Alt-BEAM Archive
Message #08916
To: "'beam@sgiblab.sgi.com'" beam@sgiblab.sgi.com
From: Wilf Rigter Wilf.Rigter@powertech.bc.ca
Date: Sat, 8 Jan 2000 09:35:40 -0800
Subject: [alt-beam] Re: Yo, Team!
Happy new year to you John (and the best to your alter egos),
(a) With rare exceptions, the data books suggest opamps and comparators are
not suitable for driving motors directly but remember this applies to HC
logic specs as well so still worth an experiment or two. The best "garden
variety" candidate is the LM324 quad opamp which has no current limiting for
current sinking and the output swing includes ground. This means you can
drive a small (30ma) motor connected between +5V and the opamp output. A
reverse diode across the motor will protect the output from back EMF. The
output voltage drop for current source with the load connected to ground is
too high, making it inefficient and likely to overheat. So a bridge circuit
with the motor between two outputs is probably not practical. The newer CMOS
rail to tail output opamps are suitable for driving small motors in a bridge
configuration. I will check the specs for a recommended part number and what
limits apply. Power audio amps such as the LM386 have been used in motor
driver designs. The CD/VCR eject motor driver chips (with those difficult
Japanese part numbers) are a large as yet untapped resource for future Beam
designs.
(a)(1) You can drive the standard h-bridge if the (CMOS) opamp outputs
driving the h-bridge swing rail to rail.
(b) You can drive an h-bridge with a LM339 open collector output with a
suitable load resistor to Vcc. You can probably forget about driving the
motor directly from the 339 comparator outputs since it has poor current
sinking capability. There are (rare) power comparators that are suitable
though.
(c) The newer CMOS opamps can do practically everything and are "extreme"
power misers to boot. (check Ben Hitchcock's Whistle Blower application for
a 1uA oscillator at 9V which can't be done with any digital CMOS circuit I'm
aware of)
regards
wilf
> -----Original Message-----
> From: John A. deVries II [SMTP:aubois@trail.com]
> Sent: Friday, January 07, 2000 11:55 PM
> To: beam@sgiblab.sgi.com
> Subject: Yo, Team!
>
> A couple-three questions:
>
> (a) can you drive a "typical" BEAM motor from the output of a "garden
> variety" 741?
> (a)(1) presuming you had an "unlimited" power source, using a couple of
> 741's to drive a "standard" h-bridge would likely end in fried parts under
> normal operating conditions (whatever the heck they are) of a BEAM device,
> right?
>
> (b) would it be better to use a comparator, something like a 339 instead
> (and you get four in one chip..)? In this case I'm supposing direct drive
> of the motor with the comparators and not even considering an h-bridge.
>
> (c) I'm not ignoring the '14s and other digital stuff but, sheesh, isn't
> there any decent more-or-less linear analog stuff that isn't too
> gosh-awful
>
>
> Anybody have any thoughts?
>
>
> Zoz
>
>
>
> p.s. Happy Bloody New Year, Y'all!
>
8917 Sat, 8 Jan 2000 12:57:21 -0500 [alt-beam] Fw: Yo, Team! "beam" "tom barker"
----------
> From: Wilf Rigter
> To: 'beam@sgiblab.sgi.com'
> Subject: RE: Yo, Team!
> Date: Saturday, January 08, 2000 12:35 PM
>
> Happy new year to you John (and the best to your alter egos),
>
> (a) With rare exceptions, the data books suggest opamps and comparators
are
> not suitable for driving motors directly but remember this applies to HC
> logic specs as well so still worth an experiment or two. The best
"garden
> variety" candidate is the LM324 quad opamp which has no current limiting
for
> current sinking and the output swing includes ground. This means you can
> drive a small (30ma) motor connected between +5V and the opamp output. A
> reverse diode across the motor will protect the output from back EMF. The
> output voltage drop for current source with the load connected to ground
is
> too high, making it inefficient and likely to overheat. So a bridge
circuit
> with the motor between two outputs is probably not practical. The newer
CMOS
> rail to tail output opamps are suitable for driving small motors in a
bridge
> configuration. I will check the specs for a recommended part number and
what
> limits apply. Power audio amps such as the LM386 have been used in motor
> driver designs. The CD/VCR eject motor driver chips (with those difficult
> Japanese part numbers) are a large as yet untapped resource for future
Beam
> designs.
>
> (a)(1) You can drive the standard h-bridge if the (CMOS) opamp outputs
> driving the h-bridge swing rail to rail.
>
> (b) You can drive an h-bridge with a LM339 open collector output with a
> suitable load resistor to Vcc. You can probably forget about driving the
> motor directly from the 339 comparator outputs since it has poor current
> sinking capability. There are (rare) power comparators that are suitable
> though.
>
> (c) The newer CMOS opamps can do practically everything and are "extreme"
> power misers to boot. (check Ben Hitchcock's Whistle Blower application
for
> a 1uA oscillator at 9V which can't be done with any digital CMOS circuit
I'm
> aware of)
>
> regards
>
> wilf
>
> > -----Original Message-----
> > From: John A. deVries II [SMTP:aubois@trail.com]
> > Sent: Friday, January 07, 2000 11:55 PM
> > To: beam@sgiblab.sgi.com
> > Subject: Yo, Team!
> >
> > A couple-three questions:
> >
> > (a) can you drive a "typical" BEAM motor from the output of a "garden
> > variety" 741?
> > (a)(1) presuming you had an "unlimited" power source, using a couple
of
> > 741's to drive a "standard" h-bridge would likely end in fried parts
under
> > normal operating conditions (whatever the heck they are) of a BEAM
device,
> > right?
> >
> > (b) would it be better to use a comparator, something like a 339
instead
> > (and you get four in one chip..)? In this case I'm supposing direct
drive
> > of the motor with the comparators and not even considering an h-bridge.
> >
> > (c) I'm not ignoring the '14s and other digital stuff but, sheesh,
isn't
> > there any decent more-or-less linear analog stuff that isn't too
> > gosh-awful
> >
> >
> > Anybody have any thoughts?
> >
> >
> > Zoz
> >
> >
> >
> > p.s. Happy Bloody New Year, Y'all!
> >
8918 Sat, 8 Jan 2000 10:16:29 -0800 [alt-beam] Re: Yo, Team! "'beam@sgiblab.sgi.com'" Wilf Rigter Hi tom,
I did not get the text in the body of your message. Can you please
re-send!
best regards
wilf
> -----Original Message-----
> From: tom barker [SMTP:crowbar@monad.net]
> Sent: Saturday, January 08, 2000 9:57 AM
> To: beam
> Subject: Fw: Yo, Team!
>
>
> ----------
> > From: Wilf Rigter
> > To: 'beam@sgiblab.sgi.com'
> > Subject: RE: Yo, Team!
> > Date: Saturday, January 08, 2000 12:35 PM
> >
> > Happy new year to you John (and the best to your alter egos),
> >
> > (a) With rare exceptions, the data books suggest opamps and comparators
> are
> > not suitable for driving motors directly but remember this applies to HC
> > logic specs as well so still worth an experiment or two. The best
> "garden
> > variety" candidate is the LM324 quad opamp which has no current limiting
> for
> > current sinking and the output swing includes ground. This means you can
> > drive a small (30ma) motor connected between +5V and the opamp output. A
> > reverse diode across the motor will protect the output from back EMF.
> The
> > output voltage drop for current source with the load connected to ground
> is
> > too high, making it inefficient and likely to overheat. So a bridge
> circuit
> > with the motor between two outputs is probably not practical. The newer
> CMOS
> > rail to tail output opamps are suitable for driving small motors in a
> bridge
> > configuration. I will check the specs for a recommended part number and
> what
> > limits apply. Power audio amps such as the LM386 have been used in
> motor
> > driver designs. The CD/VCR eject motor driver chips (with those
> difficult
> > Japanese part numbers) are a large as yet untapped resource for future
> Beam
> > designs.
> >
> > (a)(1) You can drive the standard h-bridge if the (CMOS) opamp outputs
> > driving the h-bridge swing rail to rail.
> >
> > (b) You can drive an h-bridge with a LM339 open collector output with a
> > suitable load resistor to Vcc. You can probably forget about driving the
> > motor directly from the 339 comparator outputs since it has poor current
> > sinking capability. There are (rare) power comparators that are suitable
> > though.
> >
> > (c) The newer CMOS opamps can do practically everything and are
> "extreme"
> > power misers to boot. (check Ben Hitchcock's Whistle Blower application
> for
> > a 1uA oscillator at 9V which can't be done with any digital CMOS circuit
> I'm
> > aware of)
> >
> > regards
> >
> > wilf
> >
> > > -----Original Message-----
> > > From: John A. deVries II [SMTP:aubois@trail.com]
> > > Sent: Friday, January 07, 2000 11:55 PM
> > > To: beam@sgiblab.sgi.com
> > > Subject: Yo, Team!
> > >
> > > A couple-three questions:
> > >
> > > (a) can you drive a "typical" BEAM motor from the output of a "garden
> > > variety" 741?
> > > (a)(1) presuming you had an "unlimited" power source, using a
> couple
> of
> > > 741's to drive a "standard" h-bridge would likely end in fried parts
> under
> > > normal operating conditions (whatever the heck they are) of a BEAM
> device,
> > > right?
> > >
> > > (b) would it be better to use a comparator, something like a 339
> instead
> > > (and you get four in one chip..)? In this case I'm supposing direct
> drive
> > > of the motor with the comparators and not even considering an
> h-bridge.
> > >
> > > (c) I'm not ignoring the '14s and other digital stuff but, sheesh,
> isn't
> > > there any decent more-or-less linear analog stuff that isn't too
> > > gosh-awful
> > >
> > >
> > > Anybody have any thoughts?
> > >
> > >
> > > Zoz
> > >
> > >
> > >
> > > p.s. Happy Bloody New Year, Y'all!
> > >
8919 Sat, 8 Jan 2000 10:45:37 -0800 [alt-beam] Re: SIMPLIFIED D1 CIRCUIT "'beam@sgiblab.sgi.com'" Wilf Rigter Hi Jim,
The timer in the type 2 SE is designed to reset the SE before the
charge is fully depleted. This speeds up re-charging and avoid wasting time
and energy without doing any "real work" while the SE sucks the last little
bit of low grade power from the bottom of the barrel. Now the D1 is a
"charge by day/discharge by night" kind of circuit, so some of the type 2
considerations don't apply.
However this is a good suggestion since it is possible to use for a
timer as a way of extending the operating time of the SIMD1 after it fires.
Assume that the normal SIMD1 circuit has a cap and load combination which is
good for 1 hour of continuous "after dark" operation. This could be extended
to 4 hours of intermittent operation with an additional timer which turns
the load ON for 5 minutes and OFF for 15 minutes and then repeats until the
SIMD1 resets about 4 hrs later. That way you can stretch and enjoy a pummer
light show during the whole evening instead of just the first hour. This is
one more way to optimize solar cell size, capacitor size and LED brightness
with respect to usable duration of operation. I'll draw up a little circuit
for this shortly.
Thanks for the idea!
wilf
> -----Original Message-----
> From: Jim Taylor [SMTP:ace27@icdc.com]
> Sent: Friday, January 07, 2000 1:17 PM
> To: beam@sgiblab.sgi.com
> Subject: Re: SIMPLIFIED D1 CIRCUIT
>
> Is their anyway to add a simple timer to make this a type 2 D1 SE?
>
>
>
> James Taylor
> URL: http://fly.to/springmeadows
>
> ----- Original Message -----
> From: Wilf Rigter
> To:
> Sent: Thursday, January 06, 2000 11:57 PM
> Subject: SIMPLIFIED D1 CIRCUIT
>
>
> > Hi Jim,
> >
> > I hope this circuit will shine a little light on your D1 problems.
> Consider
> > using this simplified D1 (SIMD1) type solar charger /darkness trigger.
> It
>
>
>
> > > to me. Sorry for the book, I'm just excited that someone mentioned the
> > > idea!
> > >
> > > See ya,
> > > Jim
> > > http://www.geocities.com/SoHo/Exhibit/8281/beamart.html
> > > ICQ# 55657870
> >
8920 Sat, 8 Jan 2000 12:22:25 -0800 [alt-beam] Re: Pressure Sensors "'beam@sgiblab.sgi.com'" Wilf Rigter I've found a possible cheap source of material for pressure transducers (in
fact I'm surrounded by the stuff)!
Introducing the Compression Dependent Resistor (CDR) !
The material is high voltage "semi-con" rubber, extensively used for voltage
grading in high voltage XLPE power cables. Most high voltage molded rubber
parts use this stuff too. It is also available as tape from 3M in a 15' long
roll x 3.4" wide (type 13 Scotch Electrical Semi-Conducting tape - sorry, I
don't know the cost). The stuff looks like "inner tube" rubber but is less
compressible and not as elastic.
I scrounged a little piece 1/8"x1/4"x1/4" square semi-con rubber from some
HV equipment, sandwiched it between two flat contacts (tiny 1/4" square
PCBs), with two wires solder to the contacts and some heat shrink tubing to
hold it all together. The resistance varies from 200K to 50K with zero to
2000 grams of pressure applied to the CDR assembly. The range is narrow
because the CDR is already compressed by heatshrinking the assembly. The
rate of change of resistance is somewhat slow very similar to an Light
Dependent Resistor reponse time to a change in light. The value of
resistance with the same compression applied is reasonably repeatable.I also
tried a small tube of CDR 1/4" OD and 1/8" ID on a plastic rod with copper
washers on each end for contacts and the resistance varied between 1M and
5K.
For example, a robot foot could be made from a assembly of two PCBs
separated by four CDR sensors, one at each corner. One PCB provides the
common contact for the sensors and the second PCB has 4 separate contacts
for the other side of the sensors. Now with molded CDR sensors you could
do anything.....
enjoy
wilf
> -----Original Message-----
> From: Wilf Rigter [SMTP:Wilf.Rigter@powertech.bc.ca]
> Sent: Friday, January 07, 2000 7:15 PM
> To: 'beam@sgiblab.sgi.com '; 'Sathe Dilip '
> Subject: RE: Pressure Sensors
>
> I will have to check this later but there are elastomer "o"rings and
> gasket
> strips that are used in RFI control around enclosure doors etc. and these
> may be suitable for the "rubber band sensor" application.
>
> regards
>
> wilf
>
> ----------
> From: Sathe Dilip
> To: beam@sgiblab.sgi.com
> Sent: 1/7/00 1:02:27 PM
> Subject: Re: Pressure Sensors
>
> Hi Wilf,
>
> What you say is true & such a conductive rubber band will be ideal for
> many applications. However I haven't come across any easy/cheap source
> of such material. Once I was thinking about the antistatic mats. But
> their conductivity is very low & sometimes the rubber material is bonded
> to another cloth like layer. Do you know of something that will fit the
> bill?
>
> Meanwhile, the sandwich of a zebra strip (oriented properly) should work
> better than the foam.
>
> Dilip
> -----------------------------------------------------------
>
8921 Sat, 08 Jan 2000 21:18:31 CET [alt-beam] photovore beam@sgiblab.sgi.com "johannes urke" hey guys i finally started making a photopopper photovore and when i had
glued the transistors, 1381`s,the trimpot and the photodiodes together
(usind chiu`s tutorial of course)i found out that the caps i had were 2.2uf
not 0.22uf.
now ill solder together the stuff i can without them,
and when i get some caps there will be a mail with the subject: "why the
#Ï@$ doesnt it work?" in your inbox so get ready to reply guys.
______________________________________________________
8922 Sat, 8 Jan 2000 15:46:59 EST [alt-beam] Re: SIMPLIFIED D1 CIRCUIT beam@sgiblab.sgi.com JVernonM@aol.com In a message dated 1/8/00 1:46:40 PM Eastern Standard Time,
Wilf.Rigter@powertech.bc.ca writes:
> Thanks for the idea!
No, no, that's your idea Wilf. I'm just lapping at your heals here :). I've
been testing/comparing the SIMD1 with the old, traditional D1 today and there
is just no contest. I even have the SIMD1 configured in the less efficient
diode arrangement. The fact that I can freeform a single 74hc14 into a SIMD1,
and have 5 inverters left to play with is simply awesome. I seriously think
you have solved the problems associated with night time use of these BEAM
circuits and all of my future bots will reflect that. I can't thank you
enough, I've been on cloud nine all day. As always, I look forward to your
future postings.
See ya,
Jim
http://www.geocities.com/SoHo/Exhibit/8281/beamart.html
ICQ# 55657870
8923 Sat, 8 Jan 2000 16:37:15 -0500 [alt-beam] Re: SIMPLIFIED D1 CIRCUIT "Richard Caudle" Evening Wilf (et al),
You could also use sound as an initiating method. Instead of it being off
for X number of minutes, it would be off until the sound level makes it
activate. That way the operational time could be extended even more, with
the possibility of making it through until the next morning so that the
voltage level never makes it the the 'death' point. Put a battery and solar
charger on it and you have a fully functional robot 90+% of the time!
Just another HPV thingy insinuating itself into the conversation!
Richard
Home