Electromotive XDi-2 ignition system problem

It looks to me like I have two problems and a couple of ideas to mitigate them.
1. The ECU wants a wants a higher amplitude trigger signal- Go to a smaller pickup gap say 0.020 inches.
2. Noise spikes could screw up timing - Rerout pickup lead away from the alternator as much as possible.
Do you think the 1/2 inch diameter sensor would give a taller amplitude?

 

Yesterday I closed the sensor gap to 0.020 inches. I then rerouted the pickup lead to provide about a 1.5 inch gap to the alternator housing. I also changed out the batteries but saw no improvement, so reinstalled the original one. I did find the battery terminal ground clamp connection untightened, so remedied this.

The following photo shows the new pickup mounted and the increased distance between the alternator housing and sensor lead.
Image Unavailable, Please Login

The next photo shows the resulting wave form.
Image Unavailable, Please Login

The sighal is now about 1.5 V , Noise spikes remain about 0.2 V.
I measured the pin 9 voltage at about 10.2 V when cranking. Worse than last time.

My starter quit on me. The wire to the starter solenid shows voltage when the key is in the crank position.but the solenoid does not click.

The plan is to remove the geared starter. Can anyone recommend a replacement geared starter that will spin my motor faster?
I do have the original starter and will look at getting someone to go through it before I consider putting it back on,

Do you think the starter motor could be the source of the noise?

I pulled the number 4-cylinder plug to check for spark with the improved wave form. No spark and no acknowledge by the two XDi2's of seeing the trigger signal.

 

I pulled the second XDi2 from my 26R build. It's not far enough along to run in the car, only on the dyno, so not in a position to test.
I took my Bosch starter to a local specialist. He found a sloppy drive gear so that at least will have to be sorted. It did turn over well on the bench.

 

David,
The car never started after being turned off and sitting a bit if I recall correctly. I was synching the carbs with a multibank manometer system, shut it off, and it did not start the next time I tried it. To be fair, I turned the motor over with baggies over the Webers. Effectivly a full choke condition. Eventually discovered no spark. The engine seemed generally hard to start since being installed in the car. I thought it was somehow tied to the new fuel pump in some way. Maybe it needed a one way valve somewhere?

Mark,

From the XDi2 manual:
============================================================================
7.0 Diagnostics Wiring mistakes cause a very high percentage of problems. The first step to diagnosing a problem is to check all the wiring. Also, remove the connector from the XDI-2 and make sure the pins are fully inserted. You may see a pin that is not inserted all the way to the edge of the connector. See Appendix A for more information regarding the 23-pin connector. Keep in mind that the engine will need the appropriate air/fuel mixture to operate correctly. Read your spark plugs to determine if the problem may be related to your fuel system instead of the ignition system. A simple timing light will let you verify if the XDI-2 is generating spark. The XDI-2 has a pair of status lights located next to the communication port. These lights must be visible while trying to diagnose a problem. The UPPER light will be known as the Check Engine Light. This light will be red with the ignition on (engine off). Otherwise this light will turn red when sensor failures are detected. The upper light is also used during firmware upgrade procedures. The following possible problems pertain only to the LOWER of the 2 LED’s : Problem: When I turn on the XDI-2 the lower status light is red and the car won’t start. Possible Cause: This will occur when the configuration (i.e. 4-cylinder) is not set in the software and the program file has not been downloaded to the unit. Refer to Wintec software users manual for engine settings configuration. Problem: When I crank the engine the lower status light stays solid green and the car will not start. Possible Cause: This problem is crank sensor related. The XDI-2 will flash red and green while the engine is starting. If the lower status light does not flash, the XDI-2 does not “see” the trigger wheel. The sensor could be bad. Measure the resistance between the signal wire and the ground wire of the sensor. The resistance should be approximately 620 ohms. The XDI-2 may intermittently see the sensor if the sensor wires are backwards. Refer to Section 4.3 for crank sensor wiring. A weak or non-existent crank signal will occur if you are using a 3/8” chisel point sensor with the incorrect sensor alignment. Refer to Section 5.3.2 for the appropriate alignment. Problem: The lower status light blinks during cranking but there is no spark. Possible Cause: In this situation, the coils are not getting power. Verify that the red with white stripe wire connected to pin D of the DFU connector is connected to +12V. Refer to Section 4.4 for power and ground connections. Problem: The lower status light does not turn on with the key. Possible Cause: The XDI-2 is not getting +12V and ground. Remove the connector on the XDI-2 and measure the voltage at pin 6 (+12V) relative to the main ground wire. If there is 12V at pin 6 and the unit does not turn XDI-2 Manual Version 1.3 Page 33  2015 Electromotive, Inc. on, there may be a problem with the unit. Contact your dealer. If there is not 12V on pin 6, refer to Section 4.4 for proper power wiring
=======================================================================
The pin-out for the 23-pin AMP connector. I need to map what pins are populated.

Pin Description Wire Color
1 Coil A White
2 GPO 1 White/Black stripe
3 Magnetic Sensor Shield Bare
4 GPO 2 White/Red
5 GP I/O 3 White/Green
6 +12V Switched Ignition Yellow
7 Knock Black
8 DFU 2 Coil A White
9 DFU 1 Coil B Red
10 Magnetic Sensor Ground Black
11 Magnetic Sensor Signal Red
12 GP I/O 4 White/Blue
13 External Retard Control Orange
14 Sensor Ground Black/White
15 DFU 2 Coil B Red
16 Shield for DFU 1 Bare
17 DFU 1 Coil C Black
18 Tachometer Brown
19 +5V Output Gray/Red stripe
20 MAP Signal Dark Green
21 Aux. Rev. Limiter Pink
22 DFU 2 Coil C Black
23 DFU 2 Shield Bare

 

Today I found someone to rebuild my OE Bosch starter. He's says about $100 compete but parts are about 4-5 days out. The drive gear is sloppy on its shaft. This is what we last used on the chassis dyno. I brought the whole ignition system to the dyno room and broke the engine in with it and did about 20 power pulls. Then I took my time and installed it on the car. All the movement broke the mag sensor shield wire at the end of the sensor.

Tomorrow I need to get the car higher in the rear and pull the port side rear wheel and liner. Then, pull geared starter bolts and starter. I wonder if anything came loose?

I need to take a hard look at the grounds. I run I think 8 AWG wire in shrink tubing for dedicated +12 V and ground wires direct to one relay that feeds a distribution block. This block supplies separate circuits for ecu (2 A), DFU1(10A), DFU2(10A), spare. The thing is I think I also tie in to chassis ground by hard bolting the DFU's and ground leads.

Thank you for the help.

 

Key off: No illumination of either LED
Key on: Top LED ( basically a check engine light) solid red; Bottom LED solid green.
Cranking: Top LED solid red; Bottom LED solid green.

Yes, I had to fabricate about a 2.5-ft extention to reach the ECU in the trunk.

Yes, I had to repin the extension side of the ECU connector as the shield (22 AWG) wire broke. Three new pins.

I mounted a new connector on the other end. It was provided by the sensor supplier for this purpose.
Image Unavailable, Please Login

We wrapped the connector gap with aluminum foil with an overlap of at least a couple of inches on the either side of the connectors, then grounded the foil to a cam cover bolt with allegator clips and 16 AWG wire. I did'n see any benefit on the scope, so discarded the foil shield.
The ECU side of the mag sensor shield is presumably grounded inside the ECU. I show continuity to the ECU housing. The sensor end floats. No continuity to ground. I believe this is best practice where only one end is grounded.

What do you think about a second layer sensor ground shield covering the wire near the alternator. Maybe ground this outer layer to the sensor body at engine ground?

I'm wondering if the my starting difficulty can be attributed to a defective starter motor? Could the noise also be attributed to the starter motor? I guess I'll have to wait about a week to find out.

Got car up in the air, ass high and took the driver side wheel and liner out. Now need get the failed geared starter out.

 

Pin Description Wire Color
1 Coil A White
2 GPO 1 White/Black stripe
3 Magnetic Sensor Shield Bare
4 GPO 2 White/Red
5 GP I/O 3 White/Green
6 +12V Switched Ignition Yellow
7 Knock Black
8 DFU 2 Coil A White
9 DFU 1 Coil B Red
10 Magnetic Sensor Ground Black
11 Magnetic Sensor Signal Red
12 GP I/O 4 White/Blue
13 External Retard Control Orange
14 Sensor Ground Black/White
15 DFU 2 Coil B Red
16 Shield for DFU 1 Bare
17 DFU 1 Coil C Black
18 Tachometer Brown
19 +5V Output Gray/Red stripe
20 MAP Signal Dark Green
21 Aux. Rev. Limiter Pink
22 DFU 2 Coil C Black
23 DFU 2 Shield Bare

Pins in red text are not populated
Pins in green text are not used and likely should be removed.

 

The new sensor appears nicely made. They use a shielded twisted pair, grounded on the plug end. I do have a 1/2 sensor with long lead but I'd have to drill or or change out the sensor pinch clamp. Do they make twisted pair with double shield conductor.

What about a simple low pass filter on the mag sensor line. Back in the day we shorted the input terminals of an A/D channel with a 1-2k resister to quite each thermocouple signal.

I counted about 15-16 noise spikes across 1 tooth cycle.

The electric fuel pump is disabled (pulled fuse)

I determined my cranking speed by:

(60 teeth/rev) X (2 rev/S) =120 teeth/S and generates a 120 Hz (DMM measured) saw tooth wave form.

2 rev/S =120 rpm

After I pull the starter I'll start looking at grounding and power distribution. I use only one relay for main power and distribute from there. A quick check appears to show a single point ground.

 

I don't believe we used an alternator, but not sure.
The wire you buy looks like what I should use. Twisted pair, shielded, high temp insulation. I suspect it's what is used on my new sensor.

I believe I may have found the cause of my difficulty. In disconnecting the ground lead from the battery, I noticed one of the cable clamp bolts, securing the main ignition ground, was loose and the lug could be easily rotated.
Image Unavailable, Please Login

In the above photo the large (8 AWG?) gauge black wire's lug was not tightly clamped.

The small red wire on the + terminal clamp was also loose. Not sure what this one does.
Image Unavailable, Please Login

Luckily, I haven't taken much apart yet. Won't know for sure till I get the starters swapped out. I guess I should retry the existing starter before I take it out.

 

Mark,
What other thread for the shield terminators?
I really don't like the extention Weather Tech connectors. These connectors are too bulky for the light (20-22 AWG) wires. What if I just use the pin pairs as 3-connectors then shield and seal locally? I did get a supply of Weather Tech pins from Custom Connectors.

 

I should get some of those terminators. Do I need one with and one without terminator to bridge a but connection?

I did get my geared starter out.

 

I meant to write, I should get some of those terminators. Do I need one terminator with a lead wire and one terminator without a lead wire to bridge a butt connection?