Bad stumble under load

General Topics for configuring, operating and tuning the Megajolt. Also see the <a href="http://www.autosportlabs.net/MJLJ_V4_Operation_Guide">Operation Guide</a>

Moderators: JeffC, rdoherty, stieg, brentp

nickj
Posts: 32
Joined: Sat Feb 17, 2007 7:52 pm
Location: Norway

Post by nickj »

Could RMI (interference) be the issue? I'll check over the wiring and see if I can see any
problems. It's a little untidy now, mainly due to all the wideband stuff there.

I'm running out of ideas and input, guys!

Nick

rbalmford
Posts: 33
Joined: Thu Feb 26, 2004 1:19 pm
Location: Southampton, U.K.

Post by rbalmford »

rbalmford wrote:Construct a map with 10 degrees in *every* cell, and drive the car both with and without the MJ connected. In theory the timing should remain at 10 degrees in both cases - if there is a difference in the running, then the EDIS<->MJ communication is going awry somewhere.
Have you tried this yet???
Is the power draw greater under a large load than at a smaller load (almost closed throttle). Could the 12V be faltering
under heavy load?
Don't believe so, but you require a 'better' spark at high load than at no load.

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

Post by brentp »

^^ what rbalmford said.

Also-

When you used your timing light, did you observe the engine while sweeping the RPM through the range where you're seeing your problems? You need to watch for any signs of timing instability. Watch very carefully in that range.

Even if the engine seems smooth at idle, you may observe timing instability that is *possibly* masked by the unloaded engine.

If your ignition timing is rock solid when controlled by the MJLJ at different engine speeds, why are problems observed when under load? That is the curious part.

Earlier it was suggested that you temporarily remove the MAP sensor line to have the MJLJ see a constant load. Have you tried this?

Yes, it's possible there can be electrical interference between the EDIS module and MJLJ. If the interference, or electrical noise is significant enough it will confuse whatever unit is receiving the signal- i.e. the MJLJ will get confused if the PIP signal has noise on it, or the EDIS will not interpret the SAW response correctly if noise is on that line.

PIP- MJLJ uses this as an RPM reference and a time reference when to send the SAW pulse back to the EDIS module;
SAW- the shape of this pulse commands the ignition advance the EDIS module needs to.

Are you running non-resistor plug wires?

I'll keep thinking on this end.
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

nickj
Posts: 32
Joined: Sat Feb 17, 2007 7:52 pm
Location: Norway

Post by nickj »

Ok, more testing now done.

Running a straight 10-value map appeared to be identical to running without the MJ.

Running with the vac line disconnected (and plugged!) made no difference at all.

The plug wires are brand new custom wires from these guys: http://auroraignitionwires.com/wire_features.htm

I'm not totally certain it's conclusive, as the stumbling wasn't as pronounced on this run. I tried switching in and out of
MJ, tried a couple of other maps (the very simple non-vac map and one a little more complex). It would stutter and
pop at times, but not as bad as yesterday.

I'm starting to wonder if the problem might be a moody connection to the MJ. I found the Molex connector rather fiddly
to get right, and there could be issues there. Though how that would be influenced by load, I'm not sure.

Just checked to see if a Molex-connector from a PC power supply can be used, and the PC version has all but one wire
already installed (the shift light output). I'm thinking using this and the well-fitted wires could be a more reliable option
than assembling the connector from scratch. I'll see about soldering this in tomorrow.

I'll get back as soon as I have more results, and they'll hopefully be more conclusive.

Nick

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

Post by brentp »

Interesting.

So have you observed at any time stumbling or mis-fires with the MJLJ disconnected? I'm not sure if that what you meant at the beginning of your post.

Or, are you saying the engine ran smoothly with the map set to all 10 degrees?

Also, have you tried the factory default map? It should be safe for what you're doing. In fact, a friend who ran it on a TR-6 liked it quite well! If you want to try it, you can fetch it from the Ignition Map library.
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

nickj
Posts: 32
Joined: Sat Feb 17, 2007 7:52 pm
Location: Norway

Post by nickj »

The engine appears to run pretty well in the EDIS limp-home mode. Maybe not so well at higher RPM,
but that's to be expected really, as for high RPM 10 degrees is not much advance.

As far as I could tell, the car run the same with all values in the map set to 10.

The standard map was the first I ever tried. Should be fairly ok, as long as cylinder count is set to 6
(otherwise the RPM is wrong and the map doesn't pick the right values).

I'm going to redo the connections to the MJ unit tomorrow, using a connector from a PC power supply.

Nick

Edwardatherton
Posts: 30
Joined: Sun Jul 01, 2007 9:50 am
Location: Manchester, England
Contact:

Post by Edwardatherton »

"I'm starting to wonder if the problem might be a moody connection to the MJ. I found the Molex connector rather fiddly
to get right"

After lots of messing around with maps etc, I bought a laptop so that I could see runtime data - I immediately found that the molex connector is marginal - every now and then it just stops working and I have to move it slightly to get it back on-line!
Land Rover 1957 S1 88" 2.5 petrol/LPG

nickj
Posts: 32
Joined: Sat Feb 17, 2007 7:52 pm
Location: Norway

Post by nickj »

I drove an hour today, and it's still as bad... Even under light load it's tugging (how many words can I come up
with to describe the problem?). The map I ran today was the default MJ map. On the way back home, I hooked
up the distributor system, and the car ran properly again...

Since last update I've changed the connections at the MJ end (cut down the Molex-connector from a PC power
supply and used the nicely crimped connection from that) and laid the wire by the EDIS-connector and coil pack
a little more carefully. The HT wires now go more directly to where they should go and the shielded cables to and
from the EDIS controller are tucked away from anywhere hear the HT leads.

The system really isn't that complicated, yet I must have cocked up somewhere. I'm stumped!

Nick

nickj
Posts: 32
Joined: Sat Feb 17, 2007 7:52 pm
Location: Norway

Post by nickj »

Rewired the 12V to the EDIS, MJ and coilpack. Now running three seperate wires from the same point
on the fuse panel. Also disconnected the wideband sensors and data logger. Made no change at all. It
will run fine along flat roads, but as soon as we reach an incline and theres more load on the engine,
it starts stumbling. The same route running on the distributor is without incident.

Next I'll try to change the grounding point from the bodywork to the engine block near the starter.

Nick

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

Post by brentp »

Lot's of data to consider so far. Let me know if the following is true:

1. Engine runs 'smoothly' with the MJLJ disconnected - no popping and hesitation whatsoever- just with less power.
2. With MJLJ connected and Ignition map set to 10 degrees everywhere the engine runs identically to 1. above- no exceptions.
3. Disconnecting MAP sensor has no effect.
4. default map and your own map causes misfires/popping/hesitation at certain RPM ranges.
5. Your old distributor setup works as before.

Some questions & suggestions:

1. with your timing light, can you measure the actual timing on the crank with the default 10 degrees advance? i.e. are you actually at 10 degrees on the crankshaft?

2. if the #2. question in the top section is unequivocally 'true', try bumping up the map in the problem RPM range until you observe problems.

3. Also - this is more involved, but can you compare what your distributor's advance is to the EDIS setup under varying engine conditions?

To simulate what your distributor does under load, you could disconnect the vacuum advance diaphragm so the distributor sees 'wide open throttle'. Then, with your timing light, measure the actual advance in the RPM range where you see your problems.

Repeat the above with the MJLJ and compare the measured advance with the distributor's advance.

The theme around #3 - comparing the distributor advance to the EDIS/MJLJ advance may help shed some light on your problems.
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

nickj
Posts: 32
Joined: Sat Feb 17, 2007 7:52 pm
Location: Norway

Post by nickj »

Just did a half hour run with the EDIS and MJ grounded to the engine, without any difference.

To run through Brents suggestion, truths first:

1) As far as I could tell, yes.
2) Ditto.
3) Yes, no effect.
4) Every single variation of map causes the problem. I've tried lots! From simple maps and the default
map through to maps supplied on others work (same engine tuned on rolling road, but running MS)
5) Yes. It runs very well.

And the to suggestions:

a) Yes
b) I'm quite certain the problem isn't the values in the map (as different maps have given the same problem)
c) The simple map I posted earlier has the values the distributor should have as standard. The dizzy was
calibrated to this spec 5-6 years ago when it was new. I don't think that could be the problem, same reason
as under c)

Could the issue at all be related to sensor position? What symptoms would a sensor that was too far from
the trigger wheel give?

I'm determined to get this thing running before I put it away for the winter (and time is running out, the snow
is creeping down the mountains!), but I'll admit that my enthusiasm is fading at this point.

Nick

Supercharged Nat
Posts: 43
Joined: Mon Jul 23, 2007 8:19 pm

Post by Supercharged Nat »

ive had a very quick look at your trial 1 log.. it looks like your advance is too high at around 2400-2700rpm. looking at the excel chart, youve got around 30 degrees at these rpm's. id say thats too much.

(ive never looked a v8 ignition curves before, but it would certainly be too much for a 4 cylinder).

try reducing the advance under these conditions.

what you really need to do is try datalogging and use the spacebar to mark when it first happens and stops.
that way we can see exactly what rpm and load it occurs. then you can target those areas of the maps and reduce the advance untill it doesnt do it.

Just from the simptoms you describe, id say its too advance (spitting through carbs etc)

nickj
Posts: 32
Joined: Sat Feb 17, 2007 7:52 pm
Location: Norway

Post by nickj »

Yohoooo! It's running and it's good! Finally solved my conundrum last night. On the drive in today the
car was running so well that I had a rictus grin on my face when I arrived at work. Excellent!

I'm guessing noone will want to know what the problem turned out to be, right? In the interest of
self-preservation it'll probably be best if I don't say any more. Oh, ok then (and you may all flay me
with used spark plug wires afterwards. In my state of ignited euporia, I'll bear the derision.)

When I put it all together initially, I didn't have the proper coilpack connector, so I meticulously
constructed my own connector using tiny female spade connectors. No problem there. The problem
was that I put the connections incorrectly, even though I was 100% certain it was according to the
diagram. What I did was mirror the connections, so the 12V supply was connected to one coil trigger,
and the 12V supply was being fed a trigger voltage. The effect of this (according to a smarter chap
than me) was that one coil was firing as it should, and the two others were getting very little power
and hence giving very weak sparks). Of course, once I had the proper connector in front of me, the
problem was blatantly obvious...

And now all is well! I'm very sorry for being such a pain in the posterior about this...

Nick (Ouch! Ouch! Ouch!)

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

Post by brentp »

That's great to hear- and must be a big relief for you.

I'm sure we can all relate to your situation- myself included especially!

If you're still feeling guilty you can compensate by telling all of your friends about the great crank-fired ignition system you've installed on your car! And show off your car and installation in the 'Powered by MJLJ' gallery! :D

But be sure to provide some feedback comparing it against your distributor setup, after doing some tuning and more extensive driving!

Regards,
Brent Picasso
CEO and Founder, Autosport Labs
Facebook | Twitter

rbalmford
Posts: 33
Joined: Thu Feb 26, 2004 1:19 pm
Location: Southampton, U.K.

Post by rbalmford »

Good news! There's nothing wrong with a simple solution. And in the process you've double-checked everything else.

Post Reply