[EdmondEsq]

Hey all.


Over the last few thousand miles or so, I've noticed a strange occurrence that I can't explain, and was wondering if anyone else here has experienced this as well.


When I'm at a complete stop, and with the car at operating temps, the moment I let off the brake to accelerate, the RPMs drop about 200-300. This only lasts momentarily before correcting and it doesn't happen every time, but it doesn't feel normal to me.


I haven't taken it into the dealer because without a CEL and with it being intermittent, I doubt they will be able to provide much insight into the matter.


FWIW, I also experience a stalling sensation from time to time. This is more infrequent, and it doesn't reflect in movement on the tachometer, but when I come to a complete stop with the car at operating temps, you can feel a small stumbling sensation as though it's struggling for fuel. I don't know if the two are linked, but I'd imagine that they are.


Has anyone had anything like this happen in their Elantra?

 

[newsguy99]

how many miles are on your Elantra? I'd try some fuel injection cleaner first.. And see if that takes care of the issue.
The Honda Fit I drive for the company I work for, has this issue too... Very noticeable when trying to go up a hill and pressing on the gas.
and it really is a stumble. But, putting a bottle of cleaner in, and running it, it won't do it for the next 5000-7000 miles.

 

[EdmondEsq]

I appreciate the input. I run Hyundai's fuel system cleaner (I believe it's a Techron product) in it roughly every other oil change (more like every 5k miles). I'm currently at nearly 16k miles and I run about 3750 mile oil change intervals.

 

[NicholasD]

Two in a roll, read:

http://www.hyundai-forums.com/ad-2017-elantra/606618-insight-much-desired.html

15K miles is my limit on plugs, clean them with walnut shell blaster, again at 30K, get pitched at 45K, carbon built up on the center electrode insulator shorts that spark voltage back to ground.

Another possibility is throttle by wire, EPA is forcing every one to return the PCV to the inlet of the throttle body for the latest emission control device, more carbon crap that can be built up on the TB vane causing it to stick, that servo motor is not exactly Superman. Spaying it with choke and carb cleaner helps so it works free.

Carbon in fuels, a 120 year old problem.

 

[mirageman]

Go to autozone or another chain that offers free check engine light testing and have them tell you if you have any PENDING codes. Codes that might be a problem but not detected enough to light the check engine light. See if anything interesting there so you can have some info to take to the dealer if you have to but don't let the autozone people clear the code, let the dealer see it, you know? If you don't want to go to the dealer just yet, then disconnect negative terminal on the battery for a while or overnight. Then reconnect. This will clear the computer and force it to relearn idle and manifold pressure response and fuel-air metering, etc. It won't take long, just ignore the off-behavior of the car for the next 20 miles or so after reconnecting the battery. Your issues are probably due to the changeover to winter blend gasoline. I do it twice per year in a stick-shift Toyoda (it's small and green) I got. Every few months the darn thing just starts fainting the instant I want to start going at a green light. It doesn't even attempt to stay alive, no bucking sensation, no nothing. It just totally wimps out and dies. I then disconnect the battery for a while or overnight as stated and the problem totally goes away and it starts driving the way I've always known it to drive. What's happening is, the car is used to a certain way of dealing with changes in manifold pressure, throttle position and all and now that the gas isn't the same, it behaves differently and it takes a little longer for the computer(s) to compensate to the different results you're now getting from that different fuel. "You don't need to disconnect the battery, the computer will learn to adapt by itself" the dealer tech will say. Yeah, sure...... but give it amnesia and it will start learning TODAY. Even if you are somewhere where they don't change the blend of gasoline, disconnect the battery. If it is still doing so after resetting as above, take it to the dealer since it most likely will be a sensor issue.

 

[EdmondEsq]

Go to autozone or another chain that offers free check engine light testing and have them tell you if you have any PENDING codes. Codes that might be a problem but not detected enough to light the check engine light. See if anything interesting there so you can have some info to take to the dealer if you have to but don't let the autozone people clear the code, let the dealer see it, you know? If you don't want to go to the dealer just yet, then disconnect negative terminal on the battery for a while or overnight. Then reconnect. This will clear the computer and force it to relearn idle and manifold pressure response and fuel-air metering, etc. It won't take long, just ignore the off-behavior of the car for the next 20 miles or so after reconnecting the battery. Your issues are probably due to the changeover to winter blend gasoline. I do it twice per year in a stick-shift Toyoda (it's small and green) I got. Every few months the darn thing just starts fainting the instant I want to start going at a green light. It doesn't even attempt to stay alive, no bucking sensation, no nothing. It just totally wimps out and dies. I then disconnect the battery for a while or overnight as stated and the problem totally goes away and it starts driving the way I've always known it to drive. What's happening is, the car is used to a certain way of dealing with changes in manifold pressure, throttle position and all and now that the gas isn't the same, it behaves differently and it takes a little longer for the computer(s) to compensate to the different results you're now getting from that different fuel. "You don't need to disconnect the battery, the computer will learn to adapt by itself" the dealer tech will say. Yeah, sure...... but give it amnesia and it will start learning TODAY. Even if you are somewhere where they don't change the blend of gasoline, disconnect the battery. If it is still doing so after resetting as above, take it to the dealer since it most likely will be a sensor issue.

Thanks for the response. I already pulled codes and nothing is pending. Fuel trims read good too, which made me think that it might be a misfire instead of fuel issue. It's at the dealer now to get another small issue fixed. If they can't find anything (I'm not holding my breath), I'll disconnect the battery overnight and see if that helps any.

 

[jago18]

Hey Edmond, this has happened to me once so far. I believe I was at about 25,000km. I felt the engine shudder however I could not look up fast enough to see if the tach dropped RPMs. It hasn’t happen since, I’m almost to 30,000km now.

That being said, I try to watch the tach to see if it happens again.

I don’t know if it matters, I was not at operating temperature. It was about 3km from Home with a outside temp of 10 Celsius. Running regular gas, castrol synthetic oil. Korean built.

 

[EdmondEsq]

Hey Edmond, this has happened to me once so far. I believe I was at about 25,000km. I felt the engine shudder however I could not look up fast enough to see if the tach dropped RPMs. It hasn’t happen since, I’m almost to 30,000km now.

That being said, I try to watch the tach to see if it happens again.

I don’t know if it matters, I was not at operating temperature. It was about 3km from Home with a outside temp of 10 Celsius. Running regular gas, castrol synthetic oil. Korean built.

Yeah, it's pretty unpredictable with mine as well, though it happens much more frequently. I've tried everything, changing the oil, filter, gas from different stations, etc. Nothing has worked. I have the Torque app running sometimes when I drive, but I haven't seen anything out of the ordinary (granted, I'm not a professional mechanic, just a shade-tree kind of guy).

I'll post more once I get an update from the dealership (it's staying there overnight). I don't know how much they will be able to tell me, and I am not holding my breath for them to find a fix but I will let everyone know what I'm told.

 

[NicholasD]

Never mentions spark plugs, fuel injectors are also electrically powered, a scope with a current probe on it shows a misfire in a hurry. Or one can be sticky.

Idle speed control is also done electrically to vary the opening of that vane in the throttle body, suppose to handle just air, but also the return for the PCV, that certainly can be erratic.

Ha, when codes and scanners first came out heard weird things like, your bumper is not falling off, ain't getting any codes. Misfires are type B codes, they reset themselves after the problem is gone, ha, nothing is wrong with your vehicle, ain't getting any codes. And can be very erratic. No codes for a sticky TB vane.

 

[Tooch2422]

Here's a possibility. These cars have hill assist (well, not alllll of them) which holds the brakes on for 2 seconds after releasing the brake to help prevent the car rolling back. Other vehicles I have owned that also have this (Mazda CX-5, Dodge Ram and Charger) the assist works almost seamlessly. When you release the brake, the vehicle does not move an inch. I've noticed on my Elantra the vehicle rolls just a slight bit. It took me by surprise at first, but now I'm used to it. Test and see if you are possibly feeling this and mistaking it for an engine sputter. As you don't see the sputter register on the tach, I wonder if this is what it is?

 

[EdmondEsq]

Here's a possibility. These cars have hill assist (well, not alllll of them) which holds the brakes on for 2 seconds after releasing the brake to help prevent the car rolling back. Other vehicles I have owned that also have this (Mazda CX-5, Dodge Ram and Charger) the assist works almost seamlessly. When you release the brake, the vehicle does not move an inch. I've noticed on my Elantra the vehicle rolls just a slight bit. It took me by surprise at first, but now I'm used to it. Test and see if you are possibly feeling this and mistaking it for an engine sputter. As you don't see the sputter register on the tach, I wonder if this is what it is?

That's a good thought, but I think I can rule that out. One of the most common places this occurs is at the stop sign at the exit of my neighborhood which has a slight downhill slope. Yeah, it doesn't register on the tachometer, but my scanner picks up the rpm drop and you can definitely feel it.

 

[Tooch2422]

LOL Wellllllll it was a thought

 

[EdmondEsq]

I appreciate the input. It's a real stumper. In fact, the dealership still has it. I'm forced driving around a crappy Chrysler 200, although I really shouldn't complain because I could just as easily not have been given a loaner.

 

[NicholasD]

Engine idle problems or lack of are usually the most difficult to troubleshoot, I start off with both a compression and vacuum gauge, ignition, then fuel.

You won't see 20"/Hg with a vacuum gauge with an intake air leak, one car had a defective MAP sensor gasket, others has loose intake manifold bolts.

You won't see 185 psi with a sticking valve, cam gear one tooth off, excessive carbon on a valve, this can be done in under five minutes.

Coil on plugs are easy, use a pulse generator hitting a gated bipolar transistor and a spark gap testers, expect ot see a 3/8" spark.

Can't live without a fuel pump pressure tester, on some vehicles need 55 psi, if even down to 50 psi, they won't run at all, Elantra wants to see a steady 45 psi.

Look at the plugs, if one has more carbon or is wet, fuel injector problems. All this stuff is basic.

With the new stuff, can also be firmware problems, reflash the RAM, dealers have this, for the rest of us, about 1,500 bucks per year for a subscription, one key reason I like the older stuff.

With this new stuff, they got us where they want us.

 

[jago18]

Edmond,

It happen again this morning. Full operating temperature. -16 Celsius outside.

I just came off the highway. Came to a stop. Let go of the break and the RPMs drop just below 500.

Interesting experiment i did after this happen:

At the next light, i bumped into neutral and released the breaks. The RPMs stay at normal.
Then i put it back into drive and release the brakes and the RPMs begin rise without throttle input and the car begins moving.

Odd? Normal?

 

[NicholasD]

When torque converter lockup was first introduced some would lockup when they weren't suppose to, like at a traffic light stopped in drive. At first they were all mechanical, then electronic, separate AT module with code stored in non-erasable RAM.

Then that module was incorporated into the ECU to save a couple of bucks, to save on updates, PROM was fired and replaced with flashram. just hinting on even more possibilities.

Microcontrollers depend heavily on a good power on reset to read code properly, if not, can be very erratic. All this started with EPA required OBD II specifications, other countries followed as well. In computers called a glitch, not much different than what is in our vehicles.

 

[Red Raspberry]

If the torque converter clutch is disengaged the car will stall when stopped. Just like a manual trans.

If your scanner can monitor Load that is a very active and immediate output that is directly relative to the air ingested and the rpms. It's the Mother of all power PIDs to monitor.

 

[NicholasD]

If the torque converter clutch is disengaged the car will stall when stopped. Just like a manual trans.

If your scanner can monitor Load that is a very active and immediate output that is directly relative to the air ingested and the rpms. It's the Mother of all power PIDs to monitor.

Torque converter engagement clutch will certainly kill the engine if the electronically controlled solenoid valve is opened full.

But what if that solenoid valve just kicked open for an instant due to some kind of microcontroller glitch dues to one missing bit in the instruction code? Would definitely slow the engine, but not kill it.

Just dealing with possibilities and glitches are random, hint was it does better when shifting into neutral first.

Electronic controlled automatic transmission became popular in the mid 80's, for one thing an 89 cent microcontroller is a lot cheaper than a mechanical governor and a vacuum modulator, and solenoid valves are much cheaper than that complex control valve, a Vss sensor was added, a very tiny bar magnet with a few turns of magnet would around it.

Kid was told it his had to be replaced, 4,000 bucks, does it still run, bring it home. Dirty contacts in the brake switch, also in the so called neutral safety switch, Elantra uses a rotary switch for this in manual mode, road salt corroding the transmission connector terminals. After a couple of hours of cleaning, was good to go again. His did have the separate module, pull that and touch up some marginal solder connections, always at the output terminal, use wave soldering, more mass, doesn't get hot enough to burn off the flux.

Just one example.

 

[Red Raspberry]

My guess you would get a code if the TCC was on when it should not be.

It should be something you can see too with OBDII.

 

[Red Raspberry]

A stumble is usually a lean condition. Maybe the purge valve is leaking?