Z32 Igniton Dwell Duty

[Mr Jollwah]

After looking at the ignition dwell duty table values. I googled for the conversion from dec to milliseconds, and discovered a few differing methods that yeld some very different results.

firstly there are 32 (RPM) points in the table, some believe that these are in 400 rpm blocks thus going from 400-12800 rpm (seems unlikely). 200 rpm intervals seem more likely eg. 200-6400 rpm.

next the dec values I imagine have a conversion value to convert them to a percentage, two variations on this one.

1. divide by 10

2. multiply by .33

I favor the option 1, but could be wrong.

Now to convert dwell duty percentage into milliseconds.

I used this method (1/(RPM) X 6000) X Duty % (in decimal).

This gives me a peak of around 4 ms at 200 rpm and an average value of around 2 ms for the rest of the table.

Am I on the right track with this one?

[Adriano]

Try asking ceffie, ive been playing around with this a bit lately, but ive got the same sort of results as you, and im not 100% sure and dont want to give bad imfo

[CEF11E]

Looking at the Z32 ECU I have tried to come up with a conversion table to ms and came up with zip! I have tried longer dwell on standard coils and had better burn and more torque but prone to detonation and shorter dwell and less prone to detonation but guaranteed way to fry coils.

From what I have gathered there is 2 increments per RPM scale and they seem to differ. by best guess would be use the RPM scale as a guide and devide each increment by 2 for your 32 points.

As for dwell values there was a table somewhere I will have a look and see if i can dig it up. but the results were a bit shady. there was actually a brand of coild called FET. and they gave specifications that indicate they were designed for a min of 2ms and no less.

sorry can't be much help

there are numbers of GTR's in japan making 800ps with standard coils so they know something we dont...

high on painkillers right now sorry if my explination is not the best...

[Mr Jollwah]

Here is a table of values using data at 7B60 tp 7B7F.

200 Rpm - 4.50 ms

400 Rpm - 2.25 ms

600 Rpm - 2.20 ms

800 Rpm - 2.40 ms

1000 Rpm - 2.52 ms

1200 Rpm - 2.60 ms

1400 Rpm - 2.57 ms

1600 Rpm - 2.66 ms

1800 Rpm - 2.70 ms

2000 Rpm - 2.82 ms

2200 Rpm - 2.84 ms

2400 Rpm - 2.90 ms

2600 Rpm - 2.91 ms

2800 Rpm - 2.98 ms

3000 Rpm - 3.02 ms

3200 Rpm - 3.04 ms

3400 Rpm - 3.07 ms

3600 Rpm - 2.98 ms

3800 Rpm - 2.91 ms

4000 Rpm - 2.82 ms

4200 Rpm - 2.83 ms

4400 Rpm - 2.84 ms

4600 Rpm - 2.84 ms

4800 Rpm - 2.85 ms

5000 Rpm - 2.80 ms

5200 Rpm - 2.69 ms

5400 Rpm - 2.59 ms

5600 Rpm - 2.50 ms

5800 Rpm - 2.41 ms

6000 Rpm - 2.33 ms

6200 Rpm - 2.25 ms

6400 Rpm - 2.18 ms

Seems to fit in with CEF11E info on the coils.

Maybe someone has measured dwell in ms at any of these points and can confirm if this correct.

[CEF11E]

that looks like the table i have seen before where did you get it?

Cheers

[Mr Jollwah]

I made it.

Took the values in the igniton dwell duty table, Converted them into dec.

Divided them by 10 (Conversion in to Dwell Duty %).

Then applied this formula ((1/RPM) x 60000) x Dwell Duty % (as decimal).

I scaled each of the addresses as 200 RPM Intervals.

eg. 32 addresses = 200 - 6400 rpm.

EXAMPLE:

Value at 7B60 is 0F or 15 dec.

15/10 = 1.5

((1/200) x 60000) x 0.015

(300) x 0.015 = 4.5 ms.

This is how I got those values in the table, still not sure if it is 100% accurate.

It would be good if someone has measured the dwell time at a few of these points to prove or disprove this method.

The other option of scaling the table in 400 rpm (400-12800 rpm) intervals is still possible, with smaller values.

Somewhere around 1 millisecond lower at a guess.

[Mr Jollwah]

Just thought i'd add that the reason behind all this is that I plan to use LS1 coils in the future if my Ser 2 coils give out.

I'll probably increase the dwell for these as i've heard that they are a little slow and need around 3-4.5 ms of dwell.

[T04GTR]

its an iduuctive coil. around 2mh. the norm is the higher the inductance the longer the dwell needed. the longer the better. but the longer it is the more heat is generated in the coil and driver.

generally to get a very good spark around 3-4ms is nominal. 2ms would be at tyhe very bottom of the scale without having to close down the gap.

ive been playing around with differant coil for my cdi set up. and found tha a mitsubishi gto coil pack is no good. cos its got an inductance of 6mh. where as my mazda coil are only 3mh. so they need half the charging time.

you need to workout how many ms are available at full rpm.

[Adriano]

I believe that those numbers are too high, when you put those sort of dwell numbers into an aftermarket computer the stock coils burn out, i was under the impression that the stock system runs about 1.7ms

Edited August 10, 2007 by Adriano

[T04GTR]

it may be. but someone has to messure it.

[T04GTR]

ok. 2j coils have 2.1ms dwell. and they have a 1.6mh inductance. so if we can mesure the inductance of the nissan ones it will give an idea of it.

[Mr Jollwah]

Okay,

TO4GTR - Maximum dwell duty % would be 25.5 %, That would mean that @ 6400 rpm max coil charge time would be 2.38 ms.

Adriano - The RPM Scale that I am using could be either 200 or 400 rpm i'm not sure which. If I was to use the 400 rpm scale it would give me values about 1 ms lower than those in the table.

NOTE:

Motec recommend a constant 3ms charge time for LS1 coils I hear, when I make the change i'll double check this.

[T04GTR]

cool. i was doing some research. and in the states on a efi101 forum it is sugested that 44% is the recomended max. without heating it all up to much

[Mr Jollwah]

Have attached a spreedsheet and charts of all 4 options.

Which is most likely.

Kinda like the horses, pick a favourite, run race (eg. measure dwell), see who wins.

[Mr Jollwah]

Have attached a spreedsheet and charts of all 4 options.

Which is most likely.

Kinda like the horses, pick a favourite, run race (eg. measure dwell), see who wins.

[Mr Jollwah]

Here are the charts I made up.

Doesn't seem to be much info out there about this.

I'll get it mesured and see which correction method is correct.

[T04GTR]

what ever doesnt missfire. and leave it at that

[Mr Jollwah]

True,

I'll probably use just a nistune daughter board, which will prob allow me to adjust dwell.

I'm still curious how the ecu does it.

[chockrl]

I am fairly new to all of this, but I do have the attached document that seems to show the increments are in 400rpm and the conversion is 1/3% to ms. Using the equation from earlier in the thread for a stock RB20DET ROM I am getting between 1.4 - 2.3 ms of dwell.

090318_1_.pdf

[kjb_r33]

I am fairly new to all of this, but I do have the attached document that seems to show the increments are in 400rpm and the conversion is 1/3% to ms. Using the equation from earlier in the thread for a stock RB20DET ROM I am getting between 1.4 - 2.3 ms of dwell.

This is a dwell table from my wifes 93 z32, vg30dett.