Valve bouncing detection technology
Posted: Wed Mar 17, 2021 8:02 am
Hello Speed Talk'ers,
I am planing on 90x78.5-I4-2-Liter-NA-engine spec'd to 380 hp@11.000 rpm. As I don't relay on single focused and extremely expensive measurements like a Spin-tron would offer it, I am curious if anyone of you detected valve bouncing with either a knock analyse system or a pressure indication system. In the former company I worked for we made a two step: simulation and measuring on the single cylinder and the full sized engine. I have not such capacities in my own company, but I have the two measurement systems mentioned above. What do you think or experienced, is it possible to detect it before the engine suffers damage with one of these?
The idea is to listen with the knock sensor at a window around EVC and IVC. As the window can be small as 25 °CA (with assumptions for engine speed related delays even smaller), there might be a good chance to see the seat noise at least with some mathematical extraction methods.
What's the loss in the risk calculation? Every 500 rpm lower this engine concept loses over 10 hp and the head ports would have been in need to be adapted to that lower engine speed to keep the power band wide. A safety revving band of around 1500 rpm has to be added to peak power engine speed. So I want to keep the risk lowest by lowering the probability of an unknown valve train issue on the engine dyno.
Background, it is a VTEC-less K-series valvetrain from Honda. There are valve springs and rockers available tested by aftermarket up to 11.000 rpm in a various combination, which might not the one I use, but here we will go well over this, up to 12.500 rpm for a blink of a second (slippery application sometimes).
I won't test it without the capability to see it coming. So any advice and support is welcome.
Markus
I am planing on 90x78.5-I4-2-Liter-NA-engine spec'd to 380 hp@11.000 rpm. As I don't relay on single focused and extremely expensive measurements like a Spin-tron would offer it, I am curious if anyone of you detected valve bouncing with either a knock analyse system or a pressure indication system. In the former company I worked for we made a two step: simulation and measuring on the single cylinder and the full sized engine. I have not such capacities in my own company, but I have the two measurement systems mentioned above. What do you think or experienced, is it possible to detect it before the engine suffers damage with one of these?
The idea is to listen with the knock sensor at a window around EVC and IVC. As the window can be small as 25 °CA (with assumptions for engine speed related delays even smaller), there might be a good chance to see the seat noise at least with some mathematical extraction methods.
What's the loss in the risk calculation? Every 500 rpm lower this engine concept loses over 10 hp and the head ports would have been in need to be adapted to that lower engine speed to keep the power band wide. A safety revving band of around 1500 rpm has to be added to peak power engine speed. So I want to keep the risk lowest by lowering the probability of an unknown valve train issue on the engine dyno.
Background, it is a VTEC-less K-series valvetrain from Honda. There are valve springs and rockers available tested by aftermarket up to 11.000 rpm in a various combination, which might not the one I use, but here we will go well over this, up to 12.500 rpm for a blink of a second (slippery application sometimes).
I won't test it without the capability to see it coming. So any advice and support is welcome.
Markus