Searching \ for '[EE] Clamp diode slows down solenoid switch off.' 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=solenoid
Search entire site for: 'Clamp diode slows down solenoid switch off.'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Clamp diode slows down solenoid switch off.'
2005\03\14@082717 by Peter Onion

flavicon
face
Hi,

I'm trying to use a PIC to control an old paper tape reader like this
one...

http://members.ozemail.com.au/~jacksmth/images/e803ptr.jpg

The problem I'm having is that after turning off the tape break solenoid
it seems that it takes ~15mS before the tape starts to move (as seem by
the sproket hole output going low).

I'm using a "matching unit" between the coil and the switching FET
(IRF630) as detailed in the original circuit.  This is a 18ohm resistor
and a 50uF capacitor in parallel.

I think the problem could be caused by the protection diode I've put
across the coil winding.  It seems that when the FET turns off, the
current that was flowing through the coil into the transistor now loops
back via the diode.  This looks like a short across the coil and thus
prolongs the decay of the current and hence maintains the magnetism in
the coil long after the drive has been removed.

Does this make sense ?  If so, any suggestions how to improve the
situation ?  I've tried the circuit without the diode, but the large
voltage spike that resulted seem to get into the photocell amplifiers
and cause false detection of the sproket holes.

Peter

2005\03\14@084819 by Michael Rigby-Jones

picon face


{Quote hidden}

A diode to clamp the flyback voltage certainly will slow down the
collapse of the magnetic field by allowing a current to continue
flowing.  In place of a simple diode you could use a zener which allows
the voltage to increase to some level below the breakdown of the
transistor.  Ensure you calculate the power dissiptaion in the zener
(could be surprisingly high) and rate it accordingly.

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2005\03\14@090544 by Alan B. Pearce

face picon face
>I think the problem could be caused by the protection diode I've put
>across the coil winding.  It seems that when the FET turns off, the
>current that was flowing through the coil into the transistor now loops
>back via the diode.  This looks like a short across the coil and thus
>prolongs the decay of the current and hence maintains the magnetism in
>the coil long after the drive has been removed.
>
>Does this make sense ?  If so, any suggestions how to improve the
>situation ?  I've tried the circuit without the diode, but the large
>voltage spike that resulted seem to get into the photocell amplifiers
>and cause false detection of the sproket holes.

Yes, known problem when fast relay release is required.

Two possible methods of reducing the problem.

1. put zener diode or resistor in series with flyback diode. The idea is
that the resistor value or zener voltage is calculated to allow the voltage
peak on flyback to be within the transistor spec, but the current is
promptly reduced to levels that allow the solenoid to drop out quickly.

2. use an RC snubber network similar to that used on triacs in mains
control, selected to minimise the shape of the spike.

2005\03\14@091730 by olin_piclist

face picon face
Michael Rigby-Jones wrote:
> A diode to clamp the flyback voltage certainly will slow down the
> collapse of the magnetic field by allowing a current to continue
> flowing.  In place of a simple diode you could use a zener which allows
> the voltage to increase to some level below the breakdown of the
> transistor.

Or just a resistor in series with the existing diode.  The current thru the
coil is presumably known, which will be the same current thru the resistor
immediately after switch off.  The resistor can then be appropriately sized
so that the resulting voltage is tolerable.

However, 15mS for a mechanical system to start moving doesn't sound that bad
in the first place.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\03\14@092210 by olin_piclist

face picon face
Alan B. Pearce wrote:
> 1. put zener diode or resistor in series with flyback diode. The idea is
> that the resistor value or zener voltage is calculated to allow the
> voltage peak on flyback to be within the transistor spec, but the
> current is promptly reduced to levels that allow the solenoid to drop
> out quickly.
>
> 2. use an RC snubber network similar to that used on triacs in mains
> control, selected to minimise the shape of the spike.

3. Put parallel resistor/capacitor in series with the coil.  This will put
full voltage accross the coil when it's first turned on, then fall back to a
lower voltage in steady state.  The latter is just enough to supply the
"holding current".  This doesn't fix the problem as the other solutions do,
but does make it less severe.


*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

2005\03\14@092855 by Dave VanHorn

flavicon
face

Have a look at what's done for impact printer pin solenoids. Same problem
more or less.
And the solenoid design figures into it as well.


2005\03\14@093448 by Mark Scoville

flavicon
face
Hi Peter, this may be of use to you, it is a P&B app note on "...coil
supression with DC relays".

http://relays.tycoelectronics.com/app_pdfs/13c3311.pdf

-Mark

{Quote hidden}

> -

2005\03\14@155009 by Russell McMahon

face
flavicon
face
> I think the problem could be caused by the protection diode I've put
> across the coil winding.  It seems that when the FET turns off, the
> current that was flowing through the coil into the transistor now
> loops
> back via the diode.  This looks like a short across the coil and
> thus
> prolongs the decay of the current and hence maintains the magnetism
> in
> the coil long after the drive has been removed.
>
> Does this make sense ?  If so, any suggestions how to improve the
> situation ?  I've tried the circuit without the diode, but the large
> voltage spike that resulted seem to get into the photocell
> amplifiers
> and cause false detection of the sproket holes.

Your analysis is correct.
Adding a resistor in series with the diode will decrease the
"slugging" effect on the solenoid at the cost of having some flyback
voltage occur. The voltage will  jump essentially instantaneously to
IR (I = current at off time, R = added resistance) and you can
dimension R to suit. The capacitor will also have to be discharged by
the coil - leave it out if you can. Using a reverse biased zener
across the coil also can work - voltage will jump to zener voltage at
turn off. Add an R across the zener where IR>Vzener - zener will clamp
voltage spike initially and resistor will dissipate the final energy.
(Voltage will rise to zener voltage until stray capacitance stores
reminder of 1/2LI^2 energy.



       RM

2005\03\14@161743 by Peter

picon face

Put a zener with a value of 1/2 the solenoid voltage in series with the
diode, cathode to + rail. If you switch the solenoid often calculate
dissipation on it.

Peter

2005\03\15@062459 by Peter Onion

flavicon
face
On Mon, 2005-03-14 at 13:27, Peter Onion wrote:
> Hi,
>
> I'm trying to use a PIC to control an old paper tape reader like this
> one...

Thanks for all the helpful replies.

I tried adding a zener as suggested and it has fixed the problems :-)

Just to answer a point made by Olin Lathrop
"However, 15mS for a mechanical system to start moving doesn't sound
that bad in the first place."

The only thing that actually "moves" is the paper tape.  The break
mechanism is just a metal bar that grips the tape as it passes between
the bar and the solenoid "E" shaped core.

---------------
|  BRAKE BAR  |
---------------  

-----TAPE------

|-|   |-|   |-|
| |   | |   | |
| | [COIL]  | |
| | [COIL]  | |
| |_|    |__| |
|             |
|_____________|

The brake can stop the tape from full speed (500 char/sec)
in less than one character time (2mS).  I'll take some measurements of
the time it takes to read a single char from stationary and post them
here.

Thanks again for all your helpful suggestions.

Peter

2005\03\15@070917 by Russell McMahon

face
flavicon
face
> The brake can stop the tape from full speed (500 char/sec)
> in less than one character time (2mS).  I'll take some measurements
> of
> the time it takes to read a single char from stationary and post
> them
> here.

It probably effectively stops the tape on a character by character
basis at full speed. 500 cps = 50 inches per second = 2.8 mph so it's
not QUITE as impressive as it sounds at first, but the acceleration is
probably impressive.


       RM

2005\03\15@114720 by Peter Onion

flavicon
face
On Tue, 2005-03-15 at 12:09, Russell McMahon wrote:

> It probably effectively stops the tape on a character by character
> basis at full speed.

At full speed the brake is OFF all the time.  The only program that
could read continuously at full speed was the tape bootstrap/loader
program.

>  500 cps = 50 inches per second = 2.8 mph so it's
> not QUITE as impressive as it sounds at first, but the acceleration is
> probably impressive.

I can't fault your maths, but 2.8 mph makes it sound slow, where as to
watch the tape spew out of the reader is quite impressive.

It will be nice to see the reader working under the control of original
software running on my emulator.

Peter

2005\03\18@063136 by Peter Onion

flavicon
face
On Tue, 2005-03-15 at 16:47 +0000, Peter Onion wrote:

> It will be nice to see the reader working under the control of original
> software running on my emulator.
>

I've finally got this working.  The reader read a 10,000 character tape
without errors.  The emulator was running an Algol compiler at the time
so the tape was being "checked" as it was read.

There are a few pictures on my web site

www.btinternet.com/~Peter.Onion/reader/reader2.jpg
The reader with the top cover off

www.btinternet.com/~Peter.Onion/reader/reader3.jpg
Photodiode amplifiers and solenoid drivefs (sorry it's a bit blured)

www.btinternet.com/~Peter.Onion/reader/reader4.jpg
My 16F877 development board.

www.btinternet.com/~Peter.Onion/reader/reader5.jpg
The unwound tape on the floor after successfully being read.

Peter.


> Peter
>

2005\03\18@091524 by Alan B. Pearce

face picon face
> www.btinternet.com/~Peter.Onion/reader/reader5.jpg
>The unwound tape on the floor after successfully being read.

Ahh, now that is the fun part :)))))))))

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...