• State TI Thinks Resolved
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 32 subscribers
  • Views 90 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

PGA302: Resistive bridge measurement (PGA302)

Milos Jesic
Prodigy 40 points
Part Number: PGA302
Other Parts Discussed in Thread: ADS1235, INA821, INA118, PGA855, PGA849, PGA308, PGA308EVM

Tool/software:

Hello,

I'm looking for the signal conditioning circuit for the force sensor (Wheatstone bridge) with following characteristics:

- Rated Output (RO) 0.5 mV/V nom
- Excitation (VDC or VAC) 5 V max
- Bridge Resistance 1000 Ohm nom

PGA302 seems really nice and compact but the gain and the excitation voltage are too low. For excitation voltage of 5VDC I would have +/-2.5mV at the output of the sensor. Can you recommend some other part number that would work for my requirements?

Thanks!

Milos

  • 0
    TI__Mastermind 22365 points

    Hi Milos,

    thanks a lot for reaching out with your question.

    We don't really offer devices which output a 5V excitation for a Wheatstone bridge, but we offer ADCs for bridge measurement applications which accept 5V as a reference voltage. That allows you to implement ratiometric bridge measurements.
    If you want to learn more about ratiometric bridge measurements using our ADCs, I can recommend the following application note to you: A Basic Guide to Bridge Measurements

    As your bridge sensitivity is pretty small, I would probably start looking at something like the ADS1235 for example, which is our highest resolution ADC for bridge measurements. But we have other devices available as well depending on your power, space, resolution, data rate, etc. requirements.
    Please note that those ADCs do not implement any linearization or temperature compensation as the PGA302 does. Means that would have to be implemented in an external MCU.

    In general, also don't get too hung up about the fact that you don't have as high of a gain available in the ADC as you want. What really matters in the end is the input-referred noise of the ADC and to make sure that meets your noise requirements.
    Do you have any specific noise requirements? E.g. how many noise free counts (50,000?) are you looking for? Or any other noise metric you try to meet?

    Regards,
    Joachim Wuerker

  • 0 in reply to Joachim Wuerker
    Prodigy 40 points

    Hi Joachim,

    Thank you for the quick response and sorry for the delayed response! 
    Yes, ADS1235 looks good, I might go that route. However, I was also looking at some of your instrumentation amps (INA118 or INA821). 
    Currently I don't have requirements locked so I'm exploring different options with different cost/accuracy.

    Can you recommend me some of your instrumentation amplifiers for my application?

    Best,

    Milos

  • 0 in reply to Milos Jesic
    TI__Mastermind 22365 points

    Hi Milos,

    no worries. I am unfortunately "only" an ADC expert and don't know all our latest instrumentation amplifier offerings.
    If you don't mind, could you post your question about the INAs in the amplifiers forum? You may want to let them know what supply voltages you have available and if your ADC requires single-ended or differential inputs. However the INA118 and INA821 would look good to me.
    We also have options with pin programmable gain, e.g. PGA855 and PGA849.

    Regards,
    Joachim Wuerker

  • 0 in reply to Joachim Wuerker
    TI__Mastermind 19672 points

    Hey Milos, 

    I have shifted the question to my responsibility.
    We have many instrumentation (INAs) and programmable instrumentation amplifiers (PGA) that could work. 

    Thanks for providing the following details on the bridge, off the top of my head the PGA308 would likely work. 

    - Rated Output (RO) 0.5 mV/V nom
    - Excitation (VDC or VAC) 5 V max
    - Bridge Resistance 1000 Ohm nom

    What is the expected output voltage? Would 0.5 to 4.5V work? This is a common configuration for PGA308. Or are you connecting a MCU/ADC and require a voltage clamp at 3.3V. 

    What other supply voltages do you have available in your system? The PGA855 and PGA849 require at least 8V supply voltage for the front end powersupply. 

    All the best,
    Carolina

  • 0 in reply to Caro Walter
    Prodigy 40 points

    Hi Carolina,

    Thank you for jumping in! 

    Ideally, I would like to feed the output of my signal processing to the ADC which is 3.3V rated. I could supply the bridge with 3.3V and in that case the output of the bridge would be a signal that would go from -1.65mV to +1.65mV. The amplifier would need to amplify this signal to be something like 0-3V. 

    Thank you!

    Best,

    Milos

  • 0 in reply to Milos Jesic
    TI__Mastermind 19672 points

    Hey Milos, 

    In that case, I recommend going with the PGA308, I do not believe you will have to adjust your bridge excitation, it can stay at 5V. The PGA308 has an output clamp meant to interface with ADCs as needed, here are a couple of options of how to implement this Vclamp:

    The actual output of the PGA308 is determined by programming in the desired PGA output swing. 

    Milos Jesic said:
    For excitation voltage of 5VDC I would have +/-2.5mV at the output of the sensor.

    In this scenario, I am assuming the Vdiff of 5mV is centered around a Vcm of 2.5V, like so: 

    From there you can run a calibration routine, in which I set the "Full Scale Output Target" to be 3V, but this can be set to whatever you would like. 
    I assumed an offset of 0.3mV/V (with the earlier defined span of 0.5mV/V), adjust this as needed. Once calibrated, the measured output swing range is 0.5003 to 2.999V which is less than 0.1% error (depending on your system needs this can be adjusted). 

    I did all of this with the PGA308EVM, which I recommend to purchase from here, PGA308EVM Evaluation board | TI.com

    If you decide to use the PGA308 and run into some debug needs, please post an additional question on e2e (so we do not flood Joachim with notifications). 

    If this is too complex of a solution, we also have some less sophisticated run of the mill PGAs and INAs to offer. They won't have this level of configurability though. 

    All the best,
    Carolina