€1,350
pledged of €14,900pledged of €14,900 goal
10
backers
0seconds to go
Funding Unsuccessful
The project's funding goal was not reached on Thu, August 14 2014 8:04 PM UTC +00:00
Wouter CouzijnBy Wouter Couzijn
First created
Wouter CouzijnBy Wouter Couzijn
First created
€1,350
pledged of €14,900pledged of €14,900 goal
10
backers
0seconds to go
Funding Unsuccessful
The project's funding goal was not reached on Thu, August 14 2014 8:04 PM UTC +00:00

About

USB powered, 2.5kV galvanically isolated ultra-wide dynamic range current/voltage/power (factor) meter

Based on a revolutionairy new type of current shunt, this device will measure electric current over an unprecedented ultra-wide dynamic range of 10^8. Without range switching or changing shunt values, it can measure 50nA up to 5A (bidirectional, AC and DC) with very low voltage drop (<50mV) across the shunt (the burden voltage) and is therefore uniquely suited for bursty, pulsed loads such as microcontrollers and RF transceivers operating at a low duty cycle. It will measure voltage as well, up to 500VAC, and can compute the power, power factor and phase angle. It may be used as a DMM, oscilloscope, data logger or for PASS/FAIL screening based on programmable thresholds along a production line.

For over 15 years I've been designing RF transceiver ASICs and wireless products, and I've always struggled with the reliable, accurate and reproducable measurement of the (average) current consumption of such chips and products, since they typically operate at very low duty cycles, with sub-uA consumption in sleep, but hundreds of mA during the (infrequent) bursts of activity. Average current consumption measurements are essential for predictions of the battery lifetime of the product, but also for the screening of each manufactured chip/product. At low duty cycles, the standby consumption is just as significant as the active consumption, but conventional shunts and even high-end DMMs will not be able to measure the two currents simultaneously, because they differ by so many orders of magnitude. In my experience, the readings of high-end, true RMS DMMs may be OFF by a factor 30 or more.

Products that incorporate motor actuators present similar challenges : usually, the standby consumption is many orders of magnitude smaller than the active consumption when one or more motors are running, and especially during the ramp-up, a couple of milliseconds after activation, motors will typically draw excessively high peak currents. In many a project, the standby consumption is a few uA, and the motors may draw up to 5A during the ramp-up.

You will normally use a relatively large shunt resistor to measure sub-uA standby currents. For example, a 100 ohms shunt resistor will give 0.1mV at 1uA, just sufficient to measure with a high-end DMM or oscilloscope. However, when the product-under-test then suddenly activates (microcontroller starts running, RF transmitter is activated, or a motor is switched ON), it may attempt to draw tens or hundreds of milli-amps, which would translate to a voltage drop of several volts across the 100 ohm shunt resistor, completely robbing the product of its supply, causing it to reset for sure.

If you use a 0.1 ohm shunt, the product can draw hundreds of milli-amps with no significant voltage drop across the shunt, and you will be able to measure the larger active consumption reliably, but sub-uA standby currents will translate to nano-volt shunt voltages, which cannot be accurately measured.

In this ultra-wide dynamic range current meter project, I've completely abandoned the traditional resistive shunt, and I've invented a revolutionary new type of shunt instead : a switched low-ESR capacitor. The shunt current will charge the capacitor, and when it reaches a certain threshold voltage, a low RDS-on MOSFET across the shunt capacitor will discharge it, after which the cycle repeats. The frequency at which the discharge MOSFET closes is directly proportional to the current that flows through the shunt capacitor. At very low currents, the frequency is equally low, and the update rate of the measurements would be unacceptably low. However, the (DSP) firmware will seamlessly switch over to a dV/dt measurement : the slope of the voltage across the shunt capacitor is also proportional to the shunt current, and can be measured much faster (for low currents).

The fundamental difference with a resistive shunt is that a capacitor converts peaks and discontinuous abrupt steps in current consumption into a smooth, continuous voltage across the capacitor. It will absorb the sudden changes, there will be no excessive voltage drop across the shunt, and the product-under-test will not be robbed of its supply. The capacitor will act like an integrator and intrinsically implement a 1st order CIC moving average resampler from the continuous analog domain to the discrete-time digital domain of the A/D converter and DSP algorithms, thus ensuring that even the narrowest pulse won't be missed, will be included in the (average) current measurement (no aliasing).

Furthermore, the measurement error (tolerance, accuracy) is more or less constant across the entire range. Suppose that the shunt voltage is sampled with an A/D converter that has a 0.1mV error. For a 1 ohm resistive shunt, this 0.1mV error would translate to a 0.1mA error in the current measurement, which corresponds to 1% for a current of 10mA, but to 100% for a current of 0.1mA. For a capacitive shunt, we will measure the accumulation of a certain voltage drop in a certain period of time. If we would discharge the shunt capacitor when it has charged to 10mV, the 0.1mV error in the A/D converter would translate to a 1% error in the current measurement, irrespective of the current that we measure. If we measure a current of 10mA, the capacitor would be discharged e.g. at 1 kHz. If we measure a current of 0.1mA, the capacitor would be discharged at 10 Hz. This timing has no impact on the error, so the relative error remains the same across the entire range.

Also, with a capacitive shunt, you are only interested in the slope, the increase/decrease of the voltage across the shunt over time, and therefore this measurement principe is completely insensitive to DC offsets in the analog amplifiers and A/D converter. When measuring sub-uA currents with a resistive shunt, such offsets are the toughest issue to tackle.

Risks and challenges

As far as I'm aware, this type of current shunt is unique, and not used by anyone else. In my prototype, I've demonstrated its feasibility, and I would love to develop this technology further. I feel pretty confident that I can achieve the predicted performance (50nA - 5A range) but as with any ground breaking new principle, there are also some specific challenges to overcome : compensation for the (non zero) ESR of the capacitor, the (non zero) RDS-on of the MOSFET, the non-linearity of the capacitor under (DC) bias, the aperture time of the discharge switch, the parasitic gate-source and gate-drain capacitance of the MOSFET, etc. I think I've identified all these challenges and addressed them through auto-calibration and -compensation mechanisms, but until the product is completely finished, I will not be able to commit to this range with absolute certainty, there is a small possibility that I will need to make small concessions on the range or accuracy in a certain region.

Learn about accountability on Kickstarter

Questions about this project? Check out the FAQ

Support

  1. Select this reward

    Pledge €9 or more About $10

    All the electronic documents that you require for building your own meter : Bill Of Materials, PCB layout (Gerber), firmware (HEX), PC software (EXE), by email.

    Less
    Estimated delivery
    3 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  2. Select this reward

    Pledge €49 or more About $56

    Build your own meter (BOM, firmware, software), with bare PCB and preprogrammed DSP and USB uC shipped to you.

    Less
    Estimated delivery
    Ships to Anywhere in the world
    2 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  3. Select this reward

    Pledge €89 or more About $101

    Build your own meter (BOM, firmware, software), with bare PCB, solder paste stencil and preprogrammed DSP and USB uC shipped to you.

    Less
    Estimated delivery
    Ships to Anywhere in the world
    0 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  4. Select this reward

    Pledge €199 or more About $226

    Fully assembled and tested PCB without enclosure.

    Less
    Estimated delivery
    Ships to Anywhere in the world
    2 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  5. Select this reward

    Pledge €249 or more About $282

    Fully assembled and tested product (in enclosure), ready for use.

    Less
    Estimated delivery
    Ships to Anywhere in the world
    3 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  6. Select this reward

    Pledge €499 or more About $566

    First production run : be among the first to receive the fully assembled and tested product (in enclosure), ready for use.

    Less
    Estimated delivery
    Ships to Anywhere in the world
    Limited 0 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  7. Select this reward

    Pledge €999 or more About $1,133

    Be the first to receive the fully assembled and tested product, ready for use, and receive the schematics, explanation of how it works, source code of DSP, USB uC firmware and PC software, allowing for customization, integration in your own setup etc.

    Less
    Estimated delivery
    Ships to Anywhere in the world
    Limited 0 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.
  8. Select this reward

    Pledge €1,499 or more About $1,700

    Be the first to receive the product, ready for use, schematics, explanation, source codes + have a say during the development, what additional features should be implemented in hard- and software (subject to feasibility)

    Less
    Estimated delivery
    Ships to Anywhere in the world
    Limited 0 backers
    Kickstarter is not a store.

    It's a way to bring creative projects to life.

    Learn more about accountability.

Funding period

- (30 days)