Help with High Impedance Voltmeter

Discussion in 'The Projects Forum' started by snarf, Jun 7, 2007.

  1. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Volume VI - Experiments High Impedance Voltmeter

    When trying this experiment I still get a reading of about 0.5V on the Voltmeter even when the test probe is not connected.

    The OA I'm using is an ST MC1458N because that's what I found at the electronics shop. I was told that this is the same as the LM1458. All the other parts are OK apart that instead of the four 6V batteries I'm using a DC power supply.

    I'm still very green on the subject but to me it seems like the OA is in "latch-up". The Voltage readings taken from the three test points are correct.

    Please help.

    Also how can I use this circuit to measure an external Voltage source?

    Thank you.
     
  2. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    With the input floating, the output tries to go to the negative power supply rail. I assume you are using a single (not dual) supply. You negative rail is therefore ground. The 741 op amp's output stage will only go to within about 0.5 - 1 volt of the negative rail. The actual value will depend on the manufacturer, the output load (~15k, in this case), and the individual op amp.
     
  3. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Thank you for reply.

    Yes I'm using a single supply and my negative rail is ground.

    But I didn't understand

    The 741 op amp's output stage will only go to within about 0.5 - 1 volt of the negative rail. The actual value will depend on the manufacturer, the output load (~15k, in this case), and the individual op amp.

    You mean that if using a single supply the Voltmeter reading will not go to zero?

    Can this be overcome? and how?

    Can you also try to reply to my other question?

    Also how can I use this circuit to measure an external Voltage source?
     
  4. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    You can use dual supplies (positive and negative), or you can get an op amp whose output goes (almost) to the negative rail. LM321 (single) and LM324 (dual) are common op amps with this capability. You can also search for rail-to-rail output op amps.
    To measure an external source, connect the negative terminal to ground and connect your probe to the positive terminal. You will be limited to measuring voltages that are less than your positive supply voltage. The LM321/324 will only go to within about 3 volts of the positive rail.
    You can re-scale your voltmeter by changing the 15k resistor to another value. 30k will give you half the sensitivity, 7.5k will give you twice the sensitivity.
    If you really want to measure voltages, go out and buy a cheap multimeter.
     
  5. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Once more thank you for your reply.

    I already have a multimeter, but since I have some 0-15V analog Voltmeters I would like to convert these to high impedance thereby reducing the load the Voltmeter creates and improving it's accuracy, and since it seems possible I'm also trying to learn something out from it.

    Sorry for being so ignorant but I'm a beginner and still don't understand much about electronics.

    I'm aware and familiar with multipliers that would be required to change the fsd of the Voltmeter and a 30K resistor will enable the Voltmeter to read 0-30V, but sorry I still have not understood your explenation.

    I have several MC1485N in hand and would like to make some more tests with them. Therefore you mean that it would be better to use a dual supply for the MC1458N? What is rail-to-rail output OA? Do I remove the 1M resistors from the circuit?

    Maybe a circuit diagram would help me understand better what you are trying to explain. Or perhaps an indication to where I can find such a circuit layout and description which would also include some form of circuit protection against reversed Polarity input voltage and over voltage input.
     
  6. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    This should allow you to do what you requested. I need to explain a few things.
    You need the negative supply. With a single supply, you can only measure voltages down to about 2 volts when using the MC1458 (or 741), due to the input common mode range. As shown, you should be able to measure from 0 to 15V with a 1mA full scale movement. The supply voltages don't need to be precisely what I have shown. If you have something close, let us know and we'll talk about it.
    R2, D1 and D2 provide overvoltage and reverse polarity protection.
    D3 prevents the needle from going below zero when the input is negative or floating.
    The 100nF capacitors are to prevent the op amp from oscillating.
    The unused op amp in the package needs to be connected as shown. Leaving the inputs floating can result in screwing up the internal bias network, which may be common to both op amps.
    The returns on your power supplies connect to ground.
     
  7. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Many thanks for all your detailed help.

    I'll need to get the 100nF capacitors as I don't have any of that value and I will also need to get the transformer because the DC supplies that I have are up to 24V but do not have a centre tap for the negative voltage. I also have transformers with a center tap but these are 9-0-9 Volts AC and 12-0-12 Volts AC.

    I'll try to get the parts tomorrow and let you know.
     
  8. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Hang loose, I think we can use the unused section of the op amp to get your two voltages from the 24V DC supply. I'll post a new schematic.
     
  9. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    OK, here's a new schematic. U1b creates an artificial GND, giving you +20V and -4V from the 24V supply. Don't connect the GND in the voltmeter to the GND on the 24V supply, if it has one. You can measure just about any voltage, so long as it isn't developed off the 24V supply. Actually, a wall wart or batteries would be less confusing. Of course, you can use the original circuit with 2 supplies if you want.
     
  10. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Ok but if you think it would be better with a dual supply I'll see what I can find or if you think that I should use another OA then I will use what you will recommand.

    Thank you.
     
  11. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Again many thanks.

    I'll try to put the circuit together on the breadboard and do some testing.

    I will keep you informed.
     
  12. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Since a moving coil ammeter is not exactly high precision, the 1458 should be fine. If you don't need the 24 volt supply for anything else while you are playing with the voltmeter circuit, the artificial ground circuit should also be fine. You could always split a wall wart the same way, if you can find one with high enough voltage.
     
  13. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    I’ve set the circuit up and following are the results.

    Probes not connected:
    Voltmeter reading zero
    V1 24.5 Volts 4.28mA
    Across R3 25.7 Volts
    Across R4 -3.3 Volts

    -ve probe connected:
    same as above

    +ve probe connected:
    Voltmeter reading 5.5 Volts. (Touching the probe’s cable, not the tip, varies the reading)
    V1 24.5 Volts 4.28mA
    Across R3 25.3 Volts
    Across R4 -3.3Volts

    Measuring 15V DC:
    Voltmeter reading 15 Volts. (In fact as you stated it can measure between 0 and 15V)
    V1 23.5 Volts 4.58mA
    Across R3 23.4 Volts
    Across R4 -3.3Volts

    I don’t know if the above readings are useful but I thought I’d include them just the same. Some things are still not clear to me, therefore I ask.

    1. If V1 is in the 24V range why is the voltage across R3 close to V1? I thought that having the potential divider R3 and R4 the resulting voltage across R3 should have been in the +20V region. The negative voltage across R4 seems more stable although it’s not -4V. Why are the voltages not +20V and -4V?

    2. I’m aware that the circuit makes the voltmeter more sensitive and makes it high impedance. In fact there is no voltage drop to the voltage being measured due to loading as would have been the case without the high impedance circuit. But the positive probe seems to pick stray currents which result in a reading on the voltmeter. Can this be overcome?

    3. Usually by changing the value of R1 one can change the scale of the meter. Can this also be achieved with this circuit? Or is it limited to 15V maybe because if the range is increased to say 30Vor more it would be too close to the +20V or exceed the +20V, thereby not allowing a correct reading?

    I like the idea of having a high impedance moving coil meter and since I have some of these meters I’m thinking of converting them to high impedance and put them to good use in some application or other. Before I came across this high impedance circuit I had constructed one for voltage and current with several ranges by employing multipliers and shunts, but obviously having the loading problem. If possible I would like to achieve this with high impedance. Can you help please?

    Your advice would be appreciated. If you recommend that another op amp and another circuit is required I will follow your advice.

    Again thank you for your assistance.
     
  14. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Your readings across V1, R3, and R4 don't make sense. Are you measuring them with a multimeter? What kind of power supply is V1? Is it regulated?
    I'll deal with the other questions after I understand this issue.
     
  15. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    I thought so, that’s why I posted the readings. Yes I’m measuring the voltages and current with a digital multimeter.

    It’s a 24VDC power supply 20VA. No it’s not regulated, but it’s got one smoothing capacitor 1000uF 40V.

    The electronics shop I go to has the L7824CV 24V 1.5A regulators. If required I can get one on Monday afternoon.

    Is this the right way to connect the regulator?
     
  16. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    A 24V regulator won't work on a 24V supply. It needs about 3V headroom. You could use a 20V regulator, but I don't think a regulator is necessary.
    As you should know, the sum of the voltages across R3 and R4 should equal the voltage across V1. Check those again.
    Also, V1 should not be varying noticeably as a function of what you are measuring with the probes in your circuit. Are you using both halves of one op amp chip in the circuit, as shown in the schematic? Can you post a picture of your breadboard?
     
  17. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Thanks for reply.

    I have checked the voltages several times, but I'll check them again. Yes I'm using both halves of the same op amp chip in the circuit as shown in the schematic.

    I'll check the layout again to be sure, although I've already done so several times.

    All this will have to be done some time tomorrow because it's 04:35am here and I'm a bit sleepy. Sorry.

    I don't have a digital camera so I can't post a picture of the breadboard.

    Will let you know the results tomorrow, well for me later on today as it's already Sunday here.
     
  18. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Sorry it took me some time to reply but I wanted to be sure of everything.

    In my opinion the problem could be with the power supply so I checked that as well and here are my findings.

    1. Output 26.3VAC
    2. Output from Bridge rectifier 23.4VDC
    3. Output from smoothing capacitor 29.5VDC

    Using the DC output from the smoothing capacitor and dropping it to 24VDC with a series resistance of 8K2 I get the following results.

    Probes not connected:
    Voltmeter reading zero
    V1 24.4 Volts 1.49mA
    Across R3 19.9 Volts
    From +ve rail to circuit ground 20.4Volts
    Across R4 -4.0 Volts
    From –ve rail to circuit ground -4.1Volts

    -ve probe connected:
    same as above

    +ve probe connected:
    Voltmeter reading zero unless probe’s lead is touched.
    Voltage readings same as above

    Measuring 15V DC:
    Voltmeter reading 12.5 Volts instead of 15 Volts. The meter is not reaching full scale.
    V1 17.25 Volts 2.45mA
    Across R3 15.42 Volts
    From +ve rail to circuit ground 14.8 Volts
    Across R4 -2.64Volts
    From -ve rail to circuit ground -2.86

    The voltages start to drop when taking measurements and continue to decrease when the voltage being measured is increased. The current on the 24V rail increases with the voltage increase reaching 2.6mA at 12.14V. Further voltage increase will show a current decrease. The highest voltage measured correctly was 12.5V. To me this maybe because V1 is dropping, possibly because of the series resistance I included to bring the 29.5V to 24V, therefore making V1 unstable.

    If I remove the resistance and smoothing capacitor and use the 23.4VDC all voltages remain stable and do not vary when the voltage being measured is changed. Readings are.

    V1 23.0 Volts 1.45mA
    Across R3 18.9 Volts
    From +ve rail to circuit ground 18.9Volts
    Across R4 -3.7 Volts
    From -ve rail to circuit ground -4.0Volts

    But the reading of the measured voltages are inaccurate. For example 5V reads 4.5V, 9V reads 7.5V,15V reads 11.8V. Could this be because the DC is not being smoothed or is it because V1 is 23V?

    This sounds strange to me. With variable voltages the reading is accurate up to 12.5V and with stable voltages the reading is inaccurate.

    To me the circuit layout on the breadboard is ok, it’s well spaced and no connections are shorting. I even tried replacing the op amp and still get the same results. The op amp pin connections I got from the ST MC1458N data sheet and are.

    Top view notch facing north, left hand side starting from the top pin.
    Pin 1 – output 1
    Pin 2 – inverting input 1 -ve
    Pin 3 – non-inverting input 1 +ve
    Pin 4 – Vcc -ve

    Top view notch facing north, right hand side starting from the top pin.
    Pin 8 – Vcc +ve
    Pin 7 – output 2
    Pin 6 – inverting input 2 -ve
    Pin 5 – non-inverting input 2 +ve
    Using this second op amp for the artificial ground, pins 5, 6 and 7.

    Maybe a better power supply is required. Your advice please.
     
  19. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    I didn't read the entire book you just posted, but I did read enough to trouble me:

    1. The output voltage of your bridge is not the same as the voltage on your smoothing capacitor.

    2. You can't use a series resistor to drop the voltage. It raises the output resistance of the power supply unacceptably.

    You need to post a complete schematic of your power supply, and how you have connected it to your circuit.
     
  20. snarf

    Thread Starter Active Member

    Jun 7, 2007
    56
    0
    Sorry for writing so much, I tried to explain my findings.

    1. The output voltage of your bridge is not the same as the voltage on your smoothing capacitor.

    True but but that's what it is, I didn't make it, I bought it ready. I'll replace it with a 24V capacitor if that will be better. I'll also remove the series resistor. Will that be ok?

    Schematic attached. I'm connecting the supply with two short leads from the power supply to the breadboard. An in-line on/off switch is on the +ve lead.
     
Loading...