Current probe

Go To Last Post
31 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I need to measure dynamic current consumption of MCU's (mainly for sleep modes validation). I'm thinking of some current probe that I can connect to my scope. It should measure sub uA - 100 mA (it can have several manully selectable ranges), frequency response 100 kHz, measuring at high side (1.6 - 5 V MCU's power supply).

Do you know about some ready to use solution that would cost 500 USD and less?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Oscilloscope current probes usually have a sensitivity limit in the milliampere range, if not higher. High sensitivity ones often take a special pre-amplifier or oscilloscope plugin, for plugin based scopes. DC current probes are often a LOT more expensive.

Try a pair of probes, differential scope input, and a series resistor where you measure the drop.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Hitex wrote:
To help developers optimise their software for low power consumption Hitex has introduced PowerScale - a truly innovative energy profiling tool. PowerScale has been developed specifically for embedded microcontrollers. For the first time developers can easily visualise the power usage of individual software routines

http://www.hitex.co.uk/index.php...

ka7ehk wrote:
Try a pair of probes, differential scope input, and a series resistor where you measure the drop.

There are many current-sense amplifiers from well-known semiconductor manufacturers - I would look at them to use in conjunction with my scope.

That is, in fact, how I did these measurements: http://www.antronics.co.uk/downl... (using a MAX471 - although, sadly, they're no longer available)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

or just place a 1 ohm (0.1 - 10 ohm) in the power line and use a normal probe.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

sub uA - 100 mA

Quote:

or just place a 1 ohm (0.1 - 10 ohm) in the power line and use a normal probe.

Hmmm--on a range of 100000+ ?

Let's just say that I found it tricky to measure deep-sleep of a few uA and then also measure wake current of say 80mA during RF transmit.

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Mr eevblog has a little box for the job.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Mr eevblog has a little box for the job.
http://www.eevblog.com/projects/...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The technique I described DOES use standard scope probes.

It IS limited by the attenuation accuracy of the probes. If they are 1% attenuation accurate, then one could be 1% high and one could be 1% low. That means a possible 10mv error when measuring zero current on a 5V DC level when differenced in the scope. There are ways around this if there is a "variable gain" feature on at least one of the input channels.

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Kartman -

That is an impressive box!

Jim

Jim Wagner Oregon Research Electronics, Consulting Div. Tangent, OR, USA http://www.orelectronics.net

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Personally, i don't know why people get so excited about it. To me it is an obvious solution to a simple problem. Nevertheless, it solves a (now) common problem at a low price. Kudos to Dave (in a falsetto voice!).

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gaminn wrote:
Do you know about some ready to use solution that would cost 500 USD and less?
Real-Time Current Monitor at http://www.ee-quipment.com/ appears to meet most of your requirements other than bandwidth.
A review:
Profiling power: real-time current monitor by Jack Ganssle (embedded.com; March 18, 2013)

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
Do you know about some ready to use solution that would cost 500 USD and less?

If you do not care about burden voltage and accept manual ranges then you can use a raw DMM for that. El-cheapo 1999, 0.5% would do ($3-$5).
These have a resolution of about 100nA and a maximal burden voltage of +-200mV (at full scale).
Burden voltage can be easily cancelled out in your case (that is when Vcc or GND current is measured) with a raw op-amp.

No RSTDISBL, no fun!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

IAR I-scope:

http://www.iar.com/Products/Hard...

don't see pricing, though...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

And of course on many uCs you can trigger an ADC in most power modes. Then the data can be sent to a host or buffered/logged. Advantage would be that the ranges might be switched by uC itself as it governs power modes so it knows what order of power consumption to expect.
That is the idea of STM32L-Discovery, it has a mili and microammeter (two ranges switched by a uC) on-board.
iw4ap

No RSTDISBL, no fun!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Perhaps also look at what SiLabs (formerly Energy Micro) do?

And TI for their MSP430?

I think Atmel has a lot of catching up to do here...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
IAR I-scope:
[...]
Whew! What a relief...

For a brief moment I thought that Apple introduced their own iScope oscilloscope iGadget... :P

Somebody else wrote:
If Apple would build an oscilloscope it would be called 'iScope' and would have been made from a single block of Aluminium, with only one button, and a ultra high resolution touch display with 48 finger multitouch support, It would look absolutely great, but would also have a single 542.7256 kohms input with proprietary connector which requires either special probes (in Apple white, 100MHz x10 for $579 a piece, 250MHz x1/x10 switchable for $799) or an 'iConvert' adapter ($449) which converts the proprietary input to a standard 1 Mohms BNC input. It would technically have 1GS/s real-time sample rate but Apple would call it '50G Supersampling' because that's what the scope would do in multisampling mode and this would be the only mode available (real time sampling could be unlocked after a jailbreak, though). Of course the scope would run 'iScopeOS' (which would be based on iOS) and come with the new 'iMeasure' application which would be something like iTunes but for waveforms instead of music. It would also come with its own integrated store where users can buy waveforms so they don't have to make measurements themselves any more.

A year later the new iScope 2 would come out. It would of course be even shiny-er, and come with '100G Supersampling' (still based on 1GS/s ADCs). It would also come with two or four proprietary inputs, but unfortunately the connector is new and the impedance is now 711.3423 kOhms so the old probes and 'iConvert' no longer fit. Along with the new scope a new version of iScopeOS for scopes would be introduced, now with different color scheme ('Norwegian Tundra' as default color scheme) and support for biometric user identification by licking of the single scope button which also functions as biometric sensor, but since this requires a lickable button iScope 1 would be no longer supported.

:roll:


(Sorry, I just could not resist! :lol: )

-George

I hope for nothing; I fear nothing; I am free. (Nikos Kazantzakis)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

250Euro

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Interesting idea here for a multi-range current-to-voltage converter:

http://iopscience.iop.org/0957-4...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Current-sense amp with adjustable gain - which, I guess, could be used like the "range" control on a DMM:

http://www.ti.com/product/ina223

No doubt the other purveyors in this field have similar offerings...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The April 2010 edition of Circuit Cellar magazine contains an article by Robert Lacoste on the design and construction of a DC only mA/nA/pA to voltage converter using the virtual ground technique for zero voltage burden measurement.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

daniyalmb wrote:

...the best and cheapest i've used so far both for AC/DC uptp 100A.

 

Except you haven't read the OPs requirements...

 

gaminn wrote:

It should measure sub uA - 100 mA...

 

Perhaps you could explain how your product will meet this requirement?

 

(Although given the necromancy here I suspect I'll be met with a deafening silence.)

'This forum helps those who help themselves.'

 

pragmatic  adjective dealing with things sensibly and realistically in a way that is based on practical rather than theoretical consideration.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Especially as his account has been washed off the deep end of the internet now... woosh!

 

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Can you run your micro circuit off of batteries (or perhaps you already are)...this makes hooking to a scope much simpler by eliminating most grounds loops, and even most capacitive ground effects (important at higher freqs, or for looking at fast noise spikes).  Conversely, you can get a battery operated scope & use that to beak the gnd loop.

When in the dark remember-the future looks brighter than ever.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

avrcandies wrote:
use that to beak the gnd loop.

Examples:

Image result for bird pecking

Image result for bird pecking

 

 

 

 

 

 

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Those critters showed up every time I beaked it!  wink

 

When in the dark remember-the future looks brighter than ever.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Davey Jones' micro Current is basically a (pretty nifty) USD 5 opamp in a box with a 100x amplification and a little coincell.

 

But you don't have to build it yourself and he sold quite a lot of them so I guess this product has got it's place.

It seems to be exactly what OP is asking for.

 

- It's  < USD 500 (it costs AUD 90).

- 3 selectable ranges, measures down to nA.

- Bandwith ??? Not sure, but 100kHz sounds about right.

 

For deep sleep measurement you can disconnect the power supply completely and run the uC from a capacitor.

You can calculate the current from the speed the voltage droops.

Capacitors are of the shelf prducts which cost less than USD500 (most of them anyway).

Paul van der Hoeven.
Bunch of old projects with AVR's:
http://www.hoevendesign.com

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Paulvdh wrote:
- Bandwith ??? Not sure, but 100kHz sounds about right.

EEVblog

EEVBlog

uCurrent GOLD Multimeter Adapter

https://www.eevblog.com/product/ucurrentgold/

...

Bandwidth:>300KHz (-3dB)

...

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

awneil wrote:
I think Atmel has a lot of catching up to do here...

Atmel

Power Debugger

Analog Hardware

http://www.atmel.com/webdoc/GUID-EAD481FD-28E6-4CD5-87FB-5165E7687C12/index.html?GUID-F0E592FF-579E-4828-92FA-D5699D9F5F4F

...

The 'A' channel is the recommended channel for accurately measuring low currents. It features two shunt stages which through bypass switches provide a circuit capable of measuring from 100mA on the top-end down to under 1μA. Range switching is done automatically, and it is possible to lock sampling into the high-range only should this be needed.

  1. 'A' channel high range: 100mA - 500μA, ~3μA resolution.
  2. 'A' channel low-range: 1mA - 1μA, ~30nA resolution.

The sampling rate of the 'A' channel is 62.5kHz and data is sent in 16-bit frames to the host computer. A calibrated 'A' channel has accuracy no worse than 3% down to around 1μA.

The 'B' channel is the recommended channel for measuring higher currents with lower resolution than the 'A' channel. It is based on a single shunt resistor allowing measurement of current up to 1A and down to 1mA.

  1. 'B' channel single range: 1A, ~500μA resolution.

Data is sampled at 62.5kHz and transferred to the computer in 12-bit frames.

...

Microchip Technology Inc

Microchip

Power Debugger

http://www.microchip.com/Developmenttools/ProductDetails.aspx?PartNO=ATPOWERDEBUGGER

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

gchapman wrote:

awneil wrote:

I think Atmel has a lot of catching up to do here...

 

Note that I said that in July 2014.

 

By 2015, they did have a dev board for the SAM L21 with current monitoring: http://blog.antronics.co.uk/2015/04/23/new-tricks-for-lower-power-from-atmel/

(sadly, though not untypically, the tools to support it were not available until some time later).

 

Yes, they now have the "Power Debugger" - but have you seen the price of it?!

 

surprise

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes

190USD

142.50USD with the 25% discount

11.76USD sales tax here

free shipping for CONUS and Canada

ETA 14-Sep for ship of 13-Sep; actual : shipped 11-Sep from Tempe Arizona and arrived here 12-Sep (UPS Next Day Air Saver, Express Box)

An Atmel-ICE full kit is 130USD

 

Edit : actual

 

"Dare to be naïve." - Buckminster Fuller

Last Edited: Tue. Sep 12, 2017 - 10:32 PM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Dunno if this is new, or I just hadn't noticed it before:

 

The Atmel Power Debugger Online Manual wrote:

The Power Debugger is a CMSIS-DAP compatible debugger which works with Atmel Studio 7.0 or later, or other front-end software capable of connecting to a generic CMSIS-DAP unit. 

 

http://www.atmel.com/webdoc/GUID-EAD481FD-28E6-4CD5-87FB-5165E7687C12/index.html?GUID-F0E592FF-579E-4828-92FA-D5699D9F5F4F

 

So, in theory, that means it's not limited to just Atmel processors ...

 

 

EDIT

 

highlight

Last Edited: Mon. Sep 11, 2017 - 03:24 PM