Advance not matching table

General Megajolt Questions and Answers

Moderators: JeffC, rdoherty, stieg, brentp

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Advance not matching table

Post by Jim Blackwood »

First, let me say we've been driving this car ('69 MGB-GT/455 Buick/ T5/ Jag IRS) with MJ Lite/EDIS installed for the last couple of years and it's done fine and gotten surprisingly good mileage. However:

Following a dyno pull, looking to fine tune things a bit we discovered the timing map was oriented to the wrong axis' and corrected it. However, on the next test drive with the laptop connected we discovered that the big window that shows the timing advance was not tracking the map showing the moving spot. This didn't seem right so we ran some datalogs. On one subsequent test ran the big box was showing where the dot was, then on the next it did not, however it gave advance for WOT but not for cruise.

I'd say it sounded like a bad baro sensor, except that we had the baro indications on the map. They just didn't translate to the big window.

Does this sound like a firmware issue? (How do I reload the firmware if so?)

When I looked at the datalogs they were a mess. Lots of straight verticals and horizontals instead of the curves I expected to see.

Thanks,
Jim
MGB/455 Buick/Jag IRS/MJ Lite

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

I'd add the datalogs but it says .csv is not allowed.

Jim
MGB/455 Buick/Jag IRS/MJ Lite

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

I flashed the new firmware V4.0.2, loaded the tune and tested. Again, the actual timing window reads less than that shown by the dot or the bins. So not the firmware. What then would cause this?

Jim
MGB/455 Buick/Jag IRS/MJ Lite

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

I reloaded MJLJ Configurator v4.0.2 on my laptop this morning. Removed it, downloaded a new file, and reinstalled it.

Then ran another datalog. Same results. The total advance window displays a lower number than the green block in the bins. 18 instead of 32 for example. It is always significantly lower at part throttle.

OK, so it isn't the software.
It isn't the firmware.
That leaves the tune and the hardware.

Is there any way to make the tune behave this way? I sincerely doubt it.

I need an answer to this. My follow-up means a phone call if I don't get the answer here.

Time to review the warranty information.

Jim
MGB/455 Buick/Jag IRS/MJ Lite

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Hi, can you provide a screen shot or a short video clip of the effect you are seeing?

What you might be seeing is inter-bin interpolation, where the system will interpolate the ignition advance between bins, but a screenshot or video will help confirm this.

A simplified example:

Bin 1 is 1000 RPM @ 20 degrees
Bin 2 is 2000 RPM @ 30 degrees

If RPM is currently 1500, then ignition advance will show as 25 degrees.

Also, how many cylinders is the engine? Does the Megajolt cylinder settings reflect this?
https://wiki.autosportlabs.com/MJLJ_V4_ ... Parameters

Thanks, will watch for a reply.
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

Thanks Brent,

It isn't interpolation, the spread is much too wide for it to be that. More like the processor isn't using the baro input to compute the advance, but that doesn't answer either because the bin highlights for something like say, 42 degrees advance and then the total advance output is showing maybe 24 degrees. Under WOT it tends to track pretty closely, and of course under WOT you have near atmospheric on the Baro, so little or no compensation. In effect, only "mechanical" or rpm based advance and that's what I'm seeing I believe.

It's like there is a disconnect between the bin read and the advance output. Does the advance output use a separate algorithm from the bin indication? My thought was that if so maybe it isn't getting the baro input to the advance algorithm but is getting it to the bin algorithm. That's about the only thing that makes sense to me.

I can recheck the settings and get you a screen shot or three. Probably one of the datalog might be helpful. The trace is very choppy/digitized with lots of verticals.

Jim
MGB/455 Buick/Jag IRS/MJ Lite

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

With the engine off, are you getting a correct MAP reading? it should show around 100 at atmosphere. What are you seeing at run-time?

What do you see with the factory default ignition advance? Does it operate as expected?


Also, worth asking, are you using the advance correction table to further change the ignition advance? If so, try setting these to all 0's and try observing the ignition advance again:
https://wiki.autosportlabs.com/MegaJolt ... Correction
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

Brent, I am attaching the first (2017_BritishV8.gif), 2nd and 4th datalogs as a screen shots. Notice that the advance line is showing full advance in the beginning but then does not. I can't think of anything that would have done this unless something came loose or quit working.

EOKO MAP is 98kpa.

Not sure what you are asking about the factory default advance, no distributor and a custom advance map.

Advance correction is all 0's (The bins are numbered, 0, 25, 50, etc)

What do you make of these advance traces?

Jim
Attachments
1st datalog.
1st datalog.
2017_BritishV8.gif (140.54 KiB) Viewed 14916 times
2nd datalog
2nd datalog
WTF_Lewisburg.gif (126.55 KiB) Viewed 14916 times
4th datalog
4th datalog
BL_RM0002.gif (114.63 KiB) Viewed 14916 times
MGB/455 Buick/Jag IRS/MJ Lite

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Thanks. Can you also show your ignition map? screenshot is OK.
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Also, could you attach your current configuration file as well. Thanks!
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

OK.
Attachments
Roadmaster 2017 BritishV8.mjlj
(822 Bytes) Downloaded 534 times
RM_Screen_Lewisburg.gif
RM_Screen_Lewisburg.gif (119.38 KiB) Viewed 14906 times
MGB/455 Buick/Jag IRS/MJ Lite

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

Brent, I'm guessing you'd prefer to take the weekend off so maybe by Monday your subconscious may have suggested the answer. I hope so, and have a good weekend.
MGB/455 Buick/Jag IRS/MJ Lite

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Hi Jim,

I took a look at the logs, and compared them against your ignition advance. It's not quite clear to me what exact areas of the log you're concerned about. Could you annotate the screenshots?

Also, if you can share a short video of software, showing the car idling, and then revving slowly so the advance can be correlated with load and RPM, that will help illustrate things.

Also - also, I didn't see a reply on my question regarding number of cylinders. Could you check that? See my previous post.

Jim Blackwood
Posts: 12
Joined: Tue Jun 27, 2017 5:01 pm

Post by Jim Blackwood »

Thanks Brent. It's been a confusing morning.
I checked the Global options for cylinder count and as you guessed, it was set to 4 cylinder so I corrected that and burned the change. That didn't seem to make any difference so I repeated it several times. Then I noticed the rev dial (the round one didn't match up, reading way high. ???

So I restarted and tried again. Turned on ign power first before opening the app and it read an earlier file with boosted baro bins. ??? So I reloaded the last known good file, same one I've been working from, checked cylinder count, reset that to 8. Uploaded and for now it is tracking the advance. WT??

The rpm dial still reads way high, beyond pegging. No clue about that but I'll test again later today. There for awhile I was afraid it was looking way too much like cascading failures but thankfully at the moment not so much.

As for the original problem, if you'll look carefully at the red trace (engine rpm) and the green trace (advance) on that last screen shot above, you'll see that in the first 1/3 the advance tracks engine rpm but in the last 1/3 it barely rises despite large excursions of the red trace.

I'm still expecting trouble, I don't think anything I've done could have fixed this. But I'll test it again several times and if it stays good, just try to keep an eye on it.

Any idea what could be up with that RPM dial?

Jim
MGB/455 Buick/Jag IRS/MJ Lite

brentp
Site Admin
Posts: 6274
Joined: Wed Jan 24, 2007 6:36 am

Post by brentp »

Hi,

Thanks for the update. The way those engine settings work is it's at a level below the regular ignition configuration, so it's not saved as part of the configuration file.

Try this:

Ensure the app is shut down
Power cycle Megajolt
With the tuning cable connected, re-start the app.

Check the cylinder count setting; is it correct?
Check RPM - is it what you expect?

This is to check to make sure the software and firmware don't get out of sync through the manipulation of the setting.
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

Post Reply