Powerjection III Install and Review

I guess I'm misunderstanding what you're saying. If voltage on the "Key" wire goes too low and the unit then thinks the key is off, the engine will still fire on the prime shot, even thought he PJIII now thinks the key is off. When the key wire goes back up the PJIII sees an already running engine and goes into "Auto mode" instead of idle since it never goes through the cranking mode? Also when the key wire is recharged and the PJIII see voltage again, it acts like a fresh start and puts the IAC at the top position before going back down. Long enough to kill the engine? This may be why the instruction are so specific about key wire voltage staying above 9v.

I just got new wires and connectors to run straight from the battery to the key wire, I'll let you know if it changes anything in a few hours.

Also the control voltage for the PJIII can not be tied into the coil if you use an MSD box. You also can't tie it to the coil if there are ballast resisters inline on the coil 12v supply line. You have to tap in before the resisters. Either case, a separate wire form the battery with an inline switch should work fine. It also gives a quick kill switch in the car for the fuel pump as killing the key should kill the fuel pump voltage.

 

Stroker-Porsche

I'm working the same idle issues as you and Steiner. I've been lurking but thought I would jump in. My car has a 106 centerline cam and only 6-7" vac at idle. After some huge issues last week I decided to change the Main RPM boundrary at the idle range to 400/600/1350 rather than the standard 400/700/1000. I also decided it was impossible to idle at 850 and now use 975 as my target after speaking to a tech at Comp Cams. After these changes the car suddenly idles great, even when my electric fans kick in. I think the car was changing cells too much at idle. Honestly the system is very slow to respond to real time changes so I think it was always bouncing between extremely lean and extremely rich when trying to correct. By expanding the RPM boudries it seems to stabalize the system more. Just a thought for you.

Now if I could just get the system to quit randomly deleting all data tables while I'm driving down the road and to stay connected with the USB while driving I might be able to trust the car for a road trip.

Steve

 

SUCCESS!!!!!

I ran a dedicated positive wire straight from the battery to a three way switch mounted into the dash. I ran one switch to the relay for the fan, one to the relay for the water pump and the last one to the yellow "KEY" wire for the PJIII.

Dropped my IntPw to 3500 for 3 seconds hot and 5000 for 5 seconds cold. Fired up first try and stayed running at a smooth idle from a cold start! :bla:

Obviously they key issue is causing the system to skip the idle and go into automatic mode. The PJIII wakes up from the key off moment to see the engine is already running and goes WTF??

Moral of the story??? RTFM!!! ou: The manual was very specific on keeping key voltage over 9 volts. My crappy choice of a hot lead wire made me chase my tail for 2 weeks. When things don't make sense always go back and check the simple stuff, grounds, power, vacuum leaks etc..

I'm heading out on the road for the next 3 days, be back Friday. Then it's time for a few hours of driving to load up the adaptive learning followed by some :beers2:.

 

Stroker
Glad to hear you are making some progress.

 

975 is a much healthier idle for a large cam anyway. Is it a roller cam or flat tappet? Large flat tappets need even more rpms to keep the cam oiled. Mine is a solid roller and seems to like 1050 to 1100rpms, though I may try to see if she will stay at 950 just to make it sound a little choppier... I miss the large cam with carb sound already.

 

What I'm saying with it not being a voltage issue is that if it were you would see the key signal cycle back and forth between 1 and 0 during cranking because the starter initially pulls the system voltage down but then it comes back up over 10V. You're not seeing that on the key signal....just straight zero until the car fires and the key is released back to "run" position which says control power is dropping out completely so the unit thinks the key is off. Then all of a sudden the unit is turned back on and it sees the RPM's sitting at 1000 or so instead of doing a normal crank sequence and apparently ****s itself. The unit has a power reserve after turning the key off which is why you can still read from the ECU when the key is turned off for a little while.

For the control voltage, I was referring to using the same circuit that supplies the coil at the supply, not connecting to the actual coil. This assures you're getting a signal that is hot during both run and crank positions on the key. For instance, most old GM cars have a spade terminals on the fuse block inside the car for IGN, ACC, and BATT. These are connected to the fused main feeds for each. IGN is hot with key in start and run, ACC is hot only in acc and run, and BATT of course is hot all the time. I'm feeding my HEI and PJ from the IGN spade with a 12AWG and 14AWG wire respectively. This bypasses the old resistor wire that is tied in at the engine compartment bulkhead connector. You'd do the same to supply the control voltage for an MSD box on my car....lots of people mess up and connect them to either the resistor wire or a line that's only hot with the key in "run" and wonder why the car won't start after installing the box.

 

Yay me for writing some sweet instructions! :birthday: haha

 

It's a solid flat tappet but has the cool face lifters so I'm not too worried about the oil issues. Like you I miss the lopy sound I could get with the carb but at least it's idling now. Computer is set for 975 but it's really idling in the 1050-1100 range like you.

Stiener - did anything come out of your experiment of modifying the butterflies so they allow more fuel to pass at idle?

Steve

 

I have a few questions about DC. Maybe someone will have the answers. Right now I am running 65 lb fuel pressure and have my injector flow rate at 72. At WOT I am getting some spikes in the DC up to 96%. Would it be better to up the fuel pressure again, to maybe 75 lb and set the injector flow rate to 78 so when it spikes it shouldn't go so close to 100.

I have noticed that the Map influences the IPW a great deal. With all the pulses in the intake why would they not dampen the MAP signal to smooth it out and then it would smooth out the IPW and you wouldn't get the spikes.

 

Randy, what I did to help that was change the MAP range on the main setup which then changes your ADL table and fueling. I think my full load cell is now from 88-100 instead of 90-100 because that's about where it might dip to sometimes under full throttle. If you do a full throttle log like that you can see where it's kicking back and forth between two different MAP cells and not only will you get pw cycling, you'll also get AFR cycling most likely lean when it drops out of the full load cell to the next highest cell.....like back when I set my full load cells manually to +10 fuel to make sure I had plenty up top sometimes it would have a lean spike if MAP fell below 90 since it was kicking to a cell with 0 in it. Of course after I tuned my max load cells on the dyno I went back and put a value maybe 2-3 percent lower in the next highest load cells so that there wouldn't be that big of a spike between cells.

 

Steve, it didn't really do anything that would justify anyone else wasting their time with it. Just another thing that needs to be fixed in the sw.....cold enrichment should kick out OR change to another value when TPS is over 2 since it affects the entire fuel map. I'm not going to sit in the garage until the car comes up to 140.

 

Steiner
I change my full load cell to 85 and up. My Map bounces as low as 86 at WOT. It also bounces regularly to 102 on the high end. I have seen it bounce as high as 106 once. A bounce to 102 at high RPM will put me at or close to 100% DC. I have changed my target WOT to 12.7 from 12.5 and it is a little better. I think I may put the fuel pressure up and go to the dyno next week and do some WOT pulls and let it do its learning.

 

I'll tell you right now it won't learn on the dyno. It doesn't spend enough time in the cells unless you drop the timer down low which can cause problems on down the line, at least it does in the light load cells....I think I run my full load cells around 100ms or so but can't remember. I did five dyno runs and ADL values never changed on their own. However, you can do a quick calculation using your AFR and manually enter ADL values which is what I did. It won't be exact but it'll be close to keep from doing so many runs. Just write down where your ADL table breaks are so when you look at the logs you know which cell to adjust.

(measured AFR - target AFR) / Target AFR
(12.0 measured - 12.8 target) / 12.8 target
-0.0625
So, plug -6 into the cell where you pulled the AFR value from.

 

I never changed flow rate in the fuel wizard when I changed fuel pressure haha.

Those 60 pounders are good to over 87psi.

If you get to the dyno (if its a Mustang) you can hold the car in any cell and tune each one if you wanted to...

 

I don't know if it is a Mustang but he can add a load with it so I may do that.

Steiner
After 2 or 3 WOT pulls mine is changing the ADL. Even a 2nd gear pull changes it. But I am still going a little fast for the streets. A nice long 3rd gear pull would be nice but on the street I would be in jail. Here is my Adaptive Transport.
How do you post the large screen shots.

 

Most Dynos can do a "Zero Step" test if you ask them. It's used to take out the parasitic loss due to acceleraing the engine internals. They dyno automatically stops at preset rpms, say every 250 rpms and then holds load to measure torque. They can delay how long it holds each rpm. Should work well to program under load. This is not possible on dyno jets as they are acculally just accelerometers. You need an eddy-current style like a Taylor or Mustang.

 

Look I found a picture of DJ B-ryce!

 

Look at all my fans!

Where is Steiner? He is missing the party!
.
.
.
.
.
.
.
Oh, There he is.....
.
.
.
:sleep:

 

My start, run, die after~ 5 second issue which was logging a 1100 PW shortly after it started has been resolved. I learned the SW captures and logs data from red batt wire and is independent from the yellow key wire. As some of you noticed my logs were already showing a voltage drop below the 9V min per manual V2.1. Although I connected the yellow wire directly to the fuse box I found excessive resistance in this circuit which dropped the voltage much further (1-2V) than what was seen in the logs. Using a DC power supply to send a constant and stable voltage to yellow key wire only. I found that my original issue could be reproduced when I set the voltage below ~7.0V. I then wired the yellow key a wire directly to the red batt wire and even with voltage drop to ~8.0V (from repeated start cycles) it would start, hit the PW target and idle fine. Bottom line my voltage drop is probably less than probabaly desirable and my wiring of the yellow key wire sucked.

A peak drop of more than 3V during starts seems excessive. I have SBC with denso starter and low compression. I'm thinking a peak voltage drop of 10.5V on a cold start of ~60F should be obtainable with some upgraded wiring. Thoughts?

I’m glad I found this out and fixing my wiring issues now. Car Craft Nats in MN is ~4 weeks away. I hope there is enough time left for me to obtain my V2.1 doctorate...
:3gears:

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com<o></o>