[Diy_efi] EGT detonation sensor (was Re: Timing Advance

[Mike]

Of course, I wouldnt select a slow response time, one would
need to be mindful of transient response...

good point,

rgds

mike


At 12:03 PM 12/20/02 -0800, you wrote:
>> A poor mans detonation sensor for each cylinder could be as
>> simple as an EGT sensor correctly positioned without the
>> mechanical noise issue prevalent with peizo sensors,
>> the cost of pyrometers with associated electronics could
>> well be less than the highest spec piezo sensors and
>> simultaneuously provide exhaust gas temperature - nice :-)
>
>I'm not sure how to read this, but if you're thinking of using
>just one EGT sensor for several cylinders, I don't see how, given
>typical EGT probe response times of about a second. Even if
>you meant 1 probe per cylinder, you still may be "asking for
>trouble" depending...
>
>Consider the hypothetical situation of a step change in temp from
>0 to 1000*. You might think that if the response time is 1 second,
>in 1 second you'd read 1000*, but that's not the case. The
>definition for one "tao" is 63.2% of the temp difference, so
>after 1 second you'd only read 632*, the next second would be
>63% of the difference (1000 - 632), so after 2 seconds you'd read
>632 + 232 = 864*. After 3 tao's you'd be within 10% (actually 5%)
>at 950*, but it'd take 5 seconds to get a reading accurate within
>1% at 993*.
>
>Folks familiar w/ resistor-capacitor networks will recognize the
>curve immediately.
>
>Closer to "real world", a step change from 900 to 1000*,
>immediately you'd be within 10%, after only 1 tao you'd be within
>5% at 963*, in 3 tao's you'd be within 1%.
>
>So we can see that the bigger the delta in temp, the bigger the
>error and the longer you have to wait to get an accurate reading.
>I don't have typical temp deltas for detonation, they s/b large
>and easy to detect but even if you detected knock within 1 second,
>that'd still be 50 "hammer" strokes per cylinder for an engine
>spinning 6K RPM.
>
>My point of all this is that there may be more to getting good
>readings from EGT thermocouples than folks may have thought of,
>a lot depends on "where you've been" and "where you're going",
>absolutes, deltas and taos. ;)
>
>Personally, ion-sensing is still the way to go IMO.
>
>HTH's
>philh
>(digest)
>
>_______________________________________________
>Diy_efi mailing list
>Diy_efi at diy-efi.org
>http://www.diy-efi.org/mailman/listinfo/diy_efi
>
>

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi