newbie flooding problem

Discussion in 'Injection Tech' started by Keith Miller, Nov 11, 2007.

  1. Keith Miller

    Keith Miller Farmall Cub

    I just completed GM TBI install on a Jeep 258 straight 6 stock motor. I running stock everything except a glasspack muffler. I have also installed a Motorcraft TFI coil on the stock distributor (weights and vacuum advance removed) with large rotor and distributor cap , 8.8 mm wires and am using a GM 8 pin EST. I have checked all voltages and run the Initial System Setup/Must Read checks. Timing is set a 0 with unplugged bypass, IAC is set a 500 rpm, TPS is .46 which is a little low, and fuel pressure is 12 lbs input, 1-2 lbs return. Fuel pump is an E2182 and ECM is 1227747. Donor vehicle was a Chevy Astro Van 4.3 Liter. On to the problem. The first few times I started the motor is started well and ran great. Now the motor will not start. The initial fuel sprayed into the Throttle Body during the Starter run seems to be flooding the motor. The only way I have found to get the motor started is to let the motor flood on the initial attempt to start, unplug the injectors and crank until the motor catches and then quickly connect the injectors again.

    Does anyone have any ideas as to how to correct this problem?

    Does the GM ECM have a flood circuit where flooring the accelerator will cut out the fuel flow to pass the overflow of fuel? It does seem to work in my application.

    During a couple of test drives, the horsepower seems a little weak with the current setting. I haven't built an ALDL cable yet, and will need to borrow a laptop for logging data. I need to get the starting issue fixed before trying any fine tuning.

    Bill noted in one of the threads that the GM tables are heavy on fuel at startup for an AMC motor, but I wouldn't think it was so bad that it would flood the engine.
  2. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    The ECM does have a clear flood mode.
    If the throttle is held above about 65% during cranking it will reduce the fuel while cranking.
    Have you connected pin A and B of the aldl to get the error codes?

    Have you rechecked the fuel pressure?
    You may want disconnect the return line and run it into a bucket to see if it makes a difference.

    You said it worked fine when you first started it and now it doesn't.

    Did you change anything? Try to remove the chip, change wiring?

    Have you performed the key on and engine run voltage checks?
    It could be as simple as the temp sensor connection.
  3. Keith Miller

    Keith Miller Farmall Cub

    I tried checking voltages again. Coolant temp was 3.75 Volts when the engine was cold, and 1.56V when the engine was all the way warm and idling at 700 rpm after about 14 minutes. TPS was .43 volts from B to A. Map was 4.73 from B to A. O2 sensor was .38 when engine was warm but not running. I performed the fuel return line to a bucket suggestion and noted significant fuel flow (I wasn't expecting this). I installed a pressure meter on the return fuel line and it is still reading about 2 lbs. Of note, with the pressure meter in the loop the motor did start regularily on two attempts. The return fuel line is 3/8 out of the Throttle Body, and is reduced to 1/4 hardline where I installed the meter near the charcoal canister. I seems that the motor coughs one time during attempts to start and if it catches, all is good. If it doesn't, I have to pull the plug on the injectors. The 65% WOT flood clear doesn't seem to work in my case, although I do have full travel using the gas pedal.

    The only ALDL code I'm getting is a 12.

    The only changes I made prior to this problem starting were to retard the timing to get to 0 degrees with bypass disconnected, and to cleanup the wiring and install it in split loom. I can't rule out missing a wire somewhere during cleanup, but it doesn't seem to be in the sensors and the motor runs well once it is started.
  4. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    Was the map reading with the eng running? 4.75 is basically 100 map.

    You may want to do the complete key on and eng running voltage checks as listed on sheet 1.
  5. Keith Miller

    Keith Miller Farmall Cub


    This is probably too much information, but just to show that I really am performing the checks you suggest

    Key On Cold Running 700 RPM

    12.48 V for 2 Sec. 14.0V A1
    --- --- A2
    --- --- A3
    113 mV 200mV A4 Not Connected
    85 mv 14.3V A5
    12.51 V 14.2V A6
    128 mV 300mV A7 Not Connected
    3-4.5 V 3-5 V A8
    4.99 V 4.98 V A9
    5 mV 11.3 V A10 Not Connected
    11 mV 11 mV A11
    8 mV 8 mV A12

    12.5 V 14.2 V B1
    12.4 for 2 Sec. 14.18 V B2
    .4 mV 2 mV B3
    --- --- B4
    21 mV 2 V B5
    --- --- B6
    4.99 V 4.98 V B7
    13 mV 13 mV B8 Not Connected
    --- --- B9
    3mV 14 V B10
    --- --- B11
    --- --- B12

    --- --- C1
    --- --- C2 No wire in Harness
    11.26 V .8 – 13 V switching C3
    2 mV .8 – 13 V switching C4
    11.29 V .8 – 13 V switching C5
    5 mV .8 – 13 V switching C6
    11 V 14 V C7
    --- --- C8
    0 V .1 mV C9
    3.72 V 1.44 V C10
    4.7 V 1.24 V C11
    --- --- C12
    .448 V .463 V C13
    4.99 V 4.98 V C14
    --- --- C15
    12.5 V 14.2 V C16

    .77 mV .82 mV D1
    10.1 mV 10 mV D2
    --- --- D3
    60 mV 1.2 V D4
    76 mV 4.73 V D5
    .8 mV 1 mV D6
    .39 V .5 V D7
    --- --- D8
    --- --- D9
    --- --- D10
    --- --- D11
    --- --- D12
    --- --- D13
    12.5 V 14.1 V D14
    --- --- D15
    12.5 V 14.1 V D16

    All voltages were checked at the ECM. The voltages look pretty good to me but you're the expert. Note: I was able to get ECM flood clearing to work by starting with the accelerator pedal up when engaging the starter and then go to WOT. The flood cleared and the motor started. Not an ideal way to start the motor but at least it does work without pulling the injector wires.
  6. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    There's never too much information.

    Couple thoughts....
    1. The voltage difference on D1 and D2 concerns me. It's not much but considering that they are soldered together just inside the ECM makes me think there is a bad ground somewhere.
    2. I am assuming that with the 4.98 on B7 that you have it tied to C14.

    3. Where are you getting your voltage from? Where is your 12V for B1 and C16 tied to and how are you powering A6/
    You should check your voltage drop on these during cranking.

    4. Is C9 connected? If it is, are you getting 12V during cranking on this pin?

    5. Where is your map sensor mounted?
    It should be above the vacuum port it is connected to on the back of the TBI. Make sure th hose is not holding fuel in it.
  7. Keith Miller

    Keith Miller Farmall Cub

    It's been a couple of days since I was able to troubleshoot the TBI. To answer your questions.

    1. I checks all the connections again and to only difference is that the wiring diagram I am using shows A11 tied to D2. All these are signal grounds, and the stock harness didn't have this connection. I went ahead and made the connection. I'm still getting .86mV at D1 and 11 mV at D2. Since the ground is being supplied by the ECM, and not the sensors, I don't know where the difference might be coming from. I still need to check my crimps at the engine ground, but I believe I made all the ground connections to the same point on the block.

    2. Yes B7 is connected to C14

    3. Voltage for B1 and C16 is connected straight to the Battery lead after the 15A fuse. A6 is powered from the relay, except that A6 and 8 pin module + are spliced and going to one contact on the relay via a single wire.

    Voltage is down to about 9 volts during cranking, the battery is getting weak from continuous cranking trying to get it started and to clear the flooding.

    4. C9 is connected to the starter solenoid and again is about 9 Volts during cranking.

    5. Map sensor is mounted above the vacuum port and the hose has no fuel in it.

    This problem is worst when the engine is hot. Most times I am able to crank the cold engine and get it started. I looks like the injectors are dumping way too much fuel during the initial cranking. Once started cold, or after clearing the flood the engine runs really well with no codes being generated, other than the VSS which isn't there.

    I should have data logging capability shortly after Thanksgiving. Will a data log do any good for the hot flooding problem?
  8. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    At the end of the wiring FAQ you will see where I talk about the grounds.
    D1 and D2 are soldered together inside the ECM just like A11 and A12.
    The D1 A11 connection is external in the harness for the sensor grounds.

    There is also a table in the program that will add fuel if the voltage is low.
    I would think this was added to assist with winter starts when the engine turns slow and the voltage is low.
    I normally zero these out when I set up a chip.
    Have you completely charged the battery and tried it?

    But some logged data may help.
    The BLM's along with the data log during nrmal running and then when turned off and restarted.
  9. Keith Miller

    Keith Miller Farmall Cub


    It's been a while since I have asked for any more assistance. I have the Jeep starting pretty reliability while cold, although it still floods badly when warm. I have been trying to log data, but after about 3 minutes of logging, the data stops (20) no longer pulses green. I added an LED to the cable prior to the Transitor in my logging cable to see if the data stream is being interupted, and it does not appear to be stopping, so I think the problem may be in the data actually going into WIN ALDL. I have to unplug the USB cable and shut down and restart ALDL to get data going again, and then it just stops again. I'm trying to troubleshoot other causes of the flooding and got to thinking that I had to replace one fuel injector from the orginal pair with another junk yard replacement. I have been trying to cross reference the fuel injector part numbers to determine if they are compatible with each other. One injector (from the original TBI) is GM 5235203RDP 2202 GM* and the replacement is GM 5235203 RPD 9070 GM. Could either one of these be from a V8? I have noted that the Rich Flag is toggling during times when I am monitoring WINALDL.

  10. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    the 5203 is the 4.3 injectors.

    The rich flag cycling during closed loop is normal since the ECM has to swing the fuel from rich to lean so it can always stay in the middle.

    Make sure there is nothing else running on the laptop and that there are no power savers/sleep enabled.
  11. Keith Miller

    Keith Miller Farmall Cub


    I was finally able to generate a .BLM file with over 30 minutes of running time. I have sent this file along with several .Log files that hopefully capture my troubles during warm cranking. These files have been sent to your email address.

    Thanks for the assistance,
  12. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    well definately have some issues.

    First I would recommend doing the initial setup procedures in the FAQ's.
    The TPS is too low.
    Next the IAC is maxing out at 85 counts.
    I'll have to look at your bin but all the bins I have dealt with are set at 145.
    You may have IAC problems.
    Have you verified it is working? There are a couple recent threads discussing this.

    Next you are always in clear flood mode.
    This makes me think you have a wiring issue concerning the crank sense wire C9 iirc.

    All of these will cause starting problems.
  13. lhc_cj7

    lhc_cj7 Farmall Cub

    Bill, can you post his log and bin. I'm intersted in seeing this. Thanks
  14. Keith Miller

    Keith Miller Farmall Cub


    I have reset the TPS and other initial setup procedures and now the TPS appears to be more in line with the expected values. IAC is indicating 140+ when the engine if first starting up but seems to settle into max 85 after warm up. I have swapped the IAC with another junkyard unit and the reading seem to be about the same. The Clear Flood flag seems to be intermintent, or may have to do with initial start up. It will often be set at initial start up and continue to be set for the duration of the drive, but it is often not present after a second startup. I have found that the engine will start up from cold, or after only several seconds of being turned off. Anything after a minute or two and I have the flood condition. The wiring looks like it is going directly to C9 and no where else, but I still need to ring it out just to make sure. I will be emailing you another log and BLM file. The drive was primarly highway for about 25 minutes. I know you would rather see a more all around sample of driving conditions and will try to get another BLM file soon. I was hoping you could give the log file another look and see if I'm getting closer to the baseline values
  15. Keith Miller

    Keith Miller Farmall Cub

    I have discovered on more piece to the flooding problem. After driving the Jeep for the last couple of months expecting that I will have to remove the fuel injector wires from the TBI when warm starting the engine, I finally have figured out that if I floor the gas pedal before cranking the engine, and keeping the pedal down during cranking, the engine with catch. If I floor the gas after cranking has started the damage is already done and I have to remove the injector wires to clear the flood. Once started the engine runs pretty well and I'm not showing any error codes except the VSS which is not installed. The VSS error will usually occur sometime in the next 15 minutes while decelerating.
  16. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    Got your email.
    Sorry you got lost in all my shuffle before.
    I have been doing 12hr days for a while now.

    Have you tried to remove the crank sensor wire?
    IIRC it's C9.
  17. lexum

    lexum Farmall Cub

    Wow, I finally found an active thread about the exact same problem I am having. I want to follow along here and try and add more data to help solve this TBI flooding issue.

    Some brief information about my setup:
    88 YJ 258, Howell TBI, TFI coil, Duraspark module(until now) - with resistor wire in place.

    Reasons for touching the working system on my Heep:
    I am only getting about 15 mpg. I took some WinALDL data and found that all of my BLMs were in the 112 to 118 range. The idle kept dropping low, probably because of the rich condition.

    I came across the same flooding problem this week while I tried to upgrade my ignition module to HEI. I couldn't get it working until I realized that the resistor wire needs to be bypassed. The TFI coil is pretty much useless with this resistor wire in place. Also, I read that the Howell TBI does not work properly with anything less than 12 volts from the coil?

    The engine runs wonderfully once it is started. However, it barely starts cold (fires at about 500 rpms and creeps up very slowly to a 1200 rpm cold start idle) and floods immediately on a hot start.

    I read about the "clear flood mode" in the manual, but it only worked part of the time. I now realize after reading this thread that you have to have the ECM powered before you floor it. Thanks for the info!

    What can I do to help? Should I get the same BLM data and compare? Why would the TBI start perfect every time with the original setup (Duraspark and resistor wire) and then flood with the HEI ignition module (I am assuming that Keith removed the resistor wire when he did the TFI coil)?

    I read somewhere that the Duraspark ignition module advances the timing during a hot start and the HEI module does not. Does this have anything to do with this issue?
  18. Bill USN-1

    Bill USN-1 Moderator Staff Member Moderator

    Howell FI normally is a fuel only system and you use your stock ignition.
    Choice is yours what model..duraspark or HEI.

    IIRC Kieth used a complete system with timing control so his problem should not be related to yours unless something really wierd happened.

    Now even with full efi you can chose what coil to run..but that does not mean the complete ignition system or module.

    The GM E-core and the ford TFI coils are both popular choices.

    So I recommend you start with the basics like a complete voltage check and then the fuel pressure on the inlet and return lines.
    Then try to log some data to see how the sensors are operating.
  19. lexum

    lexum Farmall Cub

    This is exactly what I am getting at!

    Has Keith solved his problem yet? I suspect it has nothing to do with his FI setup. We are having the same symptom regardless of whether our system controls spark or not.

    I have checked my fuel pressure and tried running with the return fuel in a gas container. There is no difference. Fuel pressure is not my issue.

    While I was out checking my fuel pressure I ran the following tests with a spark gap of .045", duraspark, distributor upgrade with 8mm wires:

    1. Starting with HEI ignition module, bypassed resistor wire = fires once or twice and then dies, flooded

    2. Starting with HEI ignition module, with resistor wire = fires once or twice and then dies, flooded

    3. Starting with Duraspark ignition module, bypassed resistor wire = starts with low rpm and creeps up about 1200 rpm

    4. Starting with Duraspark ignition module, with resistor wire = immediately starts and jumps above 1200 rpm

    Once running, I do not have any issues other than a rough idle with the Duraspark setup. Is there reason to believe that I should run the Duraspark IM with resistor wire and spark gap to .035". Like I said before, I think the TFI coil is useless if it is only powered with 9V.

    As soon as I get access to my wife's laptop I will log data for all of these setups to see which of these setups is running closest to stoich. Remember, my reason for going HEI is to try and correct the rich condition. If HEI works then I can setup the 8-pin module, add spark to my chip, and run Full EFI like you guys!

    Thanks for the quick response Bill.
  20. lexum

    lexum Farmall Cub

    After toying with the HEI and spark gap and timing and fuel pressure and injectors and clear flood mode... I have given up on the HEI ign mod for now. I tried 2 different HEI ign mods and the engine floods before they can fire up. I no longer believe that the HEI module will improve my mpg anyway.

    I returned to the Duraspark with resistor wire and the system start beautifully.

    So back to the rich running Howell TBI. There seems to be too much fuel all the time not just while starting. Last week, I did some datalogging with the Duraspark. The data showed that I was running in the 110 BLMs, rich.

    I wish there was a way to post the txt file from WinALDL...

Share This Page