[Lab] Help? I can't wrap my head around transistors / mosfets.

Michael Grant michael at krazatchu.ca
Tue Nov 26 13:18:59 EST 2013


An N-Channel MOSFET is best suited for low side switching, to control the
flow to ground.
It can be used for a high side switch but the gate voltage needs to be
higher than the source (heater voltage).
So provide V+ to one side of the heater, and switch the ground with the
MOSFET to turn it off and on.

Here's an example of an N-channel MOSFET as a low side switch, disregard
the current limiting resistor:
http://www.element14.com/community/servlet/JiveServlet/showImage/2-4003-4113/Heater+Control+suggestion+3.PNG

The "fast diode across the mosfet" I suggested is not show in this circuit,
it's just for added protection.
It's purpose is to catch any inductive flyback spikes when the MOSFET turns
off fast.
MOSFETs already have an intrinsic body diode but in certain cases it may
not be fast enough.

Michael Grant
--
http://krazatchu.ca/




On Tue, Nov 26, 2013 at 12:55 PM, Justin Slootsky <justin at slootsky.org>wrote:

>  It seems that what I was missing is the understanding that a mosfet
> controls the flow to ground.  I was thinking of it controlling the flow to
> the heater, and the diagrams and  reading didn't quite make sense to that.
> I'll be at the lab this week,   probably won't arrive before 8.
> I'm easy to pick out, I'm the guy with the piercings and tattoos.
>
> Michael: thanks for the parts offer. Maybe Wednesday you can tell me what
> you mean by "put a fast diode across the mosfet"
>
> On November 25, 2013 8:27:12 PM Aurelius R  wrote:
>
> I just finished a Skype call with Justin to explain a bit more about
> MOSFETs. He already was on the right track so I basically just repeated
> what he already knew.
>
> By the way Justin, here's the sample code I wrote. It's for the ATmega644
> as I mentioned, but it might help a bit.
>
> ____________________
> Peace, Love, Empathy
>
> Alexander Max Rowsell
>
>
> On 25 November 2013 20:11, Michael Grant <michael at krazatchu.ca> wrote:
>
>> I have a few FDB6060L MOSFETs I can spare.
>> They are logic level, 17 mΩ@ VGS = 4.5 V, and are a bit overkill but
>> overkill never hurt...
>>
>> You will also want a 5v regulator for the lilypad if you are running it
>> from the same battery, and a few caps to stabilize it.
>> And it wouldn't hurt to put a fast diode across the MOSFET in case the
>> heater has an unusually large inductive component.
>>
>> If your at the lab this Wednesday I'll bring a few parts.
>>
>> Michael Grant
>> --
>> http://krazatchu.ca/
>>
>>
>>
>> On Mon, Nov 25, 2013 at 7:28 PM, Justin Slootsky <justin at slootsky.org>wrote:
>>
>>> The current requirement at 3.6V is .576A
>>> The current requirement at 7.2V is 1.152A
>>>
>>> We'll likely be using a lilypad Arduino (sewable), but we do have a 3.3V
>>> and a 5V Trinket available to us if required.
>>>
>>>
>>> ------------------------------
>>>
>>> *From: *"Michael Grant" <michael at krazatchu.ca>
>>> *Cc: *"Justin Slootsky" <justin at slootsky.org>, "lab" <lab at artengine.ca>
>>> *Sent: *Monday, November 25, 2013 7:24:05 PM
>>> *Subject: *Re: [Lab] Help? I can't wrap my head around transistors /
>>> mosfets.
>>>
>>>
>>>
>>> What's the current required for heater?
>>> If you don't know, measure the resistance of the heater and apply i =
>>> V/R.
>>>
>>> This will allow you to choose a capable switching device.
>>> Likely a logic level MOSFET will be the ideal choice, switched via
>>> series resistor from the AVR.
>>>
>>> And which Arduino are you using?
>>> Most are 5v but some are 3.3 volts.
>>>
>>> Michael Grant
>>> --
>>> http://krazatchu.ca/
>>>
>>>
>>> On Mon, Nov 25, 2013 at 6:20 PM, Ken McKinnon <klmckinnon at rogers.com>wrote:
>>>
>>>> I started writing a response, but figured someone out there in google
>>>> land would have a better explanation.
>>>>
>>>> Try http://sunburst.usd.edu/~schieber/psyc770/transistors101.html
>>>>
>>>> If I recall, the arduino is a 3.3 volt device, so you may need
>>>> something like a optocoupler in between your arduino and the transistor
>>>> that would need to be at a bit higher voltage to turn on completely.
>>>> As an aside, a PWM control would probably control the heat to a much
>>>> finer degree, with a lot fewer parts....
>>>> Good luck on your project, if you need more or that link isn't precise
>>>> enough to get you on path, ping again.
>>>>
>>>> Ken
>>>>
>>>>
>>>> On 2013-11-25 4:37 PM, Justin Slootsky wrote:
>>>>
>>>>> Hey everyone, I'm having trouble understanding transistors / mosfets,
>>>>> and what I need.
>>>>>
>>>>> I'm working with Chris on a heating project, and we're looking at
>>>>> using an Arduino to control heating pads in a mitten to keep his hand warm.
>>>>>
>>>>> The heater will run on 7.2V, which will generate enough heat to keep
>>>>> warm but not enough to burn.
>>>>>
>>>>> What we would like to do is to regulate the power going to the heating
>>>>> pad based on the value returned from a temperature sensor.
>>>>>
>>>>> I forsee us turning the heater on and off based on threshold
>>>>> temperatures inside the mitten.  In the future, it might be advantageous
>>>>> for us to be able to turn the power half on if it is near the desired
>>>>> temperature but for now a binary on/off based on an output from the Arduino
>>>>> will do.  (software will determine the on/off cycle based on the value
>>>>> returned from the temperature sensor).
>>>>>
>>>>> My understanding is that we need a mosfet or a transistor or something
>>>>> for two reasons
>>>>> 1) The Arduino won't send 7.2V power. (even if we use one that has a
>>>>> direct battery line, it is too much power to send to the Arduino)
>>>>> 2) The power draw would be too high anyways.
>>>>>
>>>>> Can anyone help me with what I need in order to connect this up?
>>>>>
>>>>> Thanks
>>>>> Justin
>>>>>
>>>>> _______________________________________________
>>>>> Lab mailing list
>>>>> 1. subscribe http://artengine.ca/mailman/listinfo/lab
>>>>> 2. then email Lab at artengine.ca to send your message to the list
>>>>>
>>>>>
>>>> _______________________________________________
>>>> Lab mailing list
>>>> 1. subscribe http://artengine.ca/mailman/listinfo/lab
>>>> 2. then email Lab at artengine.ca to send your message to the list
>>>>
>>>
>>>
>>>
>>
>> _______________________________________________
>> Lab mailing list
>> 1. subscribe http://artengine.ca/mailman/listinfo/lab
>> 2. then email Lab at artengine.ca to send your message to the list
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://artengine.ca/pipermail/lab/attachments/20131126/c5c183b3/attachment-0001.html>


More information about the Lab mailing list