Subject:
|
Re: **** the Scout ****
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Mon, 15 Nov 1999 21:41:09 GMT
|
Viewed:
|
573 times
|
| |
|
|
In article <01BF2F7F.0B14FC60@northmill1>, Jeffrey Hazen
<jeffrey.hazen@northmill.net> wrote:
> The Scout will work with the standard light sensor.
Perhaps I should've clarified my initial statement about passive vs.
active sensors. According to TLG the Scout works with any passive sensor
and has an internal ADC to measure analog voltages. Presumably there is
an internal pullup resistor so that such a scheme can be used to simply
measure an external resistance. This is pretty much what the RCX does in
"passive" mode.
In active mode, the RCX (and the Control Lab) uses the same line to both
supply power to a sensor and read a voltage back from it. TLG has not
indicated that the Scout supports this mode.
The standard light sensor is an active sensor. Of course you can still
hook it up to a passive port, but you won't get 100% functionality. For
example, the brightness of the LED is going to be proportional to how much
current can be pulled through the pullup resistor. I doubt that there is
enough current to make reliable sensor readings in the absense of external
light. I tried it myself and the LED didn't even turn on (I suspect a
reasonable amount of variation between Scouts and light sensors).
The "sensor" portion of the light sensor (essentially a photo-transistor)
will still be operational. A bright enough light aimed into it should
cause a modest current to flow through the sensor - making its effective
"resistance" lower than when the light isn't present. This in turn is
interpreted by the RCX as "triggering" the sensor. Of course the "active"
part of the sensor is always pulling on the Scout, thus reducing the
dynamic range of the sensor.
> The Scout inputs float at 5 volts (presumably an internal pull up
> resistor), and when a touch sensor is depressed, drop to about 0.30
> volts....pretty standard sinking input logic. With the light sensor
> attached, the voltage drops to about 2.5v (and the led is barely lit) and
> about the same when the light sensor detects. I don't know how long the
> input will be happy with a light sensor attached, the current sourcing from
> a input like this has to be small.
>
> What's interesting (and beyond my electronics knowledge) is if you try to
> measure the resistance of the input, with the light sensor attached, you
> can trigger the input with the DVM.
I assume you had the DVM set to measure resistance in this case. In order
to measure resistance, a DVM typically supplies a known current (or
voltage) across its leads and then samples the resulting voltage (or
current). This is quite invasive to an operating circuit - the DVM is
essentially an extra current (or voltage) source. Depending on the actual
connections involved it is very conceivable that such interference would
cause the Scout to believe the sensor was triggered.
You may want to repeat the experiment but set the DVM to measure voltage
rather than resistance. In this mode the DVM is typically very high
impedance so it should have minimal effect when connected in parallel to
an operating circuit. What you should then see is the voltage level that
the Scout itself is sampling.
>
> I suspect there is more to the Scout sensors to be discovered.
I did a couple of tests with a known resistance across the sensor and
reading the voltage (thus computing the current and a hypothetical pullup
resistance):
R V I (mA) pullup (K)
open 5.0 - -
10.1K 2.44 .242 10.6
2.17K 0.88 .406 10.1
The variation is well within what I'd expect from a quick experiment like
this, and indicates a 10K pullup to 5.0V within the Scout.
Going from these numbers its hard to supply much power from the Scout.
Lets say you only need 0.7V (typical forward voltage of an LED). Then you
can drop 4.3 volts across the pullup, for a max current of 0.43mA. That's
not much.
For an extra light sensor, I'd suggest going with a completely passive
approach. One really cheap way is to use a CdS cell, which is essentially
a resistor that varries depending on incident light.
Dave Baum
--
reply to: dbaum at enteract dot com
|
|
Message has 1 Reply:
 | | Re: **** the Scout ****
|
| (...) I think that the CdS the best way to go also. I have the LEGO light sensor circuit on a page: (URL) . It is hard for me to believe that the LEGO light sensor works well enough to be useful in passive mode. I think people were looking at this (...) (25 years ago, 15-Nov-99, to lugnet.robotics)
|
Message is in Reply To:
| | RE: **** the Scout ****
|
| The Scout will work with the standard light sensor. The Scout inputs float at 5 volts (presumably an internal pull up resistor), and when a touch sensor is depressed, drop to about 0.30 volts....pretty standard sinking input logic. With the light (...) (25 years ago, 15-Nov-99, to lugnet.robotics)
|
3 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
