SpaceHiker

Just a quick note that AmazonBasics now has a possible Rotella T6 killer. It's a 5w-40 CK-4 synthetic. A few testing sites (some professional, some not) have all found the new AmazonBasics oil to be of high quality. I still have a good stock of T6 but after I've used that, maybe it's worth a try.

Good to know, thanks!

I have a Banks Brake and I don't think the Timbo will work with a Banks, otherwise I'd get one. Haven't made any adjustments, although I've read your info on how to do it properly (I tested the APPS with a meter maybe 4 years ago and everything checked out fine). Still the original APPS after 20 years and 280k miles.

Sigh, that's what I was afraid of. Ya my OBD setup (Torque Pro) just bottoms out the throttle at 9%. It's a common issue and doesn't really tell me anything unfortunately. Thanks for the confirmation.

Anyone know if there is an OBD PID for idle mode? Say for instance my truck had a weird lopey idle and I suspected that the APPS is rapidly jumping from idle mode to off-idle (because the truck also jumps out of fast-idle mode when it's cold). I know I can check it with a meter, but I was hoping I could monitor it from obd. Thanks.

Ya once you catch up on deferred maintenance they're pretty sweet. Just make sure to stay caught up lol. I pretty much agree with everything said here. I know what it's like to own a website, I operate about 7 of them at the moment, with sites coming and going all the time. Income source on all of them is Adsense at the moment, although I've used Amazon Associates as well. The big forum aggregators are just like you said, they buy sites and then let them go without monitoring much of anything.

Right, one definitive guide that solves the problem versus a whole bunch of chitchat that goes nowhere. If only people would search and read before posting.

Maybe this is all in my head, but I was wondering if anyone else has noticed this: Many of us have trucks have been around 10+ years now, and in that time I think every possible question has been asked and answered in the various online forums, particularly CF. And so, when I visit the forums now, all I see are the idiots who don't know how to search, and the trolls who answer them, usually incorrectly because they are more interested in seeing their jibberish in print than having any actual knowledge about what they are posting. All three of my vehicles are older, so I see the same pattern in the forums for all three of them. It seems like a forum has an actual useful life of about 7 years, after which it really is only needed as a search tool. I suppose something original might pop up from time to time, but otherwise I find it a waste of time to try to answer questions anymore on CF. For every knowledgeable answer that someone posts, there are 10 answers from trolls, and the OP doesn't know who to trust since they're obviously not even able to search for themselves. And so they completely misinterpret the answers they're given and just make a bigger mess for themselves. It's a pattern I see repeated a lot in forums for older vehicles. This forum is different because it's more like a place where the "already knowledgeable" go to ask about something that has them stumped, although I do occasionally see the "dur..." question here as well. But the large popular forums I visit, such as CF, NASIOC, and clublexus, have all become a bit of a joke for any vehicle older than about 7 years. ClubLexus is actually the biggest joke I've seen just because of the type of person that generally buys such a vehicle (example: "My steering wheel moves out of the way when I exit the vehicle, going to sell the vehicle if you can't tell me how to stop it from doing that?"). The other issue I think it that more people than ever before are buying old vehicles who have absolutely no mechanical inclination or experience. You can't buy a 15 year old Cummins and take it to a mechanic every time there's the slightest issue, it would cost more than a new truck within a few years. But time after time I see "My mechanic said..." on the forums. Rant over, resume normal programming.

All you have to do is leave that switch on high, and wire a rotary dimmer switch into an electrical box that you splice into the power wire. This is the switch I used on my old stove: http://www.homedepot.com/p/600-Watt-Rotary-On-Off-Dimmer-White-R50-06602-0IW/203878272

It depends on the software you are using. If you are using the torque pro app for android, it allows you to create and save a custom live data profile for each vehicle. It does not automatically switch over based on the vehicle it's connected to. You have to change the profile manually, but the procedure is obvious. Other software will behave differently. You might need to switch to a different app.

Mine clearly reports A as 1 kpa. I've confirmed it with quite a bit of driving around with live data. I guess other years report differently?

Oh yes, then there's that conversation. I forgot to mention that one. The people who were discussing that were making some important mistakes at various points in that conversation: 1. Some were thinking the ECM handles boost as psi rather than kpa. 2. Pretty much everyone assumed the ECM handles map readings as an integer, when it is almost certainly floating point. 3. People were confusing the canbus with obd. So on the canbus, the ecm is taking raw voltage readings from the output of the map sensor and converting them to a floating point (decimal) number in kpa. What we see on obd is the rounded output, because decimal precision is not necessary and the obd spec doesn't have the space for it. Logically then you would need 2 bytes on the canbus for the floating point value and 1 byte on the obd for the rounded integer.

Well, since I've taken it upon myself today to solve the mystery of my truck's poor performance on 2-stroke oil (see cumminsforum), as a side effect I solved the mystery of the boost formula. The correct boost formula that works 100% accurately at any elevation is: (A/6.8947)-14.6 This means that in any OBD software, you need to add a custom PID for boost. For the mode and pid, enter 010b or 01 0b (depending on your software's format). For the minimum value enter 0, for maximum value enter 100 or 60 or whatever you want (22 suffices for my stock truck lol). For equation, enter the above formula: (A/6.8947)-14.6 Then for decimal point precision, just use 1 decimal point. This formula does not depend on atmospheric pressure at all. And in fact, the whole discussion about atmospheric pressure is completely overblown because it really only concerns our trucks when they are at idle. You have to keep in mind the function of a turbocharger. It is by it's nature a perfect altitude compensator and will always try to achieve a certain equilibrium within it's mechanical limits. This means that as soon as we begin applying fuel with a press of the throttle, intake pressure comes up above standard atmospheric pressure anyway. It just takes slightly longer at 10,000 feet than it does at 2,000 feet. For the above reason, and since I'm at 9,240 feet, I see my boost sit at 0.0 for a little while before it starts coming up. That time where it's at 0.0 is when the turbo is reaching 14.6psi. It's a very short period of time, but it is noticeable at my elevation (lasts about 2 seconds after applying throttle). My mechanical boost gauge shows the same thing. Hopefully that clears up some of the issues that have been bubbling in this thread for a while. Just keep in mind that my truck is a 98.5, I don't know if the newer trucks report a different data format on 010b, but I can't see why they would.

That's very interesting because I see the opposite. My truck will read 100-102 for a long time before it starts picking up. That might be the difference between the early sensors and the late sensors. It seems I might have bought the wrong truck for high altitude operation lol. But after 9 years I guess I'm stuck with it.

Sorry for the disappearance, but what I'm finding with the map sensor, at least on the early ones, is that it doesn't register anything below standard atmospheric pressure. Apparently Cummins doesn't care if we're at 14,000 feet, the sensor will read sea level anyway and consider boost to be 0. If any of you have had your trucks at 14,000 feet, let the engine cool off and try to restart it, you know how badly these trucks run up in the rare air anyway, at least until they can build boost. You literally have to drive one foot on the throttle and one on the brake to get some boost going before they will run right. So readings start at 100kpa which is approximately the average sea level pressure, then go up from there as boost builds. This makes sense if all Cummins is trying to do is measure relative boost above standard pressure. It's actually the same thing our gauges do, as I described previously.

I was sitting in the visitor center lot at Arches National Park, waiting on the family, when a guy comes up and mentions how much he likes my truck and starts asking about it. He later explained that he's the Cummins engineer that has designed the software for all the electronic injection trucks, vp44 and newer. When he's in development mode, he travels the U.S. with his family in the latest Cummins, with laptop and sensor wires strewn all over the interior. He was testing software revisions to the Cummins/Aisen setup when he was at Arches. Pretty fun guy to talk to.

What that reminds me of is microphone correction. Since there is no such thing as an acoustically perfect microphone, a sound engineer that needs an acoustically flat measurement will apply a response correction to the mic's output in order to approximate a perfect response. As long as a particular microphone model can be manufactured with a consistent response, a corrective frequency curve can be applied to it. It could be the same with the MAP sensor. If the engineers came up with a simple and cheap device that changes it's output depending on air pressure, and they could manufacture it consistently, then all they would need to do is place it in a hyperbaric chamber and vary the pressure, recording the readings from the sensor and then entering them into a table in the ECM. In that case, the actual MAP readings need not be linear in any way, as long as they are reproducible.

What I've discovered is that I need to manually create custom PIDs in the Torque app for all the sensors that aren't reading correctly. So for example, my Intake Air Temp (pid 05) just reads 0 in the torque app. So I delete that gauge, then go into Settings... Manage Custom PIDS... Add New... For the pid number I use 01 05 For the formula I use (A-40)*1.8+32 to get the temp in Fahrenheit. I have to repeat this for each sensor that isn't working in torque, but for some reason that fixes it. I've tried the torque app in my other vehicles and all of them work fine, I don't know what's different about the old Dodge. The data all looks the same to me, from car to car. A strange solution, but at least I have data now. The thing that's strange about this is that right now my MAP sensor is reading 101kpa, engine not running. That translates to 29.83inHg. I run a weather station that reports to the NWS and Wunderground and I know that my current absolute barometric pressure is 21.37 and my relative pressure is 30.42inHg. I should also mention that there is no such thing as a standard correction for altitude. For instance, the average pressure at my house on a -26 degree day would be the same pressure as sea level on a 95 degree day.

I also wanted to comment on the MAP sensor to boost pressure project. I'm not sure that's it's possible to do what you want, because I have a feeling the conversion from MAP data to boost pressure requires a table rather than a formula. First of all, we have to consider how our gauges read boost: A typical boost gauge is calibrated to standard atmospheric pressure at sea level (14.7psi = 0 boost), and merely shows how many PSI above standard atmospheric pressure we are currently boosting our intake pressure up to. ************************** Now there is a way around this: Some mechanical pressure gauges have an air channel that equalizes the pressure in the gauge with CURRENT atmospheric pressure, and shows how many PSI above the current pressure we have. If we were to hook up a gauge like this to our intake, the boost numbers would be all over the place depending on the current absolute barometric pressure (primarily based on elevation and temperature). **************************** Ignoring everything in the stars for a moment, the reason a boost gauge is calibrated to standard atmospheric pressure is because our turbos are engineered the same way. They are a pressure equalizing device, so that if you supply a certain amount of fuel+air(temperature) to the exhaust vane you will get a certain amount of air pressure into the intake. And it DOES NOT MATTER what elevation you are at, the turbo will automatically compensate within it's mechanical limits (for instance at low RPM's it can not yet compensate for high elevation, but as soon as you speed up the engine, it will). That's why turbos were known as "Altitude Compensators" when they were first invented. They would allow an airplane to have the same performance at high altitude as they had at sea level, up to the mechanical limits of the turbo, because they would allow the engine to maintain the same intake pressure at any altitude. So what this means for us, is that all MAP sensor data below 14.7psi needs to be discarded, because it means nothing as far as boost is concerned. Now, if you want to translate the MAP data to boost, you would get the most accurate results if you were at a location and time when the absolute barometric pressure (not relative) was exactly 14.7psi or 29.92inHg. Then, you would drive around and look at both the MAP data and your boost gauge. At 1psi of boost, record the MAP data. At 2psi of boost, repeat. Continue on until your boost is maxxed out. Now you have a table of MAP data to boost pressure. From this, it might be possible to come up with a formula if you are a math genius. But I seriously doubt it, my guess is that there are too many variables involved in translating a MAP sensor's absolute pressure data to a relative boost reading. But at least you have a table now, and maybe you can program the table into some kind of software to give you a boost reading.

Mike directed me to this thread because I'm trying to get my OBD hardware/software to pull in as many pids as possible on my 98.5, which apparently has a different arrangement of pids than the newer trucks (01+???) I'm going to make some rough notes here to document what I'm finding, and I'll probably be updating these notes over the next few hours, so please bear with me. When I'm done, I'll clean it up so early model owners can get their OBD systems set up. I also have some comments about the MAP -> Boost project which I'll post later. atsp0 OK> at h1 OK> (turns on headers) 0100 Searching... 48 6b d1 41 00 b8 1a a0 14 4b 48 6b 15 41 00 98 3a 80 14 6f at h0 (turn off headers) Address d1 = PCM ??? Address 15 = ECM ??? Device d1: B8 1A A0 14 1011 1000 0001 1010 1010 0000 0001 0100 01 MIL status, # of DTC's 03 Fuel system status 04 Calculated engine load 05 Engine coolant temp 0c RPM 0d MPH 0f Intake air temp 11 apps (throttle position) 13 02 sensors present 1c OBD standards 1e PTO status Device 15: 98 3a 80 14 1001 1000 0011 1010 1000 0000 0001 0100 01 MIL status, # of DTC's 04 Calculated engine load 05 Engine coolant temp 0b Map sensor 0c RPM 0d MPH 0f Intake air temp 11 apps (Throttle position) 1c OBD standards 1e pto status For some reason my OBD setup does not read most of these values, probably because the truck is returning an unfamiliar format. Next I will send the individual PID's to the truck (0105 for instance for coolant temp) to see what the truck sends back. I'll do that tomorrow and update this post with the results.