[MarkDN]

I would like to know more about the amplifier requirements caused by EQ and how to model this in WinISD.

In another thread I came across this regarding WinISD and EQ:

The real power handling requirements actually shows up on the VA plot...

It was my understanding that to correctly model EQ power requirements in WinISD, you utilise the fact that for every 3dB increase, twice the power is required. Hence for every 3dB of EQ boost you would halve the signal power you enter into WinISD. (E.g. For 500w and 6dB EQ boost you would enter 125w.)

I've never fully understood exactly how the VA plot relates to amplifier requirements. The VA represents the "apparent load" not the "real load". As I understand it, the "apparent load" can never be less than the "real load". I'm assuming you can calculate the "real load" utilising the impedance phase. :nerd:
To put this in practical questions assuming:

• I have an amp that is capable of 500w RMS.
• I boost 6dB with an SOS high pass filter.
• Specifying 125w RMS as the source input in WinISD, the VA plot peaks at 278VA
• The impedance phase is +42 degrees at the VA peak.

Questions:

1. Is 278VA really the limit of my amp here? Or would the limit be 500VA?
2. Or would it be higher than 500VA because of the Impedance Phase since it is only the "apparent load"? So with a +42 degree impedance angle, this would be 672VA [500*(1/Cos(42))]?
3. Or is all this irrelevant since the amp is actually voltage limited? Hence why 500w was dropped to 125w since the voltage will be a quarter.

The ultimate reason for all these questions is that the answers will significantly effect what HPF cutoff to use when modelling. I probably just don't get it... :scratch:

It would be nice to know how to understand how WinISD works in this area as it is a bit unintuitive.
Can anyone shine any light on this?

 

[Mike P.]

Specifying 125w RMS as the source input is correct for 6 db of boost.

The ultimate reason for all these questions is that the answers will significantly effect what HPF cutoff to use when modelling

What sub, box size and tuning are you modeling?

 

[michael hurd]

6 db boost as in a parametric eq, with a center frequency and a narrow bandwidth, or a 6db / octave boost that is additive with a corner frequency? ( IE: commonly called L/T )

In the second case, with each additional lower octave, the power requirements climb. This can result in reaching the amplifier limits quickly.

6 db boost down low = 6 db less above the knee before you will run into amp headroom issues. ( this has been debated though, due to the 'spectral' content in alleged program material ) Some claim that there is no loss of headroom with lots of boost, I beg to differ.

In the end, I would look at the actual impedance in box to determine the power requirements, as well as looking at how much boost and where you plan to add this boost.

There is no real simple answer to this question, IMO.

 

[collo]

This WinISD overview page has a section on modelling boost, incl table to show how much to reduce system power by for a given dB of boost.

 

[MarkDN]

Thanks for all the feedback! I've also been giving this some more thought.

So from the gist of things and what I've previously read about what the "System input power" field in WinISD is that it is really defining the voltage input in the end. This is from the WinISD help file:
"Filter system is logically located at electrical side. 0 dB gain at filter chain means that voltage at driver terminal is equal that is specified at "signal"-tab. If filter/EQ chain gain is 6 dB at particular frequency, then voltage at that frequency is twice that is specified, and so on."

So assuming my amplifier is rated at a maximum 500w due to output voltage limitations I would have to reduce the modelled input power so that voltage is not exceeded when boost is added. Therefore, I would follow Collo's process outlined in his provided link.
onder:
This does pose an interesting thought. What about amps that are not limited by voltage, but by current? This occurs in certain configurations for example when bridged or into a 2Ω load. Judiciously boosting in the right frequency range where the "apparent load" is lower (obviously depends on the box design) it is conceivable that you could extract more headroom from your amp. This always happens with a ported box around the tuning frequency and is also typically where you would like some extra boost. When I mention "apparent load" I mean the "Amplifier apparent load power (VA)" plot in WinISD.

An example could be the EP2500 when run bridged into 4Ω. Referring to this link here the EP2500 is capable of a higher output voltage with an 8Ω load than with a 4Ω load when bridged. So in theory if you had low "apparent load" in the region you wished to boost you should be able to access extra headroom before it clips since extra voltage is available and the current requirements are still low enough. :dunno: That said looking at voltage difference, I doubt you will get much extra headroom in this case maybe like a whole extra decibel, so I doubt it is even worth worrying about when modelling.

Of course this is all theory, and only my "possibly flawed" interpretation of theory at that! :nerd:
I'd love to hear your thoughts...

 

[dougsmith]

This WinISD overview page has a section on modelling boost, incl table to show how much to reduce system power by for a given dB of boost.

Thanks for posting that link... I found it quite interesting.

-Doug