ion sensing? anyone here have a clue?

[dannyho]

I didn't see any talk over here about this, so I realize it might be a long shot, but here goes....
over on the wiki there is a good deal of talk about getting ion sensing to be incorporated into VEMS. It's a great idea, but there seems to be a consensus that you need a CDI ignition to do it, and an inductive ignition wont work.  Simply put I'd like someone to explain to me why an inductive ignition wouldn't work. Couldn't you use some fancy high voltage diodes to isolate the coil from the sensor and vice versa? 

http://www.vems.hu/wiki/index.php?page=IonSense

follow up:  has anyone actually gotten it to work and visually see where your peak cylinder pressures are, whether through VEMS or through a oscilloscope or anything else?

Thanks
Dan

[rob@vems.co.uk]

I know that theres work afoot for the next generation of hardware that will have the processing power to do the calculations and monitoring of the back EMF caused by detonation.

http://www.vems.hu/wiki/index.php?page=MembersPage%2FDamirMuha%2FIonsenzor

[MWfire]

I made ion sensor. I'am still testnig(very slow, becouse i don't have garage and it's cold).
I use 3 12kV diodes per cilinder, 1 bobbin and distributor.
On first version it's easy to see ppp(15-16 degree is in my motor).
Now i add some filter(high pass 4kHz for knock, low pass 1kHz for ppp and misfire) and amplifier. Now the signal should be "cleaner".

[dnb]

Sounds an interesting development.  Would you mind sharing your circuitry?  I'm quite keen to test this on my TVR now that I have a reasonable amount of test equipment.

[dannyho]

MWfire, I'd also like to see what your circuit looks like.  Did you use an inductive ignition or does your car have capacitive discharge ignition?  lastly, have you been able to get any output on a scope, basically what's been your methodology so far?

[MWfire]

http://www.vems.hu/wiki/index.php?page=MembersPage%2FDamirMuha%2FIonsenzor
For switcher(100-400) i don't have schematic. Only have pcb. I can add values and share.

This is pcb side(very sloppy)
Principle is like this
http://vvnet.fi/ville/ion/DIY-Ion-Sensing-2.pdf
I have inductive bobine. So far i made about 20 pictures of scope, like this

PPP is 2.3ms after spark( becouse there is no filter and amplifier it is hard to see).

ppp is 2.7ms after spark
I wait for warmer time to try measurment with filters and amplifier. I will post my results.
filter:


BTW how looks my car when i measured

[dnb]

That all looks do-able if I could only find those 22kv Vr diodes that Farnell discontinued...

[MWfire]

In idle, i measurment(i have 1:6000 sonde) only 2.5kV spark(0.7mm gap). You can use 20 1kV diodes(look for diodes with small capacitance).
http://www.voltagemultipliers.com/start_netscape.htm has 20kV diodes with small capacitance(1pF) for 4-5$.

BTW if your bobine have two secundar output you can use this method
http://delphi.com/manufacturers/auto/powertrain/gas/ignsys/ionized/

[dnb]

I think I may be able to use the Delphi approach - All my coils have 4 terminals - HT (secondary), ground for secondary, 12v for primary and ECU trigger (primary).

I might do all 8 coils if it's going to be easy

[GintsK]

BMW COP coils have 4a pin. It is other end of secondary.