[TriEmbed] Question concerning a mechanical release

Thu Sep 26 09:49:40 CDT 2013

Pete,

I just realized I was not using "Reply All".  So much for my internet
skills  I replied twice with only "Reply" so they never made the list.
 Both are copied below.

I love the Bob Pease reference on driving the solenoid using a capacitor.
(I got an autograph a copy of "Troubleshooting Analog Circuits" from him at
seminar 10 years ago).  Note that the author of the solenoid_drivers.pdf
has a typo in their schematic.  All of the reference to calculating a
resistor value are actually referring to the resistor in parallel with the
capacitor.  The resistor with the capacitor is always calculated based on
the DC resistance of the solenoid coil.  The DC resistance of the solenoid
and the resistor in parallel with power capacitor were labeled several
different ways and could confuse someone who was not familiar with the
application.

-Shane

<<Like most things in engineering, it depends.  How much weight and how
much control do you need (milliseconds or seconds between the signal and
the load dropping?) A small electromagnet could release a weight. A hobby
servo or solenoid could pull a pin to release a weight.  The hobby servo
would require minimal hardware.  The others would require at least a
transistor and diode to power the magnet or solenoid.  Below is a link for
a tutorial on the hobby servo.

http://learn.adafruit.com/adafruits-raspberry-pi-lesson-8-using-a-servo-motor/servo-motors

Is the weight load small enough to hold with a clothes-pin?  If so, you
might be able to put a cam on a hobby servo and use it to open the clothes
pin to release the load.  For that matter you could hang the load from a
pin or rod attached to a hobby servo and rotate the servo to drop the load.
 It all depend on the load and the level of control that you need.  More
detail will get you better guesses!  I hope something here is useful to
you.>>

<<The Adafruit solenoid looks very nice if you have the power to drive it.
 Use a decent  logic-level N-FET along with a kickback diode in parallel
with the solenoid and you should be all set.  I would add a 100 ohm
resistor in series with the gate of the FET just for good measure.

As Rodney mentioned, you need to be careful of noise induced or conducted
to the power rails of your microcontroller.  I recommend using a
twisted-pair cable (phone or Ethernet cable would be fine) to connect your
gate drive signal and logic ground to the FET and only connect them at the
FET with the FET mounted as closely to the solenoid as possible.>>

On Thu, Sep 26, 2013 at 10:19 AM, Pete Soper <pete at soper.us> wrote:

>  Hi Lucas,
>    Please "reply all" to this msg to let us know if the information so far
> has given you any ideas for a solution, and if so, what your implementation
> scheme will be. Also, if you're OK with sharing the details, we're eager to
> know what the string is connected to,  if the string release happens
> periodically, etc.  If circumstances have changed or something else has
> made this project of yours moot, no worries, and I don't want you to feel
> any pressure to reply. It's just that some of us are very curious. :-)
>    I'd like to summarize a few high points I remember from the msgs so
> far. Folks: I've only read a few msgs of this thread, so forgive me if this
> is partially or entirely redundant.
>       1) Raspberry Pi gpio pins are *VERY* weak, only being able to
> source/sink a few milliamperes of current, in a strictly "3.3v power
> supply" context. That is, that pin cannot tolerate negative or >3.3v
> potentials, cannot be expected to hold a large current at ground, properly
> drive a high current load, etc. Part of the "0 < voltage < 3.3" requirement
> is that no inductive load can be allowed to dump it's back EMF from a
> magnetic field collapse into the Raspberry Pi pin: that will almost
> certainly render that part of your Raspberry Pi inoperable if it doesn't
> destroy the whole chip. So, for multiple reasons an RPI can only directly
> sink/source ("provide a ground for"/"drive") a few devices such as another
> integrated circuit or bare transistor or a (low voltage) LED with a
> suitably large value resistor for current limiting.
>      2) Driving something like a solenoid, stepper or plain DC motor,
> relay, etc, requires a circuit that can "translate" the Raspberry Pi
> signal/sink into a relatively high current and usually higher voltage
> capable of running the device.
>      3) Servo motors are special. In the context of this thread (i.e in
> regards to the "hobby-type" servos Rod Radford mentioned vs an industrial
> servo), they are self-contained motors that have a very low voltage, low
> current drive requirement easily met by a Raspberry Pi and conveniently
> works in conjunction with off the shelf software capable of causing the
> servo motor to hold position, move a fixed distance one way or the other,
> rotate continuously in one direction or another, etc.
>    "String grabbing" would seem to be a special case of the "rope
> grabbing" someone described vis a vis sailboats, so emulating the sailboat
> device on a scale appropriate for your string and weight would seem
> sensible. Using a servo motor for the mechanical end of things would allow
> you to focus on your application and not get bogged down with
> hardware/software details.
>   I look forward to seeing you and the others who have contributed to this
> thread at a future meeting. The next mtg at NCSU is Monday, October 8th.
>
> Best Regards,
> Pete (home from UNC hospital and recovering from step one of a
> long-planned two part heart-remodeling project)
>
>
> On 9/23/2013 6:57 PM, Lucas Rumney wrote:
>
> I am trying to make something that would achieve the following end:
>
> raspberry pi being able to release a weight on a string.
>
> My idea was that the pi could send a signal to disengage some kind of
> clamp and release the weight, but I am not sure what kind of clamp or what
> I should use here?
>
> Does anyone have any better ideas how to release a weight on a string
> controlled by a pi?
>
>  Thanks, I can't attend this meeting because of my courseload, but I
> still want to participate in general discussions.
>
>
> _______________________________________________
> Triangle, NC Embedded Computing mailing listTriEmbed at triembed.orghttp://mail.triembed.org/mailman/listinfo/triembed_triembed.org
> TriEmbed web site: http://TriEmbed.org
>
>
>
> _______________________________________________
> Triangle, NC Embedded Computing mailing list
> TriEmbed at triembed.org
> http://mail.triembed.org/mailman/listinfo/triembed_triembed.org
> TriEmbed web site: http://TriEmbed.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.triembed.org/pipermail/triembed_triembed.org/attachments/20130926/5557950a/attachment.htm>