Sorry, an error occurred while loading the content.

## Re: [SeattleRobotics] Circuit advice

Expand Messages
• good thinking. I will keep this in mind also. ... From: Larry Barello To: SeattleRobotics@yahoogroups.com Sent: Saturday, July 31, 2004 4:17 PM Subject: RE:
Message 1 of 13 , Jul 31, 2004
good thinking.
I will keep this in mind also.

----- Original Message -----
From: Larry Barello
To: SeattleRobotics@yahoogroups.com
Sent: Saturday, July 31, 2004 4:17 PM
Subject: RE: [SeattleRobotics] Circuit advice

The power dissipation rating of a MOSFET is meaningless without knowing what
the heat sink is capable of. It makes NO sense to try to calculate the
current capacity of the device based upon the rated power dissipation!

It is probably best to design for free-air dissipation. A 4 milli-ohm part
will have roughly 6 mohm resistance at max temp of 175C. The TO220 package
has about 2 watts dissipation capacity in free air (room temp, device at max
temp). Watts = current * voltage. Voltage = resistance * current. So max
current/device is square root of watts (2)/resistance (.006) or about
18A/device. All this information is on the typical MOSFET data sheet.

I suspect the solenoid being used to modulate the NO is not variable, but
rather ON/OFF and the ratio of ON vs OFF at a particular supply pressure,
will inject a variable amount of NO into the manifold. There should be no
need for a table of PWM to flow. The PWM rate is low enough that the valve
follows the PWM, just like the current follows the PWM in a MOSFET.

Since the solenoid is on for something less than 100% of the time, it is
likely that the AVERAGE current could be used to design the capacity. In
any case two 1404's should be plenty. If you are worried, bolt them to a
small amount of metal to buffer the temperature rise for the few seconds
that the NO is flowing.

You are not planning on letting it flow for a long time, right? Even at 35A
(stated current rating of the solenoid), James could probably get away with
one FET for 10-20 seconds at a time with no or little metal bolted to the
device.

I helped design a dual 80A combat robot controller (www.robot-solutions.com
RS80D) which uses 2x1404's in each leg and has a small aluminum block for
heat-sink. It handles 80A just fine for a 4-5 minute match, but, of course,
overheats if you run 30A through it continuously. Actually, it doesn't
overheat since I fold back the drive as the temperature rises and cut it off
completely by 130C. Anyway, the point being that even at 40A/device not
much heat sinking is needed for short periods of time.

Cheers!

-----Original Message-----
From: tbrenke@...
..
mosfet:
the IRF3709Z has an on restance of .0078 at 5V 30V max power disapation
of 79W
TO262 1.87 each at digikey
2+ are needed because of the 79W rating
the IRF1404 has an on restance of .004 at 5V 40V max power dissipation of
200W
TO220 2.52 each at digikey
1 is needed because of the 200W rating can run 5A with heat sink.
36V / 200W = 5A

Visit the SRS Website at http://www.seattlerobotics.org

[Non-text portions of this message have been removed]
• Hi Larry, Kip, tbrenke, Chuck, and list Sorry for my late reply, I was laid up in hospital for a few days last week with food poisoning and haven t had a
Message 2 of 13 , Aug 9, 2004
Hi Larry, Kip, tbrenke, Chuck, and list

Sorry for my late reply, I was laid up in hospital for a few days last
week with food poisoning and haven't had a chance to get back to this.

New version at
http://www.jamesfitzsimons.com/images/ami/temp/solenoid-driver.gif

This (hopefully) is an extraction of the appropriate part of the OSMC
circuit for driving a solenoid. Do I have the solenoid part hooked into
the OSMC circuit correctly?

Also, I just re-read Larry's last email, and will probably just use two
fets although the diagram above shows three.

Cheers,
James Fitzsimons

On Sun, 2004-08-01 at 07:00, tbrenke@... wrote:
> good thinking.
> I will keep this in mind also.
>
> ----- Original Message -----
> From: Larry Barello
> To: SeattleRobotics@yahoogroups.com
> Sent: Saturday, July 31, 2004 4:17 PM
> Subject: RE: [SeattleRobotics] Circuit advice
>
>
> The power dissipation rating of a MOSFET is meaningless without knowing what
> the heat sink is capable of. It makes NO sense to try to calculate the
> current capacity of the device based upon the rated power dissipation!
>
> It is probably best to design for free-air dissipation. A 4 milli-ohm part
> will have roughly 6 mohm resistance at max temp of 175C. The TO220 package
> has about 2 watts dissipation capacity in free air (room temp, device at max
> temp). Watts = current * voltage. Voltage = resistance * current. So max
> current/device is square root of watts (2)/resistance (.006) or about
> 18A/device. All this information is on the typical MOSFET data sheet.
>
> I suspect the solenoid being used to modulate the NO is not variable, but
> rather ON/OFF and the ratio of ON vs OFF at a particular supply pressure,
> will inject a variable amount of NO into the manifold. There should be no
> need for a table of PWM to flow. The PWM rate is low enough that the valve
> follows the PWM, just like the current follows the PWM in a MOSFET.
>
> Since the solenoid is on for something less than 100% of the time, it is
> likely that the AVERAGE current could be used to design the capacity. In
> any case two 1404's should be plenty. If you are worried, bolt them to a
> small amount of metal to buffer the temperature rise for the few seconds
> that the NO is flowing.
>
> You are not planning on letting it flow for a long time, right? Even at 35A
> (stated current rating of the solenoid), James could probably get away with
> one FET for 10-20 seconds at a time with no or little metal bolted to the
> device.
>
> I helped design a dual 80A combat robot controller (www.robot-solutions.com
> RS80D) which uses 2x1404's in each leg and has a small aluminum block for
> heat-sink. It handles 80A just fine for a 4-5 minute match, but, of course,
> overheats if you run 30A through it continuously. Actually, it doesn't
> overheat since I fold back the drive as the temperature rises and cut it off
> completely by 130C. Anyway, the point being that even at 40A/device not
> much heat sinking is needed for short periods of time.
>
> Cheers!
>
> -----Original Message-----
> From: tbrenke@...
> ..
> mosfet:
> the IRF3709Z has an on restance of .0078 at 5V 30V max power disapation
> of 79W
> TO262 1.87 each at digikey
> 2+ are needed because of the 79W rating
> the IRF1404 has an on restance of .004 at 5V 40V max power dissipation of
> 200W
> TO220 2.52 each at digikey
> 1 is needed because of the 200W rating can run 5A with heat sink.
> 36V / 200W = 5A
>
>
>
> Visit the SRS Website at http://www.seattlerobotics.org
> Yahoo! Groups Links
>
>
>
>
>
>
> [Non-text portions of this message have been removed]
>
>
>
> Visit the SRS Website at http://www.seattlerobotics.org
> Yahoo! Groups Links
>
>
>
>
--
It's 5.50 a.m.... Do you know where your stack pointer is ?
• The snubber is missing, I am not certain you need it, but it is an extra margin of safety. For DC it is across the coil. I would use a 1 W 100 Ohm and a
Message 3 of 13 , Aug 9, 2004
The snubber is missing, I am not certain you need it, but it is an extra
margin of safety. For DC it is across the coil. I would use a 1 W 100
Ohm and a 200V or 400V 0.1 MFD poly capacitor.

I am not sure of the 1.5KE51CA. In the OSMB it goes between the drain and
ground. I prefer it where you have it. I will have to think about it.

Kip

At 01:57 PM 8/9/04, you wrote:
>Hi Larry, Kip, tbrenke, Chuck, and list
>
>Sorry for my late reply, I was laid up in hospital for a few days last
>week with food poisoning and haven't had a chance to get back to this.
>
>New version at
>http://www.jamesfitzsimons.com/images/ami/temp/solenoid-driver.gif
>
>This (hopefully) is an extraction of the appropriate part of the OSMC
>circuit for driving a solenoid. Do I have the solenoid part hooked into
>the OSMC circuit correctly?
>
>Also, I just re-read Larry's last email, and will probably just use two
>fets although the diagram above shows three.
>
>Cheers,
>James Fitzsimons
>
>On Sun, 2004-08-01 at 07:00, tbrenke@... wrote:
> > good thinking.
> > I will keep this in mind also.
> >
> > ----- Original Message -----
> > From: Larry Barello
> > To: SeattleRobotics@yahoogroups.com
> > Sent: Saturday, July 31, 2004 4:17 PM
> > Subject: RE: [SeattleRobotics] Circuit advice
> >
> >
> > The power dissipation rating of a MOSFET is meaningless without
> knowing what
> > the heat sink is capable of. It makes NO sense to try to calculate the
> > current capacity of the device based upon the rated power dissipation!
> >
> > It is probably best to design for free-air dissipation. A 4
> milli-ohm part
> > will have roughly 6 mohm resistance at max temp of 175C. The TO220
> package
> > has about 2 watts dissipation capacity in free air (room temp, device
> at max
> > temp). Watts = current * voltage. Voltage = resistance *
> current. So max
> > current/device is square root of watts (2)/resistance (.006) or about
> > 18A/device. All this information is on the typical MOSFET data sheet.
> >
> > I suspect the solenoid being used to modulate the NO is not variable, but
> > rather ON/OFF and the ratio of ON vs OFF at a particular supply pressure,
> > will inject a variable amount of NO into the manifold. There should
> be no
> > need for a table of PWM to flow. The PWM rate is low enough that the
> valve
> > follows the PWM, just like the current follows the PWM in a MOSFET.
> >
> > Since the solenoid is on for something less than 100% of the time, it is
> > likely that the AVERAGE current could be used to design the capacity. In
> > any case two 1404's should be plenty. If you are worried, bolt them to a
> > small amount of metal to buffer the temperature rise for the few seconds
> > that the NO is flowing.
> >
> > You are not planning on letting it flow for a long time, right? Even
> at 35A
> > (stated current rating of the solenoid), James could probably get
> away with
> > one FET for 10-20 seconds at a time with no or little metal bolted to the
> > device.
> >
> > I helped design a dual 80A combat robot controller
> (www.robot-solutions.com
> > RS80D) which uses 2x1404's in each leg and has a small aluminum block for
> > heat-sink. It handles 80A just fine for a 4-5 minute match, but, of
> course,
> > overheats if you run 30A through it continuously. Actually, it doesn't
> > overheat since I fold back the drive as the temperature rises and cut
> it off
> > completely by 130C. Anyway, the point being that even at 40A/device not
> > much heat sinking is needed for short periods of time.
> >
> > Cheers!
> >
> > -----Original Message-----
> > From: tbrenke@...
> > ..
> > mosfet:
> > the IRF3709Z has an on restance of .0078 at 5V 30V max power
> disapation
> > of 79W
> > TO262 1.87 each at digikey
> > 2+ are needed because of the 79W rating
> > the IRF1404 has an on restance of .004 at 5V 40V max power
> dissipation of
> > 200W
> > TO220 2.52 each at digikey
> > 1 is needed because of the 200W rating can run 5A with heat sink.
> > 36V / 200W = 5A
> >
> >
> >
> > Visit the SRS Website at http://www.seattlerobotics.org
> > Yahoo! Groups Links
> >
> >
> >
> >
> >
> >
> > [Non-text portions of this message have been removed]
> >
> >
> >
> > Visit the SRS Website at http://www.seattlerobotics.org
> > Yahoo! Groups Links
> >
> >
> >
> >
>--
>It's 5.50 a.m.... Do you know where your stack pointer is ?
>
>
>
>
>Visit the SRS Website at http://www.seattlerobotics.org
>
>
>
>
• Hi Kip, ... Does this look right? http://www.jamesfitzsimons.com/images/ami/temp/solenoid-driver.gif ... I have read you need the flywheel diode across the
Message 4 of 13 , Aug 9, 2004
Hi Kip,

On Mon, 2004-08-09 at 21:12, Kipton Moravec wrote:
> The snubber is missing, I am not certain you need it, but it is an extra
> margin of safety. For DC it is across the coil. I would use a 1 W 100
> Ohm and a 200V or 400V 0.1 MFD poly capacitor.

Does this look right?
http://www.jamesfitzsimons.com/images/ami/temp/solenoid-driver.gif

> I am not sure of the 1.5KE51CA. In the OSMB it goes between the drain and
> ground. I prefer it where you have it. I will have to think about it.

I have read you need the flywheel diode across the coil of the solenoid
to allow the current flowing through the coil to dissipate once the
circuit is "off".

I really appreciate your help.

Cheers,
James Fitzsimons

--
It's 5.50 a.m.... Do you know where your stack pointer is ?
• On Mon, 09 Aug 2004 23:06:56 +0100, James Fitzsimons ... [deletia] For a fun little variation on this circuit, replace the single flywheel diode with a circuit
Message 5 of 13 , Aug 9, 2004
On Mon, 09 Aug 2004 23:06:56 +0100, James Fitzsimons
<jamesfit@...> wrote:
> Hi Kip,
>
> On Mon, 2004-08-09 at 21:12, Kipton Moravec wrote:
> > The snubber is missing, I am not certain you need it, but it is an extra
> > margin of safety. For DC it is across the coil. I would use a 1 W 100
> > Ohm and a 200V or 400V 0.1 MFD poly capacitor.
>
> Does this look right?
> http://www.jamesfitzsimons.com/images/ami/temp/solenoid-driver.gif
>
> > I am not sure of the 1.5KE51CA. In the OSMB it goes between the drain and
> > ground. I prefer it where you have it. I will have to think about it.
>
> I have read you need the flywheel diode across the coil of the solenoid
> to allow the current flowing through the coil to dissipate once the
> circuit is "off".
[deletia]

For a fun little variation on this circuit, replace the single
flywheel diode with a circuit to put the back-emf to use. For some
really simple ideas, have a look at this SunSwinger Solarbotics kit:

http://www.solarbotics.com/products/index.php?scdfa-250100084-viewDetail-productzq3884zq4categoryzq31=true

...specifically pg. 4 of the PDF (pg 6 of the PDf file) linked from
the kit page:

Enjoy!

Andrew.

--

If you don't know what to do, do something.
• this looks good to me. I guess a build and a smoke test are in order now. keep in mind what was stated by another. about 18A per device for a total of 36A with
Message 6 of 13 , Aug 9, 2004
this looks good to me.
I guess a build and a smoke test are in order now.

keep in mind what was stated by another.
about 18A per device for a total of 36A with fan.

with what you stated before (low pwm duty cycles) this should not be a problem.

----- Original Message -----
From: James Fitzsimons
To: srs
Sent: Monday, August 09, 2004 3:06 PM
Subject: Re: [SeattleRobotics] Circuit advice

Hi Kip,

On Mon, 2004-08-09 at 21:12, Kipton Moravec wrote:
> The snubber is missing, I am not certain you need it, but it is an extra
> margin of safety. For DC it is across the coil. I would use a 1 W 100
> Ohm and a 200V or 400V 0.1 MFD poly capacitor.

Does this look right?
http://www.jamesfitzsimons.com/images/ami/temp/solenoid-driver.gif

> I am not sure of the 1.5KE51CA. In the OSMB it goes between the drain and
> ground. I prefer it where you have it. I will have to think about it.

I have read you need the flywheel diode across the coil of the solenoid
to allow the current flowing through the coil to dissipate once the
circuit is "off".

I really appreciate your help.

Cheers,
James Fitzsimons

--
It's 5.50 a.m.... Do you know where your stack pointer is ?

Visit the SRS Website at http://www.seattlerobotics.org