On 2017-02-09 06:04 AM, alan.b.pearce@stfc.ac.uk wrote:
>> Short comment to start the ball rolling.
>> If 10 milliohms is acceptable, as you seem to suggest, then you get 10 m=
V/A
>> or 50 mV at 15A.
>> That's usefully more than the input offset voltage of many low cost
>> comparators.
>> Unadjusted many are in the +/- 5 to 10 mV range. So about 0.5 to 1 A
>> uncertainty at max Voffset.
>> Not "precise". but you do not need precision,
>>
>> Many more modern low cost comparators have Vddma in the 5-10V range.
>> Bearable with design.
>> The very cheapest are often variants of the olde olde LM339 with 30V+
>> supply voltage ratings.
> This is why I suggested using a high side current sense chip like the Lin=
ear Technology LTC6101, LTC6102, LT6106 or Microchip PAC1921. These would w=
ork with a PCB trace at about that resistance and feeding that into an ADC =
would allow configurable current monitoring.
This sounds like a good idea.  Some nice chips here.. Just wondering if=20
I put the voltage out of these chips  into an ADC on my PIC if that=20
would be fast enough to catch a short before my FET fries. I'll have to=20
take a closer look at the time it would take to read the ADC and take=20
action. Because I am using PWM ( around 15 KHz ) I guess I would need to=20
synchronize my ADC reading with the on state.   I have read that FETs=20
can burn out fairly quickly ie. 10 uSec so the conversion and taking=20
action to turn off the FET would have to occur fast.
Jim
>
>
>

--=20
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.