Measure the sleep current of PIC10F200 micro controller?

[kimbomc]

Hi all,

i am very new to this so i hope i haven't posted in the wrong place.
I have a small circuit designed using a PIC10F200 microchip which control LEDs flashing and the data sheet of the microchips says that it should use 100 nano amps whilst in sleep mode, however it seems to be using significantly more than this. I have looked for meters that are capable of measuring nano amps but they priced at around $10,000USD each. Can anybody suggest a simple circuit i could build that would allow me to measure the current of round 100 nano amps.

I would very much appreciate some help and support with this project.

Thanks to all in advance.

Christopher

 

[Nigel Goodwin]

You could use an opamp to easily create a very low current meter - however, I would suggest you first post your entire circuit so we can see what you're doing.

 

[ericgibbs]

Hi all,

i am very new to this so i hope i haven't posted in the wrong place.
I have a small circuit designed using a PIC10F200 microchip which control LEDs flashing and the data sheet of the microchips says that it should use 100 nano amps whilst in sleep mode, however it seems to be using significantly more than this. I have looked for meters that are capable of measuring nano amps but they priced at around $10,000USD each. Can anybody suggest a simple circuit i could build that would allow me to measure the current of round 100 nano amps.

I would very much appreciate some help and support with this project.

Thanks to all in advance.

Christopher

hi,
Are you saying that the 'complete' circuit is taking far more than you expected, say 100nA.

If you have polarised standard quality capacitors for decoupling these can have many microAmps of leakage current.

Also whats the pcb material .?

What effect are you getting to make you think the current is high.?

 

[picbits]

Don't forget, if you have the comparators on or the port pullups enables these also draw current when in sleep/run mode.

Funny you should post that - I've just finished a product for a customer using a 10F200. Got it running on the mantlepiece trying to work out how long it will run on a set of SR41 cells. Hoping to get between 6-8 hours continuous out of it.

As above though - decoupling/smoothing capacitors can have a fair bit of leakage as well. Post your circuit and we'll see if there is anything obvious.

 

[picbits]

Oh and at 5v the device can take a fair bit more.

From the datasheet (you did read this didn't you ?)

Sleep mode = 2.4uA (max)
WDT = 16uA
Comparator current = 80uA
Internal reference = 195uA
GPIO weak pullup current = 400uA (!!!!)

Now I'm not 100% certain from the datasheet what actually takes current when the device is in sleep mode but if I read it right then with WDT+Internal Pullups enabled then you could be pulling a lot more current than you bargained for.

 

[Hero999]

Hi all,

i am very new to this so i hope i haven't posted in the wrong place.
I have a small circuit designed using a PIC10F200 microchip which control LEDs flashing and the data sheet of the microchips says that it should use 100 nano amps whilst in sleep mode, however it seems to be using significantly more than this. I have looked for meters that are capable of measuring nano amps but they priced at around $10,000USD each. Can anybody suggest a simple circuit i could build that would allow me to measure the current of round 100 nano amps.

I would very much appreciate some help and support with this project.

Thanks to all in advance.

Christopher

That sounds expensive for something you could easilly implement with a multimeter and a 1M resistor.

Connect a 1M resistor in series with it and measure the voltage across it, the voltage drop will be 1mV per nA drawn so it should drop around 100mV.

Bear in mind that your multimeter probably has an input impedance of 10M in which case the series resistance will be 10M and 1M in parallel so the volt drop per nA will be about 0.9091mV/nA.

 

[donniedj]

Hi all,

i am very new to this so i hope i haven't posted in the wrong place.
I have a small circuit designed using a PIC10F200 microchip which control LEDs flashing and the data sheet of the microchips says that it should use 100 nano amps whilst in sleep mode, however it seems to be using significantly more than this. I have looked for meters that are capable of measuring nano amps but they priced at around $10,000USD each. Can anybody suggest a simple circuit i could build that would allow me to measure the current of round 100 nano amps.

Christopher

No on prop $10,000 USD!

I have in my hands a button cell, a 1MOhm resistor, and an average run of the mill $10 DMM. In simple series circuit I drop the science:

I = V / R
3.089amps (calculated) = 3.07v(measured) / 994kOhm(measured)

My current measured is 3.1 microamps, very close to the calculation. A $50 meter, clean contacts of probes and test points could produce a needless better result. As you see the meter measured down to 0.1 microamps. This is equal to 100 nanoamps. So my $10 meter can measure current in steps of 100 nanoamps at its lowest current measurement setting. You do not need a $10,000 or any other outside the standard issue DMM.

You do need to try test and experiments on your own. Read up on DMM spec where you will see the sensitivity of the device. Study the datasheet for whatever microcontroller of desire. It always states what the real sleeping currents will be based on peripherals enabled, voltages applied, and clocking speed. The greater the number of peripherals turned on during sleep the greater the increased deviation from the lowest possible sleep current. The greater the voltage applied, the greater the.... you know the story.

Using nothing outside the pins attached to the silicon package to measure sleeping current first. This verifies the device. After which, slowly add components and graph the results. If leaky caps, voltage dividers, pull-ups or pull-downs are in the original test, you will have no clue to who is the culprit sucking up power.

 

[kimbomc]

Current drain on the Pic10f200

This is the test I did...

Connect a 1M resistor in series with it and measure the voltage across it, the voltage drop will be 1mV per nA drawn so it should drop around 100mV.

Bear in mind that your multimeter probably has an input impedance of 10M in which case the series resistance will be 10M and 1M in parallel so the volt drop per nA will be about 0.9091mV/nA.

I followed the instruction above instructions and the reading on the multimeter was 2.31V, converting that to Na would make that 2310 nano amps, is that right? This is a much higher reading than the datasheet gives of 100nA; this could be possible as its draining a 3.3V 30mAh battery in 7 days so we know the power drain is high. I have even been in touch with microchip about this and they say they do not have the equipment to test here in the UK.

 

[picbits]

Have you read the above suggestions on peripherals and posting your schematic ?

I can make a 10F200 sleep with 100nA current but it wouldn't be doing much peripheral wise. I can also make it sleep with 2000+nA current

 

[ericgibbs]

hi,
Is the test with the 1Mhm:connected to the positive supply pin of the PIC and nowhere else.?

 

[kimbomc]

hi ericgibbs, test was not done direct to pin 6, I will do now then post results here.

 

[mike50]

This is the test I did...

Connect a 1M resistor in series with it and measure the voltage across it, the voltage drop will be 1mV per nA drawn so it should drop around 100mV.

Bear in mind that your multimeter probably has an input impedance of 10M in which case the series resistance will be 10M and 1M in parallel so the volt drop per nA will be about 0.9091mV/nA.

I followed the instruction above instructions and the reading on the multimeter was 2.31V, converting that to Na would make that 2310 nano amps, is that right? This is a much higher reading than the datasheet gives of 100nA; this could be possible as its draining a 3.3V 30mAh battery in 7 days so we know the power drain is high. I have even been in touch with microchip about this and they say they do not have the equipment to test here in the UK.

The datasheet says the 100nA "sleep" current is for Vdd=2.0V. It appears your Vdd=3.3V so you should expect that the sleep current would be higher.

What kind of battery do you have that produces 3.3V? Or do you have a higher voltage battery and a regulator? If the latter, are you measuring the current at the input to the regulator or the input to the PIC?

Mike

 

[kimbomc]

Hi Mike,

The battery i'm using a CR927 3.3Volt 30mA. I have had the pic10f200 in sleep mode for 48 hours now, it was 3.3Volts and now is 3.12 Volts. Somthing is wrong but I'm not sure what.

Christopher

 

[Nigel Goodwin]

You've been repeatedly asked to post the circuit, how can we help if you don't?.

 

[mike50]

I assume you have
- GP3 configured as an input, not as MCLR
- Weak pullups disabled
- Watchdog Timer disabled
- all I/O pins configured as input and all tied to Vss or Vdd

Mike

 

[ericgibbs]

Hi Mike,

The battery i'm using a CR927 3.3Volt 30mA. I have had the pic10f200 in sleep mode for 48 hours now, it was 3.3Volts and now is 3.12 Volts. Somthing is wrong but I'm not sure what.

Christopher

Hi Chris,
Its just possible the 'new' batteries you have are not that new.!
Most batteries do have a shelf life loss of energy capacity with time.

Did you complete your pin 6 only test.?

As Nigel says, please post a diagram.