We’re designing a product with ADRF6518 which is used for AGC and filtering ahead AD9613 ADC. The signal source for ADRF6518 is ADL5380 quadrature demodulator.When we change modes on the transmitter, sometimes ADRF6518 locks up, i.e. there’s only some DC offset instead of transmitted carrier (sometimes in one channel, sometimes in both). When changing modes on the transmitter we observe a short spike at the ADRF6518’s input near 4.6 V, after which output of ADRF6518 has only DC offset, VPK (peak voltage detector) output has only DC offset too. This behavior does not depend on ADRF6518 register state or VGN pins voltage. In datasheet we found
1) maximum input swing 5.0 V p-p 2) absolute maximum ratings for INP1, INM1, INP2, INM2 = VPS + 0.5 V . In our case VPS = 3.3 V VPI = 5 V.
Is our input signal valid?
Now we can restore functionality of ADRF6518 by a few different ways:1) turning off and on2) pull VICM/AC to GND, but it is not right way, because ADRF6518 common mode voltage is driven by ADL5380.3) reconfigure tx again, sometimes it can help.
On attached time diagrams you can see the input (blue curve) and output (green curve) of ADRF6518 at the moment of transmitter reconfiguration.
Is ADRF6518 DC-protected when input in AC-coupled mode?
It sounds like the bias circuitry is being affected by the 4.6V spike on one of the input pins. Does this happen on both input pins (INP1 & INM1 OR INP2 & INM2) at the same time? Or just on one or the other? In other words, is is a common-mode spike or a differential spike?
There is not an ESD difference between AC-coupled mode and DC coupled mode. It basically turns on the internal bias circuitry for AC-coupled mode.
Note, that the Absolute Max Ratings specifications are the maximums you can apply to the part without damage occurring; we don't guarantee operational specifications at these max levels.
Some traces of both p and n input channels.
PBigPavel263 said:Is ADRF6518 DC-protected when input in AC-coupled mode?
I meant can we supply DC-biased input signal to ADRF6518 in AC-coupled mode?
No, you cannot supply a bias from the previous stage while having the ADRF6518 in AC-coupled mode. The internal bias is turned on in AC coupled mode (because we are assuming there are AC coupling caps between the previous stage and the input of the ADRF6518), and therefore the two bias circuits would be fighting each other.