Digital X-over Prototype

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
:D Screaming of Joy :D

I have done it !

My prototype digital cross-over is working !

It is based on :
- A CS8420 to convert SPDIF to I2S Left justified
- An AL3101 DSP doing the actual algorithm
- An UDA1328 (for now) converting the filtered signal to analog
And as last, a 16F84 as controller.

The Algorithm is just an IIR of an 4th order Linkwitz-Riley, or better a squared butterworth. Later on i hope to do some FIR filtering

Now i can start thinking of making it more mature (advanced, whatever)

Let me now what you think of it !


grtz


Simon
 
How did you interface the PIC & Alesis chips? Does the PIC load the program into AL3101, where does it store it (on PIC there isn't enough FLASH to store the whole program)?

You can use it for FIR, though only ~500 point IR for stereo and ~1000 (maybe not quite) for mono filter. You first implement either the HP or LP and then subtract the other one (single cycle). For baffle step compensation and some other EQ needed (depending on drivers) you also need some cycles. Anyway, less than 100 points is OK for bare xover functionality. Problem is the latency (1/2 of the IR length).
 
How did you interface the PIC & Alesis chips? Does the PIC load the program into AL3101, where does it store it (on PIC there isn't enough FLASH to store the whole program)?

The Program is loaded from a eeprom.
The AL3101 can do that by itself!

Problem is the latency (1/2 of the IR length).
Can also be compensated for, by a delay !?

best regards

Simon
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.