Alt-BEAM Archive
Message #05027
To: beam@sgiblab.sgi.com
From: Sean Rigter rigter@cafe.net
Date: Sun, 04 Jul 1999 20:40:05 -0700
Subject: [alt-beam] Re: UPDATED Power Smart Head page
--------------D3ECAEB13B580F9C7B661E98
content-transfer-encoding: 7bit
content-type: text/plain; charset=us-ascii
Hi Chiu,
you wrote:
> BTW. Don't know if you realize but the caps shown in the pictures are .001uF
> whereas the schematic calls for .01uF. Just something to make note of.
In this circuit, as long as the caps are the same value, the absolute
value does not matter much since it only affects the dead band and/or
overshoot between forward and reverse. Since this is a function of the
total RC of the circuit, it is also dependent on the effective R of the
PDs or LDRs. Smaller cap values are usually good for smooth operation
but try experiments with larger values which give a kind of lifelike
"double take" effect and in dark conditions may change the low power
"slow scanning" speed.
>A great description of the HLO circuit (pronounced "hello") can be found at:
>http://www.geocities.com/SouthBeach/6897/rigter/wr_3.html#mar1
Like most BEAM circuits adding just a few component changes can
radically change the circuit behaviour. The HLO (the High Low
Oscillator) circuit, like the slave Bicore, depends entirely on the
external input voltage to oscillate. The PowerSmart input circuit, like
the embedded Bicore, is an inherent oscillator which can be gated High
or Low but unlike the HLO oscillates if the external input voltage
source is removed. In addition the "Nv/Nu" output driver circuit
produces a full differential voltage across the motor when the input
stage is not oscillating and a PWM voltage across the motor when the
input stage is oscillating with a larger time constant than the Nv/Nu
output stage (low light = low power scanning). When locked on to a
"brighter" light or reflecting object, the input stage oscillates with a
time constant smaller than the Nv/Nu output stage and the differential
voltage across the motor drops to 0V. In the latter case, each motor
terminal has a square wave applied to it, but they while they are in
phase, no current flows through the motor.
The "Nv/Nu" circuit used in PowerSmart Head output is itself simply
fascinating! It consists of a single RC section but instead of the
conventional grounded R (Nv) or grounded C (Nu) this circuit has both
legs driven in anti phase (with an inverter) to combine both functions.
The two output buffers can be inverting or non inverting. The circuit
provides an output voltage duty cycle across the motor terminals which
is proportional to the ratio of the time constants of the driving
frequency (TC1) and the time constant (TC2) of the Nv/Nu circuit. At
logic 1 or 0 (TC1>>>>TC2) the output is simply forward or reverse. At
low input frequency (TC1>>TC2), the motor reciprocates forward and
reverse but with a dead band in between changing directions. At medium
input frequencies (TC1>TC2) the average motor terminal voltage is pulse
width modulated with an average voltage proportional to the duty cycle
(reduced power) and at still higher frequencies (TC1<=TC2), the output
duty cycle approaches zero and no motor current flows (minimum power). I
have attached a schematic of this Nu/Nv circuit with (hand
drawn/approximate) waveforms. Although not apparent from the waveforms,
the horizontal time base is relative and waveforms are expanded relative
to each other to a fixed length for clarity.
Like Mark T's famous XOR motor driver, it needs a sound byte name to
describe its function so I have tentatively called this Nu/Nv circuit
the PID neuron. A good name has the potential to evoke creative new
ideas and applications for a circuit. Conversely typecasting the
behaviour of a Nv or Nu neuron can seriously limit the ability to
understand the behaviour and discover new applications for these
simplest of circuits. So I think the PID tag will soon be dropped! Any
suggestions for a good name for this neuron will be greatly appreciated.
enjoy
wilf
------------------------------------------------------------------------
eGroups.com home: http://www.egroups.com/group/alt-beam
http://www.egroups.com
- Simplifying group communications
--------------D3ECAEB13B580F9C7B661E98
Attachment: PID.gif
--------------D3ECAEB13B580F9C7B661E98--
5028 Mon, 5 Jul 1999 00:13:46 EDT [alt-beam] Re: Something funny with the 1382 voltage triggers? beam@sgiblab.sgi.com JVernonM@aol.com In a message dated 7/4/99 5:34:38 PM Eastern Daylight Time, bshannon@tiac.net
writes:
> The problem is lock-up under full sun. The problem does not happen with
> indoor
> lighting
> unless the photovore is held abut 5 cm from a full spectrum desklamp.
>
Hey Bob,
I think I had a similar problem with Spinner. It uses a CD drive motor as
well. It is the long, flat on two sides motor. When hooked up to a large
cell, it would lock up if put in the sun or close to a halogen light source.
It seemed to be something inside the motor itself. I built Spinner with it,
which uses The smallest Trigger (I don't remember the letter) and two
calculator cells. The lower voltage works every time. I don't know if you are
experiencing the same problem, but it sounds similar.
Jim
------------------------------------------------------------------------
eGroups.com home: http://www.egroups.com/group/alt-beam
http://www.egroups.com
- Simplifying group communications
Home