Hi guys,

I've been doing some trouble shooting, and I think I've found what I hope is a bad ICP sensor. Before I go and spend $160 at Riff Raff for a new one, I figured I'd get your expert opinions

Truck specs:
E99 F350 with 179k miles
stock injectors
CNC Fab stage 1 HPOP about 1.5 years old
Running Rotella T6 with about 800 miles on this oil change
Hydra tuner with 1023 tunes - daily driver tune
Stock turbo with a SPTurboost 5x5 compressor wheel and turbine shaft balanced by them

The back story is, about 3 weeks ago, I was driving to work and the truck started running very poorly and I got a check engine light. I pulled codes (I can't remember the exact one right now, but it is kinda irrelevant to the story - stay tuned) and it pointed to the #2 injector not firing. Long story short, I realized that my UVCH had slipped part way apart from the valve cover gasket. I pushed it back together, did the 25 cent mod (yeah, I'm cheap and/or lazy, I haven't done the passenger side yet - I blame inflation), and everything seemed to be okay. No more codes. But it seemed to idle a bit rougher after that. I haven't really driven it much until this last weekend when I went camping. I have a slide-in camper. With the extra weight and wind resistance, I could really tell it wasn't running as good as it should. A buddy driving behind me said I occasionally puffed out quite a bit of black smoke. I use Torque Pro on my phone and when I pulled it up, I noticed my ICP and FIPW bouncing around quite a bit.

So today I started doing some diagnostics. My original thought was it had to do with the UVCH since I just had an issue with it. I thought maybe I had a burned pin or something like that from when it slid part way apart. I used FORScan and didn't get any codes other than some GEM codes that I've always had. I ran pretty much all the tests in FORScan, and again, everything passed. So I ran live data on FORScan and the ICP is bouncing a couple hundred psi at idle, and the FIPW bounces a little bit but not as much as Torque Pro showed when I was driving with my camper. I unplugged the ICP sensor and the idle is definitely smoother when it is unplugged. The FIPW also smooths out when I unplug the ICP sensor. There is NO oil in the sensor connector. The IPR tinnerman nut is on securely.

I added some csv files if anyone wants to take a look (I accidentally named them opposite of what they are - the "idle" one is actually driving around my neighborhood). What I noticed right away when I looked at those files is that the ICP bounces from 460ish psi to 70 psi back to 500ish psi back and forth. Does this sound like an ICP sensor having issues? I know that the HPO isn't really dropping to 70 psi, but that is what it is showing. I still wonder if something else is going on though since I just recently messed with the UVCH.

So, hit me with it. What do you think?

And since everyone likes pictures, this was at our camp spot on the John Day River in north central OR this past weekend.

 

ICP pigtail or wiring, maybe? I'm surprised you're not getting a low injector pressure code.

 

Nice looking campsite. Disconnect the ICP sensor and see if the truck runs any different. If it runs better with it disconnected, you have identified that as a problem. It should run a little bit off with it disconnected if it is not defective. Disconnected, your PCM uses a default code to make the engine run.

 

Well, the ICP is not reading correctly into FORScan. It appears you've selected the right one: ICP(psi), but it is putting out a spiking trace. That's why you are seeing it at 500psi briefly, then down to 70psi, over and over.




Need to get the right PID for ICP selected. I wonder if FORScan updated their PID list again. There was just ICP and ICPV available last I checked, and ICP worked.

See if you can find another ICP PID to select. Make sure the units are changed to PSI, if they are something else initially (like kPa), and log your idle and driving again. You can view the data live in

All that said, I see you've unplugged the ICP and it runs better - you've likely found your issue. Either pigtail wiring or ICP (or both).

 

I forgot to mention that I checked over the wiring. There was nothing obvious. However, I did not pull the loom apart to look inside. I may do that tomorrow just to rule it out.

Yep, I tried that, and it did idle better with it unplugged. And yes, I love that campsite. It's a really cool area with your own little beach to sit on. It was 28 degrees the first morning, so only the dogs enjoyed the water! Another pic, just cuz I love that area so much...




I'll try changing the PID tomorrow. If I remember right, there were only 2 PIDs and one showed kPa. I used the one that showed psi. I'll have to figure out how to change the value to psi. For the record, I'm not sure what you mean when you say it is "putting out a spiking trace". I'm guessing it means I used the wrong PID? Anyway, with everything I've read, it pointed towards the ICP. But it was just too much of a coincidence for me to pass up that I just recently had an issue with the UVCH. I didn't want to go down the rabbit hole chasing an ICP issue that turned out to be something else. Especially when ICP sensors are $160. I'll do a little more testing tomorrow and see what turns up.

Thanks everyone for the input! It is much appreciated!

 

Well, I'm still not sure what's going on. I changed the PIDs in FORScan and the ICP stopped jumping around. I added the csv files again. The odd thing is that it is still doing it with Torque Pro, and the FIPW jumps around in Torque Pro too. It never did that before. And it still doesn't idle as smooth as it used to, but it isn't horrible. I added a video showing me unplugging the ICP and you can see it doesn't change anything. I'm leaning towards changing the ICP sensor just to rule it out. It appears it is the stock sensor, so it is 24 years old (Jan 98 build date).


I think my next step today is to pull the valve cover off and check the UVCH for burned pins. Like I said before, this all started when the UVCH pulled apart. I also have a new International UVCH on hand that I might throw in there just for peace of mind. Then if it is still not idling like it was, I'll order the ICP from Riff Raff. Luckily, I'm not that far from Riff Raff, so I get parts the day after they ship.

 

Ignore the PW jumping around in Torque Pro. A number of us have been seeing that in the latest version. Use FORscan for this reading.

 

That idle sounds ok to me, but hard to tell the nuances without being there. So ICP unplugging was helping to smooth idle, and now it's not. I'm not seeing any significant issues in the ICP and IPR data in your latest files.




Interesting that you picked ICP(psi)_1 and that was smooth in the idle log this time. For the driving log, you picked ICP(psi) again, and it was smooth (where last time it was not). I dunno.

That EBP green trace will be smooth, not spikey, if you pick the EBP_A PID.

As old as that ICP is (original looking, I agree), not a bad idea to replace. But...idle quality is affected by the IPR as well. Cleaning it and resealing is a cheap and simple job if you'd like to try it.

Still no codes when checked with FORScan?

Good to get eyes on the UVCH of both banks, though I would expect a code thrown if they opened any injector circuits, even intermittently.

 

Good to know. That makes sense. I haven't really used Torque Pro lately until I started having problems.

In other news, I think I might have found my problem. I hope I'm an idiot and just followed the easiest route, instead of chasing the coincidence. Torque Pro, and then using the wrong PID on FORScan led me down the rabbit hole I mentioned before. This all started when my UVCH came loose while driving a couple of weeks ago. This afternoon, I swapped out my UVCH with my new one. I didn't find any burned pins, but I did find a LOT of oil in the connector. I probably should have cleaned it out when I first discovered it, but I didn't even think about it. I just shoved it together and threw a quarter in there. Today when I took it apart, oil literally dripped out of the connector. I basically broke all the tabs that hold it to the valve cover gasket, since everything was extremely brittle. After cleaning the valve cover gasket connector well with brake clean, I plugged in the new UVCH. I put it all back together, and it idled MUCH better. So that makes me happy.

However... Why does it still not change the idle when I unplug the ICP? Everything I've read says that it should idle worse when I unplug the ICP. Should I still replace it? If it was a $30 sensor, it would already be done. I'm just hesitating since it is a $160 sensor. And after changing the PID on FORScan, it shows that it is still reading pressure. So, I'm not sure what to do.



So... What do y'all think?

 

I apparently took too long to reply. My wife came home from work while I was typing and missed your reply. For the record, I cleaned and rebuilt the IPR when I did the stage 1 HPOP from CNC - about a year and a half ago (and probably only 2500 miles - I don't drive my truck a ton). And yes, still no PCM codes. At this point, I'm leaning towards just ordering the ICP sensor from Riff Raff. If it is original, it could definitely be a little off if not outright wrong. I really appreciate the help. Seeing the graph helps a visual person like myself. Unfortunately, I'm flying to Atlanta/Alabama this weekend for work for a week. Too bad I can't just take my truck to @SkiSkyJason and drop it off for some spa treatment while I'm in Alabama for the next week! I'd even pay the extra baggage fee!

Oh, and to edit, it seemed to smooth the idle a little before when I unplugged the ICP, but now there is no difference with it plugged in or not. I dunno? This is kinda frustrating.

 

Given that you've already resealed the IPR, if it was me, I'd just drive it now. If that old ICP sensor does fail at some point, just unplug it and you're good to drive it while you get a motorcraft one ordered from a reputable seller. Unplugging a good ICP sensor will not typically change the idle much, or driving characteristics in general. I've towed my 9000lb trailer up and down a mountain hwy with the ICP unplugged, after it failed. Drove very well.

Seems you found the culprit anyway - that UVCH with oil in the contacts. And darn that Torque Pro to heck...that app continues to cause trouble. Tugly and many here are able to make it dance with the rig, but I have not had that luck with the 7.3. Works ok with my Yukon and LeSabre though. FORScan is not without it's issues, but they seem more manageable.