(Previous thread, here:
Thought I'd try and canvas ideas, as this has reared it's head again.
Summary of previous thread, having intermittent - albeit fairly regularly occurring (depending on circumstances) issues with idle racing, when the engine is hot (say been running for a while, especially motorway driving). Car is a 53 plate Kia Carens 1.8 petrol, ~65k miles on the clock, fully serviced.
Turning on the air-con can drag the idle down to roughly about normal. Turning off the car for a minute or two would likely see the idle return to normal - at least for a few minutes. No codes or engine lights (have a code reader, plus a USB ELM dongle).
Initially tried cleaning throttle body and plate / butterfly, cleaning IACV, then replaced IACV with pattern part (albeit looks identical to the OEM IACV I removed).
Cleaning the IACV nor replacing it stopped the problem. I did find that if with the engine running, and the idle racing, unplugging the IACV, waiting a beat, then plugging it in again, would see the idle return to normal - this would stave off the problem for two or three days. I'd say this keeps it happy for a day or so, now.
The ELM dongle (in conjunction with an app running on my laptop) would show that the absolute throttle position would be non zero when the problem was occurring - perhaps somewhere in between 6% and 10% (or maybe slightly higher, but roughly in those bounds). Disconnecting the IACV briefly, then reconnecting it, with the engine running, and the problem happening, would see the absolute throttle position return to 0%. That kind of implies to me that when the problem starts to exhibit, the ECU is seeing the absolute throttle position, at idle, at a non-zero value - the cause of that, I suspect, is key to this - is it something happening to cause problems with the TPS over time, heat affecting the TPS causing it to do this over time, or something else, so far undetected, that's causing these symptoms.
The final thing I did was replace the TPS (I reported back to the original thread on Aug 19 that I'd done that) and that effectively cured the problem, until roughly about a week ago, and it's slowly come back, and started occurring more frequently. So maybe a little under three months, a new (OEM, from a Kia main dealer) TPS effectively cured the problem.
I guess what I'm trying to understand is why - it would appear that time and perhaps heat is causing the TPS to go bad - it's the only thing I can conclude from what happened originally - I guess it's that that I'd appreciate theories or thoughts on.
As an aside, I did seem to have a minor rattle what sounded like a heat shield, recently - which stopped, abruptly, and I was half inclined to think something had fallen off - I wasn't able to see anything awry when I'd hear the odd minor rattle, nor can see anything clearly missing now, but did make me wonder whether it might have a bearing on under-bonnet heat. I'm going to have a service at the main dealers, in the next month or so, so I'll have them look into that.
It may seem odd that I'm looking into the warm idle issue myself - but past history has taught me that intermittent and heat related problems don't seem to be things very well diagnosed by main dealers, and if there is any diagnosis, tends to be something of a vague shrug - perhaps even less likely, if there's no error codes (which there isn't).
Any thoughts, ideas or theories welcome. I'll probably buy another TPS and replace that (roughly around £23 all-in from the main dealer - so not a terribly expensive thing to do) - and if current experience goes true to form, that may last another three months - but as much as anything else, I'd like to understand what's causing it, really.