Post Go back to editing

Can AD8221BR, AD8622 improve current design?

We are redesigning a test board due to heating in a few LT1167 (LT1167ACS8) and LT1352 (LT1352CS8) devices. Due to lack of information on the specific devices that heat up, done LT Spice simulation for circuits, and have nearly narrowed down on the problem.

LT1167 analysis: In simulation seen that Is(source current) is 2.1mA, whereas datasheet for the device gives 1.3mA (maximum). So, it is probable that the LT1167 devices get heated up when the specific test is run for an extended duration. This LT1167 is used with gain of 6.18, others with unity gain.

Here we are planning to replace LT1167 with AD8221BR. Primary criteria for selection of Instr Amp was:

a) Input bias current (max) & input offset current (max) to be < 400pA

b) -3dB bandwidth not to go far below 800kHz

c) CMRR to meet or exceed 90dB/min to 95dB/typ (G=1), and 106dB/min to 115dB/typ (G=10)

LT1352 analysis: This is used to implement unity gain amplifier at output of LT1167, and at other places as low pass filter (fc=100Hz). In simulation seen two issues:

a) Device current per Op Amp stays at 0.45mA for a test case, which is more than the 250uA/min to 320uA/typ limit of LT1352. At this instant, Op Amp output is 6.9V (nearly half the supply voltage, 15V)

b) For another test case, seen that there are -15V to +7V voltage swing signals on +/- input and output lines of LT1352, though there is no differential between the +ve and -ve inputs. The datasheet only mentions that the device is not suited for continuous high swing differential inputs.

Here we are planning to replace LT1352 with AD8622.

Queries to the forum are:

1) For AD8221BR, can we consider that VOS = VOSI + (VOSO/G) would hold true with VOSI=25uV and VOSO=200uV (values from AD8221BR datasheet, page4).

2) Would there be heating on AD8622 as well, with this device also having a supply current of 175uA - 225uA. Yet to do simulation which should give insights.

3) Does Op Amp heating primarily happen when Is(supply current) is at the maximum and with output biased at half the VCC or Vee supply voltage

Thank you.

Regards

-Sathyadev

  • Hi Sathyadev,

    Thank you for considering ADI products on your applications.

    Here are my answers to your questions.


    1. The VOSI and VOSO parameters specified on the AD8221 datasheets are both tested in production and therefore we can guarantee its value. The equation that you provided is the total offset voltage (referred to the input) due to to input offset voltage (VOSI) and output offset voltage (VOSO) only. Note that there are also other components that could introduce DC error like the input bias current (IB) and  input offset current (IOS) together with the unbalanced source impedance, Common Mode Voltage, Power supply variation, etc.

    2. Please refer to this article for the computation of the junction temperature of the device: http://www.analog.com/media/en/technical-documentation/technical-articles/MS-2251.pdf . Note that if you are going to use the AD8622, make sure that the opamp is operating within its input and output voltage range. If you could share to me your specific setup or configuration then we could analyze it in more detail.

    3. By referring on the article above, we could compute and verify if the heating of the parts is expected or there might be wrong connections on your setup.

    Let me know if you need anything else.

    Best regards,

    Emman

  • Hi Emman,

    Thank you for the reply, and it was useful. I had replied a week back through the email in inbox, though dont know if you got it.

    Keeping the points brief this time:

    1) Would AD be able to provide a model for AD8221BR that can be used in ADIsimPE?

    2) Shortly would be simulating Op Amp AD8622ARZ. So, that would give an idea as to if Isy for the Op Amp exceeds the rated current - 215 to 250uA. For LT1352, as written earlier, with the LTSpice tool, Is (supply current) exceeded rating for a few test configurations. If the results are fine, we would replace LT1352 with AD8622ARZ.

    Regards

    -Sathyadev

  • Hi Sathyadev,

    1. We have the AD8221 spice model on the ADIsimPE but specs is for AD8221 A grade.

    2. Let me know the results on the simulations. I am interested to find out if our parts is working well on your configuration.

    Best regards,

    Emman

  • Hi Emman,

    We started simulating the LP filter stage using AD8622. Two stages of AD8221 precede this LP filter, and now is a good time to get the AD8221 model from you. Could you please share it?

    Regards

    -Sathyadev

  • Ok, i downloaded the CIR file for AD8221, and we will use that.

    -Sathyadev

  • Ok, i downloaded the CIR file for AD8221, and we have started the simulations.

    -Sathyadev

  • Hi Sathyadev,

    You can attached your schematic here by replying on this thread --> use advance editor --> attach (located at the bottom right of the text box). Let me know if this will work for you.

    Best regards,

    Emman

  • Hi Emman,

    Getting very unexpected responses in our simulation. Circuit is same as the one used with LT devices in LT Spice tool, but with the LT devices replaced with AD ones. Can i share with you the ADIsimPE schematic, for a quick resolution?

    And how to share files with you?

    Regards

    -Sathyadev

  • Hi Emman,

    We done simulation using the planned Instr Amp (AD8221) and Op Amp (AD8622). It uses a pulse current source, which is implemented as shown in the image  (Current_pulse) in attached zip file.

    We found that the output (taken at net COM1) does not vary when we change the low/high level of the input current pulse. We also replaced current pulse source with a voltage pulse source, but observe the same behaviour.

    Have attached the ADIsimPE schematics for both current pulse source and voltage pulse, and the COM1 output in both cases.

    Have also attached the LTspice simulation results with the existing LT devices.

    Just please point out the error in circuit or if we are not doing anything right.

    Regards

    -Sathyadev

    to_share_Jun22.zip
  • Hi Emman,

    Just now sent you an email with a zip file. It was sent by giving a reply to your email which is in my inbox, and attached the file using the usual method.

    Please check if you got the email.

    Regards

    -Sathyadev