Duty Cycle problem, i think

User avatar
vlad_rally
Posts: 50
Joined: Tue Jan 26, 2016 9:37 am

Duty Cycle problem, i think

Postby vlad_rally » Sat May 07, 2016 8:24 pm

So, my current example is: e11v2 on black top sr20, and i'm having some problems with the fueling, my injectors are wired 1-4 on inj1, 2-3 on inj2, runing semi sequential, fuel scale 100%, inj trim 0%, no current limting (sat), i',m reaching 100% duty cycle before my pw reaches 20ms, and regardless of what o write in the map, it leans out. First i was thinking of the injectors, but with this injectors i ran 0.8 bar with my previous ecu holdin a steady 11.2, now i'm leaning out at 0.4...My primary question is, whi is the duty cylce reaching 100% so early with such small pw values?
Attachments
13199313_604989679648039_1150732259_o.jpg
13161372_604989676314706_1875495616_o.jpg

the_bluester
Posts: 177
Joined: Wed Oct 24, 2007 3:19 pm

Re: Duty Cycle problem, i think

Postby the_bluester » Sun May 08, 2016 8:42 pm

Just looking at your second image there you have 24ms injection time at 6400 rpm, but two rotations of the crank at that RPM is only 18ms so you are actually well over 100% duty. I actually would not want to see more than about 15ms injector time at 6500 rpm to leave a little headroom.

The other image looks a little strange as the point where the log is hitting 100% looks like about 4500 rpm and with about 14ms injector time which is well under 100% Maybe still being influenced by a correction there like acceleration enrichment?

User avatar
vlad_rally
Posts: 50
Joined: Tue Jan 26, 2016 9:37 am

Re: Duty Cycle problem, i think

Postby vlad_rally » Mon May 09, 2016 1:09 am

All corections are turned of... An the throttle pumps are dissabled...i'm puting some 750's in it, but i wanted to make sure that it's not a tune problem, because, as i said earlyer, i managed to run double this boost before maxing out the injectors with my old ecu, in semi-seq mode.

the_bluester
Posts: 177
Joined: Wed Oct 24, 2007 3:19 pm

Re: Duty Cycle problem, i think

Postby the_bluester » Mon May 09, 2016 9:03 am

A question, I would have to fire up the tune software to look if it can be set up, have you got one or two injection events per cycle?

In both cases you seem to be hitting 100% duty at about 12ms and about 5000 rpm (It is a little hard to judge from a picture instead of a live log file) at which time one engine cycle is about 24ms. Hitting 100% duty there would make sense if you have one injection event per revolution of the crank not per engine cycle. If that is the case, anything beyond that roughly 12ms injection time (At 5000 rpm) would be over duty and the AFR will not respond to more injection time.

Why it might be running leaner to your previous setup I don't know, what type of wideband O2 sensor controller are you using and has it been calibrated recently?

User avatar
HaltechMatthew
Haltech Staff
Posts: 2578
Joined: Tue Feb 07, 2012 12:45 pm
Location: Sydney, Australia
Contact:

Re: Duty Cycle problem, i think

Postby HaltechMatthew » Mon May 09, 2016 1:53 pm

Why not full sequential? Gains you back one of the dead time pulses as available fuel, and will run better.

User avatar
vlad_rally
Posts: 50
Joined: Tue Jan 26, 2016 9:37 am

Re: Duty Cycle problem, i think

Postby vlad_rally » Mon May 09, 2016 7:12 pm

the_bluester wrote:A question, I would have to fire up the tune software to look if it can be set up, have you got one or two injection events per cycle?

In both cases you seem to be hitting 100% duty at about 12ms and about 5000 rpm (It is a little hard to judge from a picture instead of a live log file) at which time one engine cycle is about 24ms. Hitting 100% duty there would make sense if you have one injection event per revolution of the crank not per engine cycle. If that is the case, anything beyond that roughly 12ms injection time (At 5000 rpm) would be over duty and the AFR will not respond to more injection time.

Why it might be running leaner to your previous setup I don't know, what type of wideband O2 sensor controller are you using and has it been calibrated recently?


By runing semi-seq, basicly i'm injecting 4 times per cycle, and that's why i'm reaching 100% so early. The diference in the afr's, yet i don't know. But i'm using a zeitronix wideband just for mapping, and i'm calibrating it before every tune.

User avatar
vlad_rally
Posts: 50
Joined: Tue Jan 26, 2016 9:37 am

Re: Duty Cycle problem, i think

Postby vlad_rally » Mon May 09, 2016 7:18 pm

HaltechMatthew wrote:Why not full sequential? Gains you back one of the dead time pulses as available fuel, and will run better.

As we talked a couple of days ago Matt, i'm just experimenting, this SR had a MS2 stand alone, with a 36-1 trigger wheel and VR sensor. I've mounted the E11 on it in 2 hours, made a adapter harness, to show him the difference. For my car, i will have to make some sort of mod, because the NEO VVL head i'm runing has the rpm sensors where my firewall is curently sitting, i will make a trigger wheel on the crank, and i have to invent some sort of hall sensor on the cam, and manufacture a tooth of some sort on the cam wheel...it's just that i don't simply have time for my car...My question was pure teoreticly and just wanted to make sure that my thinking was ok, and indeed the current problem comes from injecting 4 times per cycle.


Return to “E11v2/E8”

Who is online

Users browsing this forum: No registered users and 6 guests