Fuel mileage going down.....?
 

My '04 has been a pretty good little van. I try to keep up on all the maintenance, changing fluids, plugs, wires, EGR, etc... No codes and runs respectively good for the amount of miles on it.

Generally it will maintain an average of about 22 mpg which is somewhat acceptable given that its always been this way and it currently has 192,000 miles on it. But recently within the past couple months or so the average mileage has started to drop to around 18 mpg. :confused:

Another strange thing is that when I'd drive down the freeway and reset the average mileage overhead display, the mileage would jump way high up in the 30's and then slowly tick down to about 25 mpg, as long as I was cruising along on the freeway that is. But now within those past couple months, when I do the same thing the overhead doesn't jump up to 30(ish) but resets to only around 21-22 and then quickly drops to about 18-19 mpg. Somethings obviously wrong.....

Could a weak fuel pump cause this? Looking for any advice you guys know of. Thank you.

What engine you got?
Could it be your board computer expects much better mpg but measures lower mpg so adjusts accordingly?
I would also calculate MPG by filling tank and set day trip counter to 0 and calculate manually when refilling tank to exclude board computer error

for reference: my 2002 2.5 crd does 33mpg (1L:14Km) in current weather conditions.. :)

22mpg sometimes 18mpg...Cough Cough...wait till I faint....Oh god...Must be Petrol and in US of A...

22mpgUS on highway/motorway for a vehicle this heavy (2.4tonsUS) doesn't sound bad to me... Oops, think I've a cough too.
Being serious, if you have that gut feeling it's not right, and the engine doesn't seem partic. down on power, see to the injectors. Petrol injectors CAN be cleaned with cleaner, unlike diesel (since they can get gummed up) but you may have to pour it in the fuel filter. Use the same as mechanics use, e.g. Forte is very popular among british mechanics. Otherwise it's injectors out and have them ultrasonically cleaned. See they go back in the same holes.
Also check your ercos for any fuelling fault.
BTW, hope your motor isn't so old it has carburetors...

Leedsman.

The engine is the 3.3, fuel injected. No codes. 25 mpg averaged just driving on the hwy is the best this van would ever accomplish too, so if thats just a US thing then it is what it is..... 22 mpg average around town and hwy is the norm. Even the newest Mopar minivan models are still very close to that in fuel mileage. Only vans like the Honda or Kia will achieve better mileage.

But that said, power sometimes seems a little lower but again, with 200k miles on this van, I'm not expecting it to charge the streets with authority. I do remember reading somewhere that if the fuel pressure drops then it will affect fuel mileage and power but I cant seem to find anything indicating this again. :confused: Over the past year and a half I've run a few bottles of injector cleaner but there's no great change in power or mileage. The main thing I did seem to notice after using injector cleaner is the fuel pump got real quiet.

There are a lot of helpful and knowledgeable members on this forum and although many drive petrol engines the most used engine here is CRD (Diesel) type of van.
I would post you problem on "the other" forum too, best try 2 sources of knowledge in cases like this :)
4th. Generation Chrysler Minivans: 2001-2007

Really, considering your partic. circumstances, high miles on the clock, old vehicle not worth much, the best you can do here is keep up with sensible maintenance.
1) Keep tyres blown up well, 42PSI. front, 35PSI rear. Check there is no feathering of tread due to wheel misalignment by gently running fingers across tread. Check if one tyre is warmer than the rest after a journey. This area can be a major waste of fuel.
2) Get rid of the junk inside.
3) Remember cold weather is always heavier on fuel. So short journeys will always be heavier on fuel than long ones.
4) Remove any brake-bind. This is common with the handbrake/parkbrake on GVs.
5) Drive like a hypermiler; if you don't know what one of them is, Google it. This method of driving involves big anticipation of near-future traffic, traffic lights etc. so you don't waste fuel DRIVING your vehicle into a stop situation, but rather coast into it in neutral. Your autobox will prob. do this for you, mine does.
6) Both spark plugs and high-voltage spark should be good. Long ago it was established that there can be a slight misfire undetectable to the driver which will ruin your mpg figs. Most modern motors have coil packs, i.e. one coil for each plug (thereby elininating the spark-voltage distributor) each coil's primary being driven by a microprocessor based system. If you lift off each coil from its spark plug carefully with engine idling (and keeping your fingers away from the high-voltage) you should be able to hear a noticeable "cracking" noise as the power jumps the gap between coil tip and sparkplug tip. If one is different i.e. a lot quieter, there is a problem there. Some systems have the coils directly on top of the plugs (like my XJ6) and some have a short connector wire for each high-voltage connexion. This will be special supressor wire to reduce electromagnetic interference being radiated. This special high-voltage supressor wire can be faulty, it is only graphited string inside instead of actual copper wire. This is not for the faint-hearted.
And so on and so on.

Leedsman.

Thanks for the input guys. I certainly appreciate it. I guess I didnt realize I'd be talking with so many people from other parts of the world who dont have the same exact thing as I do, nor the same road ways or fuel. Hard to compare under those variable differences.

But that said, just to be perfectly clear, driving habits and conditions are all the same as they have been. Vehicle maintenance is kept up too. But if a vehicle used to achieve 25 mpg on a flat freeway at 65-75 mph and within two months that same vehicle under the same conditions is now achieving only 18-19 mpg.....something significant is definitely not right.

A "lazy" upstream O2 sensor can cause a drop in mileage without setting a code.

Thats worth checking into.

Reading tech mags some even recommend changing them as scheduled maintenance around the 100K mark.

And most people recommend replacing the O2 sensor with the OEM part, which in this case would be a Mopar#56029052AA (Denso Part #2344410)

Fully agree about the lambda sensor, a bad one could easily foul up the mixture ratios, and it's working in a hostile environment too -- defo. a candidate. An exhaust gas analyser would show it up of course as unburnt hydrocarbons, but I'd hazard a guess you don't keep such an analyser in your garage. If of course you have a friendly tame mechanic with access to one, e.g. MOT testing....

Leedsman.
n.b. An afterthought -- have someone in back watch out for the slightest sign of black smoke as you floor the accelerator pedal on a quiet road. Another way is to wait until someone at night is following with headlights on, then floor the pedal. This really does show up any smoke. Nowadays there shouldn't be any smoke whatsoever from a petrol engine under any circumstances -- provided the micro is controlling the injection period properly as appropriate to data from the sensor, so there is slight free oxygen in the exhaust gas at normal running. The mixture is of course rich at idle, under acceleration and at full power. Unless all this control is working as it should, fuel will be wasted.

Before we sympathise too much with low mpg figures on U.S. vehicles, those of us on this side of the pond should remember that a US gallon is only 0.83267 of a UK gallon. :D

Hi steveatpa. There is a common issue which can cause this very problem and one I've come across before. The advise I was given before changing anything is to have a vacuum test performed and a compression test. If the compression is good then you should see 17 to 19 inches of vacuum, if its lower then you could be looking at an intake valley gasket failure. I've change it now on 3 Chrysler mini vans and all have seen an improvement however these are the before and after figures. MPG taken from the fuel/trip computer.


First time was with a friends 3.3 2001 GV, 128,000 miles, good compression. Before 14 to 16 mpg ave, 14 inches of vacuum. After intake valley 18 inches of vacuum 19 to 24 mpg

Then again with a 3.8 T&C (I think 2004 ) 187,000 miles , good compression. Before 12 to 14 mpg ave, 15 inches of vacuum. After intake valley almost 19 inches of vacuum, 17 to 22 mpg

And back here in the UK on my 98 3.3 GV 152,000 miles , good compression. Before 12 to 16 mpg ave 12 inches of vacuum, after intake valley almost 18 inches of vacuum, 19 to 22 mpg

On all 3 occasions gas mileage has returned however, something I should point out when comparing mileage between the US and UK and that is my UK spec GV fuel computer is calibrated to US gallons. I discovered this while doing mileage check and found that the figures I was seeing with a full tank to full tank didn't match the trip computer until and adjusted them for US gallons

I use the Felpro kit available from Autozone for around $20

I found these videos below very helpful

Handy rule of thumb:--
One US gallon = 3.8 litres.
One UK gallon = 4.55 litres.

Leedsman

Good info everyone..... :D I'll check the vacuum first since its easy and free.

Question though about the O2 sensor..... if the sensor(s) are faulty, can I unplug them and see how the engine/mileage reacts? I realize this will throw a code but my concerns would merely be to see if there's a large negative reaction which may point to a working O2 sensor? Throwing $100 at new sensors just to see if that does anything isnt always pleasing to the wallet.

Second question too..... After reading through the thread I realized that I failed to mention that over the past few months the engine has also developed a starting issue whereby it will crank at least twice as long before wanting to start. It used to fire right up with the touch of the key. Sometimes now I'll crank it for a half second.....let off the key.....and crank another half second so it fires up easier. Not a huge cranking and cranking issue either but just longer than normal. This is what led me to feel that the fuel pump was faulty in the beginning of the thread. Could the O2 sensors or low vacuum attribute to extended cranking? And also, it generally cranks longer with the engine hot or cold but will a little more when cold. :confused:

Replaced both O2 sensors. The CEL came on immediately regarding code p0135 and cylinder bank 1 running lean. I presume this is because the new O2 sensor is bad or incompatible.

Waiting for another sensor to replace it with. Also, the fuel mileage hasnt changed. So my next move it to check the fuel pressure.

Did you disconnect the battery when you replaces the sensor? If not try disconnecting it for a few minutes.

When I installed a new sensor upstream sensor on my '09 and the CEL returned almost immediately (within minutes) with the same code I had been getting, even though the code was a two trip code and I had disconnected the battery.

I did some research and found that while the negative battery cable is dis-connected you need to touch it to the positive cable to clear the "stored" data, in the PCM, from the old sensor. I tried it (I left the cables connected for several minutes) and the light has stayed off ever since.

It might not work for your '04, but it's free to try.

Hmmm..... I'll give it a try but I also have a scan tool which allows me to erase any codes. When the P0135 code immediately popped up after installing them, I deleted it and took the car down the street. The CEL didnt show up again until the next day when I went to town. About a mile down the road the CEL came back.

I did a little research on this and there seems to be a known issue with Bosch O2 sensors on these engines. I'm going to try another one and if that doesnt work then I'll try Denso.

Georgef has raised a pertinent issue, re. when removing the -ve battery connector, as touching it firmly to the +ve pole will discharge any capacitors wherever they may be incircuit. Now it's a feature of c-mos memories they DO tend to hold parts of their memoy even when you think those data have been dissipated. C-mos is usually a 12volt device, and I've known with volatile memories they can keep some data even if the power supply drops as low as 2volt, which could be the voltage on a semi-discharged electrolytic capacitor.
You don't know what capacitors are involved, they may not even be on the microprocessor boards, being in some cases at various points on the chassis. My Jag. had this type of design.

So I'd advise after disconnecting the -ve connector, touch it to the +ve battery pole for about a minute. This should totally clear any c-mos memories. Of course, it doesn't apply to TTL logic. The usual caution should be applied to working around any lead-acid starter battery, make sure you don't inadvertently short the battery.

Leedsman.

Maybe I'm not understanding what you're describing. Are you saying to simply disconnect the battery negative cable and touch it to the positive side of the battery? :confused:

Er...yes.

Leedsman.

Why do I feel like this is NOT something you should do. :confused:

I would not consider doing that ............. ever. If you want to take long shot old mechanics tale at face value and clear [volatile] codes that are cleanable before the caps have emptied then undo and remove both terminals and clamp them together.

As long as a least one of the battery terminal is not connected to anything there is no danger............ever.

This procedure will clear values held in the computer's memory such as Long Term Fuel Trim (LTFT), these values are not reset by clearing the codes with a code reader.


If it seems safer to you, disconnect both terminal, but it is not necessary.

So just to be perfectly clear so there's no question.....

I disconnect the negative battery cable, leave the positive cable connected to the battery, then touch the disconnected negative cable to the positive cable and post (which is still connected to the battery). Holding them together for about a minute, this will clear any code memory stored in the OBD system. And to make this even more clear, this will damage NOTHING....., nor will I wipe clean the ECM or PCM system of the information needed for the vehicle to run properly?

Last thing I want to do is see a rash of sparks and find I just destroyed most of the electronic system on my van because I did something which goes against the inner fibers of how car batteries work.

Correct.

It will clear the codes but more importantly it will clear the "learned" data in the PCM.


It will not damage anything, the PCM will be reset to factory defualts.

How can there be sparks if one terminal is disconnected?

Like I said above if it makes you feel more comfortable you can disconnect both cables and touch the ends of the cables together.

Rule of thumb for electrics (also applies to 230volt house electrics):--

If you think you are going to do something a bit dangerous, like shorting a starter battery, apply your SMALL* 12volt test lamp between the terminals you are about to short together FIRST. There are four possibilities then:--
1) The lamp will not light at all, indicating no current.
2) The lamp will light dimly, indicating a small current, less than the full current draw of the lamp. This is usually safe.
3) The lamp will light brightly, showing it's time to do another check with a bigger lamp. If the second lamp also lights brightly, excercise caution.
4) The lamp will light briefly, a 'pulse' of light. This indicates you have discharged a capacitor somewhere -- i.e. the whole point of this excercise.
Remember electrolytic capacitors can charge themselves up a little even after you think you have well-discharged them! An electrolytic is constructed like a cell/battery inside, hence they are always polarized unless it says on the label "non-pol".

Hope this relieves any anxiety.

Leedsman.
p.s. The old way to check if an electric 230volt terminal in your house was "live" was to have a 15watt lamp on wires, connect one wire end to a firm ground/earth/mass and the other to the terminal you are about to work on. If the lamp lights at all in this situation, sort it out. You can use a neon screwdriver too, but there can be complications with these. (One complication being that the finger on the screwdriver may not be earthed. It may be relying on capacitive effects, esp. if the electrician is wearing well-insulated rubber-soled shoes). Also make sure your ground is good, a copper pipe known to sunk in the ground, the supply ground, etc. are good earth points.
*Small in this case means a flashlamp size.

georgef,

It only ever worked, if at all for the old pre-CAN BUS, on modern telematics BUS's I can't see any value only an increased risk, this is a pure physics rather than a car / computer debate. Arguing that no circuit = no current ([equaliser](equalise relative to what) doesn't wash for me. All modern cars are incredibly voltage sensitive and asserting as you do its safe without telling me what is equalising ..... relative to what ......... seems an unnecessary increased risk. The static voltage in any car can be much higher than the value of the negative terminal.

You can as I stated earlier achieve the same outcome without risk - """remove both terminals and clamp them together""".

Well I appreciate everyone's advice and also looking out for me. If I do this, I'm going to disconnect both cables and touch them together, following the test lamp procedure too. Thank you. :D

I thought I'd update the thread and let everyone know that the problem was the fuel pump/fuel filter. I dont know if the fuel pump was not pumping the correct pressure or that the integrated fuel filter was just plugged up. No fuel pressure ports on this year engine.

But after the install of a new fuel pump/filter, the van runs great and the fuel mileage is back up where it should be. :D