304 TBI conversion flooding

[mstevens61]

Hello fellas, been reading the forum for quite some time, but first time posting.
I recently did a fuel injection (GM with DUI computer controlled advance from AFI) conversion to my otherwise stock AMC 304 (CJ7 - fiberglass body). The conversion went well and was completed in single weekend. The jeep was a daily driver prior to this conversion, it usually ran good but on occasion with idle very high and on other occasions it would stall for no apparent reason. I tinkered quite a bit but in the end wanted better drivability and cold weather starting. The jeep lit right off on initial crank and idled smooth but high. It didn't really except much acceleration (it would bog down and stumble) but I figured it needed to "learn" itself (according to AFIs documentation). I tried to take it for a short ride, it back fired a bit and stumbled, the progressively got worse until I limped back in the garage, (about 5 more min and 1/4 mile of driving/ running).

A couple things Im certain I had wrong right from the get go:

I did not hook up my PCV and upon arriving back in the garage had a massive oil leak from the front seal, which was already leaking a little and I latter found out it was not installed properly anyway.

I also believe on my initial running and driving I had the timing to far advanced, 10 btd, then plug the timing control wire in. (should be base 0)

It cranks, starts, and stalls in a few seconds. All the while seemingly pumping a tremendous amount of fuel, which does not seem to change between cranking and running. Plugs are fouled after several attempts (they have been pulled and cleaned several times). The inside of the tbi is wet with fuel and after several attempts there will be a puddle in the floor of the intake. It never develops good vacuum, presumably because it is running so rich it is flooding out. I do have a scan tool and a hand held vacuum gauge.
I have been able to get it to run albeit very poorly and again with little vacuum on one injector, upon plugging the second injector it revs up, smooths out and stalls, all the while gas spurting out everywhere.
The injectors are color coded green which is recommended for a 5.0. The spray pattern seems ok, but excessive.

It has been over two months since my completeing the install and with the guidance of AFI this is what I have done (btw AFI does not seem convinced that it is flooding???);

Checked and rechecked timing
Checked for spark on all wires, and visually checked spark with a plug removed
New wires and plugs (twice)
Installed an inline fuel pressure gauge (running at 12psi)
Check and rechecked grounds and ran additional grounds
Scoured the engine for vacuum leaks (removed all unnecessary vacuum lines, and plugged)
Applied an artificial vacuum to the map sensor
Replaced sensors (Water, Map)
unplugged each sensor (IAC, MAP, O2, Water, TPS)
Replaced the ignition module, coil and pick up coil in the DUI distributer
Added a relay for distributer power to be sure it was getting plenty of power right from the battery
Check compression, all between 115-126
Fixed front seal, droped oil pan, check timming chain
Pulled valve covers, all valves seem to be operating properly, i did not measure lift
Pulled the intake manifold, inspected lifters and push rods
Tried removing the crank sensor
added a ground for the distributer right to the block
Tried two diffrent ECMs and three diffrent chips, one of which I was told has a fuel curve similar to a 4 cyl.
I put the old carb back on, I got similar results, and even thought I heard some valve train noise (which is why I pulled the intake). admittedly I did a half *** job trying to get the carb back on there and running, maybe I will try again.

I have changed the oil twice through all of this and both times it was terribly gassed up. Through all of this I have seen little change and certainly nothing that corresponds to anything I have done. Occasionaly it has run a little longer but never more than 30s. It starts floods and stalls in a few seconds, so fast that I can barley get any usable data from the scan tool.
At my disposal I have the scan tool, a hand held tack, a hand held vacuum gauge, a vacuum pump, a multi meter with back probs, an IR gun, fuel pressure gauge, and a timming light.

Im stumped! I feel defeated. I cant go back, I have spent an addtional $1200 on parts and tools. My old carb sucks, my new distributer is computer advanced and my old ignition module is prob junk too. The guy at AFI is getting tiered of hearing from me and clearly hasnt helped.
This was a daily driver prior to this, and i need it to be again.
Sorry this is so long just wanted to get all the info out there, and didnt want a bunch of you guys wasting your time telling me to do something I already had.
Thanks in advance for your help.
Matt

 

[Posi]

If you believe its getting too much fuel slightly crimp the feed line and she how it runs.

 

[JeeperChris]

Kinda sounds like the injectors in the TBI are stuck open.

 

[mstevens61]

Hey, thanks for the reply. I guess I forgot to mention, I did try that, and it did improve running, but the fuel pump clear was under strain and would likely burn itself out in short time.
I believe I have this dialed down to two possibilities:
There is something wrong with the injectors (unlikely)
Or there is something wrong with the relationship to the distributer which fires the injectors via the ECM during cranking but should stop after that. I have tried two ECMs and three chips. If I wiggle the rotor back and forth with the engine off it will fire the injectors and the fuel pump. This is a DUI electronic advanced distributer and a GM ECM.
Matt

 

[mstevens61]

I don’t think The injectors are stuck open or it would be firing on initial ignition on when the fuel pump comes on. There is some gas there but they don’t fire.
Also I have tried putting the return fuel line right in a gas can, no change. And the fuel tank is new.

 

[Torxhead]

You might have a problem with the fuel return line being clogged. If you put the pressure gauge on it, it should read zero p.s.i. Installing a larger return line doe not hurt. The oem return line dia. is 1/4", which by now could have clogs or kinks on it.

 

[RiverandSand]

If an injector is stuck open you would likely hydraulic a cylinder. Fuel pressure is critical, it must be within specs for your application. Check your plugs if too much fuel they will be black sooty looking.

 

[bzahler]

I did a Diy conversion and had similar problems. Turned out the TPS was not adjusted correctly.
Go over to Binder planet and go on the fuel injection forum. Bill from Hamilton fuel injection will help you out and is very knowledgeable

 

[mstevens61]

Fuel presure is a steady 12psi as it should be. I have tried running the return into a gas can, no change. Plugs are completely sooted up and sometimes even wet.
My tps reads .5 volts on the scan tool as it should and 4.5 at wide open.so unless the ecm is not useing that value correctly I don’t think that is it. Also this thing did run for 5-10 min right out of the box, so something has changed, just can’t figure out what.

 

[Posi]

I have tried running the return into a gas can, no change.

You should put a pressure gauge on the return line to get a reading.

 

[PistolDave]

Hmm, sound like you have tried all of the obvious things, where is you pressure gauge located exactly? Have you verified the part numbers on the injectors?

5235279 - White / Green - 54pph - 5.0L / 305 Cars/Trucks

5235206 - Orange / Black - 61pph - 5.7L / 350 Cars/Trucks

17084327 - Yellow / Brown - 65pph - 5.7L / 350 Police Package

17084304 - Black / Blue - 74pph - 7.4L / 454 - 91-93

5235231 - Red / Blue - 80pph - 7.4L / 454 - 88-90

17104288 - Pink / Purple - 46pph @13psi - 7.4L / 454 - 94-95

You said you have tried different ecm/chip combos? Have you tried the ecm from a 88-92 c1500 with a 5.0 by chance(ecm #1227747)? If you have a wrecking yard nearby you might see if you could find one of those.

Does your ecm use an intake air temp sensor? Most of the passenger car based ecm's did, but the truck ecm's did not as far as I know. Also triple check the coolant temp sensor, that is probably the biggest cause of TBI running rich.

You can get a little data logger program and cable from http://www.hamiltonfuelinjection.com/ if you do not already have that or something similar. Its a big help troubleshooting TBI issues.

 

[mstevens61]

AFI is convinced that there is something wrong inside the engine, cam timing perhaps. I put the carb back on and was able to get it to idle, it was very slow 500-600 rpm, vacuum was low, 10”. So he may be partially right. So I guess i have to pull the chain cover to confirm? But wouldn’t cam out of sync show in a compression test? I ran one and it was all between 115 and 126. Not great, but it is a 40yo engine with 90k on it. Check out this video of the fuel from the tbi. It is supposedly a 4cyl fuel curve. It looks like the injectors are firing full tilt. It’s a cone of fuel not a mist. I haven’t shown that to a afi yet, just took it yesterday.
http://www.youtube.com/watch?v=QOOT0SKjEZY
Anyway if I am taking the timing chain cover off I may as well just change out the cam shaft and be done with it.
Thanks again for any feed back.

 

[PistolDave]

Also, low vacuum at the map sensor can cause this, it sees low vacuum as high manifold pressure, as if the butterfly's were open. Dumb question: is your map sensor hooked up to the port in the rear center of the throttle body? If not, that will cause major problems also. Idle vacuum at the map sensor should probably be at least 16", check really close for leaks. Just some random thoughts.

 

[mstevens61]

Hey guys, thanks for the replies. The injectors are color coded green for a 5.0l.
I have scoured for vacuum leaks. And putting an artificial vacuum on the map makes no difference in the fuel pattern.
I am running ecm 1227747, no intake temp sensor. Water temp sensor is reading accurately according to my scan tool.
I started to break down tonight to have a look at the cam chain. Before I did however I did another compression test, and had very different results. Two cilynders were higher, one significantly, two were lowere and the rest about the same. I have put virtually no run time on this thing since the last comp test. Are these readings indicative of a out of sync cam?
Thanks
Matt

 

[mstevens61]

So I pulled the timing chain cover. Cam is synced with the crank. Gear key way is still intact. I can’t pull the cam without pulling the engine because the key way in the crank gear is a little meshed up, presumably from the harmonic balancer not hitting it just right when it went on. So before I figure out the flooding problem I have to figure this out. I may just through in the towel and buy a new engine. If there is nothing wrong my cam, or it’s timing then at best I must have a head gasket leak, at worst I have a crack somewhere? Maybe I am over thinking this, but I have checked every. The engine will not buil vacuum, compression is now between 110 and 140. My high compression before was only 126!