You're welcome. I just hope we can get it all working, its so close......
Barry

 

You could try something like this:



When the knocklite pulls down the base of the PNP, it passes its current through to the NPN's base and charges the capacitor. While the NPN is active, it drains the pullup (the MS gets the low/knock signal) and is kept open for a short time by the cap even after the knocklite returns to a positive voltage (allowing the MS to grab the signal)

The graph on the right is the base of the PNP getting pulled down, the graph on the left is the output the MS would see.

 

Thanks Jeff appreciate that, could something like this be built in the proto area of the diypnp? I will try to get someone here in the UK to help me build it (give me something mechanical and I'm OK!).
Barry

 

Absolutely.

You might have to play with the values a bit to get it to hold down the signal for the correct amount of time. I used a 4.7k pullup on the MS, a 100uf cap, and a 100 ohm between the cap and ground.

 

Brilliant, knowing the knocklite offers 2 warnings, amber for light knock and red for severe knock I suppose the information I now need from TurboXS is:

What is the 5v - 0v pull down time for both warnings.

We would also need to know the minimum 5v - 0v pull down time MS needs to accept the drop in voltage.

If we know these 2 things then I assume the circuit could be built to allow 2 different timing pull rates, say 5 degrees for amber and 10 degrees for severe?

Does that make sense?
Barry

 

Does anyone know the minimum time that MS2 recognises the 5v - 0v pull down in the knock circuit to activate timing retard?

 

Not a clue, but it's definitely worth asking on MSExtra, those guys must know for sure!

 

A good safe minimum time would be 10 to 15 ms.

 

Thanks Matt. Would you know what causes the MS to pull the timing in increments, ie 3,6 or 10 degrees, is it the time delay between 5v-0v-5v or is it dependant on the level of the voltage drop? I assume it is the time that there is zero volts?
Barry

 

It keeps backing the timing off until it stops seeing knock or hits the maximum knock retard. See this link for details.

http://www.msextra.com/doc/ms2extra/...ware.htm#knock

 

Thank you Matt....

 

Better idea:

Invert the output from a 555 timer with a transistor and a pullup:


1k resistor and 10uF cap should net 10ms output.

 

Brilliant Jeff!! I will get the parts and build this on a circuit board outside of the MS to confirm it all works then I can build it in the proto area. Thanks for your help bud..
Barry

 

It Works, see https://www.miataturbo.net/showthrea...583#post826583

 

Do you guys recommend using megasquirts automatic knock retard timing feature? Retards 2 degrees at first knock then 1 degree for each additional knock

Thanks