Searching \ for '[OT] Implementing electric brake for a small DC mo' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/index.htm?key=implementing+electric
Search entire site for: 'Implementing electric brake for a small DC mo'.

Exact match. Not showing close matches.
PICList Thread
'[OT] Implementing electric brake for a small DC mo'
1998\08\17@140256 by Shahid Sheikh

picon face
Greeting everyone,

Anyone has any idea on how to implement an electric brake for a small DC
motor driven by a H bridge mosfet arrangement. The normal running current
the motor consumes is about 300 ~ 400 mA at full speed and average load.

The problem is that the motor has a small flywheel attached to it which
keeps it going for a while after the supply has been shut off.

This is for a model train decoder and I'm trying to figure out how I'm going
to implement an emergency stop. The motor's speed and direction is
controlled by a PIC using PWM. Right now I dont have any plans of using back
EMF for speed control so I'm trying to find a braking solution that doesn't
use back EMF.

Thanks,

Shahid

1998\08\17@163058 by Mark Willis

flavicon
face
Wheelchairs use electric clutches (similar to Car Air Conditioning
clutches), unclutch then apply power to the motor (so you could use that
sort of thing.)  Really a fast way to stop the motor (locks the wheels!)
but not always scalable, and eats a bit of power.  Could use a solenoid
and a friction brake?  (2 coils, 1 for Go, 1 for Brakes On, just apply
friction from the piece moved by the solenoid's core, to the flywheel?)

 Or, put a resistor in series with a "relay" (optoisolated analog
switch?) across the motor, close relay to brake, using the motor as a
generator.  Smaller resistor, faster braking.  (Could PWM the switch, to
proportional brake.)

 Not my field of expertice (obviously) but a couple thoughts.  Need to
get back current on this for one project next year <G>

 Mark Willis, spam_OUTmwillisTakeThisOuTspamnwlink.com

Shahid Sheikh wrote:
{Quote hidden}

1998\08\17@165151 by Mike Keitz

picon face
On Mon, 17 Aug 1998 13:59:28 -0400 Shahid Sheikh <.....sheikhKILLspamspam@spam@EROLS.COM>
writes:

>Anyone has any idea on how to implement an electric brake for a small
>DC
>motor driven by a H bridge mosfet arrangement.


>The problem is that the motor has a small flywheel attached to it
>which
>keeps it going for a while after the supply has been shut off.

The usual method with a PM motor is to turn on both "top" or "bottom"
switches of the bridge, so the motor sees a short circuit.  It will stop
a lot faster than just letting it coast down.  This is about the best you
can do without using power to drive the motor backwards until it stops.
If the bridge is made of MOSFETS, a capacitor could store voltage to keep
driving two of them for braking after the power is lost.  Of course you
need to be careful that the brake circuit doesn't interfere with normal
operation.


_____________________________________________________________________
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
Or call Juno at (800) 654-JUNO [654-5866]

1998\08\17@180630 by Dennis Plunkett

flavicon
face
>This is for a model train decoder and I'm trying to figure out how I'm going
>to implement an emergency stop. The motor's speed and direction is
>controlled by a PIC using PWM. Right now I dont have any plans of using back
>EMF for speed control so I'm trying to find a braking solution that doesn't
>use back EMF.
>
>Thanks,
>
>Shahid
>
>

Hmm,
Be very carful here, as there are some model trains that use a worm drive
gear, and will trend to brake on thier own flywheel or not.


Dennis

1998\08\17@191135 by paulb

flavicon
face
Shahid Sheikh wrote:

> The problem is that the motor has a small flywheel attached to it
> which keeps it going for a while after the supply has been shut off.

 This is empirically proven?  (In other words, you *have* tried
disconnecting the power and it *does* keep on running more than wanted?)

> This is for a model train decoder and I'm trying to figure out how I'm
> going to implement an emergency stop.

 Relating to a previous reply, do you need emergency stop on power
disconnection (e.g., off rails) or emergency stop when the controller is
so operated?

 Most, if not all small model trains have the characteristic that the
motor stops very quickly when the power is removed, even with a flywheel
which may be fitted to improve this.  This is so unrealistic that
controllers are specially made to simulate "inertia", and the "emergency
stop" control simply zeroes the "inertia" analog (which may be digital
or analog, the latter usually a capacitor).

 The situation you describe is very unusual, the opposite problem.
Naturally, most of us are quite surprised this would happen!

> The motor's speed and direction are controlled by a PIC using PWM.
> Right now I dont have any plans of using back EMF for speed control so
> I'm trying to find a braking solution that doesn't use back EMF.

 We don't see the connection.  Back EMF *is* used for braking,
independent of whether you want to measure it, by simply shorting the
motor.  You could use relays as suggested, but if you have a MosFet H-
bridge already in place, switching on both ground-side FETs and neither
supply-side one will do the job perfectly.
--
 Cheers,
       Paul B.

1998\08\17@230916 by russellh

picon face
Shahid Sheikh wrote:
>
> Greeting everyone,
>

<snip>
>
> The problem is that the motor has a small flywheel attached to it which
> keeps it going for a while after the supply has been shut off.
>

The standard for this doesn't specify an actual stopping distance. It
only specifies that the motor will be programmed to zero as soon as the
emergency stop code is received. I think that if you remove power, that
should be sufficient.

And a sudden stop can be almost as dangerous as a collision.

> This is for a model train decoder and I'm trying to figure out how I'm going
> to implement an emergency stop.

<snip>

Have you downloaded the Digital Command Control specifications from the
NMRA Website?

Good luck!

Russell Hedges

1998\08\18@100213 by Allen Demers

flavicon
face
I have an idea...
Use a pic to trigger a small solenoid. The solenoid will trigger a small
catapult to throw a 100gram weight off the train bringing it to a halt.
;)
(sorry for the bandwidth, couldn't help it!)

================================================================
Subject:
       [OT] Implementing electric brake for a small DC motor
 Date:
       Mon, 17 Aug 1998 13:59:28 -0400
 From:
       Shahid Sheikh <sheikhspamKILLspamEROLS.COM>


Greeting everyone,

Anyone has any idea on how to implement an electric brake for a small DC
motor driven by a H bridge mosfet arrangement. The normal running
current
the motor consumes is about 300 ~ 400 mA at full speed and average load.

The problem is that the motor has a small flywheel attached to it which
keeps it going for a while after the supply has been shut off.

This is for a model train decoder and I'm trying to figure out how I'm
going
to implement an emergency stop. The motor's speed and direction is
controlled by a PIC using PWM. Right now I dont have any plans of using
back
EMF for speed control so I'm trying to find a braking solution that
doesn't
use back EMF.

Thanks,

1998\08\19@020832 by alex dinovitser

picon face
The answer to this question depends on what sort of motor is being
used. Since it is and H bridge, I assume it is a small DC motor.
In this case, the best way to stop it is simply by shorting the motor
terminals. If you can override the H-bridge logic, you would simply
turn on both ground-connected mosfets (with both vss connected mosfets
turned off, or vice versa), otherwise use a relay that disconnects the
H bridge from the power supply and shorts the motor terminals.
This simple method will stop the motor much faster than it would stop
otherwise. Tell us how it goes....




> Greeting everyone,
>
> Anyone has any idea on how to implement an electric brake for a
small DC
> motor driven by a H bridge mosfet arrangement. The normal running
> current
> the motor consumes is about 300 ~ 400 mA at full speed and average
load.
>
> The problem is that the motor has a small flywheel attached to it
which
> keeps it going for a while after the supply has been shut off.
>
> This is for a model train decoder and I'm trying to figure out how I'm
> going
> to implement an emergency stop. The motor's speed and direction is
> controlled by a PIC using PWM. Right now I dont have any plans of
using
> back
> EMF for speed control so I'm trying to find a braking solution that
> doesn't
> use back EMF.
>
> Thanks,
>

==
------------------------------------------------------------------
Alex Dinovitser
PhD Student                            ph:  +61 8 8302 1775
Transport Systems Centre               fax: +61 8 8302 1880
City East Campus
University of South Australia
Adelaide       5000
_________________________________________________________
DO YOU YAHOO!?
Get your free @yahoo.com address at http://mail.yahoo.com

1998\08\19@031327 by Alan King

picon face
In the same vein as someone else's reply, the most sure
way I can think of is to put a solenoid at the rear of the
engine, aiming straight down.  Drive a piece of steel
between the plastic railroad ties and you should stop within
1/2" when it hits the next one..  Not sure if the rest of
the cars would stay on track though..


> This simple method will stop the motor much faster than it would stop
> otherwise. Tell us how it goes....

1998\08\19@071445 by Russell McMahon

picon face
Shorting the motor will provide good braking. This could be a relay,
or an extra FET/Transistor across the motor. A resistor can be used
in series with the shorting contact to reduce braking power.

If you want simplicity and don't mind the additional power
consumption you could place a resistor permanently across the motor.
When its powered on it draws extra power and when its off it provides
braking. Power consumption of resistor would probably need to be
similar to motor power to provide reasonable braking. Easily tried.

A solution that may work and which takes NO extra components is to
turn on only the two bottom or two top drivers. One will be forward
biased by the motor emf and one will be reverse biased. The intrinsic
substrate diode should provide a current return path. You could add a
back diode if you want to - a Schottky unit would largely keep
current out of the FET. Both eg bottom FETS need to be turned on if
motor direction is unknown. Only 1 is required if direction is known.

The same trick could be done with one additional driver in parallel
with one of eg the bottom drivers but with a small resistor in series
with this ancillary driver. The resistor will dissipate most of the
braking energy if properly sized.

-->Anyone has any idea on how to implement an electric brake for a
small DC
>motor driven by a H bridge mosfet arrangement. The normal running
>current
>the motor consumes is about 300 ~ 400 mA at full speed and average
load.
>
>The problem is that the motor has a small flywheel attached to it
which
>keeps it going for a while after the supply has been shut off.
>
>This is for a model train decoder and I'm trying to figure out how
I'm
>going
>to implement an emergency stop. The motor's speed and direction is
>controlled by a PIC using PWM. Right now I dont have any plans of
using
>back
>EMF for speed control so I'm trying to find a braking solution that
>doesn't
>use back EMF.
>
>Thanks,
>

1998\08\19@100031 by Bob Blick

face
flavicon
face
If you want to see a typical way of implementing braking within a
traditional H-bridge, look at the one on my page:

http://www.bobblick.com/bob/projects/hbridge/index.html

Even though I use Darlington transistors as the output devices, the method
I use to get the right logic for braking can be applied to other bridges.

0 - 0   off
1 - 0   forward
0 - 1   reverse
1 - 1   brake

Note that with some designs of hbridges, the 1 - 1 input causes
destruction of the bridge.

Cheers,
Bob

P.S. I don't know how far this is off from the original post, I haven't
been following the long thread or know who the original poster was.

1998\08\19@115921 by Peter L. Peres

picon face
Ok, I still don't know why you want to emulate an emergency barke with a
model train worth good money probably, but here goes my solution (which is
what is used for braking in certain servos):

Given a travel direction, when the stop order comes, invert the voltage on
the motor by driving the 'H' switch backwards (after closing the present
path). Some way to read speed (which is not necessarily RPM on the motor)
is required. You stop the motor totally by opening both lower switches in
the H switch when v == 0. This kind of braking can shear gears and have
the loco wheels turn backwards (just like the emg. brake on a steam engine
used to be). In real life, a traction loss sense system throttles the
motor while doing this to avoid slip (which causes worse braking than with
contact). You could sense the current decrease in the motor when this
happens, but that is overkill imho. If you have a free axle in the loco
you can add an opto pulse sensor to it to provide v information to the
controller while doing this. I suppose that this could be quite
spectacular. Don't forget to do something real smart about the controller
power supply as there will be showers of sparks at the wheels (looks great
in the dark - used to do it with a manual transformer when I was a kid).

Peter

1998\08\19@214341 by russellh

picon face
We are all assuming that the power to the rails will be DC. If it is AC,
then the H bridge won't really be necessary, will it? And braking by
reverse polarity for a short time could be implimented all in firmware.

Russell Hedges

Trying to bring this discussion back to the embedded controller side of
the problem.

1998\08\20@113901 by Peter L. Peres

picon face
On Wed, 19 Aug 1998, Russell Hedges wrote:

> We are all assuming that the power to the rails will be DC. If it is AC,
> then the H bridge won't really be necessary, will it? And braking by
> reverse polarity for a short time could be implimented all in firmware.

I don't think that they are using 400 Hz in model trains and at 50 Hz a
base frequency chopped switch will make the motor sing quite loudly. It
usually sings even with slightly smoothed DC from a regular transformer
and this is extremely annoying to purists who spen undescribable amounts
of money on control equipment.

I also don't think that there is room for a serious smoothing capacitor
(which must be bipolar anyway). All electronic control model trains I've
taken apart (i.e. electronics in the loco, not just a Maerklin bipolar
relay), used a bridge rectifier, a small smoothing cap, and a chopped
transistor 'H'.

Some more daring designs had the choppers built into the bridge. There is
no room for anything in there, least of them all heatsinks for anything,
no venting holes, plus everything is plastic, so if it gets hot, bye-bye
nice loco coat.

> Trying to bring this discussion back to the embedded controller side of
> the problem.

Yep. A 12 C 508 PIC could take care of such a loco with room to spare,
even some telemetry (train speed feedback to base station ?) ;).

Peter

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