[TriEmbed] Protection from Dual 12V inputs

Wed Mar 16 14:01:46 CDT 2016

----- Reply message -----
From: "Shane Trent" <shanedtrent at gmail.com>
To: "Robert Gasiorowski" <rgresume at gmail.com>, "Pete Soper" <pete at soper.us>
Cc: "triembed at triembed.org" <triembed at triembed.org>
Subject: [TriEmbed] Protection from Dual 12V inputs
Date: Wed, Mar 16, 2016 2:20 PM

An email from Brian just pointed out that my circuit does not disable both outputs if a second output pin is turned on (it only locks out the second output). Below is Roberts's circuit (with inverter wired NAND gates to match the input phase) that does correctly turn off the outputs if a second output is enabled. Simulated at http://logic.ly/demo/

Shane

On Sat, Mar 12, 2016 at 1:48 PM Shane Trent <shanedtrent at gmail.com> wrote:
Robert,
You can reconfigure the NAND gates to get non-inverted outputs. I had to use an on-line simulator to convince myself it would work. I really liked the logic.ly demo site. The image below shows the circuit locking out the second output because the first is already on.

http://logic.ly/demo/

You could also just add a single NPN transistor and a couple of resistors and allow one input to override the other input (pulling it down on the far side of that pin's current limiting resistor), forcing it's output low. The single transistor approach has one drawback/benefit by allowing you to select which input would have priority and predefine the output that would turn on anytime both inputs we driven high.

Shane

On Sat, Mar 12, 2016 at 8:53 AM Robert Gasiorowski via TriEmbed <triembed at triembed.org> wrote:

Here's another idea which uses 7400/4011. Keep in mind that outputs are inverted.

On Sat, Mar 12, 2016 at 8:42 AM, Robert Gasiorowski <rgresume at gmail.com> wrote:

Below is a better drawing of my circuit.
If you want to get fancy, you could use configurable gate, like 74LVC1G58 (you would need one for each channel.)
Use A as input, B permanently high, C opposite input.

On Fri, Mar 11, 2016 at 10:16 PM, Pete Soper via TriEmbed <triembed at triembed.org> wrote:

Hi Brian,

(Here's your original
msg for others to catch up on if they're interested. I should
also point out that folks can set their mail to "digest mode" and
get a single msg a day if this spurt of traffic gets too bad. One
person unsubscribed yesterday.)

  As one relative beginner to another, let me ask what's between the
level translator and your relays? You sort of imply below you might
be expecting the level translator to deliver a ton of current
despite it going through the 470 ohm resistor (limit with 5v source
is about 10ma). It's hard to picture the level translator operating
even the most puny relay.

  But Robert's circuit is predicated on the MOSFET gate needing very
little current (i.e. with 5v out of the level translator about
10ma). The other transistor has to be able to yank the gate on the
other side of one of these resistors to ground, so making them too
small defeats the either/or aspect.

  But it really needs resistors in the path from each level
translator to the transistor base Rob accidently connected it
directly to. Without those it's a survival duel between too high
current through the level translator or the transistor base to
emitter path and one or both parts may be destroyed instantly or at
some random, inconvenient time. The base resistors could be a few k
ohms and even the crummiest transistors would sink the 10ma to
prevent the MOSFET connected to its collector from switching on (the
470 ohm resistor is kind of like a bus pullup in this situation).

  Now that we're up to two transistors and four resistors, a single
74LV04 (Mouser part 771-LV04N112) seems
attractive.  One of it's inverters could feed a complement of a
single MCP23008 line to one relay(driver?) and the original signal
would drive the other one.  But this LV chip is wimpy. For more
drive the 74AHC04 would work as below. Or you could use the LV04 in
between the MCP and the level shifter (i.e. sticking with something
you know is working now).  

-Pete

On 03/11/2016 02:24 PM, Grawburg via
TriEmbed wrote:

I've tried this and it
doesn't work.  I did change the resistors to 220 just to see if
that made a difference.

My actual set-up does not use MOSFETs as I originally asked
about but instead uses two DPDT relays.

The input voltage goes through a level shifter since I
determined that the 3V3 through the resistor was too close to

the 2V5 that produces an unknown state in the relay modules. So
the output from the shifter is 5V which does work fine. I still
need to

be able to make sure a bad code doesn't allow both relays to
activate.  

Brian Grawburg

-----Original Message-----

From: "Robert Gasiorowski" <rgresume at gmail.com>

To: "Grawburg" <grawburg at myglnc.com>

Cc: "Triangle Embedded Devices" <TriEmbed at triembed.org>

Date: 02/24/16 05:34 PM

Subject: Re: [TriEmbed] Protection from Dual 12V inputs

Ooops, forgot to add base resistors.

On Wed, Feb 24, 2016 at 5:23 PM,
Robert Gasiorowski <rgresume at gmail.com>
wrote:

How about something simple like this:

-- 

[This was sent from a
PC running Debian 7, 64-bit Linux. No Microsoft products
were used.]

_______________________________________________
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

_______________________________________________

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

_______________________________________________

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/20160316/95d1ee98/attachment.htm>