Alt-BEAM Archive

Message #05020



To: beam@sgiblab.sgi.com
From: Bob Shannon bshannon@tiac.net
Date: Sun, 04 Jul 1999 17:18:31 -0400
Subject: [alt-beam] Something funny with the 1382 voltage triggers?


I've just been playing with a SMT SE, and noticed something odd I've never
encountered
before.

These surface mount solar engines are built on 'SurfBoards', small commercial
SMT prototyping boards that resemble a Basic Stamp SIPs, but are blank.

I build several SE's with these boards, using 1382-J voltage triggers and real
McCoy SMT 3094's and 3906's. My capacitor is 6,600 uf, and I'm using a 3733
solar cell.

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.

I changed the resistors from 2.2K to 1.8K, with no effect.

Trouble shooting with a scope gave me a supprise or two. First, the SE is not
oscillating.
Secondly, the lock up condition happens at low cap voltage, rather than just
before the trigger
point.

The main cap voltage rises to the trigger point, a motor fires and the voltage
drops down, then never rises again. The cap does not charge until the light
level drops, a lot.

I have never seen this failure mode with 1381-J's, even using identical motors,
capacitors, and
solar cell. (in fact, I'm duplicating a photovore I gave away as a gift, only
using my SMT engine boards rather than the free formed SE's used in the
original)

Has anyone notices a difference when using SMT components rather than thru-hole
parts?

I'll search the FAQ again, but I dont recall ever reading about a fails to
charge in bright light problem before, or not one where the circuit works well
in modest lighting. I can only presume that one of the SE's is not shutting
down as it should.

I hate the thought that I might need to add a resistor between the solar cell
and main cap if thats whats going on here. Is there another solution? Or is
it possible that this is an internal resistance (ESR) problem with my drive
capacitors?

This particular photovore has larger caps than many, and higher powered, but
more efficient
motors (as compared to pager motors) beacues it needs to carry some extra
cargo. This extra cargo casts a shadow over the main drive 3733 solar cell, so
I was going to add a small booster cell (2224) on the tail, facing the rear.
This will provide additional current when the 'bot is facing away from the
light, and compenstate for the shadow on the main cell.

With this high light level lockup problem, I worry that this additional cell
will only make things worse. (I'll add them in parallel, with 1N5817
schottkey's to prevent back feed)

My next question is this, is there a way to control this effect? Can I get a
photopopper to stop
once its found a really bright spot to feed? (so the cargo systems may operate
till the light level drops off and its time to go hunting again)

Last question, what is the limit of a photopoppers ability to climb over
obstacles?

How many can roll over a dime, or quarter, or the thicker US nickel? How many
can roll
over a 1/8th watt resistor and other bench clutter? Ever seen a photovore turn
a 2N3906 into
road-kill on the bench?

This has a lot to do with wheel diameter and torque.

Pager motors are efficient, but have little useful torque. There are many
motors that are just
as efficient, or more efficient, but they need a bit more power than a pager
motor to produce far more torque (walkman motors are a good example of this).

I find that these slightly higher powered, but more efficient motors will drive
wheels up to
about a centimeter in diameter without gear reduction. The result is a
photovore that covers
more ground per minute, and a greater range of terrain as well.

(this explains the 6,600 uf cap and -J type triggers I'm using with salvaged
motors from laptop
CD-ROM drives)


------------------------------------------------------------------------

eGroups.com home: http://www.egroups.com/group/alt-beam
http://www.egroups.com - Simplifying group communications



Home