Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Maximum load of Temperature monitor IC?

Status
Not open for further replies.
T

treez

Guest
Newbie level 1
Hello,

I need to load the output of this temperature monitor IC with 22K. Vsuppy is 3V3.

Is this going to be OK?.... the datasheet appears to say that the maximum load is 100uA?
The datasheet seems to say that this 100uA load is actually within the Temperature monitor(?) so me loading it with an extra 22K could be bad news.


I need to load it as on some of our PCBs the temperature monitor is not fitted, and the ADC pin interprets this as overtemperature....so i want to put the pull down on the pin, so that boards without the temperature monitor get their ADC pin pulled down.

MCP9700 datasheet
**broken link removed**
 

What makes you think the "load is actually within the Temperature monitor" ?

Do the math:

Assuming a maximum temperature of 125 deg.
Response: 10mV/deg
Vo=500mv
max voltage=.500+.01*125=1.75V
max current=100uA

Min resistance=1.75/100uA=Left as an exercise for the student.
 

1.75/100uA is 1.75E4 ie 175K, but that doesnt neccasarily mean anything.
We cannot tell from the datasheet how the ic produces its output voltage.
 

1) it's 17.5K, not 175K.

2) Why do you care how the IC 'produces its output'? The data sheet says explicitly that maximum output current is 100uA. That's all you need to know.
 

An ADC input is likely to draw more than 100uA, so i am worried now.
We are reading the temperature monitors output with an ADC input of SAM3N00B microcontroller.....the SAM3N00B datasheet does not tell how much current exactly is dran by an ADC input but im sure its more than 100uA.
I wonder if i need to pus an opamp buffer in there?
 

The datasheet seems to say that this 100uA load is actually within the Temperature monitor(?)

Datasheet actually said that self heating due to 100uA load current is insignificant.
 

i agree with you, i now am cognisant that MCP9700 is only capable of sourcing a 100uA current.

Therefore, do you think i will need to buffer its output before connecting it to the ADC pin of a SAM3N00B microcontroller?


.......We are using an MCP9700 temperature monitor IC and feeding its output into the ADC pin (AD4) of the SAM3N00B microcontroller.
The MCP9700 only provides a maximum of 100uA output current, so is that enough to supply the current drawn by the ADC pin, or do i need to buffer the temperature monitor IC's output voltage?

The description of MCP9700 on microchip wesite said that "No external components are needed" so i am surprised they didnt mention about the buffer.(?)

MCP9700 Temperture monitor IC
**broken link removed**

SAM3N00B datasheet:
http://www.atmel.com/Images/doc11011.pdf
 

The reason that they say that ""No external components are needed" are precisely that, because no external components are needed for the device to produce a voltage with a maximum output of 100uA. Do you expect them to know what your application is???

I suggest you actually spend some time reading the data sheets, all the answers are there; I found them and this isn't even my project. The input leakage current to the ADC is 1uA. There's a 14pF capacitance on the input. The sensor can handle a maximum output capacitance of 1000pF.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
We are reading the temperature monitors output with an ADC input of SAM3N00B microcontroller.....the SAM3N00B datasheet does not tell how much current exactly is dran by an ADC input but im sure its more than 100uA.

Yes you are correct, but for transient situation only.
ADC input requires more than 100uA to charge the empty sample & hold capasitor (Csample) quickly. If you dont provide such current then the capasitor will charge slowly and some time will be wasted.
But when the capasitor is charged, the ADC input will draw leakage current (1uA) only.


Therefore, do you think i will need to buffer its output before connecting it to the ADC pin of a SAM3N00B microcontroller?

No, you dont need buffer. You need a capasitor (1000pf). The capasitor will provide the necessary charging current (several times of 100uA) to charge the sample and hold capasitor (Csample).

Buffer can provide high current for any duration.
Capasitor can provide high current for short time only.
Your requirement is high current for short time. So capasitor is enough.

So the suggested connection is

MCP9700 output....then 100k pull down resistor... then 1000pF pull down capasitor...then ADC input of SAM3N00B.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
why the 100k pull down?...it will draw about 10uA..what do you need to draw 10uA out ofthe mcp9700 for?...or is it just a electrically "healthy" thing to get a bit of current flowing?..........im not mocking it, i always like to have at least a few 100uA flowing in potential divider resistors.

So we need a capacitor...right...so there are external components needed after all.....these things very commonly supply adc inputs, so for them to say "no external components" is misleading of them.
 

why the 100k pull down?...it will draw about 10uA..what do you need to draw 10uA out ofthe mcp9700 for?...or is it just a electrically "healthy" thing to get a bit of current flowing?..........im not mocking it, i always like to have at least a few 100uA flowing in potential divider resistors.

So we need a capacitor...right...so there are external components needed after all.....these things very commonly supply adc inputs, so for them to say "no external components" is misleading of them.

As I said before, you can't expect the mfr. to know what your application is. You don't NEED a cap or resistor to make it work. Are you going to complain that your microprocessor counts as an "external component", because you need that for your design to work?
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Realistically an analog output temperature ic is either going to be conected to a comparator input, or an adc.

For the adc, its going to need a capacitor, and actually a 10-100r resistor before that cap too.
I dont think that counts as "no external components"
 

Man, you're being stubborn. The temp ic doesn't need the cap and resistor, the SYSTEM needs it. And you don't need the cap (or resistor) for the opamp or comparator.
 
  • Like
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
And you don't need the cap (or resistor) for the opamp or comparator
Agreed.
The temp ic doesn't need the cap and resistor, the SYSTEM needs it.
The temp monitor ic needs it because it cant adequately supply the pulse of current to sharge up the adc cap.
Its part of the system, so yes, in that way, the system needs it
 

why the 100k pull down?...it will draw about 10uA..what do you need to draw 10uA out ofthe mcp9700 for?...or is it just a electrically "healthy" thing to get a bit of current flowing?..........im not mocking it, i always like to have at least a few 100uA flowing in potential divider resistors.

I added the 100k pulldown because you wanted a pulldown (22k) in the original post. 100k is better than 22k because it consumes less power.

So we need a capacitor...right...so there are external components needed after all.....these things very commonly supply adc inputs, so for them to say "no external components" is misleading of them.

In mcp9700 datasheet it is said that

this family is immune to the effects of parasitic capacitance and can drive large capacitive loads. This provides Printed Circuit Board (PCB) layout design flexibility by enabling the device to be remotely located from the microcontroller. Adding some capacitance at the output also helps the output transient response by reducing overshoots or undershoots.

Also
mcp9700 has built-in buffer.(AN981 DS00981A page 2) and
absolute maximum output current must not exceed 30mA.(datasheet)

Now it seems mcp9700 is capable to provide high surge current to charge sample and hold capasitor quickly, and some capasitance at output improves the transient response only. (Transient response is related to a delay in output response only. It is not related to current rating. For example High current buffer may show poor transient response at output)

Maximum output current 100uA means other parameters like output regulation, output impedence, self heat due to output current etc.. are guaranteed upto 100uA continuous output current.

So a capasitor at ADC input is not necessary. PCB track's parasitic capasitance is enough to improve transient response of mcp9700.

But if you want to put mcp9700 very close to ADC input then parasitic capasitance will be low, In that case increase acquisition time so that overshoot/undershoot decays out before AD conversion starts. Then you dont need capasitor.

Generally putting a capasitor at ADC input is a easy way to get rid of problems related to output current charging capability, output transient response, external noise at ADC input etc.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top