[ALG]

Is there any thought of adding this function. Knock detectors are quite common on turbo cars even my first gen eclipse had one 1990!

[PSIG]

No, Speeduino instead avoids the problem by using over-boost protection, so the over-boost condition does not occur.

David

[ALG]

David,
Thanks for the reply!
You said "boost control" didn't you mean boost limiting? I cant see how any of the parameter checking that's used can detect pre-ignition!I am missing something so please educate me. I know my 1990 eclipse had a knock censer . On my 2017 turbo civic i'm not sure but when you get 180 hp from 1.5 liter the boost has got to be right on the edge.
Thanks

[evo_lucian]

David,
Thanks for the reply!
You said "boost control" didn't you mean boost limiting? I cant see how any of the parameter checking that's used can detect pre-ignition!I am missing something so please educate me. I know my 1990 eclipse had a knock censer . On my 2017 turbo civic i'm not sure but when you get 180 hp from 1.5 liter the boost has got to be right on the edge.
Thanks

If you have overboost protection then you wont ever get into a situation where you will need knock detection caused by over boosting.

[PSIG]

David,
Thanks for the reply!
You said "boost control" didn't you mean boost limiting?

No, I did not use either of those terms. Evo_lucian is correct (as usual). You can do a search for knock sensor, KS, and other terms to see the viewpoints about the subject; but it actually falls to a problem of misconceptions or misuse that are a greater danger. For example, most OEM ECMS utilize knock sensing and response that works great at low and medium load and rpm, the common usage for common daily drivers. What is not commonly known is that many OEM schemes ignore the knock sensor signal above a programmed rpm. Many people think they are "protected" when they are not. Likewise, many users assume KS is automatic protection, and even if it is working well, there is a common tendency to rely on KS to protect them instead of using proper tuning techniques, fuels, control, and other ways to prevent (instead of wait to detect and react to) damaging detonation. This is similar to the common misuse of WBO2 as a real-time correction device in an attempt to force a poor tune to run better. It's still a poor and potentially damaging tune with less performance.

I appreciate and use knock sensing if properly calibrated and tested as an additional data point; but never use it alone, as KS is many times unreliable, improperly calibrated, or is not effective at high rpm (where you are wanting it to work best). Speeduino's open-source format allows the addition of external sensors and their controllers with user code changes; but does not use a dedicated one internally that is tuneable to the unknown sensor and controller, unknown engine, and unknown signals. Add one if you like, at your own risk or benefit. No problem. For other users, having no internal KS also means they must use (or learn to use) reliable tuning techniques for an overall safer and better-performing result. This is not a bad thing; safer for them, and better for Speeduino's excellent reputation to-date for not trashing engines. I hope you can see how this approach actually offers better results and greater protection to more users overall. Do your thing, add a knock input if you like, and let everyone know how it goes!

David

{EDIT] PS: I noticed you said Eclipse, and wanted to mention that a relative did a turbo swap and dyno-retuning on a good-running one a few years ago. The KS actually tried to work, but reviewing the datasheets the KS signal was tiny, and too small for the dyno operator to notice or for the ECM to trigger response. Just moments later it blew at 6200 IIRC. As you can imagine, I doubt the dyno driver used additional det detection methods, and I don't place a lot of trust in the Eclipse KS system by itself to prevent problems.

[ALG]

Dave,
Thanks for the info, that' s why I'm here.

[ecumania]

Any one experimented with Ion Sensing? I think Saab developed that.

[PSIG]

Any one experimented with Ion Sensing? I think Saab developed that.

Knock/det sensing will likely become a future feature on GPIO or possibly advanced versions of Speeduino, as it has large overhead and will be an involved project to provide reliable and tuneable sensing functions to cover many or most applications. Ion sensing would theoretically be one of the "ultimate" answers to best ignition timing control and also provide det-detection; but I don't know of anyone successfully employing it in DIY systems at this point. Being a bit immature and unreliable at this point, it is only a promising technology and something to look forward to. Knock-sensing is a valuable tool for tuning det-limited fuels, but ion sensing would allow tuning directly to best timing (any fuel), and detect det as a secondary function. The primary issue at this point is not technology, but the general public impression that knock sensing is a protective device, rather than a tuning tool.

David

[ecumania]

Sounds reasonable.
In theory, you could buy a Sparkplug with the ion/sensic inbuild and wire that to GPIO.
I'm not sure if the Connection from GPIO is reliable enough for interaction with speeduino yet.
As i understand it, the System is for selfetunig a Basic engine after a power loss or reset.
Manual tunig is better of course.