the coil thread
06-02-13, 09:00 AM
#26
Rotary Freak
06-02-13, 02:50 PM
#29
i don't know much about ignition, however there have been some coils testing going on and ts time to share the results.
first up is this https://www.rx7club.com/haltech-foru.../#post11386667
like the post says, we assume more energy is better, but have no idea about peak voltage vs spark time.
then i asked to have the FC, Rx8 and LS2 truck tested, here they are at max power
and then we retested at "factory" dwell times
finished test at 2 ms
FC
2 ms charge = 240 mV peak, 0.8 ms duration
4.5 ms charge = 510 mV peak, 1.3 ms duration
LS2
2 ms charge = 630 mV peak, 0.25 ms duration
6 ms charge = 1300 mV peak, 0.5 ms duration
RX-8
2 ms charge = 630 mV peak, 0.4 ms duration
4 ms charge = 840 mV peak, 0.5 ms duration
so under 3ms charge, rx-8 coil is the best of the three
__________________
conclusions? dwell time is important!
first up is this https://www.rx7club.com/haltech-foru.../#post11386667
like the post says, we assume more energy is better, but have no idea about peak voltage vs spark time.
then i asked to have the FC, Rx8 and LS2 truck tested, here they are at max power
and then we retested at "factory" dwell times
finished test at 2 ms
FC
2 ms charge = 240 mV peak, 0.8 ms duration
4.5 ms charge = 510 mV peak, 1.3 ms duration
LS2
2 ms charge = 630 mV peak, 0.25 ms duration
6 ms charge = 1300 mV peak, 0.5 ms duration
RX-8
2 ms charge = 630 mV peak, 0.4 ms duration
4 ms charge = 840 mV peak, 0.5 ms duration
so under 3ms charge, rx-8 coil is the best of the three
__________________
conclusions? dwell time is important!
I know your over at the Rx8 forum. What do they have to say about this info? Based on what you've posted, it would appear that there is no benefit for them swiching over to the Ls coils since they can't alter the stock dwell times.
06-02-13, 02:58 PM
#30
im running trailing coils also, but using a subaru ignitor to save some space. the FC coil captures above were using a trailing coil.. each output on the leading coil actually has slightly less duration than the trailing, probably due to the fact the outputs share a single coil. same voltage as the trailing though
06-03-13, 09:58 AM
#32
Moderator
Thread Starter
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,805
Received 2,578 Likes
on
1,831 Posts
second is that the stock coils might be more powerful, but they might not be more robust, i don't know why, but my friend is going on 90k with his coils, and the Rx8 forum people seem to only be able to go like 5,000miles. i still say the cause of Rx8 engine failures is the Rx8 forum, my non forum friends have all had really reliable Rx8s...
the third thing is that it is known that the LS1 coils work if you increase the dwell times, so that is a valid approach too, i guess, although you can turn up the dwell on the stock coils too...
so its more complex than i want more ignition power.
06-03-13, 02:05 PM
#33
Just for reference, you may want to keep these things in mind:
With the AC/Delco D585 Gen III LS flex fuel truck coil, you want to program the dwell setting for no more than about 4.2ms for the best results.
With the Mercury Marine IGN coil (external ignitor required) you want to program the dwell setting for about 7.2mS and with the IGN-1A coil (built in internal ignitior) you want to program the dwell setting for about 5.0mS, both for maximum performance.
If you want to run a multi-channel CD type ignition control box, MSD offers a special CD would version of the Mercury Marine IGN coil as #8232
With the AC/Delco D585 Gen III LS flex fuel truck coil, you want to program the dwell setting for no more than about 4.2ms for the best results.
With the Mercury Marine IGN coil (external ignitor required) you want to program the dwell setting for about 7.2mS and with the IGN-1A coil (built in internal ignitior) you want to program the dwell setting for about 5.0mS, both for maximum performance.
If you want to run a multi-channel CD type ignition control box, MSD offers a special CD would version of the Mercury Marine IGN coil as #8232
06-03-13, 04:34 PM
#34
Moderator
Thread Starter
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,805
Received 2,578 Likes
on
1,831 Posts
the thing i did do, is make an injector duty cycle calculator, because the haltech isn't smart enough to cap your duty cycle @100%, you can set it to what ever you want, which is dumb, once the thing is on its on...
anyways, just to throw some numbers around, @9000rpm, you have 6.6ms per cycle.
60 seconds / rpm number = cycles per second it converts to ms with fancy maths (x2x1000?)
the coil testing is showing that there is a small response time for the coil (.2ms?) and then there is a DISCHARGE time, about 1.5ms
so for a 4 coil system @9000rpm its 6.6 total time - 1.5ms discharge time - .2ms response time = 4.9ms of available dwell time.
for wastespark, or a distributor car @9000rpm you have the same 6.6ms - (1.5ms + 1.5ms, it fires twice), -.5 response time (twice) only 3.1ms available to charge the coils.
so keep that in mind when selecting a coil
anyways, just to throw some numbers around, @9000rpm, you have 6.6ms per cycle.
60 seconds / rpm number = cycles per second it converts to ms with fancy maths (x2x1000?)
the coil testing is showing that there is a small response time for the coil (.2ms?) and then there is a DISCHARGE time, about 1.5ms
so for a 4 coil system @9000rpm its 6.6 total time - 1.5ms discharge time - .2ms response time = 4.9ms of available dwell time.
for wastespark, or a distributor car @9000rpm you have the same 6.6ms - (1.5ms + 1.5ms, it fires twice), -.5 response time (twice) only 3.1ms available to charge the coils.
so keep that in mind when selecting a coil
06-03-13, 05:15 PM
#36
Moderator
Thread Starter
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,805
Received 2,578 Likes
on
1,831 Posts
i should have said coil on or near plug system, instead of 4 coil!
so on any engine @9000rpm you have like 4.5ms of charge time... life moves pretty fast. If you don't stop and look around once in a while, you could miss it.
06-03-13, 06:19 PM
#37
tested an LS1 coil.. i would prefer LS1 over LS2 for rotary, a lot more duration
also i was able to get the LS2 D585 to auto-discharge. I had another 8.4V battery, put it in series for about 20V, and bam they started to auto discharge at any duration past about 6ms. I ran the LS1 and the IGN-1A at higher voltage, going up to 30V and they would not auto-discharge. Its interesting at 20V, I can get the IGN-1A to output like the 10ms capture above, but only charging for 4ms.
also i was able to get the LS2 D585 to auto-discharge. I had another 8.4V battery, put it in series for about 20V, and bam they started to auto discharge at any duration past about 6ms. I ran the LS1 and the IGN-1A at higher voltage, going up to 30V and they would not auto-discharge. Its interesting at 20V, I can get the IGN-1A to output like the 10ms capture above, but only charging for 4ms.
06-03-13, 06:47 PM
#38
Senior Member
Thanks for doing the maths j9fd3s, it would have hurt my brain too much if I had attempted it.
It it an extremely helpful perspective when selecting a coil.
Interestingly enough, I didn't realise that my dwell settings may be what is limiting high rpm performance being set to 3.0ms, I guess there is a slight potential for more charge time if I increase it. As I understand the FC twin post coil can deal with much more than 3.0ms anyway.
Hah! I have also just stumbled upon the fact that my maximum spark duration is set to 0.2ms! While I understand that gives more available dwell time, it would surely be hurting performance wouldn't it? Should I bump that up to something higher? 1ms duration for 3.5ms dwell or something to that effect?
I hope this is considered on topic, I think this is all helpful/useful information for anyone setting up their ignition system/selecting coils.
It it an extremely helpful perspective when selecting a coil.
Interestingly enough, I didn't realise that my dwell settings may be what is limiting high rpm performance being set to 3.0ms, I guess there is a slight potential for more charge time if I increase it. As I understand the FC twin post coil can deal with much more than 3.0ms anyway.
Hah! I have also just stumbled upon the fact that my maximum spark duration is set to 0.2ms! While I understand that gives more available dwell time, it would surely be hurting performance wouldn't it? Should I bump that up to something higher? 1ms duration for 3.5ms dwell or something to that effect?
I hope this is considered on topic, I think this is all helpful/useful information for anyone setting up their ignition system/selecting coils.
06-03-13, 07:16 PM
#39
Moderator
Thread Starter
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,805
Received 2,578 Likes
on
1,831 Posts
06-03-13, 08:28 PM
#40
Senior Member
I just realised something. I think I misinterpreted the available time. Say I have wasted spark with 3.0ms dwell and 1.0ms max spark duration, doesn't that mean 3.0ms + 1.0ms + 3.0ms + 1.0ms plus any response time = 8.0ms+. So ideally, I want settings that fit into 6.6ms for best performance to 9000rpm. So something like 3.0ms dwell and 0.3ms max spark duration? Or just raise them both and let the ignition system battle it out?
Do we know which event gets pruned if there isn't enough ignition cycle time? does the dwell time get trimmed, or the spark duration, or both?
Do we know which event gets pruned if there isn't enough ignition cycle time? does the dwell time get trimmed, or the spark duration, or both?
06-03-13, 09:08 PM
#41
i believe the spark duration setting is used to calculate how much to dial back the charge time if rpm is too fast for full charge time
not my pic
Ignition Coil Dwell Calibration
not my pic
Ignition Coil Dwell Calibration
06-03-13, 10:00 PM
#42
i also got my hands on an MSD 6A ignition controller P/N 2600.. it uses a CDI circuit to multiple-strike a coil of your choice, this is with the FD coil
each restrike is 1.2ms apart, and each peaks at 2V with duration of only 150 us (0.015 ms), but it restrikes 8 times up to 3000rpm, where it gradually backs off to to 4 restrikes, and down to 2 restrikes at 6000 rpm. If you ran this on the factory wastespark setup, divide all those rpms by 2, so at 3000rpm it would be down to only 2 restrikes
each restrike is 1.2ms apart, and each peaks at 2V with duration of only 150 us (0.015 ms), but it restrikes 8 times up to 3000rpm, where it gradually backs off to to 4 restrikes, and down to 2 restrikes at 6000 rpm. If you ran this on the factory wastespark setup, divide all those rpms by 2, so at 3000rpm it would be down to only 2 restrikes
06-04-13, 07:09 AM
#44
Are those rpm ranges right? Its a genuine question. The most common application of the MSD 6a ignition box is with a distributor setup on a v8 engine where a single revolution of the engine is 4 ignition events. Our rotary engines are only firing half that in 360* of e-shaft movement, so wouldn't it be assuming our engines are spinning at half the given rpms if the only signal it has to determine rpms being signal from the ecu to fire?
So on a V8, assuming 4 wasted spark coils (4 MSD boxes), that would be 2 events per rotation, not 4
06-04-13, 10:08 AM
#45
tested an LS1 coil.. i would prefer LS1 over LS2 for rotary, a lot more duration
also i was able to get the LS2 D585 to auto-discharge. I had another 8.4V battery, put it in series for about 20V, and bam they started to auto discharge at any duration past about 6ms. I ran the LS1 and the IGN-1A at higher voltage, going up to 30V and they would not auto-discharge. Its interesting at 20V, I can get the IGN-1A to output like the 10ms capture above, but only charging for 4ms.
also i was able to get the LS2 D585 to auto-discharge. I had another 8.4V battery, put it in series for about 20V, and bam they started to auto discharge at any duration past about 6ms. I ran the LS1 and the IGN-1A at higher voltage, going up to 30V and they would not auto-discharge. Its interesting at 20V, I can get the IGN-1A to output like the 10ms capture above, but only charging for 4ms.
06-05-13, 01:47 PM
#49
Moderator
Thread Starter
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,805
Received 2,578 Likes
on
1,831 Posts
so you just open a spreadsheet, paste that formula in.
then you can do fun stuff, like mine has a column that divides by a percentage, like 80%, so it gives you the full cycle time, and then 80% of that. its basically like if i wanted to have an ECU map that maxed out @80% injector duty, i can just plug it in.
you could add a /2 for wastespark time...
06-13-13, 03:41 AM
#50
jalai, on EFI 101, recorded the truck coil auto discharging at 14V. As he states, it was a 1 pc sample, so your results may vary. Not a chance I'm willing to take again.
EFI University Electronic Fuel Injection Tuning :: View topic - ls2 coil recharge time at hi rpm on a race outboard
Here is his capture of the event. He has a TON of coils captured on this site.
https://picasaweb.google.com/1060588...33932691402834
EFI University Electronic Fuel Injection Tuning :: View topic - ls2 coil recharge time at hi rpm on a race outboard
Here is his capture of the event. He has a TON of coils captured on this site.
https://picasaweb.google.com/1060588...33932691402834