RoverGauge sudden drastic main voltage drop ?

RoverGauge sudden drastic main voltage drop ?

Author
Discussion

blaze_away

Original Poster:

1,502 posts

212 months

Saturday 3rd October 2020
quotequote all
Trying to help out a fellow Chim owner whose car idles for 7-8 minutes then dies.

Early 4.3 litre on Map 2 (non catalyst) no cats or lambda's.

His thread is here Titled "Help"
https://www.pistonheads.com/gassing/topic.asp?h=0&...

I want to focus on one aspect that is puzzling - Namely Sudden Main Voltage Drop

I've looked at the RG data for him and noticed that the car runs fine and all parameters appear to be OK then suddenly the 'Main Voltage' drops from 13v to 7v.

This appears to kick the TPS and MAF values to jump up (nothing touched by owner) then it all goes into melt down, fuel map column/rows start jumping all over the place then it dies.

Here is what happens in last 90 seconds of running at idle, no touching of throttle etc by owner.

At data point 30 the voltage flutters a little, MAF & TPS go haywire engine runs up to 4K then stabilises till data point ~80 when voltage flutters a little again then suddenly has major drops and engine dies.

I have suggested checking earth in cabin by gear stick and the engine to chassis earth plus battery terminals and the ECU main plug.

Whats your thoughts /


Penelope Stopit

11,209 posts

108 months

Saturday 3rd October 2020
quotequote all
Definitely onto something good or bad depending on how you look at it

Rovergauge main voltage is measured from the main relay output

Main relay is close to ECU, see black plug with relay next to blue plug with relay which is for fuel pump



Measure positive in and out voltages at main relay to determine if fault is before or after main relay, onwards and upwards once you have the readings

Is the vehicle fitted with an 80 amp main fuse close to the fuse-box?

If answer is yes, fuse holder or fuse could be playing up if the fault lies within the relay input circuit

blitzracing

6,387 posts

219 months

Saturday 3rd October 2020
quotequote all
You would have to ask, is this a genuine voltage drop, or the ECU pulling a wobbly and chucking out daft readings. Best way would be check the ignition supply on the ECU Plug. It would be very odd that the voltage rail drops at near the same point and then recovers ?


blitzracing

6,387 posts

219 months

Saturday 3rd October 2020
quotequote all
Assuming this is Simons car- the logs are odd- we get a very low RPM reading as the voltage fluctuates- far lower than the engine will actually go, so Id suspect you are loosing trigger pulses on the coil primary, that could tie in if the voltages do actually drop. I assume we don't have a restart when it goes from 0 rpm to 128 ? The Maf readings would say the engine has stopped completely ? Check the coil ignition supply and see if its stable.


blaze_away

Original Poster:

1,502 posts

212 months

Saturday 3rd October 2020
quotequote all
Yes that data set is Simon's car.

Agree something isn't behaving itself, hence asking you learned folks for help.

I have advised Simon to take a look at your comments above and start getting some readings on these.

Steve_D

13,737 posts

257 months

Sunday 4th October 2020
quotequote all
Penelope Stopit said:
.............Rovergauge main voltage is measured from the main relay output...........
I've not come across this before. There are 3 12v supplies into the ECU but I have not seen anything defining which of these the ECU measures.

Can you point me at where this is defined?
Thanks
Steve

Steve_D

13,737 posts

257 months

Sunday 4th October 2020
quotequote all
First tests would be to see if the voltage drop can be measured at 4 points.
1. the battery
2. white/green wire at the fuel pump relay pin 86 (which is also ECU pin 19 but easier to get at)
3. Brown/orange wire at main relay pin 86 (which is also ECU pin 15 but easier to get at)
4. Brown/orange wire at main relay pin 87 (which is also ECU pin 2 but easier to get at)

Now at the other end of the circuit (and a good place to start) the ECU and a number of other parts all earth at the front of the timing cover near the oil pressure switch/senders.

Steve

blaze_away

Original Poster:

1,502 posts

212 months

Sunday 4th October 2020
quotequote all
That's stuff I wasn't fully aware of, very useful.

Particularly this...
Steve_D said:
and a number of other parts all earth at the front of the timing cover near the oil pressure switch/senders.
Steve
As its happening at approx 7-8 minutes after start up and it all runs smooth till then I postulate it is heat related, thus that earth point may be a bit shaky

Steve_D

13,737 posts

257 months

Sunday 4th October 2020
quotequote all
davep said:
Steve_D said:
I've not come across this before. There are 3 12v supplies into the ECU but I have not seen anything defining which of these the ECU measures.

Can you point me at where this is defined?
Thanks
Steve
Main Voltage levels are monitored via ECU sensing input pin 2. See here: https://www.pistonheads.com/gassing/topic.asp?t=15...
Just had a quick read through that but cannot find reference to why pin 2 is the voltage reference rather than pins 15 or 19. All three pins see +12V during normal running.
Perhaps Mark (Blitz) with his close contacts with the RoverGauge gurus will be able to shed some light.

Steve

ETA looking again at the circuit I would like the sensing to be on pin19 (Direct from ignition switch) as this comes via the immobiliser and the blue loopback connector. All 3 of these could be prime contenders for disruption of supply.


Edited by Steve_D on Sunday 4th October 09:58

Penelope Stopit

11,209 posts

108 months

Sunday 4th October 2020
quotequote all
Steve_D said:
Penelope Stopit said:
.............Rovergauge main voltage is measured from the main relay output...........
I've not come across this before. There are 3 12v supplies into the ECU but I have not seen anything defining which of these the ECU measures.

Can you point me at where this is defined?
Thanks
Steve
Sure can



Overview

RoverGauge is a software utility intended for diagnostics and testing of the Lucas 14CUX engine management system, which was fitted to Land Rover vehicles produced from 1990 to 1995, as well as some sports cars produced by TVR, Morgan, Ginetta, and Marcos.

For RoverGauge to work, the computer must have a specially modified USB-to-serial converter. Instructions for building this cable can be found on the libcomm14cux wiki page.
Command line parameters

Pass these values as parameters to rovergauge.exe:

–a or ––autoconnect: Automatically connect to the ECU when the application starts. Note that this will only work if the correct serial port was selected the last time the application was run.

–l or ––autolog: Open the log file immediately when the application starts. This is most useful when paired with –a.

–f or ––fullscreen: Start in fullscreen mode. Maximize/minimize buttons will not be availble, but the application can be exited by using the File menu or by pressing Ctrl-Q.

Summary: to automatically connect and begin logging to a file, start the application with: rovergauge.exe -a -l
Keyboard shortcuts

Exit: Ctrl-Q
Start logging: F5
Stop logging: F7

Main display

Tune: The tune revision of the firmware in the ECU, such as "R3360". Note: this number was not always updated by third-party firmware sources for low-volume automakers such as TVR. In that case, multiple different tunes may share the same tune number.
Ident: An additional identifier that is updated independently of the Tune number.
Checksum fixer: A byte of data used to fix the checksum of the ROM image. This can be used to further identify a ROM image beyond the Tune and Ident values.
MIL: Lit red when the malfunction indicator lamp (MIL) is being lit by the ECU. When the engine is running, this typically means that one or more fault codes are set. The MIL is also lit under other circumstances (such as when the ECU is powered on but the engine has not been started.) Note that there are some fault codes that do not cause the MIL to light.
Communications: The green lamp is lit when the software is successfully reading data from the ECU, and the red lamp is lit when there was a problem communicating with the ECU (caused by a bad connection, disconnected cable, misconfigured USB adapter, or other problem.) Both lamps will be off if there has not been any attempt to read from the ECU.
Engine temperature: Displays the temperature read by the engine/coolant temperature sensor. The green and red colored areas (representing "nominal" and "warning" temperature levels) are approximate.
Road speed: Displays road speed. Note that some vehicles (including TVR) use a different road speed transducer setup, and the measurement on those vehicles may not be updated above a certain speed.
RPM (tachometer): Displays engine speed in revolutions per minute. The redline represents the RPM limit stored in the ECU.
Fuel temperature: Displays the fuel temperature, which is also sometimes considered as the under-bonnet temperature.
MAF reading: Displays the reading from the mass airflow sensor as a percentage of its maximum range. This display has two modes:
Linear: Shows a percentage of the maximum airflow. It will vary linearly as the sensor measures more mass.
Direct: Shows a percentage of the maximum sensor voltage (5VDC). It will vary logarithmically as the sensor measures more mass.
Throttle position: Displays the throttle position as a percentage of wide-open throttle (WOT). This display has two modes:
Absolute: Calculates the percentage by dividing the raw reading by the maximum theoretical reading. Even when the throttle is closed and resting on its stop, the "Absolute" reading will still show some positive value.
Corrected: Adjusts the minimum end of the range so that a closed throttle shows as 0%.
Idle bypass position: Displays the position of the idle air control stepper motor. 100% represents fully-open (greatest airflow) and 0% fully closed (restricted airflow).
Target idle RPM: The engine idle speed that the ECU would attempt to achieve under the current conditions. Note: Some early versions of the ECU do not store the target idle RPM in memory and it may not be displayed accurately for these units.
Selected gear: The state of the neutral switch (for automatic gearboxes), or an indicator that a manual gearbox is fitted.
Main voltage: The voltage at the main relay.
Lambda trim: Fuel trim that is being applied as a result of feedback from the exhaust lambda sensors. There are two types of trim applied by the ECU, referred to as "short term" and "long term". These lambda trim values are only relevant when the system is running in closed-loop mode, so this display changes to a MAF CO trim display when fuel map 1, 2, or 3 is selected. Note that the MAF CO trim is only updated if the engine has run.
Current fuel map: ID of the currently selected fuel map, from 1 to 5. For most North American specification (NAS) Land Rovers, this will always be Map 5. In other markets, the map may be selected with the use of a "tune resistor" in the 14CUX wiring harness. A special sixth map, Map 0, is used for fail-safe/limp-home conditions.
Multiplier: This is a value associated with the current fuel map, and it seems to be used to adjust the fueling values for the engine displacement. It will change significantly between fuel maps intended for different engine displacements.
RPM limit: Engine speed (redline) at which the ECU will cut the fuel injector time to protect the engine internals.
Row scaler: An internal computation factor used in scaling the engine load data for the current fuel map.
Fuel map: Displays the currently selected fuel map as a 16x8 matrix. Each cell is colored as a quick visual indicator of the fueling values; higher color saturation corresponds to more fuel. Columns in the matrix correspond to different engine speeds, and rows correspond to different engine loads. The cell with the value closest to the fueling value currently in use will be highlighted by the software. The numbers in the header row reflect the engine RPM threshold for each column.
Fuel pump relay: The green lamp will be lit when the ECU is attempting to run the fuel pump.
Injector duty cycle: The percentage of the time available between spark interrupts that represents the amount of time the fuel injector will be open. Once this reaches 100%, the injectors cannot flow any more fuel and any additional load on the engine may result in a lean condition. Computing this value requires that the engine speed (RPM) is also being read. Note that the injector duty cycle is read from a single memory location that is used for both the odd and even banks, so it may display unpredictable behavior during closed-loop operation (when the banks are being fueled based on different lambda feedback.)
Pulse width: Fuel injector pulse width in milliseconds. Like the injector duty cycle, this is read from a single memory location that is used for both the odd and even banks. Note that it may be possible for this value to exceed the available time between spark interrupts, which would mean that the engine is being under-fueled.

Options dialog

Serial device name: The name of the serial device connected to the 14CUX. If running Windows, this will be something like "COM2". If running Linux, it will be something like "/dev/ttyUSB0".
Speed units: Sets the preferred units of velocity for the road speed display.
Temperature units: Sets the preferred units of temperature for the coolant- and fuel-temperature displays.
Adjust road speed: Changes the road speed value displayed on the speedometer (and written to the log file) with a multiplier and/or an offset. This can be used to adjust this reading for cars that do not have a calibrated road speed sensor arrangement.
Enabled readings: These checkboxes allow the user to enable reading only certain parameters. This allows the limited bandwidth of the diagnostic port to be used for only those parameters that interest the user. If fewer readings are enabled, they will update more quickly and smoothly than if all the readings are enabled.
Periodically refresh fuel map data: When set, this causes the fuel map contents to be re-read from the ECU every few seconds. This can be useful when running a ROM emulator to tune a map on a running engine.
"Soft" fuel map cell highlight: Causes the display to show the weighted average of the four active fuel map cells by shading them in the same proportion. If this option is turned off, the display will round to the nearest row/column and show only a single cell as being active.

Idle air control dialog

This dialog allows the user to drive the idle air control (IAC) valve in the intake plenum. Because the ECU is adjusting the valve when the engine is idling, it is recommended to only test movement of the valve when the engine is off. Otherwise, unpredictable behavior may result.

The motor controlling the valve has a total of 180 steps of movement, so commanding it to open or close by 180 (or more) steps will ensure that it reaches its fully-open or fully-closed position (assuming the valve is functioning mechanically.) Note: after the IAC valve has been moved from its fully-open position, the ECU will only allow it to return to a maximum of 99% open until the next startup. This is normal behavior.
Battery-backed RAM display

Opening this dialog will read 20 bytes of battery-backed RAM from the ECU and display it in a simple tabular form. Refreshing the display can be done by closing and re-opening this dialog.


Penelope Stopit

11,209 posts

108 months

Thursday 8th October 2020
quotequote all
What on

was the outcome?

blaze_away

Original Poster:

1,502 posts

212 months

Friday 9th October 2020
quotequote all
Unfortunately the guy with the car has gone 'Radio Silence' on me so no idea what progress he has made. Hope he is OK in these worrying times.

When he does surface I'm sure info in this thread will help him.

Penelope Stopit

11,209 posts

108 months

Friday 9th October 2020
quotequote all
blaze_away said:
Unfortunately the guy with the car has gone 'Radio Silence' on me so no idea what progress he has made. Hope he is OK in these worrying times.

When he does surface I'm sure info in this thread will help him.
Oh no

Now wish hadn't asked

Yes, let's hope he's ok

davep

1,141 posts

283 months

Saturday 10th October 2020
quotequote all
Steve_D said:
Just had a quick read through that but cannot find reference to why pin 2 is the voltage reference rather than pins 15 or 19. All three pins see +12V during normal running.
Perhaps Mark (Blitz) with his close contacts with the RoverGauge gurus will be able to shed some light.

Steve

Edited by Steve_D on Sunday 4th October 09:58
Here you go Steve, pin 2 for Main Voltage sensing input:



Source: Disco Trouble Shooting Guide. Relay K116 nomenclature differs, but is commonly known as Main Relay.

Pin 2 is used by the ECM as an analogue input into its ADC chip, where its value is read and then processed by a service routine:

ADC Routine - Main (Voltage) Relay Sense - Channel 8 (8-bit conversion) (mainRelay.asm)
ADC service routines are entered with the newly measured ADC value in X00C8/C9 (only X00C9 for 8-bit readings). The A accumulator also contains the 8-bit reading.

The raw ADC value is not saved in RAM. The raw value is used in a calculation and the result is saved. The value in RAM goes up as the voltage goes down. Here is the formula:

Y = ( (100)X^2 - (189)X + 25608 ) / 4

Typically, the value starts around 1200 to 1400 (approx 12.6V) with power on, spikes to around 2000 during cranking (low voltage) and settles to around 900 to 950 (running voltage of 13+). The calculated value is the last phase of compensation during fuelling calculation.

It seems that if road speed is high, the voltage reading is saved as is, but if the road speed is low, the reading is only altered by 1 count per call.
(Courtesy: Dan Bourassa)

Hope that helps.

Penelope Stopit

11,209 posts

108 months

Saturday 10th October 2020
quotequote all
davep said:
Yet another incorrect diagram

87a is a normally closed contact

Should read as 87

blitzracing

6,387 posts

219 months

Saturday 10th October 2020
quotequote all
The logs are crazy- we have a normal idle that suddenly drops to 0 for just a few seconds, then back up to a few hundred RPM, but I dont think it can have been restarted so quickly (?) The MAF output also shows the engine is stopped when the RPM is zero. The voltage does not drop significantly until several second later. The RPM is far too low for the engine to keep running, so its possible we are loosing spark pulses, but it does not quite co inside with the drop of voltage- so whatever the ECU is reporting we still need to start with the basics of checking the 12v supply.

davep

1,141 posts

283 months

Sunday 11th October 2020
quotequote all
Penelope Stopit said:
Yet another incorrect diagram

87a is a normally closed contact

Should read as 87
Granted having an 'a' contravenes your need for Germanic compliance with DIN standards and the usual retort of DAS IST FALSCH!, but forget that for a minute and consider that Land Rover may have annotated the Main Voltage Relay and Fuel Pump Relay second 87 contact with an 'a' for troubleshooting purposes. Using the 87a annotation would allow LR technicians to easily distinguish between the relay load circuits, e.g. on the Main Relay 87 for Injector supply, with 87a for Main Voltage sense and AFM supply. Imagine telling a technician to check for continuity between ECM pin 2 and the Main Relay contact 87 or contact 87.

Penelope Stopit

11,209 posts

108 months

Sunday 11th October 2020
quotequote all
davep said:
Granted having an 'a' contravenes your need for Germanic compliance with DIN standards and the usual retort of DAS IST FALSCH!, but forget that for a minute and consider that Land Rover may have annotated the Main Voltage Relay and Fuel Pump Relay second 87 contact with an 'a' for troubleshooting purposes. Using the 87a annotation would allow LR technicians to easily distinguish between the relay load circuits, e.g. on the Main Relay 87 for Injector supply, with 87a for Main Voltage sense and AFM supply. Imagine telling a technician to check for continuity between ECM pin 2 and the Main Relay contact 87 or contact 87.
My need?

Haven't a clue what you're on about

A mistake has been made when the diagram was drawn and it managed to get past verification

Should read 87 and 87

or

87 and 87b

blaze_away

Original Poster:

1,502 posts

212 months

Wednesday 21st October 2020
quotequote all
Posted over on fb

No wonder the data looked odd.

quote
[i]
Simon Harry Firth
I gave in and sent it to my mechanic. He’s found the issue. It was a couple of frayed wires to the temp sensor. As she was heating up it caused them to short and send the temp sensor crazy. Repaired the wires and a new sensor. Problem seems to be solved. Thanks everyone for all the help. Now trying to get some time in her before I put her away for winter.
[/i]
Unquote

blitzracing

6,387 posts

219 months

Wednesday 21st October 2020
quotequote all
What gets me is the temperature readings are completely steady in the logs when it goes wrong , so I cant see how a shorted temp sensor would cause the fault. Time will tell.scratchchin