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.

LOG114EVM: EVM design differs from datasheet recommendations

Part Number: LOG114EVM
Other Parts Discussed in Thread: LOG114, REF3040, OPA365

The LOG114EVM design differs from the recommendations in the LOG114 datasheet. The differences are:

1) The LOG114 datasheet in Figure 2 uses a REF3040 to provide a 4.096V signal to generate the reference current. Additionally, on page 15 states that the advantage of dual supply include "eliminating the need for the +4.096V precision reference" This seems to imply that the REF3040 is necessary for a single-supply design. However, the LOG114EVM uses the on-chip 2.5V reference to generate the reference current.

2) The LOG114 data sheet uses the internal 2.5V reference directly to generate the required level-shifted output while the LOG114EVM uses a buffered signal generated from the internal voltage reference unity amplified by an OPA365.

3) The LOG114 datasheet shows bypass capacitors of 1000 pf and 10 μF whereas the LOG114EVM uses values of 100 nF and 4.7 μF. I suppose the values of these capacitors depend on the impedance and current requirements of the output?

So, is it possible to use the internal 2.5V reference to generate both the reference current and the level-shift so that the REF3040 and the OPA365 can be eliminated? Is there an impedance limit on the internal reference that precludes this? It does not appear that the equivalent internal resistance of the reference signal is directly provided. The short circuit current is listed as 10 mA which suggests and internal resistance of the reference source of 250 Ω. Is there something in the design that precludes the reference from being used as both the current reference and a source for the output level shift? What is the preferred design?

Also there appears to be typos in Table 2 of the LOG114EVM Users Guide. The description of C1 and C2 should be 100 nF, not 1000 pF. The description of C3 should be 1 μF not 0.1 μF.

  • First off, thanks for catching those typos.  In the long lead time world of passives we live in today, we might be forced to install whatever we have around!  I kid.  To your question, there's nothing magical about the selection of 1nF vs 10nF vs 100nF and 4.7uF vs 10uF.  It is a good practice to use two capacitors a couple decades apart in value -- generally a larger Tantalum and smaller ceramic -- to service the transient current demands over a broad range of frequencies.  

    Regarding the reference voltages and single vs dual supply configurations… In the case of bipolar operation, the Vcm provided to the two amplifiers is ground, and the internal 2.5V reference can be used to generate a current of (2.5V-0V)/R into one of the inputs, acting as the reference current.  

    In the case of dual supply operation, you have two choices

    1. Use the internal 2.5V reference as the common mode to the amplifiers, and use a higher value external reference to generate the current (Vrefext-2.5V)/R into one of the inputs, acting as the reference current.
    2. Divide down the internal 2.5V reference and buffer with an external op-amp to create a new low impedance <2.5V Vcm, and then use that voltage to create the reference current (2.5V-Vnew)/R.

    If you tried to use the single 2.5V reference for both the Vcm of the amplifiers, as well as the source of the reference current, there would be no voltage differential across R (2.5V-2.5V)/R=0.

    I hope that clears things up.

    Thanks,

    Scott