Author Topic: Precision measurement of current,voltage, energy, charge and power - Open source  (Read 636 times)

0 Members and 1 Guest are viewing this topic.

Offline SarielTopic starter

  • Regular Contributor
  • *
  • Posts: 72
  • Country: il
Hello all,

I’m planning to design an open hardware project focused on a precision measurement amplifier for DC/AC current, voltage, power, energy, and charge, with performance comparable to or even surpassing the Jetperch Joulescope.
Measurements will be done for up to 16V rails and about 3A currents for dual rails.
3rd channel will be used to monitor very precisely a single positive voltage up to 48V.

I've already selected the primary components and created a high-level hardware block design. My goal is to use high-quality parts to ensure durability (built to last and good performance).

The design will include options for measuring multiple rails, including a negative rail.
It will feature two input power source options, a USB input for programming, and data logging to a PC (with software to be developed later).
Also, three OLED displays will show data from the different channels, refreshing every 1-2 seconds.

For the initial version, I intend to develop the firmware myself. However, as I'm still a novice programmer, my choices for the MCU and ADCs are limited by the availability of compatible libraries and HAL.

What are your thoughts on my design?
Are there any aspects you would change or suggest improvements for?
What areas should I double-check or be cautious about?

Feel free to provide honest and direct/harsh feedback—I can handle it!


 

Offline PCB.Wiz

  • Super Contributor
  • ***
  • Posts: 2138
  • Country: au
I’m planning to design an open hardware project focused on a precision measurement amplifier for DC/AC current, voltage, power, energy, and charge, with performance comparable to or even surpassing the Jetperch Joulescope.

You have not defined what your precision means, but the SSR resistance is 50mOhms max and 25mOhms typ and is in series with your 100mOhm
so you have temperature and part variations of over 10% = 'precision' ?

As mentioned in the other SSR thread, you can trade some gain for better precision.
If you feed equal resistors from the 3 shunts to the PGA, you entirely avoid the SSR effects with a gain of 1/3, or you can add a second set of analog SW, to pick shunt only.
 
The following users thanked this post: Sariel

Offline Kleinstein

  • Super Contributor
  • ***
  • Posts: 15226
  • Country: de
For the relative low resistance shunt one can have the 3 shunts in series and always look at the voltage for the whole chain. The switches should only redirect the current path and not be in the voltage path.

The circuit is still missing the protection part. This could get somewhat tricky with the combined current and voltage inputs.
 
The following users thanked this post: Sariel

Offline SarielTopic starter

  • Regular Contributor
  • *
  • Posts: 72
  • Country: il
Thank you for your responses!

Could you please share a simple drawing showing how to connect the relays so that the measurements exclude their Ron values?
I completely agree that calibrating out the SSR Ron would be quite frustrating.
 

Offline PCB.Wiz

  • Super Contributor
  • ***
  • Posts: 2138
  • Country: au
Could you please share a simple drawing showing how to connect the relays so that the measurements exclude their Ron values?
I can see a couple of choices :
a) Simple feed a resistor (eg 3k3) from each shunt into the PGA input. 3k3 gives appx 0.1% error, (you can calibrate that out)  and the gain is 1/3.
b) Use cheapest/lowest C SSR's x 3 to MUX the shunts into the PGA input.  eg GAQY221 is cheap.  More space and more $$, but avoids the 1/3 Gain.

Addit : I see Panasonic have a low-C AQY221N2S, intended for signal switching, but it is very expensive. You probably do not need the low C / very fast response ?

Digikey show a marketplace APY221S that has reasonable C/price  (claims 6pF / 41c)
lcsc have a part  that looks equivalent  SUPSiC-GAQY221S C7435105, 6pF   25c/100


Addit 2: I spotted a PANASONIC  AQY221N5TY - seriously cute, a stand-up model for high density PCB.   

Lcsc have them in stock :
​AQY221N5TY   PANASONIC  SPST-NO(1 Form A) 1.14V 20V 2.8Ω 1.1pF 180mA VSSOP-4-1.8mmx2.1mmx2.9mmHi   30+ $3.8203
« Last Edit: November 05, 2024, 11:08:22 pm by PCB.Wiz »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf