muythaibxr

Not multiply AFR, *multiply MAP*.

There is an incorporate AFR setting as well but that is not what I am talking about.

Shainiac

iTrader: (7)

Sorry,
I DO have multiply MAP on. Is that what is causing the large dip in VE required?

muythaibxr

Shouldn't be. I am guessing your dead times are wrong for the new injectors or similar.

Shainiac

iTrader: (7)

I don't really know what my secondary injector deadtimes are, and I don't really have access to a scope. I have my dead times set to the default 1.0ms for low impedance injectors. My secondaries are 1600cc EV1s and ~9ohm so I imagine they are a lot higher than 1.0ms. Would it be safe to say they're probably closer to 1.5ms? There's an interesting method for testing dead times on Turbomiata.com where you datalog many different deadtimes at idle and plot MAP/AFR in excel with PW1 on the y axis and MAP/AFR on the X. Then add a trendline to find the y-intercept which should be your deadtime. I can do this for the primaries, but secondaries not so much. Should I try 1.5ms and try retuning my map accordingly?

Thanks,
Alex

Aaron Cake

iTrader: (1)

What 1600CC injectors are they?

Shainiac

iTrader: (7)

They're Fuel injector Clinic modified EV1s. IIRC, they are modified Ford white tops. They staged fine with my 550cc injectors before and my VE map was pretty flat

Aaron Cake

iTrader: (1)

Wouldn't Injector Clinic know the dead times?

muythaibxr

Worst case the best way to figure out dead time involves measuring the amount of fuel injected from 0 ms PW to 5 ms pw, then graphing the results. You'd need an injector test rig for this, but it can be well worth it. Once you know the real dead time, and you have a graph of the injector behavior at low pulse-widths, you can configure the MS (if it's an ms3) to factor both of those in. Makes for really nice behavior at the bottom of the table.

Shainiac

iTrader: (7)

Thanks, Ken.
I think I will try to find an injector service who can clean and flow my injectors and get a dead time number sometime over the winter. I tried a dead time of 1.5ms today and saw much closer numbers. Still a big dip at 5k, but less extreme in general. I may try and move the staging to something like 4k to somewhat mask this hump since the turbo doesn't really come on song till then. Now that I'm a bit more confident in the map, I've begun leaning it out and the car is really coming alive. I'll post a screen grab of my map and a datalog in a few days when it's cleaned up a bit.
Thanks again to everyone for the patience and help.

muythaibxr

Dead-times can depend on the driver being used to drive the injectors. They basically need to be tested with the ECU you'll be using to drive them. Opening time isn't a big deal but closing time can be. Both factor into total dead-time.

It's pretty easy to test with a graduated cylinder and a way of regulating "liquid" pressure to the right PSI, and a way to get said injector to squirt into the graduated cylinder.

To OP:

1.5ms sounds like a huge dead-time. It's more likely you never had it right on the old injectors so when you switched you had to correct for that.

Shainiac

iTrader: (7)

You are absolutely right about the dead time. I drove on the interstate for the first time since changing secondary dead times Tuesday. 75mph is a lot of ~3500 rpm sustained 100kpa driving because of my short gearing and hilly terrain. at 100kpa my injectors stage 75%primary 25%secondary. My AFRs would shoot from 14:1-10.0:1 as soon as the secondaries kicked in. I ended up having to add a large dip in the VE map so I could actually make it home because it would want to fart and buck from the rich AFRs at this transition. Low pulsewidth on large injectors and a big discrepancy in deadtime make for a not very smooth driving experience.

Thanks,
Alex

Aaron Cake

iTrader: (1)

True enough...Maybe they have Megasquirt data though.


Now you can go into the staging table to set the secondaries to come on higher in the load, with a much more aggressive stage (jump immediately to 35% or so) to avoid running the injectors around their dead zone.

wankelbug

did you ever solve this? i know i am digging up an old thread but this is EXACTLY what i am dealing with. waiting on new injectors to arrive so i can eliminate that factor and i am on the fence about ordering coils.

Shainiac

iTrader: (7)

In my situation, it wouldn’t rev out to redline under load because the car was pig rich. I had a couple shady connections between the AEM wideband and the O2-in on the Megasquirt that was clossing some voltage drop and for the AFRs to read leaner than actual. The AEM only goes down to 10:1, so I was seeing mid-10s AFR but was really off the scale. I removed more and more fuel until the car came back to a reasonable AFR and it ran much better.

Shortly after this, I also switched to MS3X and Bosch NGI-2 2200cc/hr injectors. These injectors are identical than ID2000s and have a kickass .55ms deadtime at 13.2V The car ran 100x better with known good injectors and sequential inj/direct fire coils. I also switched to LS coils and was able to light off 10:1 AFR with 1000cc of water injection and 25+psi of boost. With $3 spark plugs lol.

So, verify your gauge and your dashboard in TS read the same AFR. And try taking a bit of fuel out of it.
Also, Verify that timing isn’t being pulled from any other parameters. I’ve seen cars have trouble revving out because MAT timing retard was incorrect or something similar with start up timing maps.

rxFDguy

The rotary struggle, breakup after 6k.... out of curiosity what is your limiter set to?

Shainiac

iTrader: (7)

8500. With the MS3X and LS coils, I had zero high rpm misfire with as much as 30psi of boost and a lot of water injection.

wankelbug

also using the ls coils. I suspected rich conditions were a factor, I will verify my wideband connections as well.