Connect with us

Measuring ballast output

Discussion in 'General Electronics Discussion' started by Jane Panovski, Sep 13, 2011.

Scroll to continue with content
  1. Jane Panovski

    Jane Panovski

    2
    0
    Sep 13, 2011
    Hi

    I have a Tektronix TDS 210 oscilloscope and I'm supposed to measure the output current of several types of ballasts. So far I found out that I need to use a shunt ressistor because I don't have a current probe, and that I have to use both channels considering I dont have a differential input. As a shunt I tried to use a 3,7 m, 0,25 mm^2 copper wire (if I calculated corectly it has about 1.4 ohms resistance). I connected this wire in series with the lamp (ballast). The lamp is connected directly on 231 V. I used the math function Ch1-Ch2 (one probe on each end of the 0,25 mm^2 wire), and what I got was a normal sinusoidal wave. This can't be correct! No matter how high quality the ballast is, it has to change the shape of the current signal. It seems to me I'm getting the input voltage from the city grid. Could someone suggest what am I doing wrong please?

    Thank you in advance.
     
  2. Resqueline

    Resqueline

    2,848
    2
    Jul 31, 2009
    It seems you're trying to measure/calculate a mV differential signal on top of a 230V mains voltage. That is not considered good measuring technique, nor is it quite safe.
    You don't say where you put the ground clips on the probes. Chances are you left them disconnected, or else some mains circuit fuse would likely have blown.
    Or did you make sure to put the shunt in the neutral line? Nevertheless there can be several volts between neutral and ground, and high currents can flow if shorted.
    You have two options imho; either use an insulation transformer before the ballast so you can ground the shunt at one end, or get a current transformer (current probe).
     
  3. daddles

    daddles

    443
    3
    Jun 10, 2011
    The method is known as the "poor man's differential amplifier". It's an acceptable measurement method provided one knows not to connect the probe grounds to the circuit under test as Resquiline says (you only make this mistake once, and, yes, I've made this mistake :)). You don't say what current levels you expect in the ballast, but with a 1 or 2 ohm shunt, you'll likely be looking at signals with an amplitude of perhaps 0.1 to 1 volt.

    You probably should first establish that your scope is capable of such measurements. This is no big deal for a differential amplifier, but you it's possible you could be straining the scope. I'd assume you're using 10X probes, so the scope is being asked to subtract 20 to 30 volt signals. This shouldn't really be a problem.

    I've looked at the currents from 15 W fluorescent ballasts and, as I recall, they were definitely not sine waves, but looked like the sketch I've attached.
     

    Attached Files:

  4. Jane Panovski

    Jane Panovski

    2
    0
    Sep 13, 2011
    According to my calculations, I should get about 500 mV. And no I didn't connect the ground clips, I took them off :). The shunt (the wire actually) is in the neutral line. This way, when I put each probe on one side of the shunt I get exactly what I get if I put one probe in the neutral line from the grid. Don't understand why :(.
     
  5. duke37

    duke37

    5,364
    769
    Jan 9, 2011
    I would go for the current transformer solution to isolate the scope from the mains. Take a mains transformer and feed a single turn around the bobbin, put a 100 ohm resistor across the old primary and measure the voltage across that with the scope.
     
  6. daddles

    daddles

    443
    3
    Jun 10, 2011
    Well, welcome to the real world. :) Often, much of the time of making measurements is setting things up, checking them, double-checking them, then checking the results. A lot of work is like that -- for example, sometimes it can take a day or two to set up a cut to be made in a machine shop on a particular machine. Then the actual cut only takes a few minutes.

    In other words, it's easy to get measurements, but it's critical that the experimenter convince themselves that the measurements are meaningful and that they're not being fooled or making a mistake. We make the measurements and then check the results with our mental models of what we think we should have gotten along with making other efforts to verify the results.

    Have you verified that the scope can make a proper differential measurement? This is trivial to do. Just take e.g. a battery and connect each terminal to one of the channels and subtract the signals as you're doing for the AC measurements. You'd better get the DC open circuit voltage of the battery or something's wrong. Use this to estimate the measurement error too -- a critical task for any measurement. I've attached a couple of screenshots of my scope making such a measurement of a 9 volt battery, then with the leads shorted. The traces were offset slightly to show all of them. This demonstrated to me that the scope could make the differential measurement.

    Suppose you're trying to measure a 1 A current. OK, get a DC power supply and set it up to put 1 A of current through a suitable shunt. Verify the current and the shunt's resistance. Then use your scope without the probes to make your poor-woman's differential amplifier measurement of the voltage dropped across that shunt. If you can't make the measurement, then you have to figure out why. Then add the probes into the measurement situation and verify you can still make the measurement (you might want to increase the current an order of magnitude though if they're 10X probes).
     

    Attached Files:

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-