328/microplex...timing?

Mike
I am not sure which chart you have, but at certain MAP determined curves, I remember Microplex advancing to about 34 degrees for full power. In my dyno testing with a non microplex ignition, 34 degrees is about right for best power.

They are as you say hard wired and not programmable. There is a company in Germany called Lenz that makes adifferent ECU that combines with the K-jetronic that claims to produce more power.

 

Ok - I found it.
Here is the advance map for a US Microplex:
Image Unavailable, Please Login

 

This (confusing) "spec", I believe, only applies when the absolute intake manifold pressure = atmospheric pressure (i.e., under very, very heavy load). The ignition map below is from the Technical Specification 328GTB/GTS (US Version) and better matches what you are expecting as "normal" -- send me a PM with your email address, and I'll send you a copy. In the (garbled) text it says:

advance at 1000 RPM with vacuum hose connected = 0

which I take to mean, at relatively high intake manifold vacuum, the advance is 0. Of course , in the chart no curve exactly passes thru this point, but curve #9 comes the closest. So I'm guessing that the value "A" (which determines which curve to use) is the intake manifold vacuum (measured below atmospheric pressure) in units of millibars. If anyone really knows the correct interpretation of the "A" value in these charts, and it's units, please enlighten me/us!

The (garbled) text also gives the spec that you noted:

advance at 5000 RPM with vacuum hose disconnected = 28

Again, no curve exactly passes thru that point, but curve #1 comes closest, and the "A" value for curve #1 is for values near "0" so that is why I'm guessing that "A" is the amount of vacuum below atmospheric (i.e., with the hose disconnected, the pressure the MED806A unit "sees" is the atmospheric pressure).

What the curves do above ~6000 RPM seems a little unclear, but I'd guess that they just stay constant at the last value shown.

(edit: obviously, Russ already posted this same chart while I was typing, but I do want to learn how to correctly understand the value "A" -- so, someone, please speak-up if you know.)
Image Unavailable, Please Login
Image Unavailable, Please Login

 

Steve

This is how I understand it, which could be total poop. The Microplex measures the two pressures: the MAP and AMBIENT air pressure and then takes the difference between the two. I think the "A" should be interpreted as a delta P, which fits as note Marelli does not list the advance in absolute pressures, but in differentials.

<edit> so for example, curve one and two shows the minimum difference between ambient and MAP, and the curve we see is what we would expect at WOT or near WOT with max MAP available. OF course with K-jet, you'll never see that much pressure in the plenum, so curve 2 and 3 are more likely WOT advance curves. <edit>

best
russ

 

So you agree that:

A = Ambient - MAP (in units of millibar)?

 

yes

 

OBTW

I don't say this much as I am not sure, but this advance chart is what makes me wary of folks boosting their engines and keeping the Microplex. I imagine the Microplex is going to default to curve 1 under over atmosheric pressure, and over 6K you are carrying 30 degrees advance - same in the lower rpms. At higher levels of boost, I could see these curves optimized for natural aspiration pressures being problematic.
jmho

 

Curve one actually makes it up to 32 degrees and ends at 30
I think Curve two is a more realistic K-jet WOT

 

Fool the sensors...of course, don't ask me how to do it. (See what happens when one works around great developers for years? I feel like I can state the obvious and some smart engineer/programmer will simply make it happen.)

I'm actually guessing that it isn't that hard to fool the system once one figures out the specific readings the unit needs to force curve 7.

 

Here's a curve fit I did while trying to design an after-market computer (and program) to replicate these map curves. These curves are for the US edition car -- the Marelli has three connector pins (grounded or not) to tell it what model chassis it's plugged into. I presume that there are different maps for different market cars.

Image Unavailable, Please Login
As far as I can tell, the "A" (atmospheres) manifold pressure scale is in millimeters of mercury (mmHg).

I haven't examined the Marelli's electronics, but one thing bothered me about these curves. The 328 has two flywheel sensors. The "TDC" sensor produces two pulses per rev, which is sufficient for alternating between the two banks (letting the distributor choose the correct cylinder). But the "RPM" sensor, as far as I can tell, produces 144 pulses per rev, which works out to 2.5 degrees of crankshaft resolution.

These curves obviously are based on a much finer resolution, which, on my attempts to produce with modern digital microprocessors, I could only accomplish using at least 12 bits of digital time counter for both RPM counting and advance (in a sort of digital phase loop). This requires 4096 12 bit words for a map.

How Marelli could have produced them with 1985 technology is something I can only guess at.

Of course, without actually putting a scope on the ignition, I don't even know how closely the real hardware matches these published curves.

For some unfathomable reason, Marelli put the pressure sensor inside the computer box itself, under the main circuit board. This is the component which most frequently fails, and there are likely a number of Multiplex (and Digiplex) cars running around with no vacuum discrimination taking place. Given the current "street price" on a functional 806A, people are understandably reluctant to dissect a functional computer to determine the specs for a replacement pressure sensor, and Marelli appears to have abandoned all publications regarding having built this beast in the first place.

According to these curves, map 6 would be between 410 and 475 mmHg, and map 7 would be between 475 and 540 mmHg.
I think the break points are 70, 160, 250, 340, 410, 475, 540, and 600 mmHg, with map 1 below 70 and map 9 above 600.

Not knowing the implementation, I'm not sure the maps are quite that discrete in the real hardware.


(I was looking at trying to implement a plug replacement for the whole computer using a modern Renasas M32R processor. But since I haven't had the 328 and a scope in the same garage at the same time, the project got stalled in the computer model stage.

I also got sidetracked by the notion of retrofitting for individual plug mounted coils, as the 3x8 engine seems tailor suited to individual stick coils in place of the plug extenders. (That would also require converting the distributors into some form of cam position sensor, to properly determine the firing order.)

Since then, I've been too busy at projects I get paid for to get back to this.)


It's also been pointed out, on this forum, that these curves are far more complicated than an ignition map really needs to be.

 

There is a ground wire attached to various terminals for different advance curves for different markets. There is very little performance advantage by going from the US program to the BASE or most aggressive program. The max difference of the BASE to US curve is about 4 degrees at around 4000 rpm as read on the old Ferrari test unit made for that purpose. It was called a Jofotron and gave us the ability to test the various electronic advance curves. I have done it to many cars and it just doesn't do a lot. Ferrari was always very conservative about advance curves and compression until good knock sensing was available. Also pent roof chambers generally don't like a lot of advance. Many Cosworths in race tune like DFV's and BDA's only used 28 degrees but they did have much higher compression. If a 328 had decent compression I would be happy with 28. Another thing to remember is unlike most street cars Ferrari's in various places in the world really do get long hard high rpm pulls in top gear. Their tuning generally reflects an approach to making the motors live even in less than perfect tune or circumstances beyond their control so they do tend to be conservative at the cost of horsepower. Even the modern motors at top rpm in top gears at full throttle pulls they have richer mixtures and lower timing advance than would be dictated by a tune for max power. They are trying to help the cars survive.
We see it at Bonneville all the time. A motor tuned to be really good on the street melts itself down when you get to run 5 miles at max power and RPM and in some places Ferrari's regularly go further than that in top gear with someones foot in it.


If you want to go to the base curve just remove the black wire that goes straight to a ground buss. I think it is on terminal 17 on a US car.

Same applies to all microplex boxes.

 

Without having seen a euro wiring diagram, my instinct would be to guess that the base (no grounds) map would be the default for europe, where "octane" measurement comes out a bit higher than US gas.

Hmm. A knock sensor would probably be a good idea in an aftermarket conversion.

 

The "base" or "US" (or Swiss, etc) map selection is hard wired into the car's wiring harness. It's not something that can "fall back". You can only change it by modifying the car's wiring. (Or by plugging the computer into a different market car.)

 

"BASE" is actually an Italian word for "Standard". Not altogether different than the English. Ferrari uses it in many places and refers to "A standard setup" or words to that effect. In otherwords in this case the way they would have it in Italy

 

Brian, your knowledge never ceases to amaze me!!!

 

Don't get too carried away. I know a lot about a few things.

 

Some motors get better benefit from more advance than others. Hemi chambers really like a lot of advance, pentroofs don't usually.