MJLJ/E communication failure.

bobclevenger · Post by **bobclevenger** » Fri May 06, 2011 11:43 pm

There are probably other threads on this. If so, just point me to the answer, please.

I just finished installing my EDIS-8 system with MJLJ/E controller in my 1972 Jensen SP. The good news is that the car seems to run well enough.

The bad news is that I can't change anything in the controller. All I get are timeout errors. Same with any data from the controller. I purchased the USB to serial port adapter from you at the same time I purchased the MJLJ/E and that one was useless as there seem to be no drivers available to make it function under Windows 7 (which is what my laptop runs) and you supplied no drivers. I bought a new USB to serial adapter from Staples and it installs fine under Win 7 and the port is reported in device manager as COM7. However, it still fails to communicate.

Yeah, I know -- just keep buying USB to serial adapters until I get one that works. That's NOT economically feasible! Or buy a special-purpose computer with a native serial port -- same comment. I guess I can wait until you bring out the next generation MJLJ with native USB support -- that would be cool, but how long is the wait and how much MORE will I have to $pend?

Do you have any other suggestions?

Post by **brentp** » Sat May 07, 2011 12:49 am

Hello - did you purchase this adapter?

http://www.autosportlabs.com/compact-us ... -p-57.html

if so , you should install the R340 drivers on the included CD, preferrably in advance of plugging the device into the computer (ie. install the drivers with the device unplugged).

As a test, have you tried a different laptop, or a computer with a built-in serial port? Are you confident you're setting the COM port in the software to match what Device Manager configures the adapter to?

Understand it can be frustrating trying to get the technology to play nicely. Unfortunately, there are many variables out of our control- it usually comes down to the compatible hardware with the correct drivers combined with the correct settings. Thanks for working through this!

bobclevenger · Post by **bobclevenger** » Sat May 07, 2011 7:25 am

brentp wrote:Hello - did you purchase this adapter?

http://www.autosportlabs.com/compact-us ... -p-57.html

I believe this is the one I bought. Came with a DB9 to RJ-11 adapter.

if so , you should install the R340 drivers on the included CD, preferrably in advance of plugging the device into the computer (ie. install the drivers with the device unplugged).

I received no CD. Is there a way I can get this driver?

As a test, have you tried a different laptop, or a computer with a built-in serial port? Are you confident you're setting the COM port in the software to match what Device Manager configures the adapter to?

I have no computer with a built-in serial port; they don't seem to be made anymore.
Yes, when using the Staples USB to serial adapter I am confident that the software is set to the same COM port that Device Manager reports is in use (COM7). When using the adapter supplied by you I have no functionality at all.

Understand it can be frustrating trying to get the technology to play nicely. Unfortunately, there are many variables out of our control- it usually comes down to the compatible hardware with the correct drivers combined with the correct settings. Thanks for working through this!

The only parts not supplied by you are my computer (Acer Aspire running Windows 7 Starter), my Staples USB to serial adapter, and my car.

Post by **brentp** » Sat May 07, 2011 5:08 pm

Hi Bob,

send an email to sales@autosportlabs.com and we will email you an electronic copy of the driver. There should've been a driver CD slid in between the folds of the cardboard carton, but it looks like they missed it with yours.

Post by **brentp** » Sat May 07, 2011 5:19 pm

Also attached here!

Make sure the drivers are installed *without* the adapter plugged in. Then, plug in the adapter and check it's status under device manager.

bobclevenger · Post by **bobclevenger** » Sat May 07, 2011 10:10 pm

Thanks Brent.
Just out of curiosity, should there have been a Molex connector to fit the one on the MJLJ as well? I got one at a local electronic parts store.

I'll try loading the driver right away (with the adapter NOT connected).

Bob

bobclevenger · Post by **bobclevenger** » Sat May 07, 2011 11:56 pm

Well, I just tried it again.

The driver installed OK, and the adapter shows as COM8 in Device Manager. So far so good.

Under [Tools>Configurator Options] I have set the COM Port to COM8.

I have applied power to the MJLJ/E and connected the MJLJ/E to the USB to serial adapter via the serial to RJ-11Y adapter.

Under [Tools>Global Controller Options] the "Number of Cylinders" shows 8, which is correct.

Apparently the MJLJ/E is talking to the configuration program at this point.

But when I start the engine I get no readings at all and the message "Could not open port" appears below the tachometer display.

I tried it again with the same settings, except that I connected the cable and started the engine before starting up the Configurator program and this time I got "some" data, but that data did not change as I varied the engine RPM. Got the same "Could not open port" message.

As Bugs Bunny once said: "What's up, Doc?"

cng1 · Post by **cng1** » Sun May 08, 2011 8:13 am

Are you running resisitive spark plugs? If not then fit some as non-resistive plugs may well be causing interference that is stopping the ECU from communicating properly

bobclevenger · Post by **bobclevenger** » Sun May 08, 2011 7:41 pm

Affirmative. Resistor-type NGK XR5 sparking plugs. I even measured one and it has approximately 3390 Ohms from electrical connector to tip. Plug wires are helical coil core as well.

Post by **brentp** » Mon May 09, 2011 9:27 pm

Thanks for the update.

Good that it seems to be communicating. So to be extra clear:

(with the engine off, but megajolt powered up) You can:

1. Alter the ignition map
2. Write the values down to the controller
3. "Flash" the values on the controller (to store them in permanent memory)
4. Power cycle the controller
5. Re-read the ignition map, and verify your changes were persisted?

With the megajolt powered up (engine off), can you apply vacuum to the MAP sensor and observe the gauge move in the software?

If yes to both of the above, it appears the system is communicating correctly.

On to with engine running verifications:

We want to verify if the unstable operation you're observing with engine running is an issue with communications or with the controller itself.

1. With the Megajolt disconnected from PC, and using a timing light, does the megajolt adjust ignition advance smoothly, in precise increments? or, does it jump around erratically? This is to verify the controller is internally operating in a stable manner.

If yes to the above, then it would seem that electrical interference is present on the communications cable. Let's verify the above before we move on.
It'd be good to establish if

bobclevenger · Post by **bobclevenger** » Tue May 10, 2011 12:57 am

brentp wrote:Thanks for the update.

Good that it seems to be communicating. So to be extra clear:

(with the engine off, but megajolt powered up) You can:

This is with the MJLJ/E and the engine powered up, but with the engine not running

1. Alter the ignition map

Yes

2. Write the values down to the controller

Yes

3. "Flash" the values on the controller (to store them in permanent memory)

Yes

4. Power cycle the controller

If you just mean power-down the controller and then power-up the controller: Yes (by turning the ignition OFF then ON again)

5. Re-read the ignition map, and verify your changes were persisted?

Yes

With the megajolt powered up (engine off), can you apply vacuum to the MAP sensor and observe the gauge move in the software?

Yes

If yes to both of the above, it appears the system is communicating correctly.

On to with engine running verifications:

We want to verify if the unstable operation you're observing with engine running is an issue with communications or with the controller itself.

1. With the Megajolt disconnected from PC, and using a timing light, does the megajolt adjust ignition advance smoothly, in precise increments? or, does it jump around erratically? This is to verify the controller is internally operating in a stable manner.

This test will have to wait until it's dark and not hailing or snowing (maybe later on tonight). I suspect some roughness will be seen, as I feel a sudden jerk as if the ignition were rapidly switched OFF then ON again once in a while while driving the car.

If yes to the above, then it would seem that electrical interference is present on the communications cable. Let's verify the above before we move on.
It'd be good to establish if

End of previous message.

Yes, step-by-step troubleshooting works best in my experience.

BTW, I have bypassed the hard rev-limiter by connecting the ignition power wire directly to the coil pack +12VDC wires as a temporary measure so that the rev-limiter is not a factor.

Thanks for sticking with me on this.

bobclevenger · Post by **bobclevenger** » Tue May 10, 2011 5:40 am

OK, I got the timing light on the engine a few minutes ago (cold but not snowing).
The interesting thing that I noticed right away is that the light is skipping a flash every now and then.
The advance curve seems to go from zero to 10°BTDC rather nicely, but stops right there and doesn't advance any further.

Note that for some reason my installation runs at zero° BTDC in "limp mode."

At 25 kPa (a very light load such as revving in neutral) my starting ignition map should have the following advance figures:
500 RPM ->15°BTDC
1000 RPM->20°BTDC
1500 RPM->25°BTDC
2000 RPM->27°BTDC
2500 RPM->30°BTDC

I have attached my preliminary ignition map. I hope.

bobclevenger · Post by **bobclevenger** » Tue May 10, 2011 10:57 pm

Update:
I remounted my VR sensor and now "limp mode" is at 10° BTDC or nearly so (hard to read timing light in the daytime). I also mounted it a little closer to the 36-1 toothed rim of my vibration damper to try to get more stable operation (eliminate the occasional skipped flash on the timing light). Timing light still skips a flash now and then. The VR sensor bracket is very sturdy being 1/4" steel bolted to the front of the engine block. Today when I started the car up after changing the VR sensor position it communicated briefly -- about 4 or 5 seconds.

bobclevenger · Post by **bobclevenger** » Wed May 11, 2011 5:04 am

Another update.

Took the car out for a drive and it seemed to run pretty well until on the way back it started jerking and bucking again, but not so bad as earlier. I disconnected the MJLJ/E to run in "limp mode" and the bucking anf jerking was worse. Reconnected and it got better again.

Post by **brentp** » Wed May 11, 2011 3:10 pm

In limp home mode the engine should run smooth at all times. - idling or driving, but with reduced power, due to the fixed 10 degrees.

Can you post pictures of your bracket design / attachment Your bracket might be resonating at certain RPMs.

Also check your power and especially the ground connections. You should be using a "star" ground pattern, grounding the Megajolt and EDIS to one good point on the chassis.