Best of Deals Car Reviews Repair Shops Cars A-Z Radio Show

Car shudders under certain acceleration, goes away when pedal fully depressed

Hello to anybody who can help me…

I have a 2005 Kia Spectra with about 160000 km on it. It has been running fine for most of its life, but last year a stalling issue and erratic, runaway idle (along with an OBD code) led me to replace the throttle position sensor. All was well, no more stalls, and it was a relief to have the car drive “properly” again without the engine spontaneously hitting 3000 rpm at idle.

Now another issue: when the throttle is barely pressed (like the bottom 15 percent of the throttle) the engine seems to sputter and the car lurches rapidly, almost like a slipping clutch. Once I press the pedal further, the car revs smoothly. Even weirder: if I pump the accelerator to the floor, either before starting the engine or a quick pump in neutral, the issue temporarily goes away. It seems like when the pedal traverses it’s full range of motion before use, the problem doesn’t occur afterward for some time.

Just recently the stuttering issue manifested during highway driving, again, only when the accelerator is in the bottom portion of the throttle.

I’m not getting any codes or MIL, so I don’t even know where to begin. Any advice would be helpful!

Some of those throttle sensors have to be adjusted as part of the install. Perhaps the new one you have is sort of working now, but hasn’t yet been correctly adjusted. Do you have the adjustment/check procedure?

I haven’t adjusted it, mostly because I don’t see how it could be adjusted. There aren’t any other moving or exposed parts except for the semicircular hole that the throttle’s post links directly into. What would an adjustment entail?

Reading further, some of these symptoms could indicate a misfire, but again I get no codes of any kind. Can a modern engine regularly misfire without the ECU detecting it?

From what I see, for the 2.0 L engine on the 05 spectra, no adjustment is required. Any adjustment issues are ironed out by the computer apparently, after a few drive cycles. This presumes the sensor is obeying the nominal sensor specs of course. You might want to double check those.

closed — 200 to 500 mv
wide open — 4.3 to 4.8 volts

closed – 0.71 to 1.38 k-ohm (pins 1-2)
wide open – 2.7 k-ohm

1.6-2.4 k-ohm (pins 2-3), independent of throttle position

Unlikely. Misfires are detected using the crank position sensor, and if those go bad the engine run very poorly or not at all. Usually crank sensors fail when hot, and often still work ok when cold. Do you have any hot engine vs cold symptoms?

Some of what you are saying above could point to a transmission problem. What type of transmission do you have? manual? conventional automatic? CVT?

A gunked-up throttle body, faulty egr system, vacuum leak could be involved also. Presuming all the routine engine and transmission maintenance is up to date.

Thanks again for your reply. I have a conventional auto.

Before I replaced the TPS, the engine seemed happier when cold, but ever since then, it seems to happen irrespective of temperature. I had a Kia Sedona from 2003 that had a faulty crank position sensor, and the violent lurching that that caused was very different from what I’m experiencing.

Regarding the tranny, I hope it isn’t something mechanical, since at that state I’m probably better off buying another car. For what it’s worth, the problem often happens with very soft acceleration from stationary, long before the transmission will shift from 1 to 2. As I mentioned earlier, it seems most noticeable at the very bottom threshold of the pedal; if I make a semi-aggressive acceleration from stop, I don’t typically notice the issue.

Other routine steps I’ve taken is visual inspection and cleaning of the throttle body with sensor-safe cleaner.

The crankshaft position monitors the crankshafts rotation.

When a misfire occurs, the crankshaft rotation slows down.

The computer is then able to determine which cylinder misfired, and set a code for the cylinder that misfired.

So for example , the DTC P0302 would indicate that cylinder #2 misfired.

That’s because the crankshaft rotation slowed down when cylinder #2 misfired.


Engine performance and drivability problems at low throttle can indicate an overly lean mixture, often due to a vacuum leak. Suggest to measure the intake manifold vacuum at idle, and if possible obtain the fuel trims (at idle).

One of the first things I checked was the trims, and they all seemed well within tolerance, far below what a vacuum leak would suggest. However, I haven’t checked them in several months, and it might be time to do it again. It may be nothing, but upon first starting the car, the fumes smell fairly rich, richer than my 2016 HRV. Additionally, my fuel economy isn’t great; I barely get 600 km out of a 55L tank, and I drive very conservatively.

I’ll take the car out tomorrow and get the trim/MAF from my OBD reader

Small update: after about 40 minutes of driving my short trim was -0.8 and long trim was -3.1, kinda high I suppose but within tolerance.

I’ve also noticed that, when acting up, the transmission seems to occasionally shift erratically, upshifting when letting the throttle out at highway speed for instance. Again, though, about 3 or 4 full pumps of the throttle makes all the problems (temporarily) going away. Since the TPS is only months old, could this be a pedal position sensor issue?

Also, here’s a total summary of my issues:

-At very low pedal pressure (less than 10% or so) car seems to sputter, lurch rapidly until I press the pedal further, after which it works fine. Accelerated like a dream

-When letting the pedal out at high speed, occasionally odd transmission shifting behaviour such as upshifting while decelerating

-No error codes, and my trim is -3.1. Furthermore, ALL problems go away temporarily if I pump the accelerator to the floor several times before starting the car. However, they will return in about 15 minutes.

What the heck could be causing this?

One LAST update:

To clarify, I don’t mean “sputtering” in the sense of the car almost being out of gas, or that cartoon-like “pfft-pfft-p-pfft” sound. It’s almost as if, momentarily, the engine stops delivering power for a split second, only to return to normal again. It happens with such speed and regularity that it almost (at some but not all times) feels like a clutch that is rapidly grabbing and slipping.

Your fuel trim data is ok & suggest the problem isn’t a vacuum leak or overly lean operation. Still a good idea to measure the intake manifold vacuum at idle, easy enough to do. But it looks like the problem must be something else. It is starting to sound more and more like a transmission related problem. The transmission has to do a lot of fine tuning to do at low throttle, so it could be an engine or transmission sensor has failed. Usually that would throw a code. But not always. Throttle position sensor failures often don’t throw a code for example. But you’ve already replaced that as I recall, so pretty unlikely the replacement is failing. Still, if I had this problem I’d remove the TPS connector and carefully measure the ohm reading between pins 1 and 2. I’d want to verify it changes smoothly, with no glitches, from 710 ohms to 2.7 k ohms as the throttle pedal is moved from idle to wide open. Suggest to verify the transmission fluid is correct also. If a routine trans fluid change-out is due, now’s a good time to do it.

Thanks for your input, I really appreciate it.

Just took it in for oil/filter change as well as changed the trans fluid. The fluid was quite black, so it was probably high time to change it.

Unfortunately my issue still persists. One thing: when doing a rev test at the garage, I pushed the throttle slowly and smoothly and got it to about 1500, where it plateaued for a little over a second despite me increasing throttle. Finally, with a bit more pressing the engine jumped up to about 2500 instead of smoothly spooling up to that speed. The throttle doesn’t “feel” sticky, but it certainly feels like I had to clear a certain threshold before the engine snapped to the rpm I was demanding of it.

I still haven’t tested the TPS (I don’t have a voltmeter and apparently garages don’t routinely have scopes anymore) but I’ll look into getting pins 1 and 2 tested for smooth scaling up of resistance.

Sorry to bug you again but… any other ideas?

You plan is what I’d do next in that situation. Your observation that during a driveway idle test the engine rpm doesn’t increase smoothly as you press more and more on the gas pedal may be an important clue to what’s going on. I’m surprised that garages seldom use o’scopes. You’d think they’d need them more now than before all the computer electronics. I suppose the problem is techs trained to use o’scopes aren’t very common.

Update after a few weeks of inactivity…

I got tired of waiting around to check my TPS for faults, so I went ahead and found a $20 replacement sensor. After plugging it in and driving around 20 km of city driving, the car drives completely without issue.

What the hell? Not one, but TWO faulty throttle position sensors? If this one starts going as well, should I be checking to see if foreign material is poking through from the throttle body or something?

Anyways, thanks so much for all your help, I really appreciate it!