[Diy_efi] Temperature/Pressure sensors

Alex Garbino agarbino
Mon Nov 13 04:21:46 UTC 2006


That is excellent information! Thank you, it was exactly what I needed.
Although the LM335 is relatively-speaking a lot cheaper, the absolute costs
are still below the ~$20 [personal] upper limit on a sensor. They may be a
good option for a more sensor-heavy application; but these automotive
versions are certain to work in the environment, and I only need 4-5 sensors
total. Their limitation to 100C is also too low for me to use them on a
turbocharged car without worrying about cooking them...
Thanks,
Alex


On 11/10/06, Torbj?rn Forsman <torbjorn.forsman at gengas.nu> wrote:
>
> The LM335 and most other IC temperature sensors (there are also some
> with digital interface, for example I2C) have a reputation for bad
> reliability in the field, especially in severe environments as in
> vehicle engine compartments, heavy industry etc. There are good reasons
> why all mass produced engine management systems still use NTC or nickel
> resistance (PTC) temperature sensors instead.
>
>
>
> For the most common NTC sensors, there are just three basic types to
> take care of.
>
> The "Bosch" temperature sensor, also used by many other japanese and
> european manufacturers, has a characteristic which is to be found here:
> http://www.epcos.com/inf/50/db/ntc_06/LeadlessDisks__B57220__K220.pdf
>
> GM's standard sensor, used for many years on their C3 and P4 systems has
> this characteristic:
> http://www.epcos.com/inf/50/db/ntc_06/Leaded__NTC__B57962__S962.pdf
>
> Ford uses NTC temperature sensors with a resistance about 33 kohm at +25
> ?C. I haven't found any good documentation on those.
>
> Characteristics for nickel PTC sensors (seldomly found, mostly on french
> and japanese systems) can be found here:
> http://www.oiml.org/publications/R/R084-e03.pdf
>
>
> When calculating the temperature - voltage graph, one must know the
> resistance of the pull-up resistor in the ECU as well as the supply
> voltage (almost always 5 V) and if there is any resistor in parallel to
> the sensor. This has to be done by reverse engineering the ECUs.
>
> Bosch typically uses either 1,0 kohm, 2,7 kohm or 2,87 kohm pull-up in
> their Motronic and (digital) Jetronic ECUs. Ignition ECUs (EZ) as well
> as the analog L-jetronic, LE-, LU- and KE- ECUs have completely
> unstandardized pull-up resistors. In some ECUs with 2,7 or 2,87 kohm
> pull-up there may be a parallel resistor.
>
> Japanse ECUs typically have 2,7 kohm pull-up, no parallel resistor.
>
> GM uses 1,0 kohm or 348 ohms pull-up, no parallel resistor. In some
> cases, there are two different pull-up resistors (one 348 ohms and one
> in the 3 kohm range) that can be switched under software control to
> increase the measurement range.
>
> Best regards
>
> Torbj?rn Forsman
>
>
> Carter Shore wrote:
> > Check out National Semiconductor,
> > LM335 yields 10mV/degree K.
> > Cheap, tiny, easy to interface.
> >
> > Carter
> >
> > --- Alex Garbino <agarbino at gmail.com> wrote:
> >
> >
> >>Hello,
> >>
> >>I'm trying to build a data acquisition system to
> >>monitor various
> >>temperatures and pressures. I've been looking at
> >>thermocouples, RTDs, and
> >>thermistors; but they have widely different
> >>advatnages and disadvantages. I
> >>was thinking of using thermistors to monitor air
> >>intake temperature &
> >>pressure (on a turbocharged car, so I would like to
> >>monitor atmospheric,
> >>pre-intercooler, and post-intercooler/manifold) as
> >>well as coolant temp, oil
> >>pressure & temp, and EGTs, both before and after the
> >>turbocharger.
> >>Thermistors seem the best choice for air intake,
> >>since they have a fast
> >>response and are cheap; however, I cannot find good
> >>datasheets regarding the
> >>linearization coefficients.
> >>
> >>Does anyone have advice on a good way to capture
> >>data, as well as any tips
> >>on the best way to mount these sensors (a modular
> >>system would be best, so I
> >>could move sensors to different positions and switch
> >>them in and out with
> >>minimum hassle)? Also, can anyone recommend a good
> >>source of cheap sensors?
> >>
> >>The signals will be captured via a PIC18F series'
> >>A/D converters/digital
> >>inputs, and relayed immediately onto an on-board
> >>computer. Please let me
> >>know of any resources that would be helpful!
> >>
> >>Regards,
> >>Alex
> >>
> >>For a more detailed discussion on what I'm trying to
> >>achieve, please see
> >>http://www.mkiiitech.com/forums/viewtopic.php?t=903
> >>_______________________________________________
> >>Diy_efi mailing list
> >>Diy_efi at diy-efi.org
> >>Subscribe:
> >>http://lists.diy-efi.org/mailman/listinfo/diy_efi
> >>Main WWW page:  http://www.diy-efi.org/diy_efi
> >>
> >
> >
> >
> >
> >
> >
> ____________________________________________________________________________________
> > Want to start your own business? Learn how on Yahoo! Small Business
> > (http://smallbusiness.yahoo.com)
> >
> > _______________________________________________
> > Diy_efi mailing list
> > Diy_efi at diy-efi.org
> > Subscribe: http://lists.diy-efi.org/mailman/listinfo/diy_efi
> > Main WWW page:  http://www.diy-efi.org/diy_efi
> >
>
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> Subscribe: http://lists.diy-efi.org/mailman/listinfo/diy_efi
> Main WWW page:  http://www.diy-efi.org/diy_efi
>




More information about the Diy_efi mailing list