Odd no start issue 96 XJ Please Help.

1996 XJ, right hand drive, A/T.
Hey all, I am having an odd no start condition in my right hand drive 96 Cherokee. When I bought it a few weeks ago the seller told me of a few issues, one being a check valve in the fuel pump/tank that allows fuel to drain back to the tank when it sits. No biggie just turn on the key and let it prime. Which worked the first time I drove it away. Then it started to become a hassle using it for mail delivery all the time and waiting 15-20 minutes every morning before it finally fires. I've searched and searched and haven't found much to pinpoint it. Now lemme explain the problem. First off I get the NSS code, P1899, but to my knowledge that shouldn't allow it to crank, and not a crank and no start. I can crank it no matter where the shifter is. The weird problem I am noticing is when I first turn the key on to prime the fuel rail I should get all the warning lights to come on, key on engine off. The only light that comes on is the brake warning light. After a while I notice the Gen/Check Engine lights start to flicker then eventually coming on solid after turning the key on and off a few times, 25-30. Once those lights come on is when she'll start. If those lights don't come on while priming fuel pressure then I can crank all day and nothing. My question, is the Crank position sensor some how related to these lights and getting it to fire? Is the NSS related to this? Once I notice the lights coming on a few times of switching the key it will start every time. Sometimes it takes longer than others but it starts. After that I never have a problem all day. Even if I have to get out for gas, or get out at a condo and shut the engine off I turn the key back to the on position and it fires up every time. Only cold starts are like this, which throws my thinking off of most electrical problems happening when the part gets hot. Maybe there's some kind of short? Bad wiring? I've looked it over minimally for now since it sits at the office I work at and I don't have it at home to really get into it right now. I've checked fuses, relays, wiring harnesses, and they all "look" okay, but haven't removed anything yet to clean and thoroughly inspect. Anyone have any suggestions?

A bit of an update maybe someone can point me down the path of this electrical nightmare.

I replaced the crank position sensor and vehicle speed sensor. VSS mainly because the old was leaking tranny fluid and the wires on the harness side were frayed. New one comes with a pigtail and heat shrink tube. Also pulled the NSS and cleaned it, shined up the contact points. Someone had pulled it once before as the plastic side of the casing is cracked. Put it together anyway because it's what I have now. After all this it does the exact same thing, crank and no start. Warning lights flicker here and there but I haven't continued until start like I do at work. I did this morning to bring it home. I'm checking all the wiring under dash right now. The other issue I noticed is someone has hacked the OE plug for the cam sensor and has wires crimped in place running right to the distributor side and plugged in that way. Filled full of silicone to keep from backing out. I haven't messed with those yet. Any other suggestions? My belief is it is definitely electrical.

Edit: I do notice once the key is on the fuel pump runs and stays running.

Yet another update, maybe someone will chime in that has a direction towards a solution:

Replaced ignition switch, cleaned all grounds. Look for shorted, corroded, dirty, non desirable etc. wires or connectors. Found one under the dash, but after a while of trying to start it, it fired and said connection was still disconnected. So obviously not related to my problem.
Found a bad rear temp sensor connector last night, fixed and securely attached it.
One odd thing I noticed today was with the fuel pump and it's relation to this problem. When you first try to start it and turn the key on, the pump comes on and runs but stays running. Now when I turn the key and the dash lights come on finally, the pump will come on and shut back off after 3-5 seconds like it's supposed to. Now what is controlling the fuel pump directly allowing it to run forever when the dash lights don't come on as opposed to it shutting off like it's supposed to when they do finally come on?

Apologize for the long first post, but it was copied and pasted from other forums I've posted in, including my updates. This is what I've done so far.
I'm starting to lean towards PCM, but don't want to spend if I don't have to. Already spent $120 on parts to start fresh on two and take care of some bad wiring on another.

Odd thing is while running, if I unplug the NSS harness it idles up then back down to normal. When I plug it back in, it drops then back up.

Read what JNeary says post number 4, http://naxja.net/forum/showthread.php?t=33892 .

Your 96 and mine are different I have an early 96 which is mostly a 95. But I noticed when my CPS went out (though it tested good) the 5 volt feed to my cam sensor was 2-3 volts. Which lead me to believe the CPS and the cam sensor share the same 5 volt power supply, which was shorting out at the CPS. Yours may be shorting out at the cam sensor.

I'd test the cam position sensor, check the feed voltage and check for shorts. Like I said the CPS and the cam sensor share a power supply, one shorts out and the other is unlikely to work well.

My CPS had a short that killed my CEL, I really didn't notice if the other gauges were working or not, I was kind of fixated on the CEL, I really wanted to check my MIL codes.

My PCM bus went nuts one other time, it turned out to be my rear O2 sensor wire cooking on the CAT. My guess is a short in a sensor circuit and the PCM shuts down the bus or maybe in your XJ it tries to work (resets) then shuts down again and again. Maybe your short is smaller than mine was and only causes intermittent or partial issues.

Just throwing some ideas out there.

Try unplugging the CPS, then turn the key on and check your instrument cluster.

Thanks. I will check voltage at the two. But I just took her for a ride, and like always, never an issue when started. O2 sensor wiring is good, did a check on that this morn. Going to change oil in it then get back to checking things over. It sucks because I have to be back to work tomorrow so since I'm already an hour away I have to get up earlier to make sure it'll start.

Checked voltage going to cam sensor.

When dash lights come on I've got 5v. But sometimes when they are on I don't always get any voltage. Definitely don't get any voltage when the dash lights don't come on.

Disconnected crank sensor while doing this to check it and same results with voltage going to the crank sensor.

All in all at least it's not the new crank sensor I just installed.

I'd do like JNeary suggested and unplug the sensors one at a time until the CEL comes back on and the gauges act right. If that doesn't work, I'd start disconnecting the harness connectors, paying close attention to the pins, they do sometimes back out of the connector (they can be that way for years making iffy contact), wiggle the harness and see what happens.

Seeing as your cam sensor appears to be a hack job, I'd start there. I'd also inspect for rub throughs on the harness (and loose wires) on the right side of the block and the top rear of the injector harness. Check where the O2 sensor wire runs down the front of the block and down the left side (under the exhaust manifold) for burns. My starter solenoid wire was almost rubbed all the way through the insuation, it could have been any wire.

I've tied many of my wires up with cable ties on my 96, trying to avoid trouble later on. The few times I've given my 96 up to the dealership for work, I've found later they tend to cut the cable tie downs and not replace them.

^^^I understand you're trying to help too. But all this has been done and done again. With out any change to the situation.

I was thinking a sensor short is causing the PCM to shut down the bus. But what if it is a PCM power problem. The simple explanation would be something in the ignition switch, contacts making an iffy connection. The exotic would be a PCM fuse with a hair line crack (really hard to see).

I've seen it before, the fuse can act like a bi metal and flex as it it heats up. It can flex closed as well as open depending on how the crack in the fuse filament runs. I've only seen this a couple of times, but as hard as it was to find it kind of stuck in my brain.

Just trying to think out of the box a little.

Corrosion in a connector can do the same thing, sometimes you get conductivity, sometimes you don't. The metal flexes and moves as it heats and cools. Even loose pins in a connector can do this, either a pin that is partially backed out or one that has relaxed (isn't springy anymore) and makes poor contact. A tip off is a discolored connector pin, either blue or black.

Just on a hunch, pull the TCU connector and see what happens, maybe the sensor short isn't in the motor management circuit but the tranny sensor circuit and the PCM is shutting the main bus off?

looking at the wiring diagram volt supply for both the cam and crank sensor come from the same pin from the PCM. Being that when I don't see those dash lights I get zero voltage to both sensors, wouldn't that point to my PCM? I've done some procedures today but the way it's been acting today everything has worked this afternoon. This morning I got up a half hour early to get it started and barely ever got a flicker from the CEL/GEN lights. So I never tried starting except for the crank just to see. I had to call in to work, but ended up going in to help but drove my daily driver. The Cherokee is just for mail delivery and I usually leave it at the post office. Came home at around 10:45 and messed with it and same thing as this morn, nothing for an hour. Then the lights started coming on and I got it fired up and drove it for 20 minutes. I then took about an hour break and used my other truck to go and came back and the lights came on and it fired every try for the last almost 3 hours. The dash lights came on every time and so did it fire. I've let it sit for 5 to 45 minutes and it worked every time.

When you have the no instruments situation check the voltage in to the PCM as well as the voltage out at the PCM.

Ended up being the PCM. Thanks to all who contributed any USEFUL info.