• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 62 subscribers
  • Views 320 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.

REF200 significantly noisier than datasheet specs

Alexander Dingwall
Prodigy 65 points
Part Number: REF200

Hello,

I am trying to use the REF200's 100uA current source to measure the resistance across a 470 ohm resistor. We are measuring the voltage across this resistor using an Agilent precision digital multimeter as well an Agilent oscilloscope and experiencing significantly more noise than shown in the data sheet.

I am referencing figure 5 in the data sheet which shows the noise in the current never exceeding 1000pA. Correct me if I'm wrong but that means the precision in the voltage across the 470 ohm resistor should not exceed 1000pA * 470 ohm = 0.47uV. Using both the multimeter and oscilloscope we were seeing noise in the range of ~50uV.

Is there a specific way this IC needs to be integrated into a circuit? We are powering the chip with a 9V regulated voltage and there is  a 2.2 uF capacitor across the high pins and the substrate as shown in the layout schematic in the datasheet.

Thanks

  • 0
    TI__Mastermind 19370 points

    Hi Alexander,

    Can you share your schematic? I want to understand this better.

  • 0 in reply to Marcoo Z
    Prodigy 65 points

    Hi Marcoo,

    Here is how I have the chip set up. The values are as follows: VDD = 12V, C7 = 0.47 uF, C8 = 22uF, C9 = 2.2uF and R2 = 2.2kΩ.

    Thanks

  • 0 in reply to Alexander Dingwall
    TI__Mastermind 19370 points

    Hmm I think this might be related to the connections

    Pin 6 should be connected to GND.

    You can use this  TIDA schematic as an example

    This is also a bit overkill for a 100uA current source. have you thought of using figure 8 of the following app note?

  • 0 in reply to Marcoo Z
    Prodigy 65 points

    Sorry my mistake. Pin 6 is connected to ground. I just missed adding it to the schematic.

    I saw that schematic you sent in the REF200 datasheet. I'd rather not use the amplifier because I wish to measure a wide range of resistors using this source and an ADC whose max input voltage is ~2V. If we exclude the amplifier from that  schematic, then my set up is very similar to that one, just without the second current source going through R3. Would you recommend adding 'R3' to my setup to decrease the noise?

    Can you please explain why using the REF200 in this case is overkill? I'm aiming for a very low noise voltage signal across the resistor that I'm trying to measure. Why do you recommend building my own current source using the LM4041-NADJ instead?

  • 0 in reply to Marcoo Z
    Prodigy 65 points

    I marked the question as resolved by accident.

    Sorry, my issue is not resolved.

  • 0 in reply to Alexander Dingwall
    TI__Mastermind 19370 points

    Hi Alexander,

    The REF200 is very old. I looked into what documentation we have on this device. I have not found any on how the device was tested.

    The 1nAp-p is for 1/f noise. There is still wide band noise that will contribute to the output. I would add a capacitor on the output to reduce some of the noise.

    I can see a situation where the wideband noise is contributing a large amounts of noise error due to the lack of output cap. I do not have the noise spectrum output of this device so it is difficult to know what the cut off is of the device.

    At the same time I am unsure on how much rejection there is on the REF200 from the LDO noise.

    Is it possible to measure the noise before the REF200 and compare it to the noise after?

    Same test with a capacitor at the output to limit some of the noise.