AES Convention Papers Forum

This is an exciting paper because for the first time it took material that tape engineers understood but didn't talk about, and explained it carefully.

The next to last page is the exciting part.  It lists tape flux levels at different bias currents for constant signal.  This is what the meter reads when you're setting the bias on the tape machine.  You have a constant signal at some frequency going in and you are sweeping back and forth looking at the peak.  You can see from the shape of the curves that the peak is narrower (but lower) at 10 KHz than at 1 KHz.

Underneath this is a plot of the third-harmonic distortion, so you can see the point at which the distortion is lowest.  This is where the whole idea of overbias comes from.... the lowest distortion is not at the same point as the peak at 1KHz or the peak at 10 KHz, so you have to go a known amount over the mark to get to the lowest distortion point.

Under this is a plot of the third order IMD.  You'll notice that the bias level that gives lowest IMD is a little higher than the bias level that gives lowest third harmonic.

Under THAT is a plot of the modulation noise level.  You'll notice that the bias level that gives lowest modulation noise is a little higher than that.

So, what is the correct bias point for this tape?  That depends on what characteristics you want out of the tape machine, but this chart and this paper finally allowed engineers to understand what they were getting when they set the bias with a given method.

Note that the tape used for these measurements was BASF LGR30 which was a red oxide tape.  If you look at these plots on a modern tape (and you should be able to get them from the manufacturer), the curve intersection points are a little different.  But THIS PAPER was the first time when all of this information was presented out front in one place so the working engineer could get an understanding of how machine setup affected sound.