It just stopped working!

david jenkins · Post by **david jenkins** » Sat Jul 05, 2008 11:07 am

Bob,

Following on from what Brent said - is it possible that you had the engine running with the battery disconnected at any time? The reason I ask is that the battery forms part of the voltage regulation circuit in a car, and you can get very high voltages with it disconnected.

Just a (small) thought...

David

FazerBob · Post by **FazerBob** » Sat Jul 05, 2008 4:15 pm

Hi David, no to the battery, this suddenly stopped working while I was driving. I had overtaken a couple of cars with the engine pulling strongly, no problems at all, then a few minutes later went to overtake and it was like running into a brick wall! as the revs passed 3000 ish just constant stumbling and misfiring.

FazerBob · Post by **FazerBob** » Sat Jul 05, 2008 4:36 pm

Yet another update for Brent: I have soldered in a new diode at D3, and got the following voltages:-

12.06 at the Molex
D2 11.53 (band end) 12.06 at other end
V Reg 11.5 (nearest Molex) 5.03 (far end)
U1 5.01
U2 5.01
IC1 11.34
PS1 5.01
QG1 5.02

Again laptop connected after checks but no communication with MJ This got me wondering if I was mistaken about having communication between the two yesterday, so I went back and replicated the tests I had done, ie. 12v(only 9v available at the moment) feed, and no D3 in the circuit. I got figures as follows:

9.17v at the Molex
D2 8.49(band) - 9.2 (other end)
V Reg 8.49 (near) - 5.0 (far end)
U1 5.08
U2 5.09
IC1 8.48
PS1 5.08
QG1 5.03

but still no communication between laptop and MJ. what do I do now?

Post by **brentp** » Sat Jul 05, 2008 6:08 pm

What criteria were you using then (and now) to determine the MJLJ was alive and communicating?

FazerBob · Post by **FazerBob** » Sat Jul 05, 2008 6:22 pm

I opened up an 'empty' map on my laptop - zeroes in all bins, then clicked on the icon to retrieve configuration from the MJLJ. Screen flickered once, but the view stayed the same. I then opened a Zetec map on the laptop and attempted to write it to the MJ, with the same result.

Post by **brentp** » Sat Jul 05, 2008 6:43 pm

Ah- so to clarify, what behaviour did you observe before, when you thought it was working correctly?

What I usually do for testing is to fire up the software with the MJLJ powered down. As soon as you power up the controller, it will connect and read the runtime data- and the timeout message goes away. Further testing involves reading the ignition configuration and observing if the data populates into the grid.

Did anything like this happen earlier when you thought the controller was working?

FazerBob · Post by **FazerBob** » Sat Jul 05, 2008 8:31 pm

brentp wrote:Ah- so to clarify, what behaviour did you observe before, when you thought it was working correctly?

What I usually do for testing is to fire up the software with the MJLJ powered down. As soon as you power up the controller, it will connect and read the runtime data- and the timeout message goes away. Further testing involves reading the ignition configuration and observing if the data populates into the grid.

Did anything like this happen earlier when you thought the controller was working?

Sorry Brent I don't quite follow your last post - this happens when someone with the brains to invent MJLJ etc tries to communicate with someone (me) who struggles to read the morning paper!
When I was testing yesterday I had the software running on my laptop with a basic map visible. I then clicked on the icon with the blue arrow to 'get ignition configuration' The screen refreshed and then displayed the map I had been running on my car (or so I thought!) As it was towards the end of a very long day, perhaps I made a mistake and had been looking at 'my' map all along.
I have just returned from the garage where I ran the MJLJ Configurator V3.2.1, (with 'no file loaded' in the header banner) It was running with the laptop connected to the MJLJ, but no power to the MJLJ. I then supplied power to the MJLJ, and after waiting about 30 seconds I clicked on the 'get ignition config' icon. Nothing happened.

Post by **brentp** » Sun Jul 06, 2008 9:07 am

Bob,

Thanks for your testing and patience! Best to be careful, explicit and over-communicate.

I thought of two more things we can test.

1. Test if the oscillator, QG1 is working correctly.

The oscillator feeds a clock signal to the processor, which drives the execution of the firmware code. With the unit is powered up, test the pin diagonally opposite of the pin marked 1 on the oscillator. Refer to the following picture- note you'll have to test from the backside of the board.

What voltage do you see? If the oscillator is working correctly, you should see approximately 2.5V on that pin, showing the oscillator is outputting a signal.

2. The controller may not be communicating with the PC, but there's an outside chance it may still be partially alive. One way to verify this is to plug it back into your engine's harness, run the engine and see if it's altering ignition advance using a timing light.

If, by chance you can observe the MJLJ controlling timing, then we should determine why it's not able to communicate with the PC. Double check the serial port configuration (ie. you're using the right com port, etc)

If so, it seems that whatever electrical stress the controller apparently experienced might have affected the active components.

One or more of these components may be bad:

U2 (RS232 Serial transceiver),
QG1 (Oscillator for processor)
U1 (Processor)
IC1 (ULN2003 transceiver)

A bad QG1 or U1 will render the controller non-functional.
A bad or damaged U2 may affect serial communications and/or communication with the EDIS module.
A bad IC1 will affect user defined outputs; something not immediately important right now.

If it seems we've reached the end of the rope send me a PM and we'll see what we can do to get your unit working again and make you whole.

FazerBob · Post by **FazerBob** » Sun Jul 06, 2008 11:54 am

Hi Brent, I wasn't intending any critiscism, I'm glad you're covering every step, I certainly need everything spelling out.
Now for this morning's testing:
Bench test of the oscillator, QG1 shows 2.52v at the pin diagonal to pin 1
I then fitted the MJLJ back into the car, connected up the laptop, and started the car - everything is working! laptop is showing an empty bin grid ie; all zero's but the red colour is moving around. So we have:-

All zero's in bins, red highlight moving, RPM correct, advance 22 to 23(as per 'my' map) KPa 60 and advance moves when checking with timing light

Next I clicked to get ignition config:-
All bins are correct to 'my' map, red moving 22/23/24, RPM correct, advance moving correctly

I then switched the engine off, which left the laptop still showing it's last readings, idle RPM, advance etc - is this right?

Started engine again, ran through tests for a couple of mins, then suddenly laptop just showed static, empty map. No communication between laptop and MJLJ, advance not working. I checked the oscillator again, and it was still showing 2.5v at the correct pin.

Now it's time for me to go and watch the GP, and wait for your reply,
cheers Brent,

Post by **brentp** » Sun Jul 06, 2008 6:20 pm

Hey, that's good- it seems the controller is alive.

Let's clarify some things:

FazerBob wrote: Next I clicked to get ignition config:-
All bins are correct to 'my' map, red moving 22/23/24, RPM correct, advance moving correctly

So when you clicked the button to read the configuration, it populated the main ignition grid with your map and bins- all the advance numbers showed up?

FazerBob wrote: I then switched the engine off, which left the laptop still showing it's last readings, idle RPM, advance etc - is this right?

Yes, it freezes on the last information received from the controller.

FazerBob wrote: Started engine again, ran through tests for a couple of mins, then suddenly laptop just showed static, empty map. No communication between laptop and MJLJ, advance not working. I checked the oscillator again, and it was still showing 2.5v at the correct pin.

A little confused by this. When you have your laptop monitoring the controller, it periodically retrieves only small bits of information from the controller: RPM, advance, load, user output states, etc. The ignition config data (ignition map, bins, and other config values) are only retrieved when you click the 'read configuration' button. Therefore, if the config software was still up and running from the previous test, it should still show the Ignition Configuration from when you last successfully retrieved it.

So, when the laptop loses communication, the runtime gauges (RPM, load, advance) will freeze and you'll see an error message at the bottom of the configuration software window.

Also, when you say above "... advance not working." Did you mean the advance stopped working according to your timing light?

See if you can get it working again- and while you're doing it, verify both communications with your laptop *and* whether the advance is being controlled through the use of your timing light.

Keep in mind that a communications problem between the controller and your laptop does not necessarily mean it's not alive and controlling ignition advance.

I think we're getting closer! Keep at it.

When you 'read' the ignition map from the controller

FazerBob · Post by **FazerBob** » Sun Jul 06, 2008 9:45 pm

brentp wrote:Hey, that's good- it seems the controller is alive.

Let's clarify some things:
FazerBob wrote: Next I clicked to get ignition config:-
All bins are correct to 'my' map, red moving 22/23/24, RPM correct, advance moving correctly
So when you clicked the button to read the configuration, it populated the main ignition grid with your map and bins- all the advance numbers showed up? YES

FazerBob wrote: I then switched the engine off, which left the laptop still showing it's last readings, idle RPM, advance etc - is this right?
Yes, it freezes on the last information received from the controller. OK

FazerBob wrote: Started engine again, ran through tests for a couple of mins, then suddenly laptop just showed static, empty map. No communication between laptop and MJLJ, advance not working. I checked the oscillator again, and it was still showing 2.5v at the correct pin.
A little confused by this. When you have your laptop monitoring the controller, it periodically retrieves only small bits of information from the controller: RPM, advance, load, user output states, etc. The ignition config data (ignition map, bins, and other config values) are only retrieved when you click the 'read configuration' button. Therefore, if the config software was still up and running from the previous test, it should still show the Ignition Configuration from when you last successfully retrieved it. CORRECT, BUT 'FROZEN' DETAILS OF RPM,ADV ETC. I then opened the blank,empty map from my laptop, and clicked on 'get ignition config' but nothing changed and I still had the blank map showing.
So, when the laptop loses communication, the runtime gauges (RPM, load, advance) will freeze and you'll see an error message at the bottom of the configuration software window. NO ERROR MESSAGE (but that may be because I have the screen filled with the MJLJ) there may be a message below that I haven't seen.
Also, when you say above "... advance not working." Did you mean the advance stopped working according to your timing light? CORRECT

See if you can get it working again- and while you're doing it, verify both communications with your laptop *and* whether the advance is being controlled through the use of your timing light. I'll run it again tomorrow, when I get home from work, but when the laptop and MJLJ stopped communicating today my timing light showed the advance was static at approx 15 degrees, even though I was revving the engine up to 3000 rpm.

Keep in mind that a communications problem between the controller and your laptop does not necessarily mean it's not alive and controlling ignition advance.

I think we're getting closer! Keep at it.

Hope this answers a few questions Brent. I did wonder today if the problem was related to how long the car had been running (approx 15 mins) before the laptop and MJLJ stopped talking to each other.

When you 'read' the ignition map from the controller

Post by **brentp** » Mon Jul 07, 2008 3:05 pm

Hi Bob,

Any updates on further testing? Have you been able to observe the controller work for any extended period?

One thing to check is the PIP connection from the EDIS module. if you have an intermittent connection, the RPM signal would freeze in the configuration software- but granted, you'd still be able to communicate with the unit otherwise.

Your comment about a static 15 degrees is interesting. The EDIS 'limp home' mode fixes the the advance at 10 degrees (observable when the MJLJ is disconnected). Could it be that your trigger wheel is mounted 5 degrees off?

I'm curious because I'm wondering if the MJLJ is commanding that advance. One way to tell is to measure your static advance with the MJLJ disconnected.

Let us know the latest.

FazerBob · Post by **FazerBob** » Mon Jul 07, 2008 4:23 pm

brentp wrote:Hi Bob,

Any updates on further testing? Have you been able to observe the controller work for any extended period? No change from yesterday - still showing 'blank' map, and no readings for rpm, adv etc

One thing to check is the PIP connection from the EDIS module. if you have an intermittent connection, the RPM signal would freeze in the configuration software- but granted, you'd still be able to communicate with the unit otherwise. Connections checked, all appear to be ok

Your comment about a static 15 degrees is interesting. The EDIS 'limp home' mode fixes the the advance at 10 degrees (observable when the MJLJ is disconnected). Could it be that your trigger wheel is mounted 5 degrees off? No trigger wheel. I am using the Zetec flywheel, with the original sensor. Flywheel can only be fitted in correct orientation

I'm curious because I'm wondering if the MJLJ is commanding that advance. One way to tell is to measure your static advance with the MJLJ disconnected. You may have hit on something there. I've just tested again - advance with MJLJ disconnected is a steady 10 degrees. Reconnecting the MJLJ moves the advance to approx 15 degrees, but doesn't alter when engine speed increased to 3000+ rpm.

Let us know the latest.

Post by **brentp** » Mon Jul 07, 2008 5:21 pm

Interesting- so in summary, the MJLJ plugged into the engine's harness drives the ignition advance to 15 degrees, but with it unplugged, it goes to 10. That would only mean the MJLJ is sending the SAW signal back to the EDIS module in response to a PIP signal.

Where in your ignition map do you have 15 degrees configured? would it be in the upper left hand corner, (lowest load, lowest RPM)?

FazerBob · Post by **FazerBob** » Mon Jul 07, 2008 5:33 pm

brentp wrote:Interesting- so in summary, the MJLJ plugged into the engine's harness drives the ignition advance to 15 degrees, but with it unplugged, it goes to 10. That would only mean the MJLJ is sending the SAW signal back to the EDIS module in response to a PIP signal.

Where in your ignition map do you have 15 degrees configured? would it be in the upper left hand corner, (lowest load, lowest RPM)?

Yes, that's correct, I think it was only from the 3rd bin down that I specified 22 - 24 deg. Shouldn't it have changed as I revved the engine though?