Subject:
|
Re: Sensor Reliability
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Mon, 7 Feb 2000 08:47:56 GMT
|
Original-From:
|
John Barnes <BARNES@SENSORSnomorespam.COM>
|
Viewed:
|
1296 times
|
| |
|
|
I have also noted these effects and I don't think they
are due to the wire length. There are two things doing this;
1. For active sensors like the light sensor, "Lego" don't regulate
the power to the LED. So it's output amd thus the overall
response is sensitive to battery voltage. Tut tut!
2. For sensors in general, they apply the internally regulated
5 volt supply to the sensor via a 10k internal resistor, then
measure the resulting voltage. The 10k resistor is not better
than 2% accurate, so readings from temperature sensors will not
improve on that figure.
As far as "precision" is concerned, don't confuse resolution
with accuracy. Even if it isn't accurate to 0.1 degree, it can be
handy to know something changed by that amount.
Measure carefully!
JB
> In lugnet.robotics, Dave Baum writes:
> [regarding the accuracy of the LEGO temperature sensor]
> > Not very accurate - perhaps within 2 or 3 degrees F. The display uses 0.1
> > degrees precision (either C or F), but the accuracy just isn't there.
>
> My recent experience with the rotation sensor led me to two conclusions: the
> RAW values measured by the RCX are significantly affected by both battery
> voltage and the length of the connecting wire.
>
> I suspect that the *precision* (repeatability) of the inputs is very high as
> long as you control both of those variables. If you need accurate absolute
> temperature readings you'll need to calibrate with any extension wires you • plan
> to use and use the RCX on an A/C power supply. It's possible (even likely)
> that the default calibration assumes a temperature sensor with no extension
> wires (since it has one built in) and a wall power supply.
>
> BTW, was the incubator monitor that Steve Moffat was talking about ever • built?
>
> --Ben
>
|
|
Message has 1 Reply:
 | | Re: Sensor Reliability
|
| John Barnes <lego-robotics@crynwr.com> wrote in message news:3.0.32.20000207...lserver... (...) The RCX power signal is not regulated but LEGO has a current regulator that is shared between the phototransistor and the LED in the light sensor. It's (...) (25 years ago, 7-Feb-00, to lugnet.robotics)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
