• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 47 subscribers
  • Views 1512 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

XTR111 EF operation

Adnan Kurt
Intellectual 665 points
Other Parts Discussed in Thread: XTR111

I have designed a biomedical device around xtr111. Everything is fine, and working as expected and reported in the data sheet. However EF pin showed a strange behaviour. When the supply voltage is >15V, then EF does not toggle when output current is disconnected (flows through 1meg resistor, but not equal to 2mA that was set). Following is the relevant and the only description from data sheet. I think it is missing a deeper detail.

Regards,

Adnan

"The XTR111 has additional internal circuitry to detect The input voltage range for a given output current an error in the output current. In case the controlled span is set by RSET according to the transfer function. output current cannot flow due to a wire break, high Select a precise and low drift resistor for best load resistance or the output voltage level performance, because resistor drift directly converts approaching the positive supply, the error flag (EF), into drift of the output current. Careful layout must an open drain logic output, pulls low. When used, this also minimize any series resistance with RSET and the digital output requires external pull-up to logic high VIN reference point. (the internal pull-up current is 2μA)"

  • 0
    TI__Guru 64305 points

    Hello Adnan,

    Perhaps you could share a schematic of your application so we can take a look for any possible issues that may affect the performance or operation.

    Regarding the /EF operation, I've worked with the device frequently and have not observed the issue you're describing.  What do you mean the output current "flows through a 1 MOhm resistor"?  Are you using a 1MOhm resistor from IOUT to GND to simulate the effects of an open-circuit?  Does the /EF flag go low when you remove the resistor and leave a truly open-circuit?  

    Are you using an external pull-up resistor on /EF?  If so, what voltage are you pulling it up to and with what resistance?  Sharing a schematic will answer many questions.

    Thanks!

  • 0 in reply to Collin Wells
    Intellectual 665 points

    Dear Collin,

    I am attaching the schematic picture, though details not seen well with this resolution. Anyway, sorry for incomplete problem description in the previous post.

    Yes, I didnot want to leave output floating, so inserted a 1Meg resistor. And, I didnot tried to measure /EF behaviour when it is removed. Current is very low anyhow. If /EF is looking for open circuit condition only, that might be an issue. I thought it would compare set current to output current.

    Regards,
    Adnan 

  • 0 in reply to Adnan Kurt
    TI__Guru 64305 points

    Hi Adnan,

    You are correct, the XTR111 /EF flag is monitoring the ability of the output amplifier inside the XTR111 to balance it's inputs by controlling the output transistor to output the correct current.  If the correct output current can't flow then the output amplifier's inputs will not be equal and then the /EF becomes active and pulls low.  

    What current are you setting the output to when the issue occurs?

    What are ZD201 and ZD202 for, and is there any way that the zener diodes are conducting current between IS and VG? When the open-circuit condition is present, the op-amp internal to the XTR111 will drive the gate of the PMOS as low as it can to try to get more current to flow possibly allowing ZD201 to reverse bias and ZD202 to conduct.  The voltage is already internally limited to 16V below the supply voltage so I'm not sure if the Zener diodes are required there.   

    As mentioned, the internal control loop that /EF monitors is only looking at the current flowing out of IS, not the current delivered to the load, so if current is still flowing out of IS and back into the VG pin through the diodes or anywhere else, then the error condition will not be detected.  Could you put a meter in series with the IS output and see if current is in fact flowing somewhere else?  

  • 0 in reply to Collin Wells
    Intellectual 665 points

    Hi Collin,

    Yes, I think zeners are not required there, since Vg voltage is limited. However, in order to keep obeying basic rules, I've included them to make sure that the gate-source voltage never exceeds 12V. (And now I think you point to the problem! These are 12V zeners, and as soon as supply voltage gets > 12++, then there should be a current leakage). Zeners have reverse current of about 500nA by the way. And I encounter the problem when the current flow is <4mA from the IOSRC port. I didnot test higher currents, we do not need more.

    I will check this condition on Friday, and will inform about the outcome. I will remove the zeners first. If they have caused the trouble, I will use a higher zener voltage diode. We are using 18V supply, so, 20V zener should work. ref pg14 of SBOS375A – NOVEMBER 2006 – REVISED AUGUST 2007.

    Concerning your last paragraph, current flowing out of IS is equal to the load current, is that right? And MOSFET here is used for higher voltage compliance and output versatility?

    Best Regards,
    Adnan

  • 0 in reply to Adnan Kurt
    TI__Guru 64305 points

    Hi Adnan,

    Yes, current flowing out of IS should be equal to the load current.  MOS devices are specified in terms of drain-to-source current, iDS, so the current flowing into the drain is equal to the current flowing out the source (or vice-versa for PMOS)

    The MOSFET is required because the internal amplifier can not provide the continuous output current requirements of a 4-20 mA driver.  As you pointed out as well, the external MOSFET also enables higher output compliance capabilities.

  • 0 in reply to Collin Wells
    Intellectual 665 points

    Dear Collin,

    I have replaced the Gate-Source Zener with a 24V type. XTR11 and my device is working as expected without any problems now.

    Thanks for patience and support.

    Best Regards,
    Adnan 

  • 0 in reply to Adnan Kurt
    TI__Guru 64305 points

    Hello Adnon,

    Glad we could help!