Let's talk MEFI

[GaryDoug]

The data logging file (saved in the same folder as the mefi4.exe file with the extension .csv and the days date in the filename) with show the trouble codes. There are two ways to create that file. One is to click on the "Save scan to log file" blue button at the end of the scan. The other way is to check the box labeled "Auto-save scan file" before the scan starts.

[rakr]

@GaryDoug do the error codes start in Column AR?

Thanks again!!!

[GaryDoug]

BM. The first record (line) in the file has the column names . You want DTC List.

Edited May 31, 2022 by GaryDoug

[rakr]

Ok… so it looks like I got some problems… 81-12,17,20,21,24, etc.

Lots of codes, but the boat seems to run great. tuneup within the last 30 hours (albeit last summer), and injectors cleaned, all fluids recently done and seem to be at good levels. 

Anyone know of a good place to find a good consolidated list? I found some random places with a bit of info, but not a great list. Thanks!!!

[GaryDoug]

44 minutes ago, rakr said:

Ok… so it looks like I got some problems… 81-12,17,20,21,24, etc.

Lots of codes, but the boat seems to run great. tuneup within the last 30 hours (albeit last summer), and injectors cleaned, all fluids recently done and seem to be at good levels. 

Anyone know of a good place to find a good consolidated list? I found some random places with a bit of info, but not a great list. Thanks!!!

First, if any of the codes shown have an "L" at the end, it means the code was stored sometime in the past and is not necessarily current. Codes without the "L" at the end are current.

The codes prefixed by an 81 were added later by the designers and may not relate to your specific engine's ECM. The number after the 81 was added by me in the program to limit the length of the code description so that many could fit in the column. That number has no significance elsewhere as far as I know. The real description of each code is shown on the DTC tab. If this were my engine, I would probably ignore any trouble codes prefixed by an "81-". Below is a correlation list for those. It shows you the full description in the same line with the 81-xx code.

DTC81-Crank Signal Fault,81-1
DTC81-Crank Sync Fault,81-2
DTC81-Cam Signal Fault,81-3
DTC81-TAC,81-4
DTC81-Fuel Pump High,81-5
DTC81-Fuel Pump Low/Open,81-6
DTC81-Inj A High,81-7
DTC81-Inj A Low/Open,81-8
DTC81-Inj A High,81-9
DTC81-Inj B Low/Open,81-10
DTC81-Recirc (J1-32),81-11
DTC81-V5 Ref Range,81-12
DTC81-DEPSPWR,81-13
DTC81-Can bus Fault,81-14
DTC81-Oil/Cat high,81-15
DTC81-Oil/Cat low,81-16
DTC81-TPS2 Range,81-17
DTC81-TPS1 Range,81-18
DTC81-PPS3 Range,81-19
DTC81-PPS2 Range,81-20
DTC81-PPS1 Range,81-21
DTC81-Bad Checksum,81-22
DTC81-No TAC Reply,81-23
DTC81-Bus Contention,81-24
DTC81-Serial Comm Fault,81-25
DTC81-TPS12 Corr,81-26
DTC81-PPS23 Corr,81-27
DTC81-PPS13 Corr,81-28
DTC81-PPS12 Corr,81-29
DTC81-Limited Authority,81-30
DTC81-Actuation Fault,81-31
DTC81-Process Fault,81-32
DTC81-Not Tracking,81-33
DTC81-Throttle Ret Fault,81-34

 

Edited May 31, 2022 by GaryDoug

[UWSkier]

42 minutes ago, rakr said:

Ok… so it looks like I got some problems… 81-12,17,20,21,24, etc.

Lots of codes, but the boat seems to run great. tuneup within the last 30 hours (albeit last summer), and injectors cleaned, all fluids recently done and seem to be at good levels. 

Anyone know of a good place to find a good consolidated list? I found some random places with a bit of info, but not a great list. Thanks!!!

I got the same sort of thing.  Bunch of error codes that pop up for a single data point and are gone.  I attributed it to noise on the wire.  I don't really care unless something that throws a CEL shows up or something is coming up consistently.  The vast majority of my data points don't have any DTCs.

[rakr]

Awesome!!! Ill check those out! Thanks fellas! 

[rakr]

@UWSkier so are you saying that I should be looking at the entries (rows) that have a 1 in the CEL (horn/alarm ON) column? In that case, I only have one code 81-24. Thanks! 

[rakr]

@GaryDoug a nice feature improvement would be to have the "plain English" Error code output into the excel sheet, vs. just the code.

I did a Microsoft Stream recording of the Troublecode tab after running it for a while, and was able to visually capture these codes, which lined up pretty well with the excel output from the MEFI4 Application. I am going to throw up another post to not hijack this one with my troubleshooting.

 

[Ditri]

Hi mefi readers. I read whole thread. Please can you tell if im right? In our country - there is no somebody with MEFI4 reader. I have MC boat :-) with Indmar TBI and MEF4. My problem is, that the boat throw me MIL at idle. If i push more throtlle - MIL blinking stop. Im absolutely fool about it. Im chasing this problem about two years. There are no stored codes in the MEFI readable by standard blinking procedure.. As i understand, this is some kind of standard GM codes which is present only at the moment (idle). It blinks - simple signal 2Second light, pause, 2Second light pause....the video with the code is here, but it is problably not important: 

Please do you thing if this type of code is readable with the MEFI 4 for Mercruiser app. for Windows and the TTL-232 cable + diode setup?
Or if i can track the problem by this solution?

Thank you so much for some answer or idea cos im hunting this problem long time. Boat is ran normally. Oil pressure, temp, voltage are in normals... I was checked everything twice and replace many sensors..

By the way MC forum is without answer, and here? Here are the guys which are making cables and read the MEFI by yourself - respect!!

Edited July 4, 2022 by Ditri

[justgary]

@Ditri - What year is your boat?  I don't know what Indmar sold to MC, but I have a hard time believing that you have a TBI engine with MEFI4.  I would expect MPI with MEFI4.  Perhaps it is MEFI2?

In any event, yes, you can build the cable and read the data.  If you have trouble interpreting what you see, come back and post your results.

[Ditri]

Thank you so much for the info. Yes it is real - it is Prostar 205V (with OEM X-star graphic) probably 2002...it is on the transmission sticeker -> https://drive.google.com/file/d/12kzm2OzYfhtugk7FhSlw-nontK2Taz-x/view?usp=sharing

On the sticker is 5.7 TBI MEFI-4

Do you think that it will work? I see in some previous post, that somebody was writing about "Temporary GM codes." 

 

Edited July 4, 2022 by Ditri

[electricjohn]

My mefi-4 is batch fire, same as TBI.

[Ditri]

Hello guys.

I was connected the Indmar TBI and run the MEFI scanner. Everything works super nice.

But, i can not get some understable results :-( sorry that im writing here. But i was looking to the codes, and nothing is clear for me.

The problem is, that the boat write me - "General Warning 2" but i can not find, what this mean. I was searching on the internet -> and it looks, that it can be "Low oil pressure"? Im not sure. I was try to search real time  "Oil pressure" numbers in the software, but im not sure, which numbers, are data from the Oil pressure sender unit.

There is the row - Oil Pressure/CAT (V) but it is constant number - no matters on the RPM

Please can you give me some push - for the next way? Thank you so much. 

HERE are the screenshots from software and one log file. 
https://drive.google.com/drive/folders/1tZXuuc7XXT1ZUBrIPJ5Ofbbz-TwH1z5S?usp=sharing

Thank you so much for any help...

 

 

Edited July 10, 2022 by Ditri

[Steve B.]

Be interesting to see responses to Ditri's post. I have a feeling you can clear those items in green. But I am guessing the one that says "fuel pump on" should stay on. I think if the ECM doesn't see oil pressure it will not allow the fuel pump to run, so the oil switch signal should be good.

I have a constant alarm on, but all vitals are good. 

Steve B.

[justgary]

On 7/9/2022 at 5:40 PM, Ditri said:

Hello guys.

I was connected the Indmar TBI and run the MEFI scanner. Everything works super nice.

But, i can not get some understable results :-( sorry that im writing here. But i was looking to the codes, and nothing is clear for me.

The problem is, that the boat write me - "General Warning 2" but i can not find, what this mean. I was searching on the internet -> and it looks, that it can be "Low oil pressure"? Im not sure. I was try to search real time  "Oil pressure" numbers in the software, but im not sure, which numbers, are data from the Oil pressure sender unit.

There is the row - Oil Pressure/CAT (V) but it is constant number - no matters on the RPM

Please can you give me some push - for the next way? Thank you so much. 

HERE are the screenshots from software and one log file. 
https://drive.google.com/drive/folders/1tZXuuc7XXT1ZUBrIPJ5Ofbbz-TwH1z5S?usp=sharing

Thank you so much for any help...

Ditri -

From what I can tell by reading the MEFI-4 Manual, "General Warning 2" is an optional input to the ECM defined by the user (Indmar).  Check your ECM harness for a Tan/Black wire at position J1-4.  If it is present, try to trace it back in the harness to see where it comes from.  I would guess maybe the transmission oil level or temperature.  It should put the engine into limp mode (limited RPM) while it is active.

[Ditri]

Gary - Tahnk you so much!
The idea about trakcing the wire from J1-4 is looking like the solution. I will 100% try it. Please what do you think? If i will remove the wire form - J1-4 pin -> will it disconnect the general warning signal and the MIL wil stop blinking? Or it will be blinking permanently?

Thank you so much!
 

[CJR]

I am confused about how adding a diode to the transmit line from a 5v TTL serial interface keeps from pulling down the ecu serial output.  Will someone please elaborate on this?

Thanks

CJR

[justgary]

9 hours ago, CJR said:

I am confused about how adding a diode to the transmit line from a 5v TTL serial interface keeps from pulling down the ecu serial output.  Will someone please elaborate on this?

Thanks

CJR

Note that we are using the serial interface in a one-wire bus, where transmit and receive are tied together on the bus.  For normal two-wire RS232 communication, the transmit and receive lines are kept separate.

The transmit line on the serial device has a pull-up resistor to keep it at 5V while the line is idle.  The receive line cannot be tied directly to the transmit line on some devices because that will force the receive line to only follow the transmit line, which means that the device will not respond to transmissions from the ECM.  The diode decouples the transmit and receive lines on the device so that an ECM transmission can pull the receive line low.

Some devices seem to not need the diode.  Presumably this is because the pull-up resistor is higher value (lower pull up current), so the ECM can still pull the receive line low.  Adding the diode will not harm anything, so it is best to add it to your device when you build the cable instead of finding out later that you needed it.

[GaryDoug]

32 minutes ago, justgary said:

Note that we are using the serial interface in a one-wire bus, where transmit and receive are tied together on the bus.  For normal two-wire RS232 communication, the transmit and receive lines are kept separate.

The transmit line on the serial device has a pull-up resistor to keep it at 5V while the line is idle.  The receive line cannot be tied directly to the transmit line on some devices because that will force the receive line to only follow the transmit line, which means that the device will not respond to transmissions from the ECM.  The diode decouples the transmit and receive lines on the device so that an ECM transmission can pull the receive line low.

Some devices seem to not need the diode.  Presumably this is because the pull-up resistor is higher value (lower pull up current), so the ECM can still pull the receive line low.  Adding the diode will not harm anything, so it is best to add it to your device when you build the cable instead of finding out later that you needed it.

That certainly can be true but I have found another reason. I have noticed in the past that a cable without the diode may or may not work for different ECM's (for example boat versus car). That is because some ECM's have a higher value series resistor in the transmit line to protect the driver from overload damage. That coupled with a strong pull-up in the cable device can cause the same problem of inadequate received signal voltage. The diode blocks the effect of the pull-up in the cable.

I developed this app originally for F-body GM cars with the LT1 engine. I never saw an instance of any cable with no diode failing to work with those cars. But when using it on other cars (or boats), the results were not the same and some needed the diode.

Edited September 20, 2022 by GaryDoug

[CJR]

So, the FTDI chip has a strong pull up on the Tx.

I assume the ECM also has a pull up on it's Tx line.  

Do the RX(s) have a pull up or pull down resistors?

 

So here is how it goes.  

When FTDI Transmitting or idle:

FTDI outputs 5v (logic 1), Diode is reverse biased (effectively stopping current flow from FTDI pullup).  Then the ECM TX pull up keeps signal at 5v so ECM RX can receive a "1".  

FTDI outputs 0v (logic 0), diode forward biased and output set to junction voltage of diode( 0.7v).  ECU RX sees 0.7v as logic 0.  

 

When ECU Transmitting or idle:  

ECU outputs 5v (logic 1), diode barely reverse biased (but still no current flow from FTDI pullup).  FTDI RX sees logic 1. 

ECU outputs 0v (logic 0), diode reverse biased, FTDI RX sees logic 0.  

 

The diode is reverse biased in all cases except when FTDI Tx is a logic 0.  So, FTDI pull up has little to no effect on circuit.

 

Please correct this as needed, and thanks for the information.  

 

 

[GaryDoug]

1 hour ago, CJR said:

So, the FTDI chip has a strong pull up on the Tx.

I assume the ECM also has a pull up on it's Tx line.  

Do the RX(s) have a pull up or pull down resistors?

 

So here is how it goes.  

When FTDI Transmitting or idle:

FTDI outputs 5v (logic 1), Diode is reverse biased (effectively stopping current flow from FTDI pullup).  Then the ECM TX pull up keeps signal at 5v so ECM RX can receive a "1".  

FTDI outputs 0v (logic 0), diode forward biased and output set to junction voltage of diode( 0.7v).  ECU RX sees 0.7v as logic 0.  

 

When ECU Transmitting or idle:  

ECU outputs 5v (logic 1), diode barely reverse biased (but still no current flow from FTDI pullup).  FTDI RX sees logic 1. 

ECU outputs 0v (logic 0), diode reverse biased, FTDI RX sees logic 0.  

 

The diode is reverse biased in all cases except when FTDI Tx is a logic 0.  So, FTDI pull up has little to no effect on circuit.

 

Please correct this as needed, and thanks for the information.  

 

 

Correct.

Also be aware that in normal operation, both devices are not transmitting at the same time. But that can happen at the start, and then the cable has to take control of the bus and go into a primary/secondary role over the ECU by sending a "silence" request to the ECU between the data bursts. Once the request is answered, the cable takes control and sends requests for data and the ECU responds with the data. If no transmissions occur in 3 seconds, the ECU resumes normal operation.

[justgary]

59 minutes ago, CJR said:

So, the FTDI chip has a strong pull up on the Tx.

I assume the ECM also has a pull up on it's Tx line.  

Do the RX(s) have a pull up or pull down resistors?

 

So here is how it goes.  

When FTDI Transmitting or idle:

FTDI outputs 5v (logic 1), Diode is reverse biased (effectively stopping current flow from FTDI pullup).  Then the ECM TX pull up keeps signal at 5v so ECM RX can receive a "1".  

FTDI outputs 0v (logic 0), diode forward biased and output set to junction voltage of diode( 0.7v).  ECU RX sees 0.7v as logic 0.  

 

When ECU Transmitting or idle:  

ECU outputs 5v (logic 1), diode barely reverse biased (but still no current flow from FTDI pullup).  FTDI RX sees logic 1. 

ECU outputs 0v (logic 0), diode reverse biased, FTDI RX sees logic 0.  

 

The diode is reverse biased in all cases except when FTDI Tx is a logic 0.  So, FTDI pull up has little to no effect on circuit.

 

Please correct this as needed, and thanks for the information.  

 

 

As I recall, the symptom is that the FTDI chip cannot hear the ECM when it transmits, which means that the ECM Tx cannot pull the FTDI Rx line low enough.  I did not go back and look at the schematic or play electron, but the diode allows the ECM transmitter to pull the line low during transmit so that the FTDI Rx can see the transmitted 0 while the FTDI Tx is trying to pull the line high.

I agree with @GaryDoug that the primary cause is the pullup on the ECM side.  Enough current has to flow to allow the FTDI Rx to see a low.

[GaryDoug]

8 minutes ago, justgary said:

As I recall, the symptom is that the FTDI chip cannot hear the ECM when it transmits, which means that the ECM Tx cannot pull the FTDI Rx line low enough.  I did not go back and look at the schematic or play electron, but the diode allows the ECM transmitter to pull the line low during transmit so that the FTDI Rx can see the transmitted 0 while the FTDI Tx is trying to pull the line high.

I agree with @GaryDoug that the primary cause is the pullup on the ECM side.  Enough current has to flow to allow the FTDI Rx to see a low.

It's not so much the ECU pull-up resistor but rather the series resistor value in the TX line. That makes the ECU a "soft" transmitter.

Example: The cable pullup is 10k OHMS and the ECU series resistor is also 10k Ohms. So, the output cannot go below 1/2 the supply voltage without the diode being in series.

For ECU's without the resistor, the output can be damaged by a defective/shorted scantool.

 

Edited September 20, 2022 by GaryDoug

[CJR]

Some trivia...  My USB to TTL cable employs an FT232R chip.  The chip has an eeprom that allows you to configure "High Output Drive Level" for the UART I/O Pins.  If enabled, the pins will source 12 mA.  Otherwise they source 4 mA.  The default for my eeprom was not enabled (the better choice for me).  

Now waiting for cable supplies from Mouser.   Will be using MEFI3-Scan (beta1.1) from @GaryDoug.  So, thanks for the scan software Gary.  Nice to see the air cover for an engine like mine on the startup screen.  Also thanks for indulging me with the diode specific questions.