This thread reminds me of why I quit fooling with consumer decks in the first place. They just didn't make them with adequate adjustability. Once you get used to working with studio machines, which have all the adjustments you need, and adequate range on them, you get spoiled being able to get the machine dialed in properly, and no longer want to live with the compromises that the consumer decks force you into.
And I believe I was working on an A77 when I finally threw in the towel.
One of the bad engineering practices commonly recommended when pretending to calibrate these decks is to use bias trimming to get the frequency response closer, and I see that's just what you're trying to do. Sure, changing the bias will change the frequency response, and it has been recommended in countless HiFi magazines from the '60's, but think about it for a second: it's the backwards way to approach it. You wind up adjusting bias away from the optimum bias point in order to compensate for the lack of a proper frequency response adjustment.
Setting the bias for its own sake is one of the most important and critical adjustments in a recorder. The optimum adjustment point for bias is the point that yields the lowest distortion and the lowest modulation noise. Modulation noise is something you don't run into very many places in audio other than magnetic recording, so you may not be familiar with it. It isn't tape hiss, which is constant. Modulation noise is noise generated only in the presence of signal. It's one of the least appealing parts of magnetic recording, and in my opinion we should work hard to minimize it. That's why I think it's foolhardy to adjust the bias away from its optimum point just because the manufacturer failed to include a proper trimmer on the EQ circuit.
Also, let me suggest a more accurate method for bias setting: use a 1KHz tone. At 15 or 30 ips, bias 1/4 dB over the peak. At 3-3/4 or 7-1/2 ips, bias 1/4 dB under the peak. It's a long story, too long for this post, but you can find some of the story in the papers so generously provided on the MRL site.
And on another note: Tim generally gives very good advice on this forum, so I'd recommend not dismissing his suggestions out of hand. Whenever we are using measurement instruments our first job is to check the measurement setup and make sure it is really measuring what we think it's measuring. It's trickier than we think--at least until we have fooled ourselves a few times, then we gain a respect for this requirement.