LVDS to Singled-Ended - Transmission Line Balun Matching

Question

I need to convert a LVDS clock into a single-ended signal. In the past I've used a voltage balun with a center tap on the secondary but I haven't found any good past 1GHz. I need to get to 2GHz in this application and found transmission line baluns good from 4.5-3000 MHz like the TC-1-13M+ from Mini-Circuits.

Analog Devices shows an implementation of it on an eval board schematic in their datasheet for the ADL5391 (page 13). It looks like R4 is the matching for the output impedance of the IC but what is unclear is why they added R5/R6. Is this to present a 150 ohm differential impedance (R4+R5+R6) to the balun? The balun is 1:1 so doesn't that mean the single-ended output impedance is 150 ohm? The datasheet mentions the balun is 50 ohm. I feel like I'm missing something here.

I'd appreciate any clarification you can offer.

Answer 1

Notice page 3, "Differential Output Impedance", which rises with frequency; and Fig.16. Most likely they wanted a load resistor to dampen this, and then series resistors to match to 50/100 ohm outputs. Put another way, it acts as an R-pad or attenuator, which reduces signal level but stabilizes impedance.

Note that many resistors (and capacitors) are provided so the user can add/remove parts and pick whatever balance, impedance, and filtering to a certain extent, they require.

Generally speaking, transformers aren't too picky about the impedances they are used at; you simply get proportionally more loss, less amplitude, or less frequency range.

From such meager data, it's not clear how exactly the TC-1-13M+ would behave at 100 or 150 ohms.