Searching \ for '220VAC relay, Need SNUBBER help' 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/io/motors.htm?key=relay
Search entire site for: '220VAC relay, Need SNUBBER help'.

Truncated match.
PICList Thread
'220VAC relay, Need SNUBBER help'
1998\12\23@092314 by Jason Wolfson

flavicon
face
Gents,
I have a PIC controlling a 220VAC solenoid valve through a
large relay. The relay is driven with a transistor, not directly
from the PIC.

I have a clamping diode across the relay coil.
I can switch the relay all day long with out the valve plugged in,
if I plug in the valve, after a few cycles the PIC resets. It seems
I'm getting a nice big inductive kick when I hit the relay.

My question is: How do I calculate the appropriate values
for a RC snubber network? I'm not sure of the inductance of the
solenoid coil.

Thanks

Jason Wolfson

spam_OUTjasonTakeThisOuTspamlipidex.com

Lipidex Corp
781-834-1600
781-834-1601 FAX

1998\12\23@113931 by Peter L. Peres

picon face
Hello,

On Wed, 23 Dec 1998, Jason Wolfson wrote:

> My question is: How do I calculate the appropriate values
> for a RC snubber network? I'm not sure of the inductance of the
> solenoid coil.

imho, for testing only, connect a standard line filter (Corcom etc)
temporarily in series with the valve connection, and find out if it helps.
Line filters are very good at suppressing and if it does not help, a
snubber will help even less imho.

One way that I have found to help in this case, is to place a 5V6 zener in
parallel with the PIC Vdd/Vcc to prevent voltage rise during a spike, and
connecting MCLR directly to Vdd (sans R). The other thing is to use a
canned oscillator or shield the oscillator area of the PIC. I am not sure
about the latter, it seems to have worked a few times. PICs running on RC
oscillators seem to be less sensitive. Adding relatively high value
(470R..10K) R's on all the connection wires emerging from the circuit also
helps, assuming the respective attachements can work like this.

In stubborn cases only opto isolators and totally separated circuits will
help. This problem is not PIC specific, almost any CMOS and HCMOS circuit
has it.

hope this helps,

       Peter

1998\12\23@132530 by Gerhard Fiedler

picon face
At 09:19 12/23/98 -0500, Jason Wolfson wrote:
>I have a PIC controlling a 220VAC solenoid valve through a
>large relay. The relay is driven with a transistor, not directly
>from the PIC.
>
>I have a clamping diode across the relay coil.
>I can switch the relay all day long with out the valve plugged in,
>if I plug in the valve, after a few cycles the PIC resets. It seems
>I'm getting a nice big inductive kick when I hit the relay.
>
>My question is: How do I calculate the appropriate values
>for a RC snubber network? I'm not sure of the inductance of the
>solenoid coil.

what happens if you replace the diode with a zener (or a unidirectional
tvs, or put a bidirectional tvs parallel to the diode)?

ge

1998\12\23@133334 by dave vanhorn

flavicon
face
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


>>My question is: How do I calculate the appropriate values
>>for a RC snubber network? I'm not sure of the inductance of the
>>solenoid coil.
>
>what happens if you replace the diode with a zener (or a unidirectional
>tvs, or put a bidirectional tvs parallel to the diode)?

The diode or TVS may work, but scope it, you may also need some series R to
provide a place for that energy to dissipate if the coil R isn't high
enough. Remember, it's stored energy you're trying to get rid of, and you
can't just wish it away, it's got to be dissipated.


-----BEGIN PGP SIGNATURE-----
Version: PGP for Personal Privacy 5.5.5
Comment: Trust them, they're from the government!  (NOT!)

iQA/AwUBNoE1G0t+LG8vnBoNEQK18ACg84ObdQ9GinmAXihnpzrueIPSaLwAoIfQ
KAheI4Nox/aEioL6URrrpaFM
=OEdL
-----END PGP SIGNATURE-----

1998\12\23@134548 by Dwayne Reid

flavicon
face
>Gents,
>I have a PIC controlling a 220VAC solenoid valve through a
>large relay. The relay is driven with a transistor, not directly
>from the PIC.
>
>I have a clamping diode across the relay coil.
>I can switch the relay all day long with out the valve plugged in,
>if I plug in the valve, after a few cycles the PIC resets. It seems
>I'm getting a nice big inductive kick when I hit the relay.

Easiest fix is a MOV across the solenoid coil, as close to the soleinoid as
possible.  Standard treatment for us - it works well.

dwayne


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

1998\12\23@141753 by David Reinagel

picon face
At 09:19 12/23/98 -0500, Jason Wolfson wrote:
>I have a PIC controlling a 220VAC solenoid valve through a
>large relay. The relay is driven with a transistor, not directly
>from the PIC.

A somewhat unrelated question: How is your relay connected to the
transistor?  I ask because many years ago, I was working on a
project there I was driving a relay on the emitter side of a
transistor.  I had a flyback diode across the relay coil, but
what I didn't know could happen is that after snubbing the negative
transient on the relay coil, the coil voltage can actually ring
positive, and it was enough that after months of the equipment
being in the field, it would fail. The source of the failure was
the Veb voltage limit was being exceeded by this coil bouncing
positive, punching the emitter-base junction and fusing the
emitter-to-collector junction.  The fix was simple: put a diode
between the emitter and base in the backwards direction; i.e.,
for an NPN, put the anode to the emitter, and the cathode to the
base.

Now this will not fix your problems, but it could save you a lot
of pain in the future.

Oh, and have a Merry Christmas -- the remembrance of the birth of
the man Jesus that changes all history and who determines your
eternity.

David Reinagel
daverspamKILLspamcisco.com

1998\12\23@143448 by Mark Willis

flavicon
face
Peter L. Peres wrote:
{Quote hidden}

 Would a ground ring around the crystal oscillator help here, Folks?
Or would you want a complete filled-in ground plane & an above-PC Board
ground plane?  (I'm NOT a crystal expert <G>)

 I ask as I'm expecting such problems when the next big contract takes
off...  Moving 2-ton items electrically ought to generate some nice
noise.  Don't want the "Pour" circuit to trigger on same!

 Also I was planning to use a 3-terminal Vcc monitor (Panasonic etc.)
to run ~MClr, any problems with those (provided Vcc is well-decoupled
with 0.1 uF MilCer cap's) with bad noise?  Optoisolators between the
PICs and the rest of the real world, of course (Optical fibers, perhaps,
period...)

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

1998\12\23@171349 by paulb

flavicon
face
Peter L. Peres wrote:

> imho, for testing only, connect a standard line filter (Corcom etc)
> temporarily in series with the valve connection, and find out if it
> helps.  Line filters are very good at suppressing and if it does not
> help, a snubber will help even less imho.

 I kind of like Peter's idea here, essentially he is saying "use a
ready-made equivalent to a snubber", which is good, however (and I
realise he did say "for testing only",) a line filter is *not* a
snubber.  The difference is in the resistor in the snubber.

 Without this resistance, you will most certainly minimise impulse and
arcing on break, but you will *increase* it on make instead!  You might
find it very difficult indeed to figure out which is which.

1} Design of snubber.  Determine the effective load impedance of the
solenoid (hint: measure the current).  Make the resistor a plausible
fraction of that, say one-third to half.  Figure the capacitance value
to make a time constant of a fraction of a millisecond i.e., less than
1/16 of the mains cycle.  I'd hazard a guess of about 1k ohms (1W
rating) and 0.3µF, *AC* rated at minimum your line voltage.

2} Put the snubber across the solenoid, not the contacts.

3} The important stuff.  Lead dress.  Make sure the leads to the
solenoid, indeed all the mains leads, are nowhere near the control
leads, and that all returns follow actives.  Visualise this as ensuring
that there are no open loops made by the wiring.  You know that the only
ways for EMI to travel are capacitance (e-field) and inductance
(i-field), so think: space between for the first, and no loops for the
second.

4} Forget the relay coil.  You have isolated this by exclusion already,
as long as you have the snubber diode for this connected *directly
across* the coil, you are right.

 In respect of Peter's other suggestions, *do* make sure you have an
unbroken ground plane/ trace circling the board.  This is a shield.

 (*Don't*, by the way, get confused with anything you may have heard
regarding "hum loops".  It's certainly not relevant here, and frequently
a load of hogwash anyway.  My opinion if you're curious, an attempt to
explain at: http://www.midcoast.com.au/~paulb/faq_loop.html )
--
 Cheers,
       Paul B.

1998\12\23@174022 by James Cameron

flavicon
face
Dwayne Reid wrote:
> Easiest fix is a MOV across the solenoid coil, as close to the
> soleinoid as possible.  Standard treatment for us - it works well.

Yes, I agree this would work, but I worry that the problem isn't really
solved; only hidden.  What you will have done is fix the source of the
interference, leaving the remainder of the circuit susceptible to it.

--
James Cameron                                      (EraseMEcameronspam_OUTspamTakeThisOuTstl.dec.com)

OpenVMS, Linux, Firewalls, Software Engineering, CGI, HTTP, X, C, FORTH,
COBOL, BASIC, DCL, csh, bash, ksh, sh, Electronics, Microcontrollers,
Disability Engineering, Netrek, Bicycles, Pedant, Farming, Home Control,
Remote Area Power, Greek Scholar, Tenor Vocalist, Church Sound, Husband.

"Specialisation is for insects." -- Robert Heinlein.

1998\12\24@120459 by Peter L. Peres

picon face
On Wed, 23 Dec 1998, Mark Willis wrote:

>   Also I was planning to use a 3-terminal Vcc monitor (Panasonic etc.)
> to run ~MClr, any problems with those (provided Vcc is well-decoupled
> with 0.1 uF MilCer cap's) with bad noise?  Optoisolators between the
> PICs and the rest of the real world, of course (Optical fibers, perhaps,
> period...)

I have used a plain LM324 and a reference diode (LM385-2.5) to implement
both the regulator and the reset circuit. You get a stable 1/2 Vdd
reference and 2 spare op-amps almost for free.

The trick with the 324 is, to provide the amplifier that provides +5V to
the PIC with positive feedback, and use a small decoupling capacitor for
Vdd near the PIC. When the amplifier toggles, it generates a clean rise on
Vdd and MCLR. Also if the voltage drops, it drops suddenly, and the 324
sinks current from the PIC Vdd so a new cycle can start clean and cause a
proper MCLR trigger. It is very hard to make this kind of power supply
misbehave by applying and removing power randomly. Also, the 324
withstands reverse powering (!) afaik but within limits and with no
warranty.

The 324 will supply enough current for a PIC alone, and any driven
circuits are driven by sinking current from an unregulated supply (to
avoid loading the PIC power supply).

hope this helps,

Peter

1998\12\24@142334 by Michael J. Ghormley

flavicon
face
Jason Wolfson wrote:

> I have a PIC controlling a 220VAC solenoid valve through a
> large relay. The relay is driven with a transistor, not directly
> from the PIC.
>
> I have a clamping diode across the relay coil.
> I can switch the relay all day long with out the valve plugged in,
> if I plug in the valve, after a few cycles the PIC resets. It seems
> I'm getting a nice big inductive kick when I hit the relay.
>
> My question is: How do I calculate the appropriate values
> for a RC snubber network? I'm not sure of the inductance of the
> solenoid coil.

I read a *ton* of replies that all have good ideas, but I have not seen
one that addresses my first guess.  My first thought was not the back-EMF
of the coil (since he has the diode in place), but Vdd.  Could it be that
when you energize the coil it is pulling your Vdd down?

Are you using Vdd to drive the coil?  If you are, then I would look
to separate them.  Perhaps unregulated +V for the coil and regulated for
Vdd. Even then, make sure that the PS can handle the current of the coil.

Also, how about ground loops?  Maybe Vss is being driven high!?!

Just my two pence...

Michael

*************************************************************************When th
e way of the Tao is forgotten, kindness and ethics must be taught.
Men must learn to pretend to be wise and good.  --  Lao Tzu
*************************************************************************

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