Knock sensing...?
Discussion
Just checked the Emerald Manual, and it does not have an input for a knock sensor. I think you would have to use a Lambda sensor, if your worried about knock.
But if the ECU is mapped correctly and your using the correct fuel then it should never happen, you can also program the 2nd map on the ECU for use with crap fuel. and a much lower advance.
The Emerald does have the option to monitor extra sensors, so you could still use it and monitor it, but The ECU could not adjust to compensate. unless you used the knock sensor to swap maps.
But if the ECU is mapped correctly and your using the correct fuel then it should never happen, you can also program the 2nd map on the ECU for use with crap fuel. and a much lower advance.
The Emerald does have the option to monitor extra sensors, so you could still use it and monitor it, but The ECU could not adjust to compensate. unless you used the knock sensor to swap maps.
I have an accoustic knock sensor (Bosch) on my VEMS equipped Griff.
Knock is caused by many other things than simple fuelling errors and lack of octane, so it's worthwhile having the sensor, although it's a HUGE undertaking to calibrate the ECU to take account of detecting knocking and to then make the correct decision as to what action to take.
I've got some of the way now - I know roughly what the noise profile for the engine is now, so I can look for deviations from this as an indication that knock is starting to happen. (I'd put a lot more work in if it were a forced induction engine)
Knock is caused by many other things than simple fuelling errors and lack of octane, so it's worthwhile having the sensor, although it's a HUGE undertaking to calibrate the ECU to take account of detecting knocking and to then make the correct decision as to what action to take.
I've got some of the way now - I know roughly what the noise profile for the engine is now, so I can look for deviations from this as an indication that knock is starting to happen. (I'd put a lot more work in if it were a forced induction engine)
dnb said:
(I'd put a lot more work in if it were a forced induction engine)
The problem I think is that you really need a load versus revs map of the normal engine frequency/amplitude distribution, the standalone units don't know where you are on the map. I haven't got as far as using the MS one yet but I seem to recall that it is switched off outside a certain rev range, presumably because the MS people found that the engine noise characteristics tended to change too much for the simple algorithm to cope.
More or less confirms what I suspected would be the case... ie that full on realtime sensing/trimming was asking too much. Think if it could be cracked, it would be pretty damn useful on the Rv8, especially forced induction examples and/or those routinely run hard enough on higher compressions to build real heat. Whilst relying on WB lambda prolly covers the bulk of real world running for most users, it's never going to react quickly enough (or possibly at all at load extremes) to make any difference if serious detonation sets in...that'll hammer ring lands off in fractions of a second. And as has been said, there's much more to it than simple fuel ratio.
As a half-way house (perhaps triggering something like a change up light, only in this case 'back off'), there seem to be a few stand-alone sytems favoured by the Jap tuning contingent that might lend themselves to our cause (assuming they work)... might have to buy an old 300ZX to get my hands on one
As a half-way house (perhaps triggering something like a change up light, only in this case 'back off'), there seem to be a few stand-alone sytems favoured by the Jap tuning contingent that might lend themselves to our cause (assuming they work)... might have to buy an old 300ZX to get my hands on one
B&G the megasquirt guys keep teasing us with the promise of an ion sensing system which I think would be capable of detecting det.
Detonation doesn't need to be instantly destructive in order to do damage. Being able to stop it after a brief delay is better than nothing. I think that detonation provokes further detonation so if you can detect the onset early enough you may be avoid to completely avoid the more destructive heavy det.
Detonation doesn't need to be instantly destructive in order to do damage. Being able to stop it after a brief delay is better than nothing. I think that detonation provokes further detonation so if you can detect the onset early enough you may be avoid to completely avoid the more destructive heavy det.
Gassing Station | General TVR Stuff & Gossip | Top of Page | What's New | My Stuff