atmeha32 takes 100 ma is it normal

[aamir]

I am working on ATMEGA32 development BOARD .IT takes 100 Mili amp current when active is it normal ? As i think it should take approx 50 Ma current AVR is working not getting hot . so any problem with board

[pickit2]

depends on what you also have connected onboard. MAX232 or other chips, any relays,lcd or LEDS fitted.

Data Sheet for Atmaga48 is at 1v8
• Power Consumption at 1 MHz, 1.8V, 25°C
– Active Mode: 0.2 mA
– Power-down Mode: 0.1 μA
– Power-save Mode: 0.75 μA (Including 32 kHz RTC)

[scrts]

Are You sure You're measuring ONLY atmega current, or the whole board? Also set the unused I/Os to output set to ground.

[aamir]

I have connected max232 and LCD and some 74 ic i am asking for whole board 

[microkid]

I have connected max232 and LCD and some 74 ic i am asking for whole board 

Then you should list out all the components on the board

[engamor]

The actual current of the whole board shall be checked against the spec. But you shall take into account the clock frequency at which you are operating the system. Also, in my experience, the reliability of the data sheet at Atmel is not very good. A 100% variation between the "typical" and the actual current drain is possible. Of course first consider the other components. By the way, the variability of current consumption was one of the most important reasons for us to dismiss the atmel microprocessors from our production.

[hate]

LEDs drain current. If u have a couple of LEDs sourced with 10mA-20mA u can easily have 100mA of current. But if u don't have anything lighting up that's not normal. Max232, LCD and other logic stuff shouldn't drain that much current, there may be some leak somewhere or u may have some of ur AVR pins defined as output while connected to some output pin with reverse logic. I mean AVR pin may be at logic 0 level while connected to some logic 1 level pin which is like connecting the + supply pin to GND. Try removing the mcu and measure the current again.

Regards...

[scrts]

The actual current of the whole board shall be checked against the spec. But you shall take into account the clock frequency at which you are operating the system. Also, in my experience, the reliability of the data sheet at Atmel is not very good. A 100% variation between the "typical" and the actual current drain is possible. Of course first consider the other components. By the way, the variability of current consumption was one of the most important reasons for us to dismiss the atmel microprocessors from our production.

There is picopower However the code has to be optimized as required for the best results... Unused pins must be outputs tied to GND, the code must be completely interrupt driven and so on... Its a really hard task.

[metal]

May be there is some way, I don't want to call it special, but when clocking is involved, may be there are concerns and restrictions on using a normal multimeter the ordinary way. I have never tried using a multimeter to see the current consumption of an MCU based project. But when you think about the way current is consumed, may be normal meters don't measure correctly under certain circumstances. I will dig more about this. Seems an interesting topic to look at.

[engamor]

may be there are concerns and restrictions on using a normal multimeter the ordinary way
...  may be normal meters don't measure correctly under certain circumstances. I will dig more about this. Seems an interesting topic to look at.

This is for sure, The current drain can vary in the operating cycle. As we are very much concerned with current we check the average current and the peaks as well. One simple way is to place a reasonable resistor in series with the power supply and look at the voltage drop with an oscilloscope. That will not necessarily be easy, as you want to keep low the voltage drop, but it can be done and give you a reasonable idea of what is going on. Choose the resistor as to get something between 10 and 100 mV drop. You may want to have a quick short circuit path to the resistor ( a jumper) in order to let the microprocessor start up, in certain cases this is necessary.