MG-Rover.org Forums banner

1 - 16 of 16 Posts

·
Registered
Joined
·
7 Posts
Discussion Starter · #1 · (Edited)
I need advice from some K series gurus. My Elise S1 is suffering an odd malaise that I have searched for but not found anywhere.

When the engine is at idle and the throttle is blipped, there is a loud gasp and the engine stumbles, then picks up. This only happens when the car is warm - at cold, with a rich mixture, it is perfect. I am engaged in a thread on SELOC at the moment trying to get some ideas but it seems to have run its course. So far I have cleaned & tested the IACV (idle is perfect so it's not likely to be that), tested the lambda sensor (with an oscilloscope and by looking at the ECU data), tested the TPS (with a meter, and by looking at the ECU data), searched for air leaks (checked MAP and fuel pressure regulator valve line, listened for hisses, sprayed brake cleaner round the manifold and throttle body while the engine is idling). Probably done some other stuff too - reset the IACV several times (ignition / 5 throttle pumps per the service manual), tweaked settings in the ECU (mostly ignition offset and long term fuel trim).

Nothing I try is helping.

The engine is a standard K16 with MEMS 1.9 as fitted to the MGF. It has a metal throttle body, which is the standard 48mm size. The air filter is standard and was renewed when I did a B service last month. Same for the distributor cap and rotor arm. Plugs and gaps checked, all good. The car is very low mileage - it's a 1999 S1 Elise with 10k on the clock - and it is exceptionaly clean and unmolested. Starts, idles, runs normally (with the exception of this annnoying glitch). Engine pulls hard all the way to the red line.

So, has anyone on the forum encountered this problem before, and what was the cause?


PS< the SELOC thread is here: SELOC Forums and includes some discussions of the MEMS diagnostic data, with a link to a CSV file captured with the ROVER MEMS Android app. There are some throttle blip events visible which you can see if you load the file into Excel and graph the throttle percentage column against other columns. Some of these graphs are shown as pictures in the thread. So far it all looks normal.
 

·
Registered
Joined
·
244 Posts
I am no expert you seem to have covered the 'usual suspects' (air, fuel, spark). Have you had a look at the accelerator cable tension...and that the TB is operating correctly, butterfly fully cleaned out with proper cleaner, Stepper/IACV and hoses a-ok; I had similar symptoms years ago with the plastic 48mm TB (you do say your's is metal) and changed it out for a 52mm version, I would do that on your's in any case as gives better throttle response.

Have you messed about with the MEMS1.9 and parameters?If so return to OE spec with base idle settings using T4 or PSCAN

HT leads new...squirt of contact cleaner/WD40?

and I will whisper it......some have said in the past this may also be the start/symptom of HGF....my TF was certainly iffy when it started going South....losing coolant?
 

·
Registered
Joined
·
6,259 Posts
Try a new coolant temperature sensor, these are known to give symptoms like you are experiencing.

Good luck.
 

·
Registered
mg_tf
Joined
·
1,604 Posts
I'm no K-series guru but a stumble when the throttle is blipped suggests to me either (a) underfueling at that instant* or (b) the timing being way out momentarily (heavily retarded?).

* Many moons ago I had an Escort Mk II and the mpg was only about 30 around town. I decided as an experiment in fuel economy I'd disconnect the linkage on the carburettor that dosed in a pulse of excess fuel to enrich the mixture when the accelerator was depressed suddenly. The engine seemed to run just fine driving along but if I tried to pull away from a junction sharpish the mixture just went too lean as the inrush of air from the throttle body outpaced the fuel delivery and the engine stumbled. Having found it was a good way to put myself into awkward situations I went home and reconnected the offending carb linkage...
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #5 ·
Thanks for your replies!

From the MEMS_diag output I can see that the coolant temperature sender looks OK. I think I'll revisit that one. Best I can do at the moment is compare the MEMS reading with the dashboard, as they use different sensors. I'll see what a new one will cost, but I am a little sceptical as it does run perfectly at other points in the load cycle. Plugs are clean with a light brown deposit on the insulators which suggests a good mixture.

I've experimented with timing offsets (in MEMS_diag) but it made no difference to the problem. Besides, if it was timing we'd see it stumbling when cold, so I think it's safe to rule that out.

As to the throttle body, well, it's clean as a whistle (though more cleaning wouldn't harm it I suppose). I'm not going to fit a 52mm body as that will have a disproportunate effect on the insurance. I pay £150 fully comp for the car; the moment I declare a mod like a sports exhaust, it goes up by £100! Besides, it does need to be kept standard (bar a few sensible upgrades) as it is a special low mileage beastie, so I want to keep it original.
 

·
Registered
Joined
·
21 Posts
Just a thought have you changed the TPS on the end of the butterfly it could have a 'dirty' spot and gives false values at that point.
Last point has the car been OK if yes then what did you do to it just before it went wrong if you can go back and put back what you changed.
As a rule having been bitten in the rear before I never throw away old parts until the new have at least lasted 6 months to a year as there are so many cheap crap parts sold as OEM on the market that just dont quite cut it ....boxes look nice😊!
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #8 ·
Please refer to the attached series of graphs, which show various MEMS parameters while the engine is running. Most are labelled correctly and should be self explanitary. The bottom two (both labelled "Series 1") are MAP sensor Kpa and coolant temperature respectively (Kpa is the last but one, highlighted with a pink bar). This data was logged while the car was idling at normal operating temperature. You can see several throttle blip events on the throttle graph and line them up with any of the other parameters shown. To me it looks like it is all normal.

136721


Regarding the question about the TPS, this has been checked with a meter and it is operating smoothly with no jumping around. The ECU is reading it fine, and also shows a smooth progrtession when opening the throttle.

The car has not run properly during my period of ownership. I bought it knowing about this problem but reasoned that I'd be able to sort it out myself (ha ha). Before anyone asks "Why'd you do that?", here's why:

136722


With 9500 miles on the clock, it's about as close to brand new as it's possible to get.
 

·
Registered
Joined
·
6,259 Posts
OK, so that looks like some progress. Next on my list would bd the inlet manifold gasket.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #10 · (Edited)
Replace or check? If the latter, how best to do it? And if former, which one is a known good quality one? (Actually scratch that, I just ordered one for £2.60).

I assume they crack over time leading to leaks.
 

·
Registered
Joined
·
1,187 Posts
Elise 1 - Not to highjack your thread, I have tracked down the App that you are using - I see that it is for Android and PC. It looks really interesting as it works with 2J and earlier MEMS. How have you found it? Have you used Pscan as a comparator?

Until I have used this APP, I have no idea whether it picks up the number of sensor readings that Pscan does. If you used Pscan, you may be able to pick up more data in order to identify your problem....
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #12 ·
I don't know what PSCAN is, but I presume it's a Rover tool. The apps can log everything that PSCAN can see, because they are getting the same data packets out of the MEMS ECU.
 

·
In the Garage
MG TF
Joined
·
33,178 Posts
PSCAN can go deeper into the RoverMEMS cable and software but it is £100+ more expensive and in this case I don’t think it will help over and above what you already have.

One of the main drawbacks with RoverMEMS is that you don’t get any extra sensor info such as fan speed etc but I can see that being added in.
 

·
Registered
Joined
·
1,187 Posts
RoverMems appears to come out with less readings than Pscan but I have not gone live as I am awaiting the right cable to be delivered (usb and micro-usb leads).

Rovermems however works on a mobile phone as well as Windows, so given more dev time it may well equal or even better than PScan - very interesting, a fantastic byproduct of your post!

Regarding your problem, have you checked the number of steps the IACV is set at hot idle (oil temp must be at 85C + and idle at circa 875rpm)? On my F VVC, the number of **steps is set at 21 but I think that anything around 25 is OK. In my case, I encountered a very severe stumble (more extreme than yours I suspect - the Stepper at hot idle was at 50 steps before adjustment but idle often was at 1600 rpm - ie very unstable) so I checked that the throttle at idle was at *4 to 5 degrees (doing a physical calibration) and then made the adjustments such that my idle settings ended up at 4-5 degrees open (physically), steps at 21 and idle at 875rpm. There was a fair amount of faffing around but I got there in the end.

I note that your idle is rock steady but when I had my problem, the idle could equally be rock steady, albeit not all the time, but the stumble on acceleration was still there. Personally, I would be checking ALL the data for clues.

* You appear to be showing in the graph 10 degrees at idle. This cannot be right.
** Pscan can read the number of steps but currently, I cannot see RoverMems detecting this reading.

Ps, Your vacuum pressure is on the nail.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #15 ·
Thanks Julian.

Could anyone suggest reasonable values for the following: ign advance offset, long term fuel trim and idle decay, please? I'll go check what I currently have and report back.
 

·
Registered
Joined
·
7 Posts
Discussion Starter · #16 · (Edited)
OK, just checked and the values are (as reported by ROVER MEMS app):
  • Ignition advance offset = 126 (range 116-140)
  • Idle decay = 31 (range 10 - 60)
  • Fuel trim 135 (range 0-254)
  • d2-d3 2 (no range given)
..so I'm looking for a known good set of values for these parameters to use as a starting point.
 
1 - 16 of 16 Posts
Top