch sensor that's configured in the S2000 base map. I think Link should update their base map to avoid confusion in the future. 

I have a GM IAT sensor and before the engine is running both IAT and ECT match now 

Well that’s good to hear! How do we get LINK to update this?

[Vaughan]

3 hours ago, Ucf120 said:

Well that’s good to hear! How do we get LINK to update this?

I update the sample map and put it in the help manual once we are confident in the change.

[Adamw]

8 hours ago, mpetrillo said:

I confirmed yesterday on my AP1 that the settings above are much more accurate then the Bosch sensor that's configured in the S2000 base map. I think Link should update their base map to avoid confusion in the future. 

I have a GM IAT sensor and before the engine is running both IAT and ECT match now 

This does not confirm anything.  We need the resistance of the S2000 sensor measured at some known temperature that is above 80°C - I suggest boiling water if you are near sea level.  

Below 75°C the Honda and bosch calibrations are near identical so comparing to IAT is pointless unless IAT was above 80°C when you done it.  

[Mikel58]

1 hour ago, Adamw said:

This does not confirm anything.  We need the resistance of the S2000 sensor measured at some known temperature that is above 80°C - I suggest boiling water if you are near sea level.  

Below 75°C the Honda and bosch calibrations are near identical so comparing to IAT is pointless unless IAT was above 80°C when you done it.  

I'm going to pull my sensor tomorrow and test it in some boiling water.

[Mikel58]

After seeing many other s2000 guys post their frustration on the Facebook s2000 group about the s2000 linkecu g4x Bosch ect calibration on a recent post from mpetrillo on there. I figured I'd try the different values first before pulling the sensor. Turns out s2000 guys that use g4x seem to all make a new calibration ect table because the default calibration on the s2000 base map is not right. Seems like all the s2000s guys chase that issue. So it should probably be changed.

I changed my ect calibration to the honda k20 calibration. Temps look similar to what I'm seeing from my stock ecu now. 190s while driving. And the car actually can idle now without the temps rising into the 220s and 230s while the fans are going. 

[Adamw]

7 hours ago, Mikel58 said:

Seems like all the s2000s guys chase that issue.

The reason I keep insisting on a resistance measurement is there are 3 sources of error at play depending on how these "S2000 guys" are concluding there is a coolant temperature problem:

1. The scaling of the gauge on the dash is user configurable in our ECU, in our base map it is setup quite different to the factory ecu to give better resolution over the temp range that most users would be interested in.  It will appear hotter in many cases as the factory ecu scaling displays "3 bars" all the way from 160F to 234F.  From memory ours would be displaying 6 bars at 234. 
2. The fan temp activation temp and hysteresis is probably different to the factory ecu.
3. The un-confirmed temp sensor calibration.  

So it would be nice to confirm what colour sensor your car has and its resistance as some known temp.  

[Mikel58]

The sensor is black.