Post Go back to editing


I'd like to get a better understanding of the noise and distortion properties of the ADG4612 part.  This is a 4:1 analog MUX rated at a bandwidth of 193 MHz, but the data sheet treats it as an audio part, with the THD+N characterized with an AP audio analyzer.  The published THD+N specs are relatively poor, on the order of 1% or -20 dBc. 

However, when I measured the part's HD2 and HD3 performance on your eval board at various frequencies from 100 kHz-20 MHz, I'm seeing much better HD2 numbers in the -60 dB to -70 dB range as long as I keep the level around 0 dBm.  The odd-order distortion appears to be negligible, as does additive noise.

So the question is, what's up with this part's THD+N specs?  These were measured at very high signal levels (3V peak) into 110 ohms according to the data sheet so that might account for the harmonic performance.  But I'm looking at using it as an RF part in a 50-ohm system, and it seems to be a good performer from what I can tell.  I'm a little nervous at the lack of RF distortion measurements on the data sheet, so I'd like some confirmation that I'm not missing anything before I use it to drive a 14-bit RF ADC.  Is there some reason why it wasn't characterized as an RF part?

  • Hi John,

    Something to consider here when looking at the THD+N performance is what does the Ron look like over the same supply and signal range.  You can refer to Figure 4 and Figure 5.  You can see over large signal ranges (Vs Vs) the Ron varies and this in turn is a function of the supplies (either single supply or dual supply).  The THD results on the datasheet are typical use cases for larger signals going though the switch.  Note, the lowest variation occurs for the higher +/-5.5V dual supply case and the higher single supply voltage ranges and this follows for the difference in results in Figure 19.

    Your CW signal through the switch is very small at 0dBm (0.224Vrms) so you should not see that much variation in Ron from your voltage changing, that is why you are seeing better HD performance.

    This part was not primarily targeting RF applications due to its modest bandwidth and  off isolation at higher frequencies.  But if these are adequate for your application and our other more RF orientated switches don't fit your application then it maybe enough for you.