Multiple LMH6554 Vcm Reference

Hello Victor,

I am not sure what accuracy you want for the Vcm.

I think that for many applications a resistor divider with a bypass capacitor of 0.01uF would be a good reference for the Vcm.  If you use two 500 Ohm resistors it will be able to drive 5 LMH6554 devices.  The maximum input bias current on the Vocm pin is 18uA  for 5 devices it should be less than 90uA.  Although the distribution is random it is not centered on 0 so the currents do not cancel, but it is unlikely for all 5 devices to draw the maximum current. 

If the 90uA bias current will create too much error then you could consider using a precision amplifier to buffer the Vcm voltage. 

For best linearity the output voltage should be biased to the middle of the supply voltage.  For a single 5V supply this would be 2.5V. 

Because this is a single ended to differential application on a single 5V supply I would recommend that you AC couple the signal path. 

If you want to run simulations of your circuit you can use our TINA-TI software.  The LMH6554 model is built into the software.  There is only one bug, the symbol of the LMH6554 has the feedback pins mis labeled. 

I have attached a TINA file that will help to show how to set up the model. 

Loren

Hello Victor,

The TINA schematic is attached to this post.  The file timed out on the previous post.

Loren

1134.LMH6554_AC_SE2Diff.TSC

Hello Victor,

I looked at your schematic.  It looks good. 

I removed the 2.5V DC bias on VG1.  It will cause large currents in the ideal transformer primary winding.

I also put a 10M Ohm resistor to ground on the transformer primary to eliminate design check errors.  It does not change the circuit operation. 

Loren