Alt-BEAM Archive

Message #05041



To: "'beam@sgiblab.sgi.com'" beam@sgiblab.sgi.com
From: Wilf Rigter Wilf.Rigter@powertech.bc.ca
Date: Mon, 5 Jul 1999 23:00:46 -0700
Subject: [alt-beam] Re: Something funny with the 1382 voltage triggers?



content-type: text/plain;

HellO Bob,

I doubt that SMD parts are the problem except that the longer lead length of
TO92 packages in freeforming may make the circuit more sensitive to EMI or a
small amount of feedback which causes some small oscillations when the latch
is on the edge of the linear operating region and which can help initiate
the latch turn off. I suggested the increased resistor value to reduce the
"gain" of the latch decreasing saturation and increasing the sensitivity to
feedback. The side effect of this lower gain is potentially more (not less)
oscillation during turn on with some motor loads because the 2N3904 (NPN)
may not fully saturate. With an efficient (higher resistance) motor, a 4.7K
base resistor for the 2N3906 (PNP) is enough to saturate the NPN during turn
on but ought to cause the NPN to come out of saturation earlier to turn the
latch off at the end of discharge. A 5K pot for the PNP base resistor
allows fine tuning. However all that PNP base resistor fine tuning may not
be the best solution since it optimizes the circuit for a narrow operating
condition and does not solve the more fundamental problem of the 0.7V drop
out voltage.

As Steven Bolt points out, the voltage on the cap has to drop below 0.7V for
the 2 transistor SE to reset. Ouch! That wastes a lot of energy since the
motor probably stalls before that point and the dumps the rest of the
capacitor charge without doing any work. Looking at the capacitor voltage in
a bit more detail, it's not the voltage on the capacitor but the voltage
across the PNP base resistor that determines when the latch resets. When the
voltage on the capacitor drops to " 0.7V", we are really looking at the
effect of that voltage on the PNP and NPN base currents. At that capacitor
voltage, the voltage across the PNP base resistor drops to some value around
100mV and depending on the gain of the transistors and motor resistance, the
base currents are no longer sufficient, the NPN comes out of saturation and
a little feedback causes the latch to reset.

While the 1381 SE will trigger reliably at 2.7V, the turnoff voltage is
dependent the characteristics of the solar cell, light level, the motor and
the motor load. As shown in the attached, adding a LED and 1K resistor
across the motor gives a visual indication when the 1381 fires and since the
LED is also in series with the (now) 1K PNP base resistor, the voltage drop
across the LED will raise the minimum capacitor voltage required to keep the
latch on to about 1.7V Since the cap no longer discharges to a low voltage,
the motor continues to do useful work until the latch turns off and the cap
recharges more quickly. While this may not be a BEST solution, it's simple
and worth a try.

regards

Wilf Rigter mailto:wilf.rigter@powertech.bc.ca


<<1381ledSE.gif>>


> -----Original Message-----
> From: Bob Shannon [SMTP:bshannon@tiac.net]
> Sent: Monday, July 05, 1999 6:27 AM
> To: Bumper314@aol.com; beam@sgiblab.sgi.com
> Subject: Re: Something funny with the 1382 voltage triggers?
>
>
> Bumper314@aol.com wrote:
>
> > > Has anyone notices a difference when using SMT components rather than
> thru-
> > > hole
> > > parts?
> >
> > yes...I have had this problem, twice, I have made 2 robots with SMD
> parts,
> > and I have another SE from SMD's in my to-do tray, but I have the same
> > problem with BOTH SEs. I dont know why, i corrected this by putting a 5k
> > resistor in series with the motor/coil, and it worked fine, but i know
> this
> > really lowers your power output, so you might just try putting a small
> value
> > in series with the solar cell, i found that if you have it out side and
> its
> > locked up, and you cover the cell for a second and uncover..it will
> fire, so
> > try those and tell me what you get...i am anxious to see what others
> say.
>
> Your exactly correct, if I cover the cell, and wait a long while then the
> SE will
> begin to cycle again.
>
> You say that you have had this problem with 2 out of 2 SMT SE's? Thats
> very
> discouraging. Have you tried the same cells, cap and motor on a 'normal'
> SE?
>
> Adding a resistor in series with the motors is a problem for me, I'm using
> these
> SE's to drive a photovore thats got to carry around some extra cells and
> circuits,
> so
> I really need the power to the motors during the discharge cycle.
>
> Wilf suggested either raising the resistor levels beyond 2.2k, to 4.7K.
>
> I've often found (with these same motors and standard parts) that the SE's
> usually
> work
> better with a bit less than 2.2K, and I began using 1.8K. Now it seems
> that I've
> got to
> go the other way, and raise this value.
>
> I'm thinking that rather than place a resistor in series with your motors,
> you
> might do better
> to place a resistor in series with the solar cell. Charge time will go
> up, but
> you will still
> have full motor current when it fires.
>
> But I'm not at all sure just how sensitive this problem is to changes in
> the
> charge current.
>
> I'm thinking my next SE's will be PIC's, which dont have these problems at
> all!


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

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





Attachment: 1381ledSE.gif

Home