Thread: MSD 72lb injector help. -- Injectors bad or tune?

I got a resistor box from a Honda Accord at the junkyard over the weekend, but have yet to wire it up. My testing w/an ohm meter suggests all the current wiring is correct. If I can figure out what the real issue is, I may wire in this resistor box for a more permanent solution for the future.

Yes, I'm sure the plugs are going to the right injectors. I double checked this a few times thinking I was going crazy but they are going to the right places.

-G

So after some digging in the memory banks, I remembered that the resistors are supposed to be wired to the ground side. I am not exactly sure why, I believe it has to do with the injectors sharing a common power source.

hmmm, which wires are the ground side??? I do notice that all of one side of the injector clips, be it ground or power are all going back to only 1 pin on the injector harness. When I check for continuity I get the same signal on 1 pin for one side of the clips. The current resistor setup is wired into all the 'other' pins that have individual signals back into the injector harness.

-G

You have it wired correctly then. The ECU grounds the injectors out. This is why the injectors share a common power, and 4 ground signals to the ecu.

When you swapped out your injectors, how did you retune your ecu? Is your afpr set to 3bar or 4bar? Is your ecu tuned to match?

Fuel pressure was turned down from 4bar to 3bar once car was turned on. ECU tune I started with was from my old STI tune but with the k-value/inj latency/TP scales adjusted for the new injectors. I'm using Calum Basic ECU. I did richen up the k-value from 215 to 238(215 was based off a Calum tune I found for my setup) because I wanted to start out w/a richer tune as a base, which is why I'm shocked only a change of 22 in the k-value could dump so much more fuel that it would flood the engine if that was the problem and that was only at idle. I kept the injector latency from what the Calum tune was, which is .55. My intention is to actually lean this tune out slightly to a k-value of around 210 and see what the car does. Perhaps it was just the tune, but a difference of 22 in k-value shouldn't have done that which is why I want to rule out the injectors as a problem as well.

The other idea I had is to clear the self-learn function of the ECU, but I left the power unplugged for at least 1 day, maybe more which should have already done that. I was thinking maybe the ECU has some old information on the smaller injectors and was dumping too much fuel. Doubt this was the problem though.

I doubt your k value is the problem here. It seems more like the latency. Start dropping it little by little at idle, you should notice a difference everytime you drop it. Also, bring your k value back to where it should be, and just add a pound or two to your fuel pressure. Once you get decent afr at idle, pull back the fuel pressure and start adjusting from there.

What I thought a little strange was the info from this chart:

Fuel Injector Lag Time

Correct me if I'm wrong, but the injectors run at 12v, right? According to that chart latency should be at .97 But that wouldn't make sense because that'd make it a lot richer at idle in theory. Unless the value is so low that the injectors are not firing at the right time or something..

That seems high.

Try this formula and see what values you get.

Stock injector latency/stock inj. CC's * (new inj. CC) = new rough latency

Then

(new rough latency - stock latency)/2 + stock latency = New Latency.

Also, aren't MSD peak and hold?

hmm:

0.70/259*756 = 2.04

(2.04-0.70)/2+.70=

1.37 Latency

In my quest to find what the stock latency is I actually ran across this post:

http://www.sr20forum.com/turbo/116141-msd-72lbs-injector-lag-time.html

These numbers would be correct if stock latency is .70 for the 259cc injectors.

Now if we go off my last tune which were 605cc injectors (sti 520 @ 4 bar) then we have...

0.69/605*756 = .8622

(.8622-0.69)/2+.69= .77

I figured the numbers would be higher. But, are MSD injectors peak and hold(low impediance, duh)? Take a look at the rc engineering 750 cc on the chart you posted. Note the latency.