Hi, dialog. Recently I am measuring the power consumption of my custom board which has some sensors on it. I double check that I have set the correct sleep mode of all sensors, and disabled uart and jlink interface also. But when I measure the current with one 10 Om resister using oscilloscope, the result is as high as 160ua. Could you please give me some suggestions on how to find the wrong place? Thanks.
Device:
Hi Summer, is the 160uA a constant, average, or peak ? is it noisy or clean ... If you probe around the DA14580, do you see any activity on any GPIO etc ? Also, 10Ohm might be a bit high : can you take a look at section 4.1 of the below document ?
http://support.dialog-semiconductor.com/system/files/resources/AN-B-015_...
BR JE_Dialog
Hi, the 160uA current is constant which is the offset of the measured result. I have read the doc before, but I'm still confused about the high measured current. Maybe there is something wrong with my board? But I double check the schematic all day long yesterday and unfortunately, don't make any progress.
Hi There, you mention that hte 160uA is an offset : so is the 1.6mV there even if you disconnect the supply completely ? i was wondering if you are seeing 1.6mV of offset/noise on your measurement circuit. On a 10 Ohm resisors, what looks like 160uA is only 1.6mV which could be offset/noise. BR JE_Dialog
Thank you for your reply, but I measure another board which is the BLE product of one company China, there is no offset current.
OK.. no problem. I will send you a message and have one of the team take a look at this : we'll need your schematics and a scope waveform of what you are seeing. BR JE_Dialog
OK, I have emailed you and I am appreciated about your work, thanks.
It seems that the current is caused by i2c interface, but I am sure I have set correct low power mode of the sensors attached to the i2c bus. What do you think about it ?
Hi Summer,
the I2C pull-up resistors (1K) connected to Vbat3 could be the source for your leakage current.
When in sleep, are the SDA and SCL lines indeed high-impedance?
Are you able to measure the Vbat3V current of the DA14580 device alone?
How much current are the SPI Flash devices drawing?
Is the buzzer really completely switched off when not used?
Best regards, BB_Dialog.
Hi, thank you for your suggestions. After I removed the SPI FLASH and the Vbat3V supply of pull-up resistors, the current was 90ua, and after I removed the Vbat3V supply of SPI FLASH, lis3dh, bme280 and pull-up resistors, the current dropped to 1.5ua. I also did more tests, but I am not going to describing the tedious details. Now I have one question, how can I set the SDA and SCL line indeed high-impedance ? As far as I am concerned, the high current may caused by one of the following two sources: the i2c bus or lis3dh, but I can not figure it out now. So please be kind to help me push it forward, thanks.
Hi Summer,
the I2C lines should be in high impedance state when the DA14580 is in sleep mode, they will not draw current.
Best regards, BB_Dialog.
但是当我测量卷tage of the SCL and SDA pin, the result is as high as the Vbat3V level. And what's more, it is said that DA14580 will latch the pin status when in sleep mode, and i2c bus is idle high, so the result I get may be correct. So is there anything that I am wrong ?
Hi Summer.
when measuring SCL and SDA levels as high as Vbat3V, it indicates there's no current throught the pull-up resistors. Pins are high.
All is okay in our opinion.
We think you have to look for current leakage in the other devices.
Best regards, BB_Dialog.
好吧,我只是想确认,即使基于“增大化现实”技术e no pull-up resistors, the SCL and SDA levels as high as Vbat3V, is it true ? Because I have removed the pull-up resistors already.
Hi Summer,
yes, we would expect that.
In this state, you do mesasure the 1.5µA sleep current?
Best regards, BB_Dialog.
It's weird to me. After I remove the pull-up resistors and the Vbat3V supply of LIS3DH without initializing the i2c interface, the current is about 13ua. But if I enable the i2c interface, the current is as high as 100ua.
After I remove the LIS3DH from my board completely, it seems that the current returns to normal. So my question here is, what should be taken care of when I use LIS3DH ? It seems the LIS3DH worked well before in spite of the high current.
I changed the accelerometer LIS3DH with KX022, and the current is below 10 ua now.
Hi Summer, thanks for letting us know that you fixed the issue. I'm surprised the LIS3DH doesn't work as its very popular product and we have multiple customers using it with the DA14580.. The KX022 looks like a good device ! BR JE_Dialog