24V Dimmer

[Josh Koffman]

Hi all. I might be called on to help out with a project. The part I'm
most unsure about is that the intensity of a 24V 250W light bulb will
need to be controlled, preferably by a PIC. It's being supplied by a
24V battery.

 I'd love to be able to find a compact, ready made module that would
do this part for me. Digital (ie serial data, etc) control would be
ideal, but analog might be possible. Alternatively I could use the CCP
module to generate some PWM, but I'd need a driver. The reason I'd
like a module is that in the event of a problem it would be easy to
swap out. I've done some searching and what I'm finding are a ton of
smaller modules meant for dimming LEDs.

Alternatively I could use PWM and some sort of home built driver. The
circuit here: http://www.solorb.com/elect/solarcirc/pwm1/ suggests the
use of a FET. Having never controlled this large a DC load before, I
have no idea if this is a good or bad idea. Any thoughts?

Thanks for the help!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Listas de Correo-2]

Hi Josh,
        I would like to recommend you to take a look at the IPS (Intelligent
Power Switch) that IR was to offer. It is under the automotive section on
www.irf.com

        An IPS is basically a MOSFET with a driver and some protection
features like short circuit, battery reversal protection and over current
among others. You can even measure the current consumption and detect open
load (load removed).

        There is a big spectrum of solutions, but I think that you need
something that can support 24V (the 70V devices are for 24 inductive loads,
since your load is resistive, you can use the 40V devices). The device
should also be able to handle 11A or more and support a switching frequency
of, at least, 100 Hz to make sure the PWM is not seen.

        There are high and low side drivers.

        I have chosed one for you, the IPS6021SPBF
(https://ec.irf.com/v6/en/US/adirect/ir?cmd=catSearchFrame&domSendTo=byID&do
mProductQueryName=IPS6021SPBF). This device can withstand up to 12A
continuous output at 85°C on the TO220 model with a heat sink. The internal
current limit is set to 32A.

Regards,

> -----Original Message-----
> From: piclist-bounces@... [mailto:piclist-bounces@...] On
> Behalf Of Josh Koffman
> Sent: Wednesday, November 04, 2009 21:55
> To: Microcontroller discussion list - Public.
> Subject: [EE] 24V Dimmer
>
> Hi all. I might be called on to help out with a project. The part I'm
> most unsure about is that the intensity of a 24V 250W light bulb will
> need to be controlled, preferably by a PIC. It's being supplied by a
> 24V battery.
>
>  I'd love to be able to find a compact, ready made module that would
> do this part for me. Digital (ie serial data, etc) control would be
> ideal, but analog might be possible. Alternatively I could use the CCP
> module to generate some PWM, but I'd need a driver. The reason I'd
> like a module is that in the event of a problem it would be easy to
> swap out. I've done some searching and what I'm finding are a ton of
> smaller modules meant for dimming LEDs.
>
> Alternatively I could use PWM and some sort of home built driver. The
> circuit here: http://www.solorb.com/elect/solarcirc/pwm1/ suggests the
> use of a FET. Having never controlled this large a DC load before, I
> have no idea if this is a good or bad idea. Any thoughts?
>
> Thanks for the help!
>
> Josh
> --
> A common mistake that people make when trying to design something
> completely foolproof is to underestimate the ingenuity of complete
> fools.
>         -Douglas Adams
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Richard Prosser]

Hi Josh

Yes, a MOSFET main switch is an easy way to do this (compared to a
bipolar) but for PWM control you are going to need someting to drive
the fet. You can either roll your own or use a driver chip to do it
for you. Depending on he frequency of operation, an IGBT may be
another option for the main switch - but again it will need a good
driver.

For a one off / limited production project I'd suggest a driver chip.

You'll really need to use PWM to control the brightness, otherwise you
need to be able to dissipate the excess heat somewhere. It's likely to
be in the 180W area for a linear controller.

One thing to watch with incandescent lamps is that the cold current
can be 10x what you'd otherwise expect. For your 250W load at 24V the
current will be around 10.4 A when  hot - it could be approaching 100
A as a turn-on surge when cold.. This means that your MOSFET will have
to be able to handle this, and the driver circuit must be able to
provide a useful amount of drive current to counter the miller
feedback effect.

RP

2009/11/5 Josh Koffman <joshybear@...>:

> Hi all. I might be called on to help out with a project. The part I'm
> most unsure about is that the intensity of a 24V 250W light bulb will
> need to be controlled, preferably by a PIC. It's being supplied by a
> 24V battery.
>
>  I'd love to be able to find a compact, ready made module that would
> do this part for me. Digital (ie serial data, etc) control would be
> ideal, but analog might be possible. Alternatively I could use the CCP
> module to generate some PWM, but I'd need a driver. The reason I'd
> like a module is that in the event of a problem

 it would be easy to

> swap out. I've done some searching and what I'm finding are a ton of
> smaller modules meant for dimming LEDs.
>
> Alternatively I could use PWM and some sort of home built driver. The
> circuit here: http://www.solorb.com/elect/solarcirc/pwm1/ suggests the
> use of a FET. Having never controlled this large a DC load before, I
> have no idea if this is a good or bad idea. Any thoughts?
>
> Thanks for the help!
>
> Josh
> --
> A common mistake that people make when trying to design something
> completely foolproof is to underestimate the ingenuity of complete
> fools.
>        -Douglas Adams
> --
> http://www.piclist.com PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jinx-4]

> Alternatively I could use PWM and some sort of home built driver. The
> circuit here: http://www.solorb.com/elect/solarcirc/pwm1/ suggests the
> use of a FET. Having never controlled this large a DC load before, I
> have no idea if this is a good or bad idea. Any thoughts?

Hi Josh. Controlling a light bulb of that wattage is pretty straight-forward.
Much less problematic than a DC motor of that size for example (resistive
vs inductive load). The only consideration of the above circuit might be
whether the FET is turned on and off quickly enough to avoid unnecessary
heating. It's not hard to find or design a simple gate driver though, which
should include gate protection. AIUI the gate is what often fails. Usually a
small (10 ohm) series resistor and zener is OK

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Dario Greggio (in giro)]

Josh Koffman ha scritto:
> Hi all. I might be called on to help out with a project. The part I'm
> most unsure about is that the intensity of a 24V 250W light bulb will
> need to be controlled, preferably by a PIC. It's being supplied by a
> 24V battery.
>

Is it AC or DC?

I used 220V dimmers for 24VAC with success, or simple PWM switcher for DC.
I have a PIC-based board which accepts commands via RS485.

Dario
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jo Scherpenisse]

A FET like the IRF540N can do the job. But to be sure you need to raise the
gs voltage to about 7 Volts instead of the max 5 directly from the PIC.
And be aware of the fact that the current with a low filament temperature is
much higher than when the lamp is already heat enough.

----- Original Message -----
From: "Josh Koffman" <joshybear@...>
To: "Microcontroller discussion list - Public." <piclist@...>
Sent: Thursday, November 05, 2009 1:55 AM
Subject: [EE] 24V Dimmer


> Hi all. I might be called on to help out with a project. The part I'm
> most unsure about is that the intensity of a 24V 250W light bulb will
> need to be controlled, preferably by a PIC. It's being supplied by a
> 24V battery.
>

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Forrest W Christian-2]

Josh Koffman wrote:
> Alternatively I could use PWM and some sort of home built driver. The
> circuit here: http://www.solorb.com/elect/solarcirc/pwm1/ suggests the
> use of a FET. Having never controlled this large a DC load before, I
> have no idea if this is a good or bad idea. Any thoughts?
>  
Basically, where I'd start is to drive a fet with the PWM output from
the PIC, and use that.  The schematic you reference above basically does
the same thing but with an opamp instead of a PIC generating the PWM.

Depending on the exact FET you use, and frequency, etc....You may need a
FET driver or similar circuit, if you can't find a fet which will switch
the 11 or so amps fully on when driven directly from the PWM output of
the PIC.   And 'driver' might be as simple as another transistor.

-forrest


--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jake Anderson-2]

Josh Koffman wrote:

> Hi all. I might be called on to help out with a project. The part I'm
> most unsure about is that the intensity of a 24V 250W light bulb will
> need to be controlled, preferably by a PIC. It's being supplied by a
> 24V battery.
>
>  I'd love to be able to find a compact, ready made module that would
> do this part for me. Digital (ie serial data, etc) control would be
> ideal, but analog might be possible. Alternatively I could use the CCP
> module to generate some PWM, but I'd need a driver. The reason I'd
> like a module is that in the event of a problem it would be easy to
> swap out. I've done some searching and what I'm finding are a ton of
> smaller modules meant for dimming LEDs.
>
> Alternatively I could use PWM and some sort of home built driver. The
> circuit here: http://www.solorb.com/elect/solarcirc/pwm1/ suggests the
> use of a FET. Having never controlled this large a DC load before, I
> have no idea if this is a good or bad idea. Any thoughts?
>
> Thanks for the help!
>
> Josh
>  

take a look at robot powers simple-H its probably in the ball park, feed
it with a PWM from the PIC and you should be set.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Dario Greggio (in giro)]

Josh Koffman ha scritto:
> Hi all. I might be called on to help out with a project. The part I'm
> most unsure about is that the intensity of a 24V 250W light bulb will
> need to be controlled, preferably by a PIC. It's being supplied by a
> 24V battery.
>

Is it AC or DC?

I used 220V dimmers for 24VAC with success, or simple PWM switcher for DC.
I have a PIC-based board which accepts commands via RS485.

Dario

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Alan B. Pearce-2]

> I'd love to be able to find a compact, ready made module
>that would do this part for me. Digital (ie serial data,
>etc) control would be ideal, but analog might be possible.
>Alternatively I could use the CCP module to generate some
>PWM, but I'd need a driver. The reason I'd like a module
>is that in the event of a problem it would be easy to swap
>out. I've done some searching and what I'm finding are a
>ton of smaller modules meant for dimming LEDs.

I would be tempted to use a small PIC 10F or 12F to do a software PWM,
driving a Microchip FET driver, driving a suitably rated FET, as a module.
Then you could use whatever interface to the PIC you wanted.

Alternatively a slightly bigger PIC, in the 16F627/28/48 has a hardware UART
that you could use for the external interface, instead of using a software
UART with the smaller devices.

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Josh Koffman]

On Wed, Nov 4, 2009 at 9:09 PM, Mauricio Jancic <listas@...> wrote:

>        I would like to recommend you to take a look at the IPS (Intelligent
> Power Switch) that IR was to offer. It is under the automotive section on
> www.irf.com
>
>        An IPS is basically a MOSFET with a driver and some protection
> features like short circuit, battery reversal protection and over current
> among others. You can even measure the current consumption and detect open
> load (load removed).
>
>        There is a big spectrum of solutions, but I think that you need
> something that can support 24V (the 70V devices are for 24 inductive loads,
> since your load is resistive, you can use the 40V devices). The device
> should also be able to handle 11A or more and support a switching frequency
> of, at least, 100 Hz to make sure the PWM is not seen.
>
>        There are high and low side drivers.
>
>        I have chosed one for you, the IPS6021SPBF
> (https://ec.irf.com/v6/en/US/adirect/ir?cmd=catSearchFrame&domSendTo=byID&do
> mProductQueryName=IPS6021SPBF). This device can withstand up to 12A
> continuous output at 85°C on the TO220 model with a heat sink. The internal
> current limit is set to 32A.

Sorry for the delay, naturally I got pulled onto something else for
awhile, and now it looks like the timeline on this project is getting
shorter. Thank you to everyone for their suggestions!

Mauricio, I like your suggestion a lot. It looks like these little
guys can take logic level inputs with no problem and I like all the
protection built in. It even has a way to tell me if it has gone over
temp.

Since this device has an internal current limiter, does this mean I
don't have to worry about cold filament current? Will the internal
limiter just limit to its max (in this unit's case, 32A typical) until
the filament has warmed?

I'm also unsure about heatsinking needs. I can't figure out how much
heat I will have to dissipate. Do I base it on the on resistance? How
does one heatsink a D2Pak anyway? Should I just stick with T0220?

Also, it seems that the IPS6011 part is similar but with a higher
current capacity. Comparing the temperature vs output current graphs
seems to show that at my current levels I could get away with a much
higher operating temperature with the IPS6011. To me this would mean I
could use less heatsinking. Is this a bad idea?

Thanks for the help!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Josh Koffman]

On Thu, Nov 5, 2009 at 3:28 AM, Dario Greggio <adpm.to@...> wrote:
> Is it AC or DC?
>
> I used 220V dimmers for 24VAC with success, or simple PWM switcher for DC.
> I have a PIC-based board which accepts commands via RS485.

Hi Dario,

It's DC. Is your board something I could see for driver ideas?

Thanks!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Josh Koffman]

On Thu, Nov 5, 2009 at 8:33 AM, Jake Anderson <jake@...> wrote:
> Josh Koffman wrote:
>> Hi all. I might be called on to help out with a project. The part I'm
>> most unsure about is that the intensity of a 24V 250W light bulb will
>> need to be controlled, preferably by a PIC. It's being supplied by a
>> 24V battery.
>>
> take a look at robot powers simple-H its probably in the ball park, feed
> it with a PWM from the PIC and you should be set.

Interesting, I'll have a look!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Josh Koffman]

On Thu, Nov 5, 2009 at 11:37 AM, Alan B. Pearce
<Alan.B.Pearce@...> wrote:

>> I'd love to be able to find a compact, ready made module
>>that would do this part for me. Digital (ie serial data,
>>etc) control would be ideal, but analog might be possible.
>>Alternatively I could use the CCP module to generate some
>>PWM, but I'd need a driver. The reason I'd like a module
>>is that in the event of a problem it would be easy to swap
>>out. I've done some searching and what I'm finding are a
>>ton of smaller modules meant for dimming LEDs.
>
> I would be tempted to use a small PIC 10F or 12F to do a software PWM,
> driving a Microchip FET driver, driving a suitably rated FET, as a module.
> Then you could use whatever interface to the PIC you wanted.
>
> Alternatively a slightly bigger PIC, in the 16F627/28/48 has a hardware UART
> that you could use for the external interface, instead of using a software
> UART with the smaller devices.

Hi Alan,

I'm starting to think that my "module" idea isn't really applicable.
There isn't going to be that much extra stuff in this project, so it
might just make the most sense to have it all on one board. If it
goes, repair/replace the whole board. I'm also worried that there just
won't be that much extra room in the box. That said, if I do go with
that IPS chip Mauricio pointed me to, I have to figure out how to
heatsink it. I might still want to use a module approach so that I
have more control of the heatsink and its relation to the outside. For
instance, if I made the wall of the box a heatsink and attached the
PCB to it, they could be swapped out as a single unit.

Thanks!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Dario Greggio (in giro)]

Josh Koffman ha scritto:

> On Thu, Nov 5, 2009 at 3:28 AM, Dario Greggio <adpm.to@...> wrote:
>> Is it AC or DC?
>>
>> I used 220V dimmers for 24VAC with success, or simple PWM switcher for DC.
>> I have a PIC-based board which accepts commands via RS485.
>
> Hi Dario,
>
> It's DC. Is your board something I could see for driver ideas?
>

Oh, just a basic PIC->resistor->Darlington (say BDX53C)

I can post a schematic but it's really no more than this, apart from a
fuse and the firmware
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Josh Koffman]

2009/11/15 Dario Greggio <adpm.to@...>:
> Oh, just a basic PIC->resistor->Darlington (say BDX53C)
>
> I can post a schematic but it's really no more than this, apart from a
> fuse and the firmware

I'd be interested. So far I've decided to start with a HEXFET and
driver from IR. But I'd love to see what others are doing.

Thanks!

Josh
--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Dwayne Reid]

At 04:22 PM 11/17/2009, Josh Koffman wrote:

>I'd be interested. So far I've decided to start with a HEXFET and
>driver from IR. But I'd love to see what others are doing.

Our DC dimmers are a 12F675 or 12C508 driving a IRF3708 MOSFET via a
47R resistor.  Brightness control is a rotary pot using a RC-time
type measurement (12C508) or DC voltage from 0 to 5V (12F675).  The
MOSFET is a logic-level device and does not need a driver.  We do PWM
at 500Hz or 977Hz or 4 KHz, depending upon which version we are shipping.

There is also a clamp diode (mbr1045) connected across the load to
deal with inductive spikes.

The 3708 MOSFET is good for at least 10A without heat-sinking and we
routinely drive 30A loads by using the aluminum case as the heat-sink.

We've got code in both assembler and PICbasic (again - different
versions for different customers).

Pretty simple stuff.

dwayne

--
Dwayne Reid   <dwayner@...>
Trinity Electronics Systems Ltd    Edmonton, AB, CANADA
[phone] voice          [phone] fax
www.trinity-electronics.com
Custom Electronics Design and Manufacturing

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Dario Greggio (in giro)]

Josh Koffman ha scritto:
> 2009/11/15 Dario Greggio <adpm.to@...>:
>> Oh, just a basic PIC->resistor->Darlington (say BDX53C)
>>
>> I can post a schematic but it's really no more than this, apart from a
>> fuse and the firmware
>
> I'd be interested. So far I've decided to start with a HEXFET and
> driver from IR. But I'd love to see what others are doing.


Basically the same as Dwayne wrote.

Only I'm using the darlington, with some 470ohm at its base. And a
frequency of around 200Hz-1KHz.

In the end, it's an heritage from a NE555 circuit driving a brush
(glass-wiper) motor @12Vcc, in the 70s and 80s :)

Dario
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist