• State Resolved
  • Locked Locked
  • Replies 19 replies
  • Answers 5 answers
  • Subscribers 47 subscribers
  • Views 4070 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.

Starterware/ADS1298: 12 lead derivation for ADS1298

Vishnu Pradeep
Intellectual 505 points
Part Number: ADS1298
Other Parts Discussed in Thread: ADS1296

Tool/software: Starterware

Hi,

I understand from the section to derive 12 lead values from ADS1298. But for the chest leads I should know WCT. How can I calculate WCT from this? Is there a separate register I need to read or can I calculate from the 8 channel values? Please help. 

Thanks,

Vishnu 

  • 0
    TI__Mastermind 30365 points
    Moving this thread to the Precision data converters forum.
  • 0
    TI__Mastermind 38745 points
    Hello Vishnu,

    The ADS1298 has an integrated WCT module that will find the average of electrode voltages you chose and then output it on the WCT pin. That pin's output can then be wired to the INxN pins of channels where you want to measure the chest leads. You can find the details of the WCT amplifiers on page 39 of the ADS1298 datasheet.

    Regards,
    Brian Pisani
  • 0 in reply to Brian Pisani
    Intellectual 505 points

    HI Brian,

    Thanks for the reply. Can you please verify whether the settings are correct or not? I am getting the data now. but just want to make sure the settings are correct or not. I am using an external 4V reference with 5V supply.

    CONFIG3 0x40

    CONFIG2 0x02

    CONFIG1 0x86

    WCT1 0x0B

    WCT2 0xD4

    CHnSET 0x60  (for all 8 channels)

    This is the circuit I am using

    Thanks & regards,

    Vishnu

  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points
    Vishnu,

    One change that you should make in the CONFIG3 register: you will need to enable both the reference and the 4V reference setting. I would change that register to 0xE4.

    Regards,
    Brian Pisani
  • 0 in reply to Brian Pisani
    Intellectual 505 points

    Hi Brian,

    Thanks for the reply. I'll make the change accordingly.

    But one doubt is If we use external reference, why would we need to enable both? and does the ADC consider external reference if we enable internal reference?

    Also can you please verify the circuit ?

    regards,

    Vishnu 

  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points
    Vishnu,

    My apologies, there is no need to enable the internal reference - you do not need to change the registers. I mistakenly thought you planned on using the internal one. Your schematic looks good.

    Brian
  • 0 in reply to Brian Pisani
    Intellectual 505 points
    Thanks for your time Brian.

    Cheers
  • 0 in reply to Vishnu Pradeep
    Intellectual 505 points

    HI Brian,

    I am getting a lot of noise in the graph. What is the recommendable filter to use in the software side ?

  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points

    Hello Vishnu,

    In a lot of biomedical applications there is  a lot of interference at the mains frequency from overhead lights. Make sure you have enabled the ADS1298's RLD amplifier which cancels common mode signals by setting bit 2 of the CONFIG3 register. You can then chose which input to use for RLD feedback in the RLD_SENSP and RLD_SENSN registers. Details about the RLD amplifier are given in section 9.3.1.7.6 in the datasheet. There is also an app note called Improving Common-Mode Rejection Using the Right-Leg Drive Amplifier which discusses the subject in detail.

    However, for most applications, even this is not enough to totally eliminate mains frequency interference. You may also have to develop a digital notch filter that attenuates signals at that frequency even more.

    Regards,

    Brian Pisani

  • 0 in reply to Brian Pisani
    Intellectual 505 points
    Thanks Brian,

    I haven't enabled the RLD function. Which channel should go to RLD_SENSP and RLD_SENSN? I am a little confused.
  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points
    Hey Vishnu,

    First I'd advise you to read that app note I linked to in my last post. It is very helpful for understanding the behavior of the RLD circuit.

    Now to answer your question. Every channel that can be routed to the RLD amp, if you want. The more channels that are input to the RLD amplifier, the higher its gain will be. In you notice in Figure 43, there are 220 kOhm resistors at the output of the PGA that connect them to the RLD inputs. Since we wish to use that amp to amplify the common-mode signal, those resistors can be thought of as "in parallel". The more inputs we use as inputs to the RLD amplifier, the more of those resistors are "in parallel" and the smaller their equivalent resistance will be.

    Since the formula for gain in an inverting op-amp is -Rf/Ri, the gain will get larger when Ri gets smaller i.e. when you use more device inputs as inputs to the RLD amplifier. The only issue with having too much gain is in the case where the RLD amplifier output saturates. If this is the case, the amplifier becomes non-linear and won't do a good job of cancelling common-mode signals. The exact right amount of gain will depend on the amplification since the impedance of the cable and of the electrode-patient connections changes on a system-to-system basis. Does this make sense?

    Regards,
    Brian Pisani
  • 0 in reply to Brian Pisani
    Intellectual 505 points

    Hi Brian, 

    I am using an ECG simulator to test my device. I printed out individual channel values using UART after filtering the adc values. I am getting LEAD1 values as below which is fine for now. But none of the channels except LEAD1 are coming correctly with the register configuration I mentioned earlier. Is the WCT settings are correct according to the circuit diagram? 

    and the remaining channels are coming as below.

     

    Regards,

    Vishnu

  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points
    Vishnu,

    Your WCT settings look like they match the circuit. Is Lead II coming back ok? That channel should not be affected by the WCT settings. Ostensibly it should behave the same as Lead I. If not, can I see the portion of your circuit in front of the ADS1296?

    Brian
  • 0 in reply to Brian Pisani
    Intellectual 505 points

    Hi Brian,

    We are not getting Lead II also. It is coming like all other channels. We tried connecting LL to IN2P and LA to IN3P and took reading. we got same result. Only Lead I was coming correctly and Lead II was with noise. I have attached the circuit in front of AFE. In this we have used only the resistors(22K and 10K) and  TVS didode. Rest of the components are not connected for all the channels.

    Please verify the circuit. 

    Regards,

    Vishnu Pradeep

  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points
    Hey Vishnu,

    If I understood you correctly, you are saying that the electrode is the difference between the measurements. That is to say that even when swapping LA and LL, you are still only able to measure ECG between LA and RA. I believe your schematic looks fine. Let's try to rule out the electrode as the source of the issue. Are you able to measure ECG if you place the LA electrode in the spot that you would place LL?

    Brian
  • 0 in reply to Brian Pisani
    Intellectual 505 points
    Hi Brian,

    By saying electrode I meant snap lead. I switched the LA and RA leads in the ECG simulator and got ECG only for channel 2. Only channel 2 was working at that time.
    I'm seeing some improvement in the measurement now. when I bypassed the input resistors R46 and R44 I could see the expected signals with a good amount of noise which can be filtered out. Is there anything I need to worry about the input resistance? 22k and 10k along with a cable resistance of 1k?

    regards,
    Vishnu
  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points
    Hello Vishnu,

    I still find it strange that channels 2 and 3 are almost identical but you are only able to get a clean signal on channel 2. I don't think the resistance should affect it since the resistance is the same for both channels. Do you have multiple board assemblies that exhibit this behavior? I'd like to rule out assembly issues.

    Brian
  • 0 in reply to Brian Pisani
    Intellectual 505 points
    Hi Brian,

    It was an issue with PCB for the 3rd channel. I took another board and connected the simulator. I checked the noise by input short and then generated internal square pulse . Both were coming correctly. But the normal electrodes were not taking. I got only noise signals. Then I replaced the 22K resistor with 10K for channel 2 and got the LEAD I signal. Same happened for channel 2. Later I changed all the 22K values to 10K and all the 8 channel signals were coming. That is why I asked the importance of input resistance. When I put 22K I am getting nothing but noise.

    Vishnu
  • 0 in reply to Vishnu Pradeep
    TI__Mastermind 38745 points

    Vishhnu,

    If I understand you correctly, the channel 2 input is similar to the others in that it only works when the resistors on the inputs are reduced. If this is the case, then I am certain that you are seeing mains interference coupling on to the electrodes. The reason it gets better with lower resistance is because the CMRR of the input RC filters improves for the mains frequency when the cutoff of the filter is moved higher in frequency. Refer to my previous posts in this thread regarding the rejection of mains frequency interference.

    Brian