>Was wondering if anyone in the group has ever heard of or tried the
>mod described below and if so could you share your comments about
>wheter it was worthwhile doing.
There is nothing intrinsic in the 3053/3028 that would require
biasing the input terminals to 2/3 Vcc rather than 1/2 Vcc -- it's
just a differential pair with an on-chip current source feeding the
emitters. The external circuit, together with the LO and "RF" drive
levels, will set the linear operating range. You need to ensure
adequate voltage overhead for both the differential pair and the
current source, and the differential pair base bias trades these off
against each other.
The Omni V uses U3 as a Gilbert cell mixer with the LO (BFO in this
case) fed to the base of the current source, and it appears that T-T
may not have the BFO injection and "RF" (in this case, IF) levels
quite right. I'm not very fond of the fully bypassed emitter
resistor on the current source (C22) when the base is used for LO
injection, but the more unbiased emitter resistance there is at that
point, the less voltage headroom the current source has. I'd rather
see the Gilbert cell run on 15-20 Vcc so these limits weren't so close.
TenTec mailing list