EVAL-AD2S1205 - unexpected excitation buffer outputs

Question

When I connect scope to buffered excitation outputs - EXC and /EXC - I get unexpected signal ( sin/cos expected): 
 
 There are basically only halves of sin() signals. When I put probes in various parts of board I get these values - please note that EXC output from AD2S1200 looks correct (yellow wave):

Question: Is that buffer output correct? (I'm expecting normal sinusoidal output as is shown right at EXC PIN34 of AD2S1200). 
 This behavior was observed in both extreme cases: 
 
 Just powered evaluation board with 9V power supply and only connected scope. No other connections made. 
 Board powered, connected Resolved simulator card (to both Excitation output and SinCos inputs), connected via USB to PC and run AD2S1200_05 
 
 When I combine differential EXC and /EXC outputs together (using differential input on scope) they resemble sin() function, but not exactly - there is noticeable distortion. 
 
 Details on board and chip: 
 Board is: EVAL-AD2S1200/05SDZ Rev A 
 Chip on board is: AD2S1205 YSTZ

Human · Accepted Answer

Hello, all! 
 Finally I was able to fix my decoder card replacing these parts: 
 
 Q1 and Q13, BC846B bipolar NPN transistor, Package SOT23 - type and package kindly provided by ADI support 
 R47, 4.7Ohm,at least 0.1W (the higher the better), Package 0603 - type and package is my guess (I was too impatient) 
 
 Here are both EXC and /EXC outputs as seen on scope now (LK7 and LK8 in B position, also connected AGND): 
 
 If you encounter same situation I recommend to: 
 
 replace corresponding output transistors - Q1 or Q2 or Q13 or Q14 (depends which channel stopped working, and which half-waves you will see - if positive or negative) 
 if it does not help also try to also replace corresponding resistors - R13 or R47 or R14 or R48. 
 
 Please be aware that in my case there was no visible damage on these parts although they were no longer working. 
 I'm currently thinking of protecting these 2 outputs with fuse or low-voltage bulb (say 12V, 50mA bulb). Because according to Ohm law - shorting output to ground (which may easily happen while prototyping) - will likely blow NPN transistor(s) (Q1, Q13) and/or resistors (R13, R47) with current over 1A.)

Partner Forums

Partner Libraries

Products Mentioned

Associated Drivers

Associated Drivers

Associated Drivers

EVAL-AD2S1205 - unexpected excitation buffer outputs