We are using the TAS5731 in a 2.1 channel design. The subwoofer signal path only has 2 bi-quads available and we need another 1 or 2 to meet our goals. I've been staring at the block diagram trying to find a novel solution and may have found one if the diagram is showing what I think it is.
So the question I have concerns the apparent signal path from the left channel (between the 6 BQ filters and the single BQ filter) to the switch before the bi-quads in the sub channel. Does this path really exist and what was it's intended purpose?
What I would like to do is use the first bi-quads in the left and right channels to create a hi-Q, HP filter (to give the sub some low end boost), then use mixer 55 and BQ 5E to low-pass the right channel signal before mixing it back in with the left channel with mixer 53. So then this summed signal would be sent to the sub channel using the path mentioned above. Then BQs 59 and 5D would be high-pass filters to remove the low frequency (and thus right channel signal).
Would this work as I think it will? Thanks!
Your plan could work, but if you tell me more about what your objectives are, I might be able to give you a better answer. This would probably meet your subwoofer needs, since you would essentially have a couple of extra filters to shape the signal with. One of my concerns is that you might not be able to tune the left and right channels as much as you might want to. Your manipulations to get the best subwoofer signal might alter the left and right channel content enough so that one filter isn't enough to correct it.
The original purpose for the signal path from the left channel to the MUX in the subchannel was to allow for a 2.0 two band DRC application. This allows you to run the DRC algorithm for low and high frequency content separately.
It is a very small, self contained 2.1 system and only having 2 bi-quads (one high-pass, one low-pass) has not been enough to optimize our bass response. Essentially, if we tune the HP low enough (with a high Q to give us some extra extension) then we end up with a "saddle" in our acoustic response between that filter and the LP filter. If we move the HP up enough to flatten our response then our needed LF extension goes away.
So our acoustics engineer would like another filter (or two) in the sub woofer channel. If there is another way with this IC I'm all ears.
I'm not super concerned about tuning the L & R signals as we only need a couple of EQ filters to optimize that acoustic response (and the HP to filter the sub frequencies). So using some of the other BQs in the L & R signal paths will be OK and shouldn't effect what we're trying to do with the sub. I know this will likely require a little more trial and error as we overcome any interactions between the desired signals.
Is there a way to implement this in the GUI? Thanks!
Jon,
I don't see another way to give you an extra filter for the subwoofer, still have separate left/right channels, and have both the left and right channels' content in the subwoofer. Since what you are suggesting isn't what this part was designed for, I can't guarantee that you'll be able to tune the part enough to meet your needs. If you really need the extra filters for the subwoofer, then your scheme is worth a try.
One more thing to consider before going down this road is a different filter design. Maybe try using a bass shelf followed by a low pass filter. The current version of the GUI doesn't design that type of filter, but you can input the filter coefficients manually.