LS2 Coils - Any change to cranking dwell, spark duration, dwell/battery correction?
 

Hello all, I read up on the megathread and it stated that the nominal dwell should be 4.2-4.5ms

But does anything else need to change?

https://www.miataturbo.net/attachmen...ine=1447280173

This is a stock LS2 (d585, fat truck) coil dwell table. My MS2 uses percentage based voltage correction, so I did math to set the dwell table right.

doublepost

Thanks! Care to share? MS3x also uses voltage/percentage.

Did you change crank dwell and spark duration?



Here's what I have:
https://www.miataturbo.net/attachmen...p;d=1447280835

https://www.miataturbo.net/attachmen...p;d=1447280835

https://www.miataturbo.net/attachmen...ine=1447281134

I use 4.6 as nominal dwell, and 2.0ms as spark duration. I have no idea what spark duration is for, but I don't think it matters on our sub-10krpm miata.

You're awesome

in the updated msextra, they revised the dwell to 3.5 for the LS2 coils, down from 4.5 on page 83 of the pdf

http://www.msextra.com/doc/pdf/MS3ba...rdware-1.4.pdf

Any change to cranking dwell and spark duration?

I've been trying to find some actual values for the MS Dwell Battery Correction table for these coils.

gross_polluter at LS2 coil / megasquert signal wiring question - Turbobricks Forums states:
"The dwell time for an LS2 coil at 13 volts is 3.15ms, taken directly from an LS2 ecu.

Your dwell battery correction table should look like this (again, values taken directly from an LS2 ecu):
6.0v 193%
8.0v 182%
10.0v 141%
12.0v 113%
14.0v 89%
16.0v 75%"

The Megasquirt MegaManual at Megasquirt Sequencer Coils mentions the following (Nominal Dwell at 14v):
Maximum Dwell: 4.5 ms
Maximum Spark Duration: 2.0 ms
Net Voltage Dwell Compensation
8.0 Volts 2.4 ms
10.0 Volts 0.9 ms
12.0 Volts 0.0 ms
14.0 Volts -0.5 ms
16.0 Volts -0.9 ms

I created the attached spreadsheet in OpenOffice Calc (but had to save it in .xls format to attach it which screwed things up a little) with both these tables for comparison. The grey cells are the numbers from the data above. The yellow cells are calculated fields. The Green cell allows the table to be re-calculated with the specified Nominal Dwell value.

If you set the MegaManual Nominal Dwell time to 3.5ms (as recommended now in the MegaManual) in the spreadsheet (highlighted in green), the two tables are very close (ie. the "Calc Dwell(ms)" values are VERY close) and the 12-16V range is very close to deezums' values (but quite different below that).

Am I over-thinking this too much?

Should I just plug in gross_polluter's numbers, my calculated values, or has someone else got a more accurate table?

deezums mentioned he calculated his values but I couldn't re-create the curve when I tried initially from the LS2 table values he posted (would actually like to see the calculations, which is why I posted this spreadsheet for others to comment what's wrong with my calculations, if indeed they are flawed).

The table I used there is a D585 coil, not a car ls2 table. I guess they are different now? Last year I seem to recall everyone suggesting running the same settings for both.

If I recall, I picked 4.6 as nominal dwell. On that table I posted, I used the 800 rpm cell at 13 volts ~4.6ms.

For the low value, 6V, I need 22.1ms of dwell according to the table. 4.6 * 4.80 = 22.08ms of dwell. So, when the battery voltage is 6V, megasquirt will add 480% to base dwell. Same with the 16 volt high point, 4.6ms * 0.88 = 4.08. Not exact, but close. 16V should be something you don't ever really need to worry about much anyway. You need the lower voltages for after-cranking and electrical load switching, center on those if you have to do anything.

If you change base dwell you should not need to change the dwell correction table too much, the curve should be largely the same. If you feel like the car is misfiring at low voltages, you might bump the dwell correction percentage up.

I can't find any car LS2 coil dwell maps, but if you go over to ls1tech you might be able to sweet talk someone into posting them up for you. If you have the big round d585 coils, I'd just use 4.5ms.

Hope that helps!

Thanks for that deezums.

I've also posted this at Megasquirt Support Forum (MSEXTRA) ? Dwell for LS2 D514A Coils (View topic), hopefully someone is good enough to share.

In case you've missed it in your searches, see Savington's post, #373, at https://www.miataturbo.net/build-thr...t-84579/page7/ for D585 Dwell Battery Correction and Dwell Table (or post #141 at https://www.miataturbo.net/diy-turbo...d-82744/page3/