• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 2 answers
  • Subscribers 46 subscribers
  • Views 1686 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.

ADS1198: False DC lead off detection in ADS1198

Bhairu Patil1
Intellectual 570 points
Part Number: ADS1198

Dear TI Team,

         I am using ADS1198 in our design. I have selected Current Source DC lead off method for lead off detection. The register settings are as per below.

WREG LOFF                          0x0F      // Comparator threshold at 95% and 5%, Current source // DC lead-off
WREG CONFIG4                   0x02       // Turn-on dc lead-off comparators
WREG LOFF_SENSP            0xFF       // Turn on the P-side of all channels for lead-off sensing
WREG LOFF_SENSN            0xFF       // Turn on the N-side of all channels for lead-off sensing

We are facing problem with lead off detection, when all the chest leads V1-V6 & RL lead is disconnected from the patient's body( only RA LA & LL leads are connected to patients body). I have verified this in 2 devices(same hardware), both of these boards are showing different values only for these particular case. as shown below.

  95%-05% threshold
leads combinations connected
to patients body		 EXPECTED result Device_1 Device_2
LOFF_
SENSP		 LOFF_
SENSN		 LOFF_
SENSP		 LOFF_
SENSN		 LOFF_
SENSP		 LOFF_
SENSN
RA LA LL  0XFF 0X00 0xF9,
0xF9,
0xF9		 0X00,
0X00,
0X00		 0XF9,
0XF9,
0XF9		 0X00,
0x70,
0xF6

            As shown in above table, the LOFF_SENS values in DEVICE 2 are changing continuesly(0x00, 0x70,0xF6). if any one or all  of the Chest leads are connected, along with the RL lead, both the devices are showing the expected result.

The complete  results are shown in following table with respect to 2 devices as below.

  95%-05% threshold
leads combinations connected
to patients body		 EXPECTED result Device_1 device_2
LOFF_
SENSP		 LOFF_
SENSN		 LOFF_
SENSP		 LOFF_
SENSN		 LOFF_
SENSP		 LOFF_
SENSN
RL RA LA LL V1 V2 V3 V4 V5 V6 0X00 0X00 0X00 0X00 0X00 0X00
RL RA LA LL V1 V2 V3 V4 V5  0x01 0X00 0x01 0X00 0x01 0X00
RL RA LA LL V1 V2 V3 V4  0x41 0X00 0x41 0X00 0x41 0X00
RL RA LA LL V1 V2 V3  0x61 0X00 0x61 0X00 0x61 0X00
RL RA LA LL V1 V2  0x71 0X00 0x71 0X00 0x71 0X00
RL RA LA LL V1  0x79 0X00 0x79 0X00 0x79 0X00
RL RA LA LL  0xF9 0X00 0xF9 0X00 0xF9 0X00
RL RA LA  0XFD 0X00 0XFD 0X00 0XFD 0X00
RL RA 0XFF 0X00 0XFF 0X00 0XFF 0X00
RL  0XFF 0X06 0XFF 0X06 0XFF 0X06
RA LA LL  0XFF 0X00 0xF9,
0xF9,
0xF9		 0X00,
0X00,
0X00		 0XF9,
0XF9,
0XF9		 0X00,
0x70,
0xF6
RA LA  0XFF 0X06 0XFF 0X06 0XFF 0X06
RA  0XFF 0X06 0XFF 0X06 0XFF 0X06
LA LL  0XFF 0X06 0XFF 0X06 0XFF 0X06
LL 0XFF 0X06 0XFF 0X06 0XFF 0X06
RA LA LL V1 V2 V3 V4 V5 V6 0xFF 0X00 0xFF 0X00 0xFF 0X00
RA LA LL V1 V2 V3 V4 V5 0xFF 0X00 0xFF 0X00 0xFF 0X00
RA LA LL V1 V2 V3 V4 0xFF 0X00 0xFF 0X00 0xFF 0X00
RA LA LL V1 V2 V3 0xFF 0X00 0xFF 0X00 0xFF 0X00
RA LA LL V1 V2 0xFF 0X00 0xFF 0X00 0xFF 0X00
RA LA LL V1  0xFF 0x00 0xFF 0x00 0xFF 0x00

Please, clarify on the following.

  • Why both the devices are showing wrong results when only LA, LL, RA are connected to the body ( RL lead and chest leads V1 to  V6 disconnected).
  • How the Lead off detection for RL lead is depends on the chest leads V1- V6. ( that means if any one or all of the chest leads is connected along with LA, LL, RA lead,  I am getting expected result- LOFF_SENSP = 0xFF, LOFF_SENSN = 0x00)

   The schematics for the same is attached below.

Thanks and Regards,

Bhairu

  • 0
    TI__Mastermind 38745 points

    Hello Bairu,

    The operation of the lead-off circuitry is relatively simple. On each input, there is a current source and a comparator. The RL electrode serves as a regulated voltage which can sink any lead-off current. Understanding how any of the various scenarios should play out is a matter of applying Kirchoff's current law with the patient as the net you are analyzing. 

    The one thing you must look out for, though, is that the current sources cannot bias the patient like the RLD amplifier can. So in oyur first example where RA, LA, and LL are connected, you see at first that those electrodes are connected, but after a while, the voltage at the electrodes begins to drift since RL is disconnected, and eventually they start to indicate lead off.

    One way to accelerate this behavior is to only turn on the lead-off current source for one of the inputs connected to RA (i.e. either IN2N or IN3N, but not both). That way, when RA is off, the extra current from LA and LL will cause at least one of those inputs to go high quicker.

    In addition, the inputs connected to WCT do not provide any actual information about the status of lead-off. I recommend disabling the current sources for those inputs.

    To summarize, I recommend changing LOFF_SENSN to either 0x04 or 0x02.


    Do you understand my answer?

    Brian

  • 0 in reply to Brian Pisani
    Intellectual 570 points

    Dear Brian,

    Thank you, for your quick reply. It has worked as expected with the changes, you suggested( LOFF_SENSN = 0X04), for most of the cases except for two cases which are highlighted below( Chest leads V1- V6 are disconnected from the patients body, The leadoff status of chest leads are detected as expected but not taken into considerations in table),

    95%-05% ,, LOFF_SENSN = 0x04

    leads combinations connected

    to patients body

    		 EXPECTED device1 device2
    Lead_OFF Lead_OFF
    detected    		 Lead_OFF
    detected
    RA LA RL LL NO NO NO
    RA LA RL LL LL LL
    RA LA ALL LEADS LL LL
    RA  ALL LEADS ALL LEADS ALL LEADS
      ALL LEADS ALL LEADS ALL LEADS
    LA ALL LEADS ALL LEADS ALL LEADS
    LA RL RA LL RA LL RA LL
    LA RL LL RA   RA   RA  
    RL LL LA RA  LA RA LA RA
     LL ALL LEADS ALL LEADS ALL LEADS
    RA  LL ALL LEADS LA    LA  
    RA  RL LL LA   LA LA
    RA  RL  ALL LEADS ALL LEADS ALL LEADS
    RL  ALL LEADS ALL LEADS ALL LEADS
    LA LL ALL LEADS ALL LEADS ALL LEADS
    RA LA  LL ALL LEADS ALL LEADS ALL LEADS

          Please, suggest a suitable solution to fix this issue.

    Thanks and Regards,

    Bhairu

    95%-05% , LOFF_SENSN = 0x04
    leads combinations connected EXPECTED device1 device2
    Lead_OFF Lead_OFF Lead_OFF
    RA LA ALL LEADS LL LL
    RA  LL ALL LEADS LA    LA  
  • 0 in reply to Bhairu Patil1
    TI__Mastermind 38745 points
    Hello Bhairu,

    The problem here is that LA and LL are providing current to the patient and RA is sinking current. When RL is removed, the current will maintain the voltage at a steady state for a little bit. Eventually, though, lead-off should be indicated for those electrodes as well, but not right away.

    Brian
  • 0 in reply to Brian Pisani
    Intellectual 570 points

    Dear Brian,

             Thank you for your quick reply. I am using ADS1198 for 3 different variant application as follows.

    • 12 lead( LA,LL, RA, RL,  V1,V2,V3,V4,V5,V6)
    • Limb leads(LA,LL,RA, RL) and
    • Single leads (LA,RA, RL  or LL,RA,RL) 

          I have done the changes as per your suggestion on LOFF_SENSN  (0x04 or 0x02).  and the results are as below. This is works well for All leads and Limb leads application but failed for Single lead application for detection of RL.   

           I have tested with LOFF_SENSN  (0x00) this is works well in all the cases..  the test results are as per below.

    LOFF_SENSN
    register settings    		  RL detection
    All lead Limb Lead Single lead 
    0xFF PASS Fail Fail
    0x04 or 0x02 PASS PASS Fail
    0x00 PASS PASS PASS

     The LOFF_SENSP & LOFF_SENSN values are as per below.

    Lead combinations used leads combinations connected
    to body    		 LOFF_SENSN
    register settings    		 EXPECTED Device 1 Device 2
    LOFF_
    SENSP    		 LOFF_
    SENSN    		 LOFF_
    SENSP    		 LOFF_
    SENSN    		 LOFF_
    SENSP    		 LOFF_
    SENSN
    Limb leads
    (LA, LL, RA, RL)    		 LA, LL, RA 0xFF 0xFF 0x00 0x00 0x00 0x00 0x00
    0x04 or 0x02 0xFF 0x00 0xFF 0x00 0xFF 0x00
    0x00 0xFF 0x00 0xFF 0x00 0xFF 0x00
    Single lead
    (LL, RA, RL)    		 LL, RA 0xFF 0xFF 0x00 0x00 0x00 0x00 0x00
    0x04 or 0x02 0xFF 0x00 0x00 0x00 0x00 0x00
    0x00 0xFF 0x00 0xFF 0x00 0xFF 0x00
    Single lead
    (LA, RA, RL)    		 LA, RA  0xFF 0xFF 0x00 0x00 0x00 0x00 0x00
    0x04 or 0x02 0xFF 0x00 0x00 0x00 0x00 0x00
    0x00 0xFF 0x00 0xFF 0x00 0xFF 0x00

                   

              The  test is carried out with LOFF_SENSN = 0x00, for all the combinations of LA, LL, RA & RL by disconnecting all the chest leads V1  to V6. The test results are as per below. The lead off detection is works very well except for one case i.e only RA is conneccted to body as shown in below highlighted..

    95%-05% , LOFF_SENSN = 0X00, 
    leads combinations connected
    to body    		 EXPECTED Device 1 Device 2
    LOFF_
    SENSP    		 LOFF_
    SENSN    		 LOFF_
    SENSP    		 LOFF_
    SENSN    		 LOFF_
    SENSP    		 LOFF_
    SENSN
    RA LA RL LL 0xF9 0X00 0xF9 0X00 0xF9 0X00
    RA LA RL 0xFD 0X00 0xFD 0X00 0xFD 0X00
    RA LA 0xFF 0x00 0xFF 0x00 0xFF 0x00
    RA  0xFF 0x00 0xFF 0x06 0xFF 0x06
      0xFF 0X06 0xFF 0X06 0xFF 0X06
    LA 0xFF 0X06 0xFF 0X06 0xFF 0X06
    LA RL 0xFD 0X06 0xFD 0X06 0xFD 0X06
    LA RL LL 0xF9 0X06 0xF9 0X06 0xF9 0X06
    RL LL 0xFB 0X06 0xFB 0X06 0xFB 0X06
     LL 0xFF 0X06 0xFF 0X06 0xFF 0X06
    RA  LL 0xFF 0X00 0xFF 0X00 0xFF 0X00
    RA  RL LL 0xFB 0X00 0xFB 0X00 0xFB 0X00
    RA  RL  0xFF 0X00 0xFF 0X00 0xFF 0X00
    RL  0xFF 0X06 0xFF 0X06 0xFF 0X06
    LA LL 0xFF 0X06 0xFF 0X06 0xFF 0X06
    RA LA  LL 0xFF 0X00 0xFF 0X00 0xFF 0X00

                Please clarify the following.

    • How The Leadoff detection is working for RA, even though the leadoff detection for RA is disabled( CH2N & CH3N). 
    • Is this okay to use LOFF_SENSN = 0x00 for ADS1198.  will this effect the performance of the ADS1198( ECG signal).

    Thanks and Regards,

    Bhairu

  • 0 in reply to Bhairu Patil1
    TI__Mastermind 38745 points

    Hello Bhairu,

    • Even if the LOFF_SENSN bit corresponding to the RA electrode is disabled, the comparator is still enabled meaning it can still transition. However, I have no explanation for why LOFF_STATN would continue to behave as expected even when the current source is enabled. It is possible that it has something to do with the order that you are taking the electrodes on and off. In any case, I do not think this will ultimately be repeatable or reliable.
    • The ECG performance should not be affected by any lead-off setting, so yes it is ok from that point of view. However, please note that I do not know how it is indicating correctly and I can't be sure that it will work every time going forward.

    Brian

  • 0 in reply to Brian Pisani
    Intellectual 570 points

    Dear Brian,

             Thank you for your quick reply.  I have consider the following changes, as you suggested. and it is works well.

    • changing LOFF_SENSN to either 0x04 or 0x02.

     

    Thanks and Regards,

    Bhairu