That's the same method that Tascam cited for adjuting the bias on the 388 and it only really works if you are using the tape type that the machine was designed for...are you using 456?
If not then I suggest you do the overbias technique (i.e. back off the bias trimmer all the way, advance until you find the peak and then continue until you are down as many dB's as specified by the tape manufacturer). This can be a bit of a pain on a two head machine which is one reason why Tascam guides you to setup the bias by measuring the output voltage of the bias amp, but, again, that is only a relatively reliable way to do it if you are using the tape that the machine was spec'ed for which was 456 for the TSR-8.
Example: The 388 was designed for 457, and the manual specifies to adjust the bias amp output to 150mV. I'm using LPR35 and 150mV wasn't cutting it. I talked to Jimmy at Tascam and he said that the bias requirement between 457 and LPR35 is significantly different and to setup the bias using an oscilloscope for minimum distortion. I haven't done this yet, but I will soon and when I do I'm going to setup each track for minimum distortion and THEN measure the bias amp output levels, average them and then I can post that up as a target when using LPR35.
That doesn't help you.
So are you using 456? If so, use the 35mV spec in the manual. If you aren't then get the spec sheet for the tape you are using and it should indicate how much "overbias" to employ. You could do a couple tracks using the overbias technique and then measure the bias amp output voltage on those and then set the rest up by just dialing the voltage into those bias amps.
Other methods: the LF distortion technique (i.e. the "rocks")...record an LF tone on your tape like 20Hz or 30Hz and listen for the LF distortion artifacts which sound kind of like rocks clacking together...big rocks. Keep dialing the bias level up in small increments during tracking and when the sound of the rocks is at its minimum then that's the optimum bias level. Same thing at that point to save some time: do a couple tracks using the "rocks" method, measure the bias amp output on those channels and then dial that in to the others.
The be-all/end-all test is "how does it sound"?
Remember, the whole point of biasing is to hit the tape with that HF signal which basically excites the oxide and prepares it for audio signal. When properly biased your tape will track with minimum distortion. Neither setting the bias amp output blindly nor using the overbias technique actually measure how effective the bias level is in terms achieving minimum distortion. Minimizing the "rocks" or actually measuring distortion with a scope is a true feedback method of setting the bias. But using the first technique works just fine if you are using the tape for which the machine was spec'ed, and the second technique works just fine as well and is probably the most common.
Then there is the whole subject of custom biasing to elecit certain performance characteristics (i.e. pruposely over or under biasing for certain sonic effects)...a whole 'nuther topic.
