Searching \ for '12C509 to drive 12V latching relay from 24VDC?' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power/batterys.htm?key=12v
Search entire site for: '12C509 to drive 12V latching relay from 24VDC?'.

Truncated match.
PICList Thread
'12C509 to drive 12V latching relay from 24VDC?'
1998\10\23@045613 by James Cameron

flavicon
face
I'm a bit clueless on this ...

I want to drive a latching relay from two outputs of a 12C509, but I'm
not sure what the simplest and most reliable method is.  I fear I may
need a driving transistor because of the supply differences.

The specifications of the relay are;
       - 12V DC 400mW nominal coil power,
       - two coil latching design,
       - DPDT,
       - Minimum set and reset power 180mW,
       - Nominal set and reset power 360mW,
       - Switching current maximum 2A,
       - [Jaycar SY-4060 for the Aussies here].

This is for a wheelchair addition ... a swing-away chin control arm,
being driven by an 8RPM 24VDC 0.2A motor.  I'll be configuring the DPDT
relay with limit switches, to keep the limiting function away from the
software.

Unfortunately the main traction supply is 24VDC, and the relay is 12VDC,
so I'm wondering whether I should use some form of capacitive discharge
into the relay coil rather than add a 12V supply.  Since the relay
doesn't need to be triggered often, this seems appropriate.

There's no requirement for PWM speed control on this one, otherwise I'd
consider a H-Bridge design.  The main purpose of the PIC is to detect
the "right" control pattern from the client's microswitches and to avoid
spurious triggering.

--
James Cameron                                    (spam_OUTcameronTakeThisOuTspamstl.dec.com)
Digital Equipment Corporation (Australia) Pty. Ltd. A.C.N. 000 446 800

1998\10\23@084849 by Peter L. Peres

picon face
On Fri, 23 Oct 1998, James Cameron wrote:

> The specifications of the relay are;
>         - 12V DC 400mW nominal coil power,
>         - two coil latching design,
>         - DPDT,
>         - Minimum set and reset power 180mW,
>         - Nominal set and reset power 360mW,
>         - Switching current maximum 2A,
>         - [Jaycar SY-4060 for the Aussies here].

A small question: Why use a latching relay for an application that can
certainly benefit from a self-resetting relay and where there is plenty of
power available.

A DPDT 5V relay can manage more than 3A and you can still use your
micro-switches to limit throw. Two such relays are cheaper than a latching
one and you can cunningly wire them (in series) such that they can't
auto-destruct even if both are turned on at the same time. If I were you
I'd use two micro-switches for each end, in series, each with its own
diode, just to be sure. Smoke from a cooking motor coil does not benefit
the health of a person that cannot move away from it.

For an existing latching relay, you can probably use 2 DTC124 digital
transistors and 2 kickback stopper diodes (1N4148) and that's it. It's a
good idea to fuse the relay coil power line with a micro thermo fuse.
Latching relay drive circuits are dimensioned for momentary operation and
have been known to 'cook' quite hot for a long time if failed.

my $0.02

Peter

1998\10\23@091911 by Mike Keitz

picon face
On Fri, 23 Oct 1998 08:54:11 +0000 James Cameron <.....cameronKILLspamspam@spam@STL.DEC.COM>
writes:
>I'm a bit clueless on this ...
>
>I want to drive a latching relay from two outputs of a 12C509, but I'm
>not sure what the simplest and most reliable method is.  I fear I may
>need a driving transistor because of the supply differences.
>
>The specifications of the relay are;
>        - 12V DC 400mW nominal coil power,

Ok, from that I compute the coil current is 33 mA and the resistance is
360 ohms.

You will need a driver transistor for each coil to handle switching the
12V.  It's nothing to be scared of.  Ordinary NPN transistors like 2N2222
will do fine.  Connect the emitter to ground, base to PIC output through
a resistor (1K should be good) and collector to the relay coil.  Also you
must have a diode in parallel with the relay coil to absorb the
"inductive kick" when it is switched off.

>        - two coil latching design,
>        - DPDT,

Consider two standard relays instead of the more expensive latching one.
The PIC would engage one relay or the other for a long enough time to be
sure the motor has moved to the limit, then turn it off.


>Unfortunately the main traction supply is 24VDC, and the relay is
>12VDC,
>so I'm wondering whether I should use some form of capacitive
>discharge
>into the relay coil rather than add a 12V supply.  Since the relay
>doesn't need to be triggered often, this seems appropriate.

Just place a resistor in series with the supply to limit the current so
that only 12V appears across the relay coil when it's on.  Since you want
to cut the voltage in half, the resistance should be the same as the
relay coil's -- 360 ohms.  The same resistor can be shared by both coils
since only one will be active at a time.

The main supply is likely to have a lot of spikes when the motors are
active.  Place an RC filter and maybe an MOV before your regulator if
deriving the PIC power from the 24V line as well.

___________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
or call Juno at (800) 654-JUNO [654-5866]

1998\10\23@115123 by Harold Hallikainen

picon face
       Since it's a latching relay, it seems to me that it would be safe
to pulse it at the full 24 volts.  The relay coil should act like an
inductor and the current would just ramp up twice as fast as it would at
12 VDC.  By limiting the pulse width, you limit the the current,
protecting the relay.  You may want to include a single PTC thermistor
(like a Raychem Polyswitch) in the +24V supply to the relay to protect
it, just in case.


Harold



Harold Hallikainen
haroldspamKILLspamhallikainen.com
Hallikainen & Friends, Inc.
See the FCC Rules at http://hallikainen.com/FccRules and comments filed
in LPFM proceeding at http://hallikainen.com/lpfm


___________________________________________________________________
You don't need to buy Internet access to use free Internet e-mail.
Get completely free e-mail from Juno at http://www.juno.com/getjuno.html
or call Juno at (800) 654-JUNO [654-5866]

1998\10\23@125343 by Quentin

flavicon
face
How about just using a Resistor to drop the voltage to the Relay?
Relay coils are not as critically voltage depended such as semiconductors.
Most of them can handle up to more than 50% of rated voltage.

Quentin

1998\10\23@134628 by Mark Willis

flavicon
face
James Cameron wrote:
>
> This is for a wheelchair addition ... a swing-away chin control arm,
> being driven by an 8RPM 24VDC 0.2A motor.  I'll be configuring the DPDT
> relay with limit switches, to keep the limiting function away from the
> software.

 The setup of this sort I saw (For one of Robin's friends' wheelchairs)
initially used a 28 VDC aircraft motor for arm movement, had a pair of
headswitches and a bank of LED's on the arm (Left switch = select, right
switch = increment, one selection was for Arm deploy, the other was Arm
retract <G>)  Other LEDs were for Radio on/off, Headlights, and a few
spares (Robin remembers that they did leave 3 or 4 slots for future
expansion.)

 For convenience, you could have it go back to the top of the list each
time you complete an action;  the wheelchair shop that wired this one
had it just stay in place, so the chair's owner could leave the system
in "deployed" mode & lean on the left headrest without the system moving
the arm - so she could look left, sleep, etc.  For some with better head
control, with all those open slots that's annoying (So maybe jumper them
disabled until used! <G>)

 That system was entirely electromechanical, AFAIK.  Could have been
7400 series TTL for all I know, but I suspect not.

 Nothing at all wrong with just running the current through a pair of
limit switches at each limit, in series, and enabling the motor run
current for a set length of time to allow the motor to run through it's
full range.  (Making the limit switches easily replaceable is good, as
that way Murphy won't break them <G>)  Good data from other folks here!

 Mark Willis, .....mwillisKILLspamspam.....nwlink.com

1998\10\23@143019 by Thomas McGahee

flavicon
face
James,
You can run the 12 VDC relay coils direct off the 24 VDC if you
do the following:
1) Use a transistor to drive the coil.
2) Use a current limiting resistor in series with the coil.
3) Use a reversed diode across the coil to reduce spikes.
4) *PULSE* the appropriate coil for just long enough to
  ensure proper latching.

Note that the wattage rating of the resistor can be considerably
less than the calculated wattage P=E*I since it will only be
on for a few hundreds of milliseconds. You should be able to
get by comfortably with a 1/4 watt resistor.

Another alternative is to use an electrolytic or other high value
cap in parallel across the resistor. This will allow a high
initial turn-on current that will then decay to a small
current by the time the coil has initiated the latching action.
Actual values depend on the specifics of the coil being
activated.

*****
If you want to save parts you can probably get away with
leaving out the current limiting resistor altogether and just
relying on the short activation time to prevent transistor
problems.

Hope this helps.
Fr. Tom McGahee




----------
{Quote hidden}

1998\10\23@194121 by James Cameron

flavicon
face
Okay, thanks for the input guys ...

Mike & Peter asked "Why a latching relay?"  I want the device to
remember current position without having to save it in the
microcontroller.  We don't want the arm to move if the power is cycled,
or if the microcontroller resets due to bad shielding or something.

[There are recent stories about wheelchair traction motors engaging
while the user waits at pedestrian traffic lights after their
insufficiently protected logic encounters an over-powering signal from a
passing mobile telephone equipped car!]

Mike added that I should add an RC filter and a MOV before my PIC's
regulator.  In the past I've just used a 1N4004 diode to a 100uF
electrolytic capacitor before the 7805.  Would this be sufficient?  If
using a MOV, what specification should I look for?

I'd love to get into integrated H-Bridges, as they seem so obvious a
solution, but I'm an amateur and don't often have the guts to order from
Farnell, RS, Digikey, or other professional suppliers.  ;-)  They list
so many parts I'm at a loss to know which is the best fit.

--
James Cameron                                    (KILLspamcameronKILLspamspamstl.dec.com)
Digital Equipment Corporation (Australia) Pty. Ltd. A.C.N. 000 446 800

1998\10\24@013434 by Dwayne Reid

flavicon
face
>I'm a bit clueless on this ...
>
>I want to drive a latching relay from two outputs of a 12C509, but I'm
>not sure what the simplest and most reliable method is.  I fear I may
>need a driving transistor because of the supply differences.

<big snip>

You are definitely heading in the right direction.  I would use small
transistors (2n4401) with 2K2 from base to '509 and another 2K2 from base to
emitter (emitter grounded), collector to relay coil and clamp diode, other
side of relay and clamp diode to 220uF cap with 2K2 resistor from cap to 24V
supply. Note that you probably need only 1 cap / R combination to power both
relays so long as you allow enough off time to allow the cap to recharge.
Play with the value of C to ensure that the relay sees enough energy to
fully switch every time.  You can do this by making C smaller and smaller
until the relay does not switch, then multiply that value by 5 or 10.

The fact that the relay coil is monentarily over-voltaged should not be a
problem.

dwayne


Dwayne Reid   <RemoveMEdwaynerTakeThisOuTspamplanet.eon.net>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
(403) 489-3199 voice     (403) 487-6397 fax

1998\10\24@063904 by Russell McMahon

picon face
I would be tempted to use something like the ULN2803 driver IC.
'Tis cheap and has 8 drivers. These cam be paralleled for extra drive
if desired. Each is a Darlington Transistor with base resistors so
can be driven directly by PIC output. Catch diodes are built in and
returned to a single pin which is then connected to supply or
resistive snubber or whatever as appropriate.

I would be tempted to use a 12volt relay and a regulator.
Most TO220 regulators are 1 amp rated and a minimal heatsink will
suffice in this application.

Total drive circuit then becomes one cheap drive chip and a cheap
TO220 regulator plus 1 resistor perhaps for catch diode energy
dissipation (plus of course the normal "glue" components around the
regulator (2 caps).

regards

           Russell McMahon


{Original Message removed}

1998\10\24@083657 by James Cameron

flavicon
face
Here's my draft design.

http://ftp.digital.com.au:6153/arm.gif

The dots within circles are PCB termination points for cabling to the
various bits and pieces that won't live on the PCB.

As you may see, I'm using the simple switching transistor technique
suggested, with a resistor to drop the voltage seen by the 12V latching
relay to "about" 12V.

--
James Cameron                                    (spamBeGonecameronspamBeGonespamstl.dec.com)
Digital Equipment Corporation (Australia) Pty. Ltd. A.C.N. 000 446 800

1998\10\24@164857 by paulb

flavicon
face
James Cameron wrote:

> Here's my draft design.

 The diodes across the Arm Travel Limit Switches don't do anything.
Specifically, they can *not* absorb inductive kickback from the motor in
that position.

 Probably just as well since a 1N4148 is a *signal* diode anyway, and
not rated to take inductive kickbacks - you need a 1N4004 power
rectifier for that purrpose.

 I'm not sure what the reverse diode across the 12V supply is for
either, but I'd hazard a guess it will burn out well before the 2A fuse
if the battery is connected backwards.  If it was a 1N5404 it should
however survive the fuse.
--
 Cheers,
       Paul B.

1998\10\25@084024 by paulb

flavicon
face
Russell McMahon wrote:

> I would be tempted to use something like the ULN2803 driver IC.
> 'Tis cheap and has 8 drivers.

 *Cheap!*  You say?  Could you *please* explain to us where to get it
cheap?  List price in the Dick Smith catalogue is $4.95 or $3.96 in
tens.  I think I can get a 12C509 for that price!  The ULN2003 which you
probably meant to suggest for this particular project is $3.95 or $3.17
in tens.

> These cam be paralleled for extra drive if desired.  Each is a
> Darlington Transistor with base resistors so can be driven directly by
> PIC output.

 The PIC ability to source as much as 20mA means you don't really need
a darlington.  A BC337 should switch about 500mA from about 2mA and
cost 35 cents, plus 10 cents for the catch diode and 8 cents for the
(base) resistor (all one-off prices from the same catalogue, *not* that
I'd do that myself!) and we, or at least James, want two of these so I
total $1.06 (AUS).

> I would be tempted to use a 12volt relay and a regulator.

 *I* wouldn't when regulation is not required and an 8 cent resistor
will do!

> Total drive circuit then becomes one cheap drive chip

 Cheap?  *Please* Russell, where do you get ULN2803s for $1.00 or less?
Assuming the postage is reasonable I'll have a couple of dozen.

> and a cheap TO220 regulator

 Which draws 5 to 10mA with no load...

> plus 1 resistor perhaps for catch diode energy dissipation

 What's that?

> (plus of course the normal "glue" components around the regulator (2
> caps).

 Yeah, got plenty.
--
 Cheers,
       Paul B.

1998\10\25@120505 by Peter L. Peres

picon face
On Sat, 24 Oct 1998, Russell McMahon wrote:

> I would be tempted to use something like the ULN2803 driver IC.

Yes ! Moreover, the ULN2803 can take 50V on its outputs (check ! I'm not
sure) and also has integrated kickback diodes. So, it can drive a 24 V
relay directly. Even bipolar. You can also afford to drive a few lights
off 24V in the same manner using spare ULN outputs.

By removing the need to use 12V you can use a 78L05 to regulate 24V down
for the PIC. It only needs to supply the PIC current. The ULN drive
current will add another mA or so per output driven.

Peter

1998\10\25@152943 by Mike Henning
flavicon
face
>> I would be tempted to use something like the ULN2803 driver IC.
>> 'Tis cheap and has 8 drivers.
>
>  *Cheap!*  You say?  Could you *please* explain to us where to get it
>cheap?  List price in the Dick Smith catalogue is $4.95 or $3.96 in
>tens.  I think I can get a 12C509 for that price!  The ULN2003 which you
>probably meant to suggest for this particular project is $3.95 or $3.17


You can get ULN2803's or ULN2003's from Mouser Electronics for about .50
cents each!  And probably about the same from Digi-Key Electronics.

1998\10\26@053200 by Russell McMahon

picon face
From: Paul B. Webster VK2BZC <TakeThisOuTpaulbEraseMEspamspam_OUTmidcoast.com.au>

>Russell McMahon wrote:
>
>> I would be tempted to use something like the ULN2803 driver IC.
>> 'Tis cheap and has 8 drivers.
>
>  *Cheap!*  You say?  Could you *please* explain to us where to get
it
>cheap?  List price in the Dick Smith catalogue is $4.95 or $3.96 in
>tens.  I think I can get a 12C509 for that price!  The ULN2003 which
you
>probably meant to suggest for this particular project is $3.95 or
$3.17
>in tens

I did mean the ULN2803 but other similar chips in the family would do
also.
I THINK the price I got my last lot of 2803's for was between $NZ1
and $NZ2  each in tube quantity - these were from an AVNET VSI
sub-supplier who deals in smaller quantities. My price recall may be
wrong :-).
Yes, this is dearer than discrete transistors for 2 channels (as are
required here)
. For 8 (which its capable of), depending on current, its a fairly
close comparison).

>  The PIC ability to source as much as 20mA means you don't really
need
>a darlington.

Yes, the darlington isn't essential but it comes as part of the IC
(as does the input resistor and catch diode).

>> I would be tempted to use a 12volt relay and a regulator.
>
>  *I* wouldn't when regulation is not required and an 8 cent
resistor
>will do!


Yes. I agree. For some reason I think that I was thinking in terms of
switching the motor itself directly when I wrote this - it was
specified as 200ma so would drop 2.4W if it was a 12 volt one.
I would hope that TO220 regulators would typically cost $A1 from a
reasonable source (but of course you can always go to DSE :-)).

>> Total drive circuit then becomes one cheap drive chip
>
>  Cheap?  *Please* Russell, where do you get ULN2803s for $1.00 or
less?
>Assuming the postage is reasonable I'll have a couple of dozen.


As above - check with AVNET re tube quantities - you MAY be
pleasantly surprised (but maybe not).


>> plus 1 resistor perhaps for catch diode energy dissipation
>  What's that?


Depends on circuit and timing - if you place the back "anti-spike"
diodes directly across the relay coil, when the relay opens the
current will "circulate " through the coil and diode until the energy
dissipates. With low resistance in the circuit this can slow the
relay release time (time constant is L/R where L is coil inductance).
If you place a resistor in series with the diode the energy will
dissipate faster due to higher R and the relay will turn off more
crisply. This affect can be significant in some applications. The
"downside" is that the voltage spike will rise to "IR" above the
supply voltage (I is relay current, R is total resistance including
coil and resistor). The diode forward voltage drop complicates all
this only slightly.

regards

   Russell

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...