DEBUNKED ! Cold start , stop , restart = No knock

The "theory" was you start up cold, shut down in a matter of seconds, then the car sits, say for a further 24 hours. Then you get the knock. This is is what happened with ours, just the once.

 

No, this was the once, pre-can. Yeah I think one of the parameters was the goodly delay to second start. I'm afraid to experiment, lol. I'm pretty biblical about avoiding short start up and shut down. If you're game to try, it'd be good to see how that goes.

 

What catch can are you running?

 

I started believing in this oil catch can and I think I will buy and install one when I start doing the routine maintenances.
My car is at 143k. I got it last month at 142k. It is clean and runs good but it did this cold-start knocking on me once for about 5-10 seconds (and another time for like a split second). I was surprised when it happened but then I realized it is a common issue with Gen 3 models.
I asked the previous owner. He said it never done it to him. Well I still love this car. It is an awesome daily driver.

 

the catch can may be the answer, not sure how toyota would have implemented it as a fix, or why they dropped a bundle on improved intake manifolds that didn't work.

the only way to prove it is to do as mendel said, and for a long enough period of time to be scientific. it doesn't always happen, which is even more frustrating because it's hard to show the dealer. conditions have to be just right.

 

Well, I can't really say much but he told me it didn't do it to him. If he is lying, then he is lying. He just didn't seem that kind of guy to me. A nice husband; a dad to two kids. If he had said "Yes he experienced it", nothing was going to change anyways. I bought the car as-is.
When you are buying a used car, there is always a risk. I looked for one accident-free (no body repair and paint job). You can't expect everything to be top notch.
We will see how it goes with the car. So far, it has been running great (except the knocking I experienced). The car has been serviced at the Toyota dealer during the previous ownership. I got the service records and checking what has been done to the car in the past few years.
Then I will devise a plan and start doing certain maintenances (like IM, EGR, PCV cleaning, Spark plug etc...)
I have a spare car; that would take the pressure off me when doing maintenance...

 

Correct. I read the entire 46+ page Knocking issue thread and found out that this can happen to any Gen3 owner (probably less the ones that use catch can and doing routine maintenance) at any time. It doesn't have to be cold, or hot,, or warm-up cycle interrupted, or anything. In that thread, there are many different scenarios where people experienced this issue.
I wish Toyota would just install a similar device that would catch oil instead of spending money on that useless IM upgrade.
I am pretty sure that they damn well know what the issue is. They just couldn't find a easy and economic route to get out of it. Instead they elected to ride it out. Now with the Gen4 is out, I am sure they feel more comfortable and let the time cure the wounds this embarrassing issue caused to company.
But there are still a lot of Gen3 Prius out there and it will take many more years for this model to be completely out of the market.

 

i had it happen once in summer, after driving for awhile and pulling off of a 40mph road into a parking lot. i don't know if it was the same thing, but my father even asked me what all the racket was. shut it down, went shopping, got back in and it was gone, and never happened again.

 

i wonder if there are any reports on the gen 4 engine yet, it will be interesting to watch.

 

That has a scientific sound, and got me started thinking about just what steps would be needed to approach this in a scientific way.

1. We would need to know, replicably, how to make the knocking happen in a stock Prius. It doesn't happen to everyone, even everyone who makes short trips in cold weather, so there's already work needed just to be able to say "do these things under these repeatable conditions, and you will observe this when the car is stock." That's the "control."
   • It's possible that, even under strictly stated conditions, the problem can only be made to happen some fraction of the time in a stock Prius, and that's ok; experiments can deal with probabilistic behavior. The control result would end up being given as, roughly, "do these things under these repeatable conditions, and you will observe this with probability p", where p is some fraction less than one.
   • There is no observing p directly: p itself is a parameter of the underlying behavior, that can only be estimated experimentally. If you create the stated conditions and get the knock m times in n starts, then m/n is an estimate of p, but if you repeat that whole trial several times, you'll end up with several such estimates of p. That's what has to happen in order to reckon the tightness of the estimate of p. So the final statement of the control result would look like "do these things under these repeatable conditions, and you will observe this with probability p, which (we have c% confidence) lies somewhere between pmin and pmax."
   • That was all just about the control.
2. Once a good statement like that can be made about the control, evaluating the OCC is straightforward; it just needs the same kind of repeated trials to be done in the same stated conditions, enough times to come up with a new estimate pocc of the probability of seeing the same behavior with the OCC in place. There will also be some range of confidence around pocc.
3. Once such experiments have produced estimates of p and pocc, then if pocc < p, that suggests the OCC has some benefit, and the strength of the suggestion depends on how far pocc falls outside of the range of expected variation for p. In particular, the higher and tighter the range for p in the control case can be shown to be, the easier it is to show a benefit of the OCC. Conversely, if the estimate of p in the control case came out either (a) quite low, or (b) loose enough that rare or no knocking could be an unsurprising outcome for a particular trial even in the control case, then even if knocking with the OCC turns out very rare or unobserved, it's harder to show that means much.

All of that's good, straightforward work that could be done, and would be interesting to read here; from the threads I've seen so far, there's still a need for digging into it, starting right off with (1).

-Chap

 

I imagine if you can get a Prius without an OCC to knock by doing that, you could get one with an OCC to knock by doing the same thing. That's ok; we're not actually trying to explain the behavior of cars whose owners pour oil into their manifolds, or to improve the behavior of cars whose owners do that, if there are any.

The real trick is to come up with a concise statement of actual operating conditions, encountered by people merely operating their cars, not opening them up and pouring stuff in weird places, under which we can say the knock will occur with some high-ish probability.

After that, it's just a matter of showing that the proposed remedy lowers that probability, under those same actual operating conditions, and lowers it by "a lot" relative to the confidence intervals around the two probability estimates (which implies taking observations quantitatively enough to have those estimates and those intervals).

That would really give the thread some strength.

-Chap

 

That test could include both oil-guzzlers and non-guzzlers, if I may introduce another variable.

 

I'm new to this thread but ^this^ is definitely not toyota's path. If anything they are advertising against the future.

 

New to prius chat. This forum has been invaluable for me.I prchased a 2010 Prius (base) with 250K with a CEL on constant not blinking, the owner (friend) advised me that the check engine light was diagnosed by his mechanic and the mechanic suggested he change the plugs. Ran the codes with an OBDII scanner. P0301 and P0302 came up. Misfiring in cylinder one and two. Some research and everything pointed to EGR system needing to be cleaned. Low and behold the EGR pipe, cooler and intake were completely clogged. I could not believe how clogged. I was able to get the cooler relatively clean (able to see a healthy stream of water when blowing powerwasher through cooler) and pipe along with the intake. Oh, I did replace the plugs before this mammoth task of taking the EGR system apart.
Plugged OBDII scanner back in and erased codes. Shortly after P0301 came up again. I assumed the system was still getting used to being clean and expelling any additional toxins that may have been in the system. Erased codes again. Fast fwd two weeks and no more CEL but the engine shutter comes back periodically. I like your explanation that in theory keeping the oil out of the intake may make the shutter/knock go away and it's a relatively inexpensive fix. Any other wisdom out there anyone want to bestow?