DDL - faster Nernst cell sensing

[bcroe at juno.com]

I guess you could just "hunt" for a .45 volt crossing
like an ordinary ECU does.  But you will still need to
consider the sensor delay in your software, to build
a stable closed loop.  You will not need anything
like 1 mhz; 15 khz is probably plenty.

I would think with a single crossing point detector
and no knowledge of error size, you would need 
to considerably degrade the sensor feedback 
loop response.  Probably just run a simple (and
slow) integrator.

If you came up with a "negative" pulse Ip generator,
you might be able to dispense with the virtual
ground.  The negative pulse could be generated 
with capacitive coupling from all positive circuitry
since it is just a fixed digital pulse.  Not so easy 
to do with a straight analog drive.

Bruce Roe

On Sat, 10 Nov 2001 11:47:44 +0800 (WST) Bernd Felsche
<bernie at innovative.iinet.net.au> writes:
> The following was published on the DDL mailing list about a week ago.
> 
> >From bernie Thu Nov  1 15:14:06 2001
> >Subject: DDL - faster Nernst cell sensing
> >Date: Thu, 1 Nov 2001 15:14:06 +0800 (WST)
> 
> [snip]
> 
> >One of the previous limits in DDL was the ADC delay for determining
> >the Nernst Cell voltage. It came to mind during that discussion 
> that
> >the AVR has a perfectly-adequate analogue comparator on-chip and
> >that this could be used to detect a swing from stoichiometric
> >(0.45V) almost instantaneously. The deviation from the nominal
> >voltage, and its direction could be used to determine the pump
> >current direction required to re-balance the Nernst Cell.
> 
> >Circuit complications arise in requiring that the nominal reference
> >voltage float above the cell's virtual ground as that pin has to
> >be raised above the "signal" level when measuring the cell's
> >resistance for temperature feedback (and hence heater control).
> >The virtual ground voltage also rises to pump ions "out"; the pump
> >and Nernst cells share one connection.
> 
> >At first these complications appeared onerous.
> 
> [snip]
> 
> See http://bernd.felsche.org/tech/EFI/DDL/DDL.html for the few
> circuit changes required to implement the fast switching detection,
> largely eliminating the bandwidth limitations imposed by ADC and
> subsequent numerical comparisons. Response bandwidth for the
> controller now approaches 1 MHz compared to the 15kHz or so limit of
> the ADC strategy for Nernst cell voltage detection.
> 
> Furthermore, the detection method lends itself to a simpler control
> strategy for applying pump pulses (e.g. bang-bang control relying on
> sensor hysteresis), as well as the ability to measure intrinsic
> sensor characteristics such as ion migration delays. Cell
> contamination and other age determinants can thus be quantified.
> 
> 
> Bernd Felsche 

----- End of forwarded message from owner-diy_efi at diy-efi.org -----
----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org