56508Re: Sensor types
- Aug 3, 2007Good example of the difficulty of reaching consensus on standards.
It is all a matter of defining the organization a complex system into
meaningful layers or subsystems.
In biological systems this is incredibly difficult just think of how
long it took to get to the current understanding of the organization
of biological functions of the human body and this knowledge is still
only the tip of the iceberg of how it actually functions.
Because BEAM systems are so simple it should be possible to do a
useful definition of the organization of subsystems on which many of
these simple autonomous electromechanical designs are based.
Flip flops, delays or memories should probably be considered part of
a deeper part of the control circuit. However you can broaden the
sensor subsystem definition to include these functions or define a
separate sensor processing or "signal conditioning" layer as distinct
from the motor control layer.
Perhaps a diagram that follows the signal flow will describe the
model I am using.
--- In firstname.lastname@example.org, "Mike Robertson" <mikerobe@...> wrote:
> how bout i start with temp sensor
> could we use a comparator with a volage divider to set the
referance on the negative and a lm335z on the positive
> or would a window compartor connected to a flip flop work
better...at least with the flip flop you can easily select if you
want a high or low output to trigger the bots function
> if that works then simply changing the sensor type and divider
should work for different detection types
> ----- Original Message -----
> From: wilf_nv
> To: email@example.com
> Sent: Wednesday, August 01, 2007 12:12 PM
> Subject: [beam] Re: Sensor types
> Good idea!
> To provide interchangable sensor modules you can normalize all
> outputs to proportional voltage outputs which can be compared and
> then generate an open or closed "switch" contact.
> For example, use a series load resistor or potentiometer to
> resistive sensors such as potentiometers, PDs, thermistors and
> to generate a voltage output.
> Next use an opamp to compare the sensor voltage output to a
> voltage or to another sensor's output voltage. When a sensor
> is above a threshold or is greater than the output voltage of a
> second sensor the comparator output goes high.
> Each opamp output signal controls two 4066 type analog gates.
> Finally, the analog gate I/O pins connect like simple switch
> to the rest of the robot control circuit to provide signal
> change time constants, discharge delay capacitors, multiplex
> signals (for reversersing) or implement wired AND/OR logic gates.
> These modules can be a complete subsystem that can be
> and plug interchangably into a protoboard or PCB.
> Here is a suggested form factor and connector layout to make it
> compatible with protoboard or a PCB. I assume standard through
> Each sensor module should be 1"x 1" with two IC sockets for
> a LM358 and a CD4066 plus 4 resistors. External connections are
> two 8 pin connectors made from single inline rows of IC machine
> sockets with 0.1 inch spacing between pins and positioned along
> top and bottom edges of the module with 0.9 inch spacing between
> The top 8 pin connector is used to connect two passive or active
> sensors and two threshold adjustment pots to the (+/-) inputs of
> dual opamps and also provides pins for Vcc and gnd. The bottom 8
> connector provides 4 independent 4066 switch contacts to be
> connecteted to the main part of the control circuit.
> This allows the module to be plugged into a proto board with the
> 8 pin machine sockets pins and the socket acting as female input
> connetors or as feedthroughs for external sensors leads and
> jumpers for the control output signals.
> The module can be raised up above the protoboard to provide
> the hidden contact strips underneath the module by stacking up
> another pair of 8 pin machine sockets on the the two rows of
> connector pins.
> The module circuit does not have to be a standalone daughter
> but can also be integrated into the main circuit protoboard or
> desired to interchangably accomodate a wide variety of sensors.
> A standardized sensor subsystem would require more discussion and
> consensus on the internal circuit and connector pin assigment. If
> common interface is agreed on, it may go some way to creating a
> of circuit modules that can be connected together to provide plug
> play capability.
> --- In firstname.lastname@example.org, "Mike Robertson" <mikerobe@> wrote:
> > i need some ideas/schematics for sensors for my new bot....i
> to have modular types that i can plug and swap to change the
> of the bot.
> > some i thought of so far are
> > hot/cold
> > light/dark
> > batt full/low
> > edge detect
> > tilt
> > proximity
> > and any other range by ir or emf
> > oh and naturally the default touch switches....but they're
> > i need them to have high/low outputs to interface with the rest
> the bot design
> > thanks Mike
> > i dont ask for much do i?? lol
> > ----- Original Message -----
> > From: Jesse
> > To: email@example.com
> > Sent: Tuesday, July 31, 2007 8:14 PM
> > Subject: [beam] Re: how to find capacity of solar cell?
> > hello Bill. (ever seen the movie kill bill? JK)
> > hooking up you meter in series is a way to read current, but
> > remember that the load determines the current draw, well up to
> max in
> > this case.(im an electrician and you arent dealing with live
> > here)
> > a way to calculate(someone correct me if need be)the current
> > is to use I = E/R, which is current equals voltage divided by
> > resistance. the current will be small, say in the miliamp
> > one, measure the resistance of the solar cell. Step two,
> > voltage. Step three, do some math: voltaged divided the
> resistance to
> > get the current. Step four, wish you had better solar cells
> > peace out
> > Jesse Meers
> > just build it!
> > --- In firstname.lastname@example.org, "Bill" <wjl82871@> wrote:
> > >
> > > Hello,
> > >
> > > I have a few little solar cells laying around. I know how to
> find the
> > > voltage of the cell, but how do I find the amp output? Do I
> need to
> > > make some sort of load and then put the VMM in series with
> > circuit?
> > >
> > > Thank you for your help and time!
> > >
- << Previous post in topic