Post Go back to editing

AD8628 vs ADA4528-1 Output Current Capability


Looking at Figure 11 (Ouput Voltage to Supply Rail vs. Load Current) of the AD8628 datasheet it can be seen that you can get up to 10 mA at the expense of about 400mV drop from the supply rail.

From Figure 15 of ADA4528-1 it seems that you can get the same 10mA with a bit less than 200mV drop, not only that, it appears that the part can handle almost 20mA if it sacrifies 400mV drop. This is correct?

I'm looking to use the ADA4528-1 or AD8628 (3.3V single supply powered) as a buffer for a 3V voltage reference to use with three AD7685 but at the same time use this output to power a wheatstone bridge sensor that could consume max 8mA. I'm not going to require 20mA but probably I'll need about 12mA to 14mA which seems that the ADA4528-1 can manage.

  • CRFA,

      It all depends on whether you want to build one system, or a thousand a week.  All of the mins and maxes in the

    spec tables are 100% tested in a production environment unless there is a footnote that it is based on simulation

    or characterization.  Typicals are NOT tested or guaranteed.

      For the AD8628, the Voh is 2.67V into 10k with a Vcc of 2.7V.  This is only 267 microamps.  The short circuit current

    is 10 mA minimum, but would not be useful for your system, because the Vout would be zero volts.

      For the ADA4528-1, Voh is 2.44V over temp, with a 2k load CONNECTED to Vcm, which is 1.25V with a Vcc of 2.5V,

    so the current would be 595 microamps.  The short circuit current is a TYPICAL of 30 mA, with no guarantee.

    The typical graphs on semicoductor datasheets are usually generated from a small sample of parts (say five or less)

    from the first wafer run.  (A lot of pressure from management to release new products.....)  A wafer run six months

    or a few years later may be a little better or a little worse., so you can only rely on the tested limits.

    With the low supply voltage available, for a high volume, production-worthy design, I would have the op amp drive

    a discrete, small signal, PNP to dirve the bridge.


  • Thanks a lot Harry, that solution will do for my purposes, and actually will serve a lot if at some point I'll have to change the sensor.


  • This question has been closed by the EZ team and is assumed answered.