• State TI Thinks Resolved
  • Locked Locked
  • Replies 13 replies
  • Answers 5 answers
  • Subscribers 46 subscribers
  • Views 574 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.

ADS1263: Data converters forum

Esther Lee
Prodigy 70 points
Part Number: ADS1263
Other Parts Discussed in Thread: INA828, ADS1256, , ADS1262, ADS127L11, ADS124S08, ADS1258, ADS131A04, ADS131E06, ADS1261, ADS1260

I am looking for an ADC for INA828 which is taking inputs from a loadcell. I would like to be able to have 12 single-ended inputs ranging from 0-5V and a sample output speed of >1kSPS per channel. I am currently using the ADS1256 and am wondering if there is a higher performance ADS I could use. May I check whether ADS1263 is a suitable ADC and what the performance differences would be for ADS1262 AND ADS1263? 

  • 0
    TI__Mastermind 35111 points

    Hi Esther,

    To answer the last question you asked, both the ADS1262 and ADS1263 are 32-bit ADCs. However, the ADS1263 has an additional 24-bit ADC inside - that is the only difference between these two ADCs.

    For your specific application: will your signal chain be External MUX --> INA828 --> ADC? Or were you planning on having multiple INAs per ADC?

    If you have the INA828 in front of the ADC, do you need an ADC with integrated gain? I assume you will set the system gain using the INA, not the ADC.

    If you just need a 1-ch, higher-speed ADC with no gain, I could point you to the brand new ADS127L11: https://www.ti.com/product/ADS127L11

    The ADS127L11 is a very small, very high performance delta-sigma ADC that can sample up to 1 MSPS. Right now it is being offered in a TSSOP package, but it will also be available in a 3x3mm QFN package if your system is sized constrained. Samples and EVMs are currently available for this ADC at the link I provided, and it will release later this year.

    If your signal chain will be different than what I have described here, please let me know so we can identify a suitable solution.

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 70 points

    Hi Bryan,

    Thanks for your reply. I am planning on having multiple INAs per ADC if possible and will be setting the gain with the INA.

  • 0 in reply to Esther Lee
    TI__Mastermind 35111 points

    Hi Esther,

    The ADS1262 / ADS1263 do not allow for 12x single-ended inputs unfortunately (10x max), so either way you would need a mux.

    We also have the ADS124S08 that does allow up to 12x single-ended inputs, but it has a max data rate of 4 kSPS. Therefore there is no way you could get 1 kSPS/Ch with this device unfortunately.

    That leaves the ADS1258, which is a 24-bit, 16-ch, 125kSPS delta-sigma ADC. Since you will have 1x INA per input channel, this ADC would be ideal for your application. Also, the noise of this system will be dominated by the INA, since the ADC noise will be divided by the gain. So the ADC noise should be less of a concern.

    Finally, there is always the option of using a MUX + INA + 1-ch ADC such as the ADS127L11 that I mentioned previously.

    Let me know if you have any questions about these options.

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 70 points

    Hi Bryan,

    Thank you for the suggestions. Could I ask how you determine the output data rate per channel for the ADS1258? Also, would simultaneous sampling ADCs would be an alternative choice and what factors I would have to consider if I were to use a simultaneous sampling ADC instead?

  • 0 in reply to Esther Lee
    TI__Mastermind 35111 points

    Hi Esther,

    Table 7 in the ADS1258 datasheet gives the effective data rate in Auto Scan Mode assuming different DLY times. Note that the effective data rate is for one channel.

    For example, if DLY = 000 and DRATE = 11, then the effective data rate is 23,739 SPS for a single channel. If you wanted to measure 12 channels, then your effective data rate is 23,739 / 12 = 1,978 SPS / Ch. Or, about 2x what you need.

    Also, the ADS1258 has a MUXOUT / ADCIN feature that allows you to put 1x INA in between the MUX and ADC. This might save having to put 12x INAs in your system, assuming the INA can drive the ADC inputs to the resolution you need.

    You could use a simultaneous sampling ADC as well. This would allow you to sample a bit slower since you would be sampling in parallel, not in series as you would with a MUX'd ADC. If you were definitely going to use 1x INA per ADC channel then this might be a better choice. We do not have any ADC with 12x simultaneous channels however, so you would need multiple devices. I would consider something like the ADS131A04.

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 70 points

    Hi Bryan,

    I have some concerns regarding the input range of the ADC1258/ADC131A04. As my inputs would range from [0,5)V, according to the ADC1258 datasheet, it seems that the full scale range is 1.096*VREF where VREF = 4.096 which doesn't cover the range I require. Please correct me if I calculated wrongly. Also, the ADC131A04 also seems to be limited to around 3.3V range which doesn't suit my requirements. 

    Would you have other suggestions if I were to have 2 ADCs reading 6 inputs each instead?

  • 0 in reply to Esther Lee
    TI__Mastermind 35111 points

    Hi Esther,

    For the ADS1258, you can simply bias the AINCOM pin to something like 2.5V. Your applied signal of 0-5V will then be measured against this voltage. So an input of 0V will read as 0V - 2.5V = -2.5V, while an input of 5V will read as 5V - 2.5V = 2.5V. This will actually use more of the bipolar coding scheme than a true single-ended signal (that only uses half of the code range). If VREF = 2.5V, then you are going to use a majority of your full-scale range with this device.

    The A04 operates similarly, and can operate on a 5V supply with the charge pump disabled (see section 11 in the datasheet). Also, the INA828 has a VREF pin that will set the output signal common-mode to whatever is necessary for the ADC. And the gain resistor RG will allow you to tune the amplifier gain to the input range of the ADC. Again, I believe either of these options should work for you.

    You could also consider the ADS131E06 if you are looking for a 6-ch simultaneous sampling delta-sigma ADC. This device has an integrated amplifier however that cannot be bypassed.

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 70 points

    Hi Bryan,

    Thanks for the suggestions. To clarify, if I were to bias the AINCOM pin to 2.5V, the VREF pin should be set to 0V? Also, would you have suggestions for multiplexed ADCs that can sample at least 6 channels?

  • 0 in reply to Esther Lee
    TI__Mastermind 35111 points

    Hi Esther,

    No, the VREF pin still requires a valid reference voltage or the ADC will not work. If you use VREF = 2.5 V then you could also use this voltage to bias AINCOM.

    For the 6-ch ADC: do you need 6 single-ended or differential inputs? And do you still require 1kSPS per channel? Or is the sampling speed different?

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 70 points

    Hi Bryan,

    I realised my question might have been a little confusing. I meant to ask what the VREF for the INA828 should be set to. I need 6 single-ended inputs. I would still require at least 1kSPS per channel preferably. 

  • 0 in reply to Esther Lee
    TI__Mastermind 35111 points

    Hi Esther,

    Thank you for clarifying. Typically you would connect REF on the INA to mid-supply, as this puts the output common-mode signal at the mid-supply voltage of the ADC. But this of course depends on the output signal from the INA. Can the output of the load cell swing both positive and negative (that is, compression and tension)? Or are you just looking at the positive direction e.g. for weight? In either case you just want to consider what the output span of the INA will be. If the output is 0-5V relative to the INA REF pin, then it makes sense to tie REF (and AINCOM) to 2.5 V. Then the INA output of 0-5V looks like a -2.5V to +2.5V input signal to the ADC. If the INA output is something else, let me know and we can work through that issue.

    We don't have too many 6-ch ADCs, but the ADS1256 could be an option. This is a 24-bit delta-sigma ADC with 8 single-ended channels. If you look in this ADC's datasheet, Table 14 identifies the multiplexed throughput for each data rate. For example, setting the data rate to 2 kSPS provides a multiplexed throughput of 1438 SPS, which means you would get data out of each channel at this data rate (assuming your SCLK  and master clock speeds match the values in the datasheet, and excluding any external analog settling). This device could be a good option for your application.

    -Bryan

  • 0 in reply to Bryan Lizon86
    Prodigy 70 points

    Hi Bryan,

    I'll be looking only at the positive direction. Would it affect the operation of the INA if I connected the INA REF to +2.5V since I'm running a dual supply on it to get the full range from 0V? I am open to 8-ch ADCs too if there aren't many 6-ch ones. I am currently using the ADS1256 but I was wondering if there were any newer options available since it has been around for a while. 

  • 0 in reply to Esther Lee
    TI__Mastermind 35111 points

    Hi Esther,

    As long as you adhere to the INA datasheet guidelines about input, output, and common-mode signals given the supply range, this should not affect the INA or ADC performance.

    Indeed you did mention that you were using the ADS1256. We have released a next-generation version of this device, the ADS1260 (5-Ch) or ADS1261 (10-ch), that could be of interest to you. The ADS1261 just has more channels than you requested, so I did not mention it earlier. But if you wanted the latest and greatest device in this family, then the ADS1260 / ADS1261 are it. They are also smaller and lower cost compared to the ADS1256, and include a 2.5V VREF that could be used to bias the REF pin on the INA and the AINN pin on the ADC.

    Let me know if you have any questions about these devices.

    -Bryan