LT5537 Vout starts at 0.9 V

Question

Dear all, 
 
 I created a simple circuit like shown in the LT5537&lsquo;s datasheet just without the R1 and C6 from figure 13. 
 I tested two different LT5537 without any RF input and nothing connected to the output. I even grounded both input pins. The Vout stays at about 0.9 V. I tried it with the simple input circuit from figure 13 and with the matching network. 
 Disabling the LT5537 does only slightly reduce Vout. 
 Why is my initial Vout that high? It should be around 0.4 to 0.5 V right? 
 Maybe this is also the problem why I&lsquo;m not even close detecting 900 MHz at -70 dBm. 
 
 Any hint is welcome. 
 
 Best, 
 Michael

ADIApproved · Accepted Answer

Hi Michael, 
 Sorry about the delayed response. 
 You're correct, LT5537 Vout with no input should be around 0.4V, but there is no min or max spec, so there could be a relatively wide variation from part to part. 
 I'd suggest using one of the external component value sets from DS table 3. Pick one design number for test; avoid combining or altering of the designs. 
 Poor layout could cause an oscillation, raising the Vout for no input. These devices have exceptionally large gain*bandwidth product. A good low-inductance ground plane for the entire circuit is essential, along with proper Vcc decoupling capacitors and a low noise Vcc (not a switcher). 
 The only other parts I could recommend for 900MHz, -70 dBm or better would be perhaps AD8313 or MAX2016. It's probably better to use a less sensitive detector, along with an LNA and RF bandpass filter in front of the log detector. 
 Hope that helps. -Bruce H.)