• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 282 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.

ADS1255: ADS1255 - noises on output

Leszek C
Prodigy 30 points
Part Number: ADS1255

Hello

We've been using the ADS1255 in microvolt level voltage measuring device (EEG amplifier) for many years without any problems. The output signal was always ok, free of interference and the noise level was low. However, for last several months we have been observing some strange phenomenon. The last few batches of this IC that we bought and installed on PCB work differently than all previous batches. A significant amount of noise, strange bands in the FFT spectrum, and signal aliases appeared in the output signal. The systems began to behave as if it were a "sigma-delta idle tones" problem. We've never seen it before. We still use the same hardware and software without any changes and without any problem. Sampling takes place at a speed of 500sps, and in one device we have up to four ADS1255 ICs driven by the same clock signal.

Has anyone observed such behavior recently and knows a reason?

  • 0
    TI__Mastermind 44550 points

    Hi Leszek,

    Welcome to the TI E2E Forums.

    I'm not aware of any changes that have been made to the ADS1255 anytime recently. Have the devices been sitting on the shelf for a awhile (> 2 or 5 years) or are these recently purchased devices? If they have been on the shelf for a long time then baking can performed to remove absorbed moisture from the IC package.

    It is possible that you could be seeing a delta-sigma idle tone, but I would recommend trying a couple of different tests to help confirm...

    • I would first recommend shorting the inputs of the ADC together (and to ground or any valid common voltage), and looking at the noise performance of the ADC. Shorting the inputs together removes input signal noise and reference voltage noise, and may help determine the noise source.

    • If you have any front-end circuitry prior to the ADC, then try the same shorted-input test with the inputs of the signal conditioning circuit-shorted to test if the noise from this circuitry may be the cause.

    • Probe the reference voltage with an oscilloscope to check that the reference voltage is stable.

    • Do an "A-B" swap test. Take a older version of the board that does not exhibit this behavior and a new version that does...swap the ADCs on these boards and see if the issue follows the ADC or if the issue is still observed on the newer PCB.

  • 0 in reply to Christopher Hall
    Prodigy 30 points

    Hi

    Thank you for your answer. We buy all ICs on a regular basis and we don't store them on the shelves. We also control the humidity and dirt level on PCBs after assembly. Since we noticed the problem, we have already done a number of things and tests, including those you mentioned. We also tried to swap devices. However, all traces always lead to ADS1255.

    We noticed a small dependency. We can't confirm it for sure, but in the attachment I put a photo of two ICs. In our view, those ICs that look like device on the left always behave correctly. Those ICs that look like device on the right usually behave differently. They are different physically: they have a different texture and a different, less legible print. They are definitely from a completely different batches. However, we don't have enough old PCBs and ICs to compare more series, or old photos to compare.

    In the past we bought these ICs at Rebound. A few old PCBs that we have from that time have a layout like the "good" one on the left of the photo. When we started to notice the problem, the ICs probably already looked like the "other" on the right (and look like it till now). We recently started buying these ICs at DigiKey; once we got IC looking like this "good", other time like this "different". And both these series differed from each other in behavior.

  • 0 in reply to Leszek C
    TI__Mastermind 44550 points

    Hi Leszek,

    Thank you for the image.

    Would you be able to share the lot trace code from one or more of the "other" devices?
    ...Unfortunately, we couldn't make out this value from the image.

    The lot trace code from the "good" device appears to have come from Digi-Key.

  • 0 in reply to Christopher Hall
    Prodigy 30 points

    Hi

    Yes, a "good" device on the left side of the photo, with the code 85DEF5K, was bought at Digikey in october this year. However, last month we bought, also in Digikey, ICs with lot code 88CHZ5K, which worked and looked "bad" (like the one on the right side of photo).

    The IC on the right side of the photo has lot code 82CC4QK and was bought by us over the past year in Rebound. Another lot from this year, also bought from Rebound and operating "badly", was 82CC4RK.

    An example of a lot we bought in Rebound in the past and worked correctly is 4CAR3YK. Unfortunately, we probably won't find more older ICs to compare.

  • 0 in reply to Leszek C
    TI__Mastermind 44550 points

    Hi Leszek,

    Thank you for the additional lot trace codes. I'm still waiting for our quality team to get back to me once the'y tracked down the origin of these devices. I'll provide you with an update as soon as I can.

  • 0 in reply to Christopher Hall
    TI__Mastermind 44550 points

    Hi Leszek,

    Our quality team got back to me and indicated that we don't have any record of sending those materials to Rebound.

    Unfortunately, because these devices were not purchased through a TI Authorized distributor (reference: http://www.ti.com/info-store/distributors.html) and they are not traceable to us, we are not able to provide any warranty coverage or customer support for them.

    You will need to contact and work through your supply chain to resolve any issues. If you have any further inquires for us about these devices, please send them to ti-counterfeit@list.ti.com.