no doubt good advice in a general sense.
In this particular situation, @neo004 is using Hypex FA amps. I have used 8 to 10 dB 2nd order shelf to boost woofer output with my Hypex FA253 amps. The hardware/software of these amps seem unfazed by big EQ over 0 dB.
They use a adau1452, which has 42dB of headroom.
But don't forget the DAC, preamp etc as well.
I forget if they have their buffers running at 15V or 18V rail, but that gives you about 24dB or 25dB of headroom, referenced to a 1Vrms signal.
Those Hypex amps have a gain of around 20x (18.8 to be precise).
So we need at least 1Vrms to be able to deliver 100W@4ohm
Last edited:
The Hypex has multiple input options, analog balanced (XLR), analog single ended (RCA), digital SPDIF (coax), and digital AES. I only use the RCA analog connection. I imagine that this might be much more forgiving of operating above 0 dBFS than the other inputs, or maybe not.... So just be aware that my experience is only valid with analog RCA inputs.
I am not just talking about input.
The Class-D NCore modules also have a buffer stage on board.
How forgiving things are depends on rail voltages, not on the type of input.
So for @neo004, keep this all in mind. I doubt the hypex HFD software would let us build a filter which pushed the Hypex DSP over its limits. But it might be better for you to start conservatively, especially if you are using something other than RCA inputs.
When you build the filters in HFD, you can always adjust the overall gain, or the gain of the individual drivers, down so that the max gain on the woofer is at 0 dB.
j.
When you build the filters in HFD, you can always adjust the overall gain, or the gain of the individual drivers, down so that the max gain on the woofer is at 0 dB.
j.
I don't doubt that, I have done it.
Well okay that was just with a separate adau1452 into a ncore module
I don't know if they changed something in their own firmware to prevent these kinds of things.
They do have a "clip indicator" pin onboard, so I assume they use that.
But that is just an assumption, I am not that familiair with their software, rather prefer SigmaStudio (understatement)
They do have some circuitry to prevent bad things from happening, but can't completely recall the schematic anymore.
Btw, this is not just a "Hypex problem" but always be something to lookout for even if you use other components.
This is always a trade-off between more gain in the power stage, but also adding more noise because of that gain.