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.

  • TI Thinks Resolved

INA149-EP: INA149-EP vs INA149

Expert 3175 points

Replies: 2

Views: 103

Part Number: INA149-EP


Good day.

Our customer wants to know the details of what  has been enhanced in the INA149-EP compared to INA149 aside from wider temperature range. This question arises because our customer noticed that it seems that INA149  has a better performance than INA149-EP based on datasheet. They are referring to the electrical characteristics at V+ = +15 V, V– = –15 V.

Please refer to the image below. The parameters being asked are the: CMRR, Gain error, Offset Voltage, and Temperature Range.

I appreciate your help on this matter. 



  • Hi Cedrick,

    Most of these differences in specs can be chalked up to the way they are reported. In that table CMRR, gain error, and VOS are all specified at +25C for the commercial part, whereas they are given over the entire -55C to +125C temperature range for the EP.

    Figure 31 in the INA149-EP datasheet and Figure 30 in the INA149 datasheet show approximately identical CMRR vs temperature curves, both down to -55C. As you approach -55C (on both plots), the CMRR distribution changes such that larger absolute values in uV/V are reported, which means a lower minimum CMRR in dB. Note that the "typical" specs are the mean CMRR plus one standard deviation, so reporting CMRR over this wider range will shift the mean and thus slightly alter the typical value.

    Figures 33 (EP d/s) and 32 (regular d/s) are another example of this: as the temperature continues to decrease for both parts, the gain error changes seemingly identically. Note once again that for the regular part gain error is only reported at +25C, whereas for the EP the gain error is reported over the -55C to +125C range. The maximum spec is used as a hard pass/fail for devices, so to account for shifts in this gain error over that temp range, a more "loose" maximum value is utilized.

    Finally, the offset voltage spec in the table is again reported over the entire operating range for the EP, whereas the regular part is specified at +25C. Compare Figures 26 (EP) vs 25 (regular part) and it's clear the typical +25C offset voltage is the same for both devices. Compare Figures 30 (EP) vs 29 (regular) and you'll see both devices see approximately the same offset shifts across temperature. Once again, a more relaxed maximum specification is used to account for shifts in VOS over the wider temperature range.

    As for the maximum operating temperature, there definitely seems to be some discrepancy. I'll be reaching out to some other team members and will report back with our findings.



  • In reply to JMac:

    Regarding the max operating temperature, the max junction temperature is defined as +150C for both the commercial and EP variants of the INA149. To understand the safe operating range, we can look at how the junction temperature Tj will change during normal operation. We need to know the power dissipation Pd, the junction-to-ambient thermal resistance Θja, and the ambient temperature Ta to use the formula Tj = Ta + Pd*Θja.

    Now using the datasheet tables, for supplies of +/-15V the maximum quiescent current is 1.1mA, so the maximum power dissipation due to the quiescent current is Pq = 33mW at +125C. However, if we look at Figures 37 (EP datasheet) and 36 (regular part datasheet), we can see that even up to +150C the quiescent current is unlikely to go above around 1.05mA, so using this value for Iq we could calculate a Pq closer to 31.5mW. Note that this is not the true worst-case max Pq, but rather an approximation. Depending on manufacturing differences, different parts might experience different Iq values, and thus this number could theoretically be even lower. However, we'll use the worst-case Pq of 33mW for our purposes.

    The total power dissipated Pd is the sum of Pq and Pmax, a term which depends on the load resistance and is found as VCC^2 / (4*RL). However, if you are not sourcing a lot of current (for example, if the INA149 output is being fed into an ADC), this Pmax term will be small enough relative to Pq that we can ignore it.

    Using Pd ~= Pq and Θja = 110 degrees C/W (from the datasheet), we can start plugging terms into the formula mentioned earlier. We can find the maximum allowed value of Ta in order for Tj to stay within its safe operating limit of +150C:

    Tamax = Tjmax - Pd*Θja = 150C - (33mW * 110C/W) = 150C - 3.63C = 146.37

    What does this mean? Well, it means as long as certain criteria are met, the amplifier can actually be operated at ambient temperatures as high as around +146C without exceeding its maximum junction temperature. If the amplifier is operated under ambient temperature conditions of +150C, the junction temperature could rise to +153.63C, which is outside the absolute maximum ratings and could damage the device. This applies to both the EP and commercial versions of the INA149 because they share the same Iq and package characteristics. Again, however, each individual unit will differ slightly in its Iq characteristics and some units could theoretically survive hotter ambient temperatures than others. Note too that in all of the characteristic waveforms in the datasheet involving testing over temperature, the parts were tested up to +150C.

    The answer, as you can see, is not very black-and-white. No, the part should probably not be operated at an ambient temperature of +150C; but on the other hand, as long as certain criteria are met, it can be operated past +125C. The customer should consider what load they will be driving in order to calculate their exact margin of safety. The +125C term from the INA149-EP datasheet is perhaps a somewhat conservative estimate for maximum ambient temperature, but in my opinion the +150C recommendation from the INA149 datasheet is too high and should not be followed without keeping some margin in mind.

    TL,DR: both parts have the same maximum junction temperature, packaging, and Iq characteristics, and therefore neither one has any inherent maximum temperature advantages over the other. The given maximum ambient operating temperature specs are a matter of conservative vs aggressive estimates - the important thing is to keep the junction temperature below its maximum of +150C. Above all, however, remember that these specs concern the operating temperature range rather than the specified temperature range, which for both parts only goes to +125C. Once the specified temperature range is exceeded, the other amplifier specifications are no longer guaranteed. Therefore, for the best performance, stay under the +125C threshold. 

    I know that's a lot of information but I hope this was helpful. Let me know if there are any other questions.



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.