• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 47 subscribers
  • Views 5537 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Low cost high side currrent measurement

Tyler Brooks47813
Prodigy 40 points
Other Parts Discussed in Thread: LM2902, INA148, INA332

I am designing a light dimmer for the US market that has power measurement capabilities.  I am looking for a low cost solution.  I want to be able to control/measure loads from 0W to 300W in 5W steps (or so) so accuracy is less important than cost.

My first thought was something like I have attached.  I have built it on the bench with mixed result.  The current sensor is 'touchy' as you might expect.  I have also tried a couple traditional instrumentation amplifier configurations with a little better result.  I find that in all configurations if I increase the AC voltage much past 120VAC, then the opamps become unstable and slam against the rails.  I take this to mean that I am struggling with common mode voltage.

3716.HighSide.pdf

 I suppose this is an application for the INA148.  However, it is pretty expensive for this application and probably over-kill.  Could I use the INA332 for this application?  Is there a better (and cheap) opamp that TI sells that could replace the LM2902?  Any thoughts are greatly appreciated.

 

  • 0
    Prodigy 40 points

    Bump.

    I am still looking for suggestions on the current sensor.  I have put the attached schematic in spice and it is indeed 'touchy'.  Even 0.1% resistors don't appear to be good enough.  If I set the resistor values to their error extremes, I can get large differences in the input.

    5850.HighSide_schm.pdf

    Would the INA148 solve this issue?

     

  • 0 in reply to Tyler Brooks47813
    TI__Guru**** 152966 points

    To get a high side current monitor to work with 120V AC will require very high common mode rejection ratios.
    The voltage across the sense resistor is quite low compared to the main AC voltage.
    It will be much easier to get a low side current sense implemented.

     

  • 0 in reply to Ron Michallick
    Prodigy 40 points

    Understood.

    However, given the current state of wiring in most American homes, I am forced to consider putting the shunt on the high side.  The reason for this is that in 'gang' switch boxes (switch boxes with multiple light switches) the neutral is often buses together at a single wire nut.  Only the hot side is broken.  I could put the shunt on the low side but then I am forcing the installer (maybe just a homeowner with no training) to find the correct neutral and rewire the box.  This is something we would like to avoid if possible.

    I downloaded the spice model for the INA148.  It works pretty good in the simulation.  I have not tried it on the bench yet.

    Another approach we are considering is the MSP430FE42x.  These are commercial grade metering devices that appear to work on the highside.  Can you verify that for me?  I am in the process of selecting an eval board to verify.  Notice that the MS430 is much cheaper than the INA148 in Q1K or more.  However, I am short on space so the (1.5" x 2.5", double sided) so the 64-pin QFP of the MSP430 might be too much.

  • 0 in reply to Tyler Brooks47813
    TI__Guru**** 152966 points

    I took a look at the MSP430FE42x. The circuit examples showed transformer coupled or a low side shunt resistor sensing. The MSP430 is a computer on a chip.  

     

    If safety rules allow, you could put both the current sense resistor and low voltage power supply reference (‘ground’ point) on the hot line. I would not be surprised if this practice was prohibited.

  • 0 in reply to Ron Michallick
    Prodigy 40 points

    Are you sure the MPS430FE42x can only have the current shunt on the low side?  In what document did you find the circuit example to which you refer?

    In 'Implementing An Electronic Watt-Hour Meter with MSP430FE42x(A)/FE42x2' (slaa203c.pdf)  Figure 1 shows the shunt on the high side.  I took this to mean that they had solved the large common mode voltage problem. 

    Unfortunately, the reference board to which they refer can not be purchased so I guess I will have to build something up and try it myself. 

  • 0 in reply to Tyler Brooks47813
    TI__Guru**** 152966 points

    This is a case where we are both correct. Figure 1 in 'slaa203c' does show a shunt on the hot line. The local DC supply is not described; however it is obvious from the way the line voltage is reduced that the local "ground" for the meter is at line (hot) potential. All inputs to the MSP430FE42x must be between the device’s ground and VCC pins per datasheet’s recommended operating table.

    The example in the MSP430 data sheet. http://focus.ti.com/general/docs/lit/getliterature.tsp?genericPartNumber=msp430fe423&fileType=pdf shows in figures 14 and 15 that use transformer and low (neutral) side sensing. In this circuit the local DC ground is at low (neutral) potential.