CIRCUIT TOPOLOGY TO REDUCE CROSS TALK BETWEEN SELECTED AND UN-SELECTED PRE AMP INPUTS
Original Publication Date: 2000-Feb-01
Included in the Prior Art Database: 2003-Jun-20
The typical cross talk number from selected to unselected preamp inputs for existing preamp modules is between 55 to 70 dB. The topology described below and illustrated above will give cross talk isolation of at least 85 dB. This isolation is very important in the tape products because of the noise conditions that exist on the deselected head inputs. The high noise conditions 1 on the deselected readers is due to the fact that active write head lines are next to the deselected read lines in the head cable. In current high end tape products, either separate desecrate FET switch devices are used to short out the deselected preamp inputs or separate preamp modules are used. Either of these solutions involves added cost and complexity at the card. The new topology described eliminates the added complexity and cost by not requiring external FET devices or separate preamp modules. The above topology has a number of unique features that will not be found in any other tape or disk drive preamp module. The first is the addition of the block called "Short Node" to each of the preamp inputs. The purpose of this block is to short the node to ground when it is not being used. This is done with an FET transistor. By shorting the node, the signal going to the deselected preamp input is reduced. The second is the fact that each preamp differential input (pins 2/3, 37/36, and 34/33) have separate "First Gain Stages". Each of the first gain stages have a typical gain of 33 when it is active. If the first gain stage is inactive, then the gain is a very small number (much less than one). By having separate first gain stages, interference on a deselected input will have less chance of effecting the selected first gain stage. The first gain stage is designed to eliminate feedthrough of signal from its input to its output when it is deselected.