GregH Posted November 27, 2007 Share Posted November 27, 2007 Ehhh... Got a "problem" vehicle. Hot or cold, you get in the truck, hit the key, and it has to crank over for 2.5 seconds before it starts. Not a problem, right? The truck belongs to the office manager that signs my paycheck... Yeah, it's a problem. Codes - U0306 for software incompatibiity in the FICM. Truck is programmed, knew that. Restore to stock using the programmer, and flash the three modules to the latest level with the IDS. No codes have come up since then, and I've gotten rid of the P1000. Ok, so looking at some PIDs now. Mean cranking RPM is in the 180's. ICP/IPR are by the book. Plenty of ICP right off the bat. Stabilizes around 1500psi during cranking before it starts. (Voltage also correlates, but I can't remember it right off the top of my head.) FUELPW remains 0 until FICMSYNC toggles to "Yes." Then FUELPW goes to around 2ms and the engine immediately starts. 40,000 miles on the truck, meticulously maintained (by me). The 2.5 seconds of crank time doesn't vary at all - hot, cold, soak time, whatever. I had the opportunity to work on another '03 for a high pressure leak. 209,000 on the clock, and after clearing the air out of the high pressure system, all you have to do is bump the key and it starts up. Hotline says as long as the truck doesn't exceed 7 seconds of cranking, then there is no issue. Wish I could tell the owner that.... Any help, guys? Thanks... Quote Link to comment Share on other sites More sharing options...
Bruce Amacker Posted November 28, 2007 Share Posted November 28, 2007 Mean cranking RPM is in the 180's. Warm or cold? Most F-model 6.0's crank about 215 on the IDS compression test when warm. It seems E-vans /forums/images/%%GRAEMLIN_URL%%/smirk.gif crank about 175 or so warm, I'm guessing because of the battery location. When someone calls me with a hard start and says, "It cranks fine", I always make them physically check the cranking RPM with a scan tool. Slightly low cranking RPM is undoubtedly the #1 reason behind a hard start and very few guys can hear it audibly (I can't). Does this truck have the original batteries in it? (sidebar: 7.3's crank at about 175 RPM warm, 6.0's 215 warm, and 6.4's 245 warm. What do you think the engineers have found?) (sidebar 2: the cranking RPM listed in the diag sheets is BULLSHIT! 100RPM cranking will barely get these trucks started on a warm summer day in Miami, much less a 0F day in Cleveland.) If the cranking RPM is excellent, I'd compare graphs of similar '03 trucks cranking and see if there are any subtle differences. Didn't the HPOP volume output go up on the similar '04 pump? Maybe you could use this as leverage to tell the guy that's why '03's don't start as fast, or install an '04 pump on a lark. (Will it fit?) Why isn't it starting immediately? If I were to guess, I'd say a slower rise of ICPV is the most likely item, so I'd be looking at graphing the ICPV vs. RPM graph very closely. You could do an air check of the system (yech) as I know of at least one truck that has a slight audible air leak yet still starts fine. If everything is dead on as far as ICP/RPM graphs went, I'd scope the CKP/CMP and make sure they weren't noisy, and check the glow plug amperage on a lark. (170amps initial cold) Try a heavy dose of fuel lubricity additive in it and see if it makes a difference. Also, what's MFDES at a hot idle? I'm thinking 12-14 is about right. If it's higher than this, it's starving for fuel (bad injectors). I've seen plenty of hard start trucks up here that a set of injectors fixed, but they were usually hard start cold. Good Luck! Quote Link to comment Share on other sites More sharing options...
Keith Browning Posted November 28, 2007 Share Posted November 28, 2007 Quote: I've seen plenty of hard start trucks up here that a set of injectors fixed, but they were usually hard start cold. Under warranty? Surely you jest! That claim will bounce so high the truck will probably start before it comes back down!!! You had better have a better reason than 2.5 seconds cranking time. We had a couple that received complete sets that even our Rep couldn't back us on. Quote Link to comment Share on other sites More sharing options...
GregH Posted November 28, 2007 Author Share Posted November 28, 2007 By the looks of the data, the engine has got everything it needs to start. ICPV is well above .8, RPM is sufficient (although I did "borrow" a pair of batteries from parts and gained about 20rpm - no difference), system voltage stays above 10.5V, and the FICM powers look good. It really looks like the PCM is not recognizing the crank and cam signals right. Every time, the SYNC and FICMSYNC PIDs will wait about 2.5 seconds before toggling. FICMSYNC goes first, then FUELPW changes from 0 to around 2ms, then SYNC goes to yes. These last three things happen within less than 1/10th of a second. Then a few tenths of a second later, then engine starts. I'll make some recordings of the '03 I fixed yesterday (high pressure leak) and post the two graphs here for you to peruse, if you don't mind... Thanks, guys... Quote Link to comment Share on other sites More sharing options...
DwayneGorniak Posted November 28, 2007 Share Posted November 28, 2007 Take his licence plate and install it on a new V10 Superduty and hand him the keys. Problem solved. /forums/images/%%GRAEMLIN_URL%%/laugh.gif Quote Link to comment Share on other sites More sharing options...
GregH Posted November 28, 2007 Author Share Posted November 28, 2007 Here's some screenshots... First one is the sick truck. Note the amount of time the engine is cranking before the sync's turn to yes. And this one is the one I fixed yesterday. First start after sitting outside all night in below freezing temperatures. Quote Link to comment Share on other sites More sharing options...
GregH Posted November 28, 2007 Author Share Posted November 28, 2007 Well, finally got some solid information. Using the sync.pdf I found on here, I hooked up to graph the crank and cam signals. (by the way, I couldn't make the cam signal work until after I grounded CMP-) Looks ok, I guess. I see some amplitude variation in the CKP signal, like a wavy tone ring. During the time that I couldn't get a CMP signal, I decided to remove the CMP sensor. Found the cam timing peg had hit the sensor. Replaced with a new one, got my signal by grounding CMP-, and graphed my signals. Called the hotline with a simple question - that they weren't able to answer - so I'll pose it here. At what point in the crankshaft rotation should the camshaft sensor signal peak? In other words, is this engine "slightly" out of time, like a tone ring on the crankshaft slipped slightly? Now, back to the cam sensor - I took the new one back out, and found a slight mark on the tip of the sensor. Timing peg hit the new one too... Short block time. I wonder if the crank tone ring is wavy? I wonder if the peg holding the crank tone ring is intact? Wonder if that will fix the 2.5 second crank time, or if all of this is a red herring? Such is the life of the diesel tech... Quote Link to comment Share on other sites More sharing options...
Bruce Amacker Posted November 29, 2007 Share Posted November 29, 2007 I'm pissed. I KNOW I have a good capture of a 6.0 CKP/CMP waveform and I CAN'T FIND IT. But I do have a known good 6.4 and it "might" be the same. It's exactly like yours- the CMP wave is 14-15 teeth to the left of the CKP identifier. (For those who don't see it, there's a red wave in the background that barely shows up on the screen against the black until you zoom it.) I enhanced it a bit: There's a big question though- what's with the "double whoopee" for the CKP identifier? It should clearly show a gap and a substantially higher AC voltage like the one I have. When I zoom in on yours, it shows some kind of double bump instead of a nice oversized AC sine wave. It's kind of grainy when I zoom in on it. Could you zoom it and repost it? I do have a known good 6.0 CKP without the CMP in it: Taken from pin 30 to ground idling. I'm thinking your big problem isn't your cam pin hitting your CMP, it's your CKP trigger wheel or something wrong in there. Maybe try a new CKP sensor and some kind of a CMP shim (?!) and check the waveform? I'd be less concerned with your "wavyness" and more concerned about the identifier (or lack of). /forums/images/%%GRAEMLIN_URL%%/grin.gif Quote Link to comment Share on other sites More sharing options...
Keith Browning Posted November 29, 2007 Share Posted November 29, 2007 If the timing pin on the camshaft has contacted the cam sensor the camshaft must be replaced. This is what engineering and several documents tell us to do. The timing pin is not supposed to back out. /forums/images/%%GRAEMLIN_URL%%/wink.gif I recommend reading this topic ----> cmp to ckp sync and just in case you missed the link I PDF'd the really good diagnostics for this the HotLine faxed me ----> sync.pdf Good thread guys! /forums/images/%%GRAEMLIN_URL%%/popcorn.gif Quote Link to comment Share on other sites More sharing options...
Steve Mutter Posted November 29, 2007 Share Posted November 29, 2007 the other diesel tech has one doing the same thing,long crank on a 03,same thing we did not think it was that bad,but he had codes for the cam sensor so he went to pull it out to try a new one,he could not get it out and it broke off in the block,you can move it back and fourth but it is like the head is mushroomed,so the cam pin can slide out? if so that suck's Quote Link to comment Share on other sites More sharing options...
GregH Posted November 29, 2007 Author Share Posted November 29, 2007 Yep, cam trigger pins can back out and even fall completely out. The hotline guy was telling me about an incident where he advised the tech to stick a pencil into the cam sensor hole and roll the engine over slowly. The pencil slipped furthur into the block when the pin hole came 'round. Keith - I scoured that thread while reaserching this problem, and I stole a copy of your sync.pdf. Did it have sample waveforms attatched to it at one time? Bruce - thanks for the 6.4 graph. I didn't have my settings optimized on the oscilloscope, and I apologize for that. When I get time, I'll scope it again. Either way, the ball is rolling for a short block. The cam sensor getting hit was enough to justify an engine for me... Again, if this truck had belonged to anyone else - instead of our office manager - I would have kicked them to the curb... Oh, and I had another first - The first call to the hotline resulted in me getting a complete tool for an engineer. Treated me like I had never seen a 6.0 before, and fixated on some mysterious high pressure leak (that the IDS graphs showed didn't exist). I couldn't make him understand that it was a sync issue, even when I finally coaxed him into looking at the data I had uploaded. ("well, did you select all of our recommended PIDs?" "No, I didn't know you had a list. I do have the pertinant ones, though." "There's no point in me looking at the recordings if you don't have all the data we request." "You haven't requested any data yet!!") Second hotline guy finally got the idea. It's inspiring to see (hear?) the lightbulb turn on in their minds. "lets see.. Ohhhh.. there is a long delay before the sync pids turn to yes.." bingbingbingbingbing! motor will be here monday. we'll see.... Quote Link to comment Share on other sites More sharing options...
Bruce Amacker Posted November 29, 2007 Share Posted November 29, 2007 I sent your scope wave to my PicoScope guru, Matt Fanslow. He agreed it was a problem and further examination of the tone wheel is a good idea. I'm thinking it's a sensor problem myself, because the only way a tone wheel would cause that would be if it had a tooth in the empty spot. Highly unlikely, but possible. Let us know when it comes out. Good Luck! Quote Link to comment Share on other sites More sharing options...
G. Bedford Posted November 29, 2007 Share Posted November 29, 2007 Greg, if Hotline didn't already tell you, the PID list they want is located under the Powerstroke tab on PTS. Go to Diesel Repair Aids, the heading is 6.0 Diesel Diagnosis/Repair Aids and you will see listed: PID's to select when making WDS/VDR recordings. I always use this list when I record and have never found it to slow down the rate enough to miss intermittant events yet. Keeps the Hotline from getting their panties in a twist, too. Quote Link to comment Share on other sites More sharing options...
snw blue by you Posted November 30, 2007 Share Posted November 30, 2007 A boroscope works wonders when verifing the cam peg or the tone ring. Been through this before with an '05 E350. Consistent long crank (2.5 to 4 sec)since new, very intermittent no start, only set 2614 when no start. Turned out to be a wire harness. Navistar worked with me on that one, because a good waveform was not to be had from the CMP and you'd swear it was a problem with the peg, this was verified by the FSE prior to getting a tech assist from Navistar. Although the build harness passed all tests, when it came down to it, he made the call for a harness and the vehicle has been fine for over 50K since. Go figure and take it for what its worth. Quote Link to comment Share on other sites More sharing options...
Keith Browning Posted November 30, 2007 Share Posted November 30, 2007 /forums/images/%%GRAEMLIN_URL%%/hahaha.gif "Keeps the Hotline from getting their panties in a twist, too." I never had that impression but it is good not to waste your, and their time by not recording enough relative data for them to look at. I printed out the list and keep one in my diagnostic cart. I ran into "lack of data" when working on that 6.4L they insisted had two bad fuel pumps? Yeah. Anyway, I asked if there was a list for the 6.4L yet... not yet... they haven't decided on what is important yet! Quote Link to comment Share on other sites More sharing options...
GregH Posted December 17, 2007 Author Share Posted December 17, 2007 Heh.. Installed my short block. No real problems there... Started it up, matted the pedal, and the engine won't rev past 2600 rpm. Uhhh, that's a little slow.. Anyway, a little testing, and a hotline call later, and another short block was on the way. Recieved short block #2, inspected and rejected based on a nick on the crank trigger wheel. Ordered another short block. Recieved short block #3, inspected and approved. Off with the cab, out with the "old" and in with the new, slam the cab back down and spin the key - now it won't rev past 3200 rpm. Progress, I guess. Yeah, anyway, FICMSYNC starts to flicker from yes to no at 3200 rpm, and the rpm pid goes erratic. Feels like it hits a rev limiter. This condition wasn't present with the original engine. A new engine control harness and FICM harness were installed with the second short block (block #3). New crank and cam sensors have been installed and swapped a few times. Ehhh, field service engineer is on the way... Not that I mind slinging blocks, I just kinda wish this one would go away... Did I mention this one belongs to the husband of the woman who signs my paychecks? Quote Link to comment Share on other sites More sharing options...
Bruce Amacker Posted December 17, 2007 Share Posted December 17, 2007 Yeah, anyway, FICMSYNC starts to flicker from yes to no at 3200 rpm, and the rpm pid goes erratic. Feels like it hits a rev limiter. This condition wasn't present with the original engine. A new engine control harness and FICM harness were installed with the second short block (block #3). New crank and cam sensors have been installed and swapped a few times. Do you have time to scope the CMP/CKP? Quote Link to comment Share on other sites More sharing options...
GregH Posted December 17, 2007 Author Share Posted December 17, 2007 Yes, I'll do it after I finish this last vehicle. Since I don't have the 104 to PTEC adapter, it'll be a cranking pattern... But I'll see if I can backprobe the connectors and getting a running pattern. Also, I've still got your crank sensor set aside. I'll mail it as soon as I get a chance. Quote Link to comment Share on other sites More sharing options...
GregH Posted December 17, 2007 Author Share Posted December 17, 2007 Ok, got some data. First, second and third images are scope patterns of the engine running at about 3100 rpm, and dropping FICMSYNC every few seconds - kinda like just barely bumping a rev limiter. As you can see, it looks like there is no problem with the signal, except for some noise on the CMP line. Now, we have the CMPO and CKPO signals. However, it appears that I didn't get a good connection to one of them. Backprobing is sometimes hit or miss on this connector. However, the line I did make good contact with looks good: That was at a low RPM. When I boosted the RPM to the trucks cutout speed, I noticed this: and this, even more pronounced: I'm thinking the PCM is taking SYNC information and processing it incorrectly at higher frequencies. The FICM doesn't know what to do with the long square wave, and looses sync, therefore cuts off the injectors. I'm thinking PCM.... Quote Link to comment Share on other sites More sharing options...
Bruce Amacker Posted December 18, 2007 Share Posted December 18, 2007 As you can see, it looks like there is no problem with the signal, except for some noise on the CMP line. I'm not so sure I agree with that. Good news and bad news: On your last engine, your CMP wave looked good and your CKP wave identifier had a weird "double bump" where a distinctive rise and fall should have been. Now your CKP looks great but your CMP looks like $hit! Your CMP is showing excessive hash and this may be the source of your trouble. Normally, an AC signal waveform is taken from the positive side of the AC wave to battery ground. This would be pin 30. If you look at the schematic closely, it shows the + and - signs at the PCM. Being that your CKP identifier wave starts with a downward slope instead of an upward slope, I'm betting you took it from pin 41, which is the negative side of the sensor. You may want to verify that, 'cause if you took this CKP from pin 30, your sensor is wired backwards. If it is wired backwards, it will cause MAJOR GRIEF with the way the truck runs. I saw this once on a gasser and it had a lot of new parts hung before someone figgered it out. My CMP looks like this idling: This one looks better: Now, we have the CMPO and CKPO signals. Where? I only see a CMPO at high RPM in all of the following captures. However, it appears that I didn't get a good connection to one of them. Backprobing is sometimes hit or miss on this connector. However, the line I did make good contact with looks good: That was at a low RPM. Are you sure about that? Do you have your captures mixed up? With one pulse about every 40ms, it would be a CMPO signal at 20ms per crank revolution, which translates to about 3000RPM. When I boosted the RPM to the trucks cutout speed, I noticed this: and this, even more pronounced: I'm thinking the PCM is taking SYNC information and processing it incorrectly at higher frequencies. The FICM doesn't know what to do with the long square wave, and looses sync, therefore cuts off the injectors. I agree with this to a point. I'm thinking PCM.... I'm not sure I agree with you here. To me it looks like the CMP In signal might be getting dirty and the CMP Out reacts accordingly. Did you scope the CMP in at high rpm? I might put channel 1 on pin 31 (CMP+) and channel 2 on pin 20 (CMPO) and see if they both act up at the same time. Were both CMP and CKP sensors replaced when you stuffed the short block? How about putting the old CMP back in (if you have it handy)? The high RPM misfire/cutout didn't occur with the original engine and the same PCM, right? Wouldn't you try to relate the problem with something that was changed and not the PCM, which wasn't changed? Here's my CKP out idling, note there is one pulse about every 1.5ms. I wish I had taken a longer capture showing the identifier. Please check my mathematics: at 1.5ms per pulse, divided into 60,000 (the number of ms in a minute) gets 40,000. 40,000 divided by 60 (CKP teeth, which are actually 58 but at a 60 tooth pulse rate) gets 667 RPM, which the engine was idling at when the waveform was taken. Here's my CMP out idling, note the time base is in seconds, not MS and there is about one pulse every .2 sec, which would work out to about 600RPM: My head hurts. I need a beer. /forums/images/%%GRAEMLIN_URL%%/drinkingdude.gif Quote Link to comment Share on other sites More sharing options...
LARRY BRUDZYNSKI Posted December 18, 2007 Share Posted December 18, 2007 Not that I know any better than you guy's but it seems to me that the "new" sensors are probably bad. Or maybe a crossed wire or bad connection. You also stated that you replaced the engine and FICM harness'. It goes back to the "old saying" Being new doesn't always mean being good. Need to check with known good parts. I would see if the old harness' and sensors are still around or swipe another off of a truck in the shop, not with the same problems "of course". By the way thanks guy's I think I'm starting to understand what your talking about!!! /forums/images/%%GRAEMLIN_URL%%/thumbup.gif Bruce I appreciate you explaining the graphs. Quote Link to comment Share on other sites More sharing options...
Jim Warman Posted December 18, 2007 Share Posted December 18, 2007 Greg... there is an old (are you ready for this?) TSB for adjusting the synchronizer on the Taurus if you don't have the positioning tool.... I'll see if I can find the time later to see if I can find the number. This TSB has a very good graphic of roughly what you should be seeing for sensor traces.. Your CKP and CMP (aside from the backasswards crank 'signature')look pretty good.... Recently, I nearly got "hung out to dry" for the third time in the last 10 years... and I have a new rule of thumb. When I am given a set of circumstances where my diag begins with "Maybe it's the....", I'm going to reach over the motor and unplug the alternator... Grasping at straws? Perhaps.... Easy and quick to do? Without a doubt.... What I have learned... it isn't necessarily the amount of ripple... it is the frequency of the ripple that matters. Back when automotive processors were much slower, it was rare to see any concerns.... Now - baud rates are much higher and all of those network circuits festooning current vehicles can be nothing more than a fancy antenna array. It's a long shot.... but what do you have to lose? Quote Link to comment Share on other sites More sharing options...
Bruce Amacker Posted December 18, 2007 Share Posted December 18, 2007 Greg... there is an old (are you ready for this?) TSB for adjusting the synchronizer on the Taurus if you don't have the positioning tool.... I'll see if I can find the time later to see if I can find the number. This TSB has a very good graphic of roughly what you should be seeing for sensor traces.. Sounds interesting! Your CKP and CMP (aside from the backasswards crank 'signature')look pretty good.... Note on some of my traces I also tapped the wrong wire. Some of these 6.0 patterns were taken a few years ago before I knew the difference between the + and - side of an AC signal. Recently, I nearly got "hung out to dry" for the third time in the last 10 years... and I have a new rule of thumb. When I am given a set of circumstances where my diag begins with "Maybe it's the....", I'm going to reach over the motor and unplug the alternator... Grasping at straws? NO! Perhaps.... Easy and quick to do? Without a doubt.... I agree totally and have also been burned by this. What I have learned... it isn't necessarily the amount of ripple... it is the frequency of the ripple that matters. Back when automotive processors were much slower, it was rare to see any concerns.... Now - baud rates are much higher and all of those network circuits festooning current vehicles can be nothing more than a fancy antenna array. It's a long shot.... but what do you have to lose? To expand on this, many things can corrupt a sensor signal and piss off the PCM. Some of the things I've ran into are: *Alternators with either a bad diode rectifier or "brush bounce" from being parked a long time with the brushes in one spot. Like Jim says, unplug the alt and it goes away. The charging system may have no other complaints like a dead battery, but on a heavy load test the output amperage may be low. *Wiring harnesses with "green death" allowing voltage to bleed between circuits in a connector. *Wiring harnesses where the vinyl insulation has become conductive, allowing voltage to bleed between circuits. (I've proven this one FOUR times with a scope, much to my own disbelief!) *High amperage circuits running parallel with a sensor circuit, inducing voltage via magnetic induction (EMI/RFI). IH recently released a TSI to move a battery cable 1.5" to prevent a problem. *Spikes from two way radio transmissions. ("It runs fine, except when I key the mike") *"Drunk" auxiliary processors corrupting data bus circuits. (Anthony just fixed a DT that drove some dealer techs to drink. Complaint: PCM controlled engine fan runs all of the time. After trying many parts, the problem turned out to be J1939 data bus corruption from the Allison tranny processor babbling on the bus line) *Power supply circuits with green death that caused voltage drop on heavy loads. These won't always show up on a DVOM because of the speed of when it happens. *Bad grounds causing weird stuff. You think voltage dropping a ground with a DVOM is enough? Boy, can I show you some neat sensor waveforms caused by bad grounds! Always hook your scope negative to the battery negative post. Some of the best trainers in the country do not check grounds with a DVOM. Being proficient with a scope is the only way you'll find some of these problems. /forums/images/%%GRAEMLIN_URL%%/grin.gif Quote Link to comment Share on other sites More sharing options...
GregH Posted December 18, 2007 Author Share Posted December 18, 2007 Good thoughts on the alternator - It was unplugged the whole time (first thing pulled, been burned there before). But, the truck was hooked to a battery charger running wide open. I didn't notice the backwards CKP waveform. Good call. I'll verify the proper wiring before continuing. As for the first CMPO graph, your right. That was at high speed, just below cutout. My bad. The noise present on the CMP line in the first graph may be misleading. I had adjusted the scope settings between the first and second graphs, and the y-axis has twice the resolution in the first graph than in the second and third graphs. I had adjusted the time base too... A field service engineer called me back last night and advised me to swap a known good FICM onto this truck. So, I'll do that, and then make some more readings when that doesn't change anything. The suggestion was for a CKP(+) graph to verify wiring, and a CMP(+)/CMPO graph. Anything else? Oh, and remove the battery charger while graphing... Thanks for the assistance, guys. I don't know what I'd do without it.... Quote Link to comment Share on other sites More sharing options...
GregH Posted December 18, 2007 Author Share Posted December 18, 2007 Even more graphs for you guys... First, the CKP wiring issue... This is from pin 30: And this is from pin 30, after swapping pins 30 and 41 - this one is a "proper" CKP waveform: Unfortunately, the truck requires an extremely long crank time to start, and it runs like shit. The tach also jumps alot at idle. So, I swapped the pins back. Evidently on this vehicle, the first CKP waveform is correct. Now, the CMP vs CMPO at below cutout speed: And again: Now, just barely bumping cutout: And at WOT, so the cutouts happen more frequently and violently: And again: It looks to me like the CMP input is good. However, it is quite obvious that the CMP out is bad. I did swap the FICM off a known good truck, per the field service engineer's suggestion, and it did not make a difference. Also, the noise that we see/saw on the CMP line - it's still present with the battery charger off the vehicle. The alternator field and output are still disconnected. The noise is not nearly as bad when the engine is at KOEO, but still there. Anyway, there you go.... Again, thanks guys.... Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.