Searching \ for '[PIC]: Mosfet speed control of 12V DC motor - Par' 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/power/batterys.htm?key=12v
Search entire site for: 'Mosfet speed control of 12V DC motor - Par'.

Exact match. Not showing close matches.
PICList Thread
'[PIC]: Mosfet speed control of 12V DC motor - Par'
2001\10\09@085839 by Gary Neal

flavicon
face
Guys,

        Thanks for the help.  Flyback Diodes!  Daaaa!  Completely forgot
that.  I put a 1N4004 (only 1A, but the biggest one I had lying around) on
the motor and it worked like a charm.  I'm running at about 20khz.  This
diode only gets hot when the duty cycle is less than about 80%.
        I don't have a whole lot of experience sizing these flyback
diodes.  Someone said it should be rated for as much as the full current of
the motor (~30A).  A 30A diode is HUGE!  These big diodes are all Stud
mount types it seems.  This doesn't seem very compatible to mounting to the
leads of a motor.
        Can anyone give me some insight on what to use?  Potential sources
for the product?  I'd like to keep it as cheap as possible.  Don't want to
spend $20 on a single diode when I've only got about a total of $10 in all
the other parts combined.
        Also, any other methods to help this reverse current from the
motor?  Slower PWM frequency (audible noise doesn't really matter to
me)?  Some configuration of caps?  I'm all ears when it comes to you
experts  :-)

Thanks,

Gary Neal


At 08:32 AM 10/8/01 -0700, Bob Blick wrote:
{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@093234 by Allen Mahurin

picon face
If you need the diodes to handle more current, but
want to keep the size and cost down, what about
paralleling (is that a word?) more small diodes?  You
say the one you're using doesn't even get warm until
you're below 80% ... what about putting 4 or so of
these diodes in parallel across the motor?

Good luck,

ATM

--- Gary Neal <.....gln103KILLspamspam@spam@MAIL.PSU.EDU> wrote:
{Quote hidden}

__________________________________________________
Do You Yahoo!?
NEW from Yahoo! GeoCities - quick and easy web site hosting, just $8.95/month.
http://geocities.yahoo.com/ps/info1

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@093852 by Gary Neal

flavicon
face
        Ya, I thought about paralleling the diodes too.  Just didn't know
if that was something that was good to do or commonly done.  Any comments
on that suggestion?

Gary




At 06:31 AM 10/9/01 -0700, Allen Mahurin wrote:
{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@101753 by Olin Lathrop

face picon face
>          Thanks for the help.  Flyback Diodes!  Daaaa!  Completely forgot
> that.  I put a 1N4004 (only 1A, but the biggest one I had lying around) on
> the motor and it worked like a charm.  I'm running at about 20khz.  This
> diode only gets hot when the duty cycle is less than about 80%.
>          I don't have a whole lot of experience sizing these flyback
> diodes.  Someone said it should be rated for as much as the full current
of
> the motor (~30A).

Yes it should.  For this purpose the motor is an inductor and whatever
current is running thru it when it gets shut off will continue somewhere -
at whatever voltage it takes.  You want that somewhere to be the diode so it
needs to handle the maximum motor forward current.

> A 30A diode is HUGE!  These big diodes are all Stud
> mount types it seems.  This doesn't seem very compatible to mounting to
the
> leads of a motor.

As close as you can manage though.

>          Can anyone give me some insight on what to use?  Potential
sources
> for the product?  I'd like to keep it as cheap as possible.  Don't want to
> spend $20 on a single diode when I've only got about a total of $10 in all
> the other parts combined.

Including the motor!?  I don't think so.  At your voltage you should be able
to get away with a shottkey diode, which cuts the dissipation about in half.

>          Also, any other methods to help this reverse current from the
> motor?  Slower PWM frequency (audible noise doesn't really matter to
> me)?  Some configuration of caps?

20KHz is way higher than the motor needs.  At that frequency you have to be
careful to drive the FET fast enough to make sure it spends most of its time
either fully on or fully off.  The motor is a large mechanical object with
inertia.  It probably won't "notice" the individual pulses above 100Hz or
so, but you want to avoid excessive vibrations in the winding wires.  If
audible noise isn't an issue, I would run it around 500Hz to 1KHz.  That's
plenty slow enough for efficient switching, but fast enough so that the coil
windings don't have time to move appreciably.

There's another issue you might want to think about.  The flyback diode
causes the motor to run like a shorted generator in the off phase,
especially if this is a permanent magnet motor.  This means the motor won't
be free wheeling during the off phase but actually be "electronically
braked".  This could be undesireable.  Unfortunately the cure is a lot more
complicated than a simple diode.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, .....olinKILLspamspam.....embedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@104453 by Spehro Pefhany

picon face
At 08:57 AM 10/9/01 -0400, you wrote:
>Guys,
>
>         Thanks for the help.  Flyback Diodes!  Daaaa!  Completely forgot
>that.  I put a 1N4004 (only 1A, but the biggest one I had lying around) on
>the motor and it worked like a charm.  I'm running at about 20khz.  This
>diode only gets hot when the duty cycle is less than about 80%.

It needs to be MUCH bigger, and MUCH faster. A 1N4004 is not much use at
20kHz, you'll be putting huge current pulses through your MOSFETs during
the LONG reverse recovery of these molasses-like antediluvian rectifiers.
You can also use a similar number of parallel MOSFETs, which may have less
power dissipation than the diode, but will be more costly.

>         I don't have a whole lot of experience sizing these flyback
>diodes.  Someone said it should be rated for as much as the full current of
>the motor (~30A).  A 30A diode is HUGE!  These big diodes are all Stud
>mount types it seems.  This doesn't seem very compatible to mounting to the
>leads of a motor.

They don't have to go right on the leads, right by the MOSFETs is fine.
You'll need a heat sink. The wires will radiate RFI in any case.

>         Can anyone give me some insight on what to use?  Potential sources
>for the product?  I'd like to keep it as cheap as possible.  Don't want to
>spend $20 on a single diode when I've only got about a total of $10 in all
>the other parts combined.

Look at: http://www.onsemi.com/pub/Collateral/MBR3045WT-D.PDF

How about an ONsemi or IR MBR3045WT 30A, 45V Vf=0.60V. in TO-247 case for
only $1.10 in  100s (USD). Worst case is probably around 50% (15A average),
but I have not worked it out in detail, if that's right you'll need to
get rid of around 11W. Parallel the two diodes in the '3045.

>         Also, any other methods to help this reverse current from the
>motor?  Slower PWM frequency (audible noise doesn't really matter to
>me)?  Some configuration of caps?  I'm all ears when it comes to you
>experts  :-)

No, you need to direct this current through the motor. It's what you
need to maintain decent speed control (it's like applying a variable
DC voltage to the motor). You can slow down the PWM and you'll get more
noise and less heat loss in the MOSFETs (maybe not all that significant)
but you'll still need a diode or an equivalent active device, and it
will get about as hot.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
EraseMEspeffspam_OUTspamTakeThisOuTinterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
     /.-.\
    (( * ))
     \\ //     Please help if you can:
      \\\      dailynews.yahoo.com/fc/US/Emergency_Information/
     //\\\
    /// \\\
    \/   \/

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@112018 by Douglas Butler

flavicon
face
The flyback diodes shouldn't need to be any bigger (or more expensive)
than the MOSFETS you are already using.  For convienient mounting (at
the expense of cost) you could use another of the same MOSFET, using its
inherent reverse diode as your flyback diode.  Real diodes should be
cheaper, and Schottkeys would run cooler.

Sherpa Doug

> {Original Message removed}

2001\10\09@112826 by Alan B. Pearce

face picon face
>> A 30A diode is HUGE!  These big diodes are all Stud
>> mount types it seems.  This doesn't seem very compatible
>>to mounting to the leads of a motor.

>As close as you can manage though.

At these currents you will also need to make sure that the wires go to the
diode then to the motor. Do not have the diode on a  side spur off the wires
going to the motor. Make sure the wires go to the diode first then the
motor, not the motor then the diode. The stray resistances and inductances
will play havoc and cause little spikes that will be hard to see on any
oscilloscope and will take you back to the havoc you had. :)

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@113902 by artstar

flavicon
face
No, actually there are plenty of high current diodes out there in TOP-3
packages which would suit nicely too, but you're right, they're big but
not as cumbersome as the stud mount diodes.

Adios,
LarZ

---------------  TAMA - The Strongest Name in Drums  ---------------

{Original Message removed}

2001\10\09@114706 by Olin Lathrop

face picon face
> If you need the diodes to handle more current, but
> want to keep the size and cost down, what about
> paralleling (is that a word?) more small diodes?  You
> say the one you're using doesn't even get warm until
> you're below 80% ... what about putting 4 or so of
> these diodes in parallel across the motor?

If you're going to do that put a small resistor in series with each diode.
It should drop a 1/2 volt or so at max current.  Otherwise one diode has a
slightly lower on voltage so it gets more current so it gets warmer so its
on voltage drops a even more so it gets even more current so it gets even
warmer ... poof.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olinspamspam_OUTembedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@115734 by Gary Neal

flavicon
face
        OK, let me get this straight.  I could use the same type of
MOSFETS I am currently using for the drive side as the flyback
device.  Just use the internal diode in the MOSFETS?  Would I just connect
the gate straight to ground?  I've got extra MOSFETS lying around and I've
got extra room on my heat sink.  Should be able to just wire them up and
try it out.  That would save me an order to Digikey/Newark :-)
        IRF says the IRLZ44N can handle a "Continuous Source Current (Body
Diode)" of 47A.  Does that mean the internal diode in this thing can handle
up to 47A (with proper heat sinking)?
        I'm a little confused about the efficiency and corresponding
output power of the entire system with this flyback diode in there
now.  This is a portable application, so efficiency and power are a concern
(lesser of a concern than learning stuff and making it run though).
        Anyway, the way I see it, if I use a MOSFET for the flyback diode,
there's ~1.3V drop across it's diode.  That means more heat dissipated in
the MOSFET.  If I use a Schottkey diode, ~.6V drop across diode, less heat
dissipation in diode, but that extra energy that's not dissipated in the
diode has to go somewhere right?  Does that mean that the motor would
correspondingly see more reverse current and heat up more (and also not put
out as much power)?  Does it mean that the overall system efficiency is
"about" the same no matter what device I use in there?

Gary




At 11:13 AM 10/9/01 -0400, Douglas Butler wrote:
>The flyback diodes shouldn't need to be any bigger (or more expensive)
>than the MOSFETS you are already using.  For convienient mounting (at
>the expense of cost) you could use another of the same MOSFET, using its
>inherent reverse diode as your flyback diode.  Real diodes should be
>cheaper, and Schottkeys would run cooler.
>
>Sherpa Doug
>
> > {Original Message removed}

2001\10\09@120947 by Spehro Pefhany

picon face
At 11:54 AM 10/9/01 -0400, you wrote:

> If I use a Schottkey diode, ~.6V drop across diode, less heat
>dissipation in diode, but that extra energy that's not dissipated in the
>diode has to go somewhere right?  Does that mean that the motor would
>correspondingly see more reverse current and heat up more (and also not put
>out as much power)?  Does it mean that the overall system efficiency is
>"about" the same no matter what device I use in there?

No, the energy that the diode dissipates is lost forever, the energy that goes
into the motor is, in part, converted to work at the shaft. Less drop =
more efficiency - you'd otherwise have to crank the PWM up, and thus use more
energy total. At 12V, the difference in losses is not dramatic, however.

Best regards,

Spehro Pefhany --"it's the network..."            "The Journey is the reward"
@spam@speffKILLspamspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
     /.-.\
    (( * ))
     \\ //     Please help if you can:
      \\\      dailynews.yahoo.com/fc/US/Emergency_Information/
     //\\\
    /// \\\
    \/   \/

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@121814 by Gary Neal

flavicon
face
        OK, I was thinking the energy that was going back into the motor
was restricting the motor from turning.  I guess this isn't the case.  That
energy is kind of keeping the motor turning?

Gary



At 12:14 PM 10/9/01 -0400, you wrote:
{Quote hidden}

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@122858 by cision Electronic Solutions

flavicon
face
Gary,
Although I don't know the particulars (check sites on electric vehicles),
you can use the diode and large capacitors to capture the energy.  A good
friend of mine used this technique with a three phase motor used in a energy
efficient drive system.  He had size constraints too, so used arrays of high
capacitance ceramic capacitors for the energy storage.  The energy was used
to supply a dc/dc converter used for, among other things, the 3-phase
generator/ controller electronics.  It would be interesting to use the
second set of MOSFETs as synchronous rectifiers which turn on as the main
set turns off.  You would still need diodes to handle the crossover time so
that no cross conduction occurred, but the internal diode may serve that
purpose.  And, yes, you might need a floating gate drive.

Have fun.

Ed

{Original Message removed}

2001\10\09@131122 by Douglas Butler

flavicon
face
Generally the voltage and current ratings of a MOSFET reverse diode are
the same as those of the FET.  Tie the Gate to the Source so the FET is
never accidentally turned on.

Now we get to the part I have to think about, and I might get wrong.
For a common permanent magnet motor current is roughly proportional to
torque.  The current that flows when the PWM switch is off flows with a
negative voltage so it is braking torque.  If the PWM switch was
mechanical and you just switched it off regardless of voltage there
would be no current, no braking torque, and the motor would just coast
to a stop.  If you mechanically shorted the motor the running current
would be diverted through the switch, the motor would see its running
torque in reverse and slow down asymptotically to zero.

Assuming the PWM happens so fast that speed is constant, torque is power
and torque x time is energy.  This means you want to stop the current
quickly for high efficiency.  If the source of the current is the
inductance then storage time should be R x L.  Therefore reducing R, or
reducing diode voltage drop, should shorten the period of braking torque
and increase efficiency.

I am open to corrections on that last part...

Sherpa Doug


> {Original Message removed}

2001\10\09@132343 by Jay Hanson

flavicon
face
>It needs to be MUCH bigger, and MUCH faster. A 1N4004 is not much use
>at 20kHz, you'll be putting huge current pulses through your MOSFETs
>during the LONG reverse recovery of these molasses-like antediluvian
>rectifiers.

I am fairly new to this stuff, but can't one use fully avalanche rated power
mosfets such as the 140 amp IRL3803
http://www.irf.com/product-info/datasheets/data/irl3803.pdf without external
diodes?

If one needs a really big diode for voltage clamping, how about a "Transient
Voltage Suppressor"?  Isn't that what they are for?  For example, this one
costs 74 cents and can dissipate 1,500 watts peak.
http://www.crydom.com/assets/tvs/1.5ke.pdf

Jay

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@142603 by Spehro Pefhany

picon face
At 12:14 PM 10/9/01 -0400, you wrote:
>         OK, I was thinking the energy that was going back into the motor
>was restricting the motor from turning.  I guess this isn't the case.  That
>energy is kind of keeping the motor turning?

Yes, it does. As long as the PWM frequency is high enough that the ripple
through the motor inductance can be ignored it's like running the motor
from a voltage source of Vin * ton/(ton + toff). The motor current flows
through the MOSFET during ton and through the diode during toff, and is
essentially constant (though the voltage at the motor terminal wags
around). Ignore the comments that imply there's something wrong with this,
it's how PWM motor control works.

Best regards,
Spehro Pefhany --"it's the network..."            "The Journey is the reward"
RemoveMEspeffTakeThisOuTspaminterlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com
     /.-.\
    (( * ))
     \\ //     Please help if you can:
      \\\      dailynews.yahoo.com/fc/US/Emergency_Information/
     //\\\
    /// \\\
    \/   \/

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\09@145636 by Douglas Butler

flavicon
face
Ahh! So the PWM frequency should be high enough so that the current
through the motor windings is nearly constant (DC + ripple).  You don't
want to stop the current at all, just give it a place to go without
loosing too much energy in diode losses.  That makes much more sense to
me than my previous ramblings.

Sorry if I threw mud in the water.

Sherpa Doug

> {Original Message removed}

2001\10\09@155059 by Brent Brown

picon face
> Ahh! So the PWM frequency should be high enough so that the current
> through the motor windings is nearly constant (DC + ripple).  You
> don't want to stop the current at all, just give it a place to go
> without loosing too much energy in diode losses.  That makes much more
> sense to me than my previous ramblings.

Yes (jumping in to the tail end of the discussion). It's difficult to
think through all the theory required here, and very easy to come to
a wrong conclusion.

So just to reinforce what you said...during the "off" phase you want
the armature current to carry on flowing as if nothing has happened.
The flyback diode provides a path for this current to flow in a loop.
Ideally the diode would have no voltage drop and the armature
would have zero resistance. As you know, when current is flowing
in a cuircuit and there is zero resistance there is no power
dissipated. Kind of like the current in the coil of a superconducting
magnet, for a short period of time.

Brent Brown
Electronic Design Solutions
16 English Street
Hamilton, New Zealand
Ph/fax: +64 7 849 0069
Mobile/text: 025 334 069
eMail:  spamBeGonebrent.brownspamBeGonespamclear.net.nz

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads


2001\10\10@061517 by John

flavicon
face
Hello Olin & PIC.ers,

>Date:    Tue, 9 Oct 2001 09:41:10 -0400
>From:    Olin Lathrop <TakeThisOuTolin_piclistEraseMEspamspam_OUTEMBEDINC.COM>
>Subject: Re: [PIC]:  Mosfet speed control of 12V DC motor - Partially
>         Solved

<snippo>

I'm a wee bit confused by this paragraph....

>
>There's another issue you might want to think about.  The flyback diode
>causes the motor to run like a shorted generator in the off phase,
>especially if this is a permanent magnet motor.  This means the motor won't
>be free wheeling during the off phase but actually be "electronically
>braked".  This could be undesireable.  Unfortunately the cure is a lot more
>complicated than a simple diode.
>
>

Surely the diode just allows a path for continuation of the armature
current in the `off phase', at least until it dies away to zero.
The current direction in the armature is unchanged, so the rotor is
still motoring, producing torque in the forward direction (gotta be..?).

Thereafter, the armature will act as a generator,  with the output
polarity in the same sense as it is during the `on phase'.
How can the diode conduct at all while the armature is generating
e.m.f. reversed across it?


       best regards,   John

e-mail from the desk of John Sanderson, JS Controls.
Snailmail:          PO Box 1887, Boksburg 1460, Rep. of South Africa.
Tel/fax:            Johannesburg  893 4154
Cellphone no:   082 741 6275
email:                RemoveMEjsandspamTakeThisOuTpixie.co.za
Manufacturer & purveyor of laboratory force testing apparatus, and related
products and services.

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


2001\10\10@074404 by Olin Lathrop

face picon face
> >There's another issue you might want to think about.  The flyback diode
> >causes the motor to run like a shorted generator in the off phase,
> >especially if this is a permanent magnet motor.  This means the motor
won't
> >be free wheeling during the off phase but actually be "electronically
> >braked".  This could be undesireable.  Unfortunately the cure is a lot
more
{Quote hidden}

Yeah, you're right.  That's what I get for knee-jerk answering without
thinking it thru right.  At least someone on the PIClist is guaranteed to
catch mistakes like that.

I ran into a problem like I described once, but the differences is this was
a multi pole motor with separate leads for each stator field pole.  Each
winding was driven 120 degrees out of phase using PWM.  The problem was that
the flyback diodes of the *other* windings were making them act like shorted
generators.  One or two windings were producing torque, and the remaining
winding or windings were essentially causing viscous friction.

However, none of this applies to a DC motor, as you explained.


********************************************************************
Olin Lathrop, embedded systems consultant in Littleton Massachusetts
(978) 742-9014, olinEraseMEspam.....embedinc.com, http://www.embedinc.com

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics


More... (looser matching)
- Last day of these posts
- In 2001 , 2002 only
- Today
- New search...