Newbie question on REW / Eq-APO for a 2.1 system with a 'stereo dac'

[bjamk]

Apologies if this is a newbie question that has been asnwered before - but if so, I couldn't find the answer either on this forum or elsewhere.

The DAC I have (SMSL d6s) only has one set of stereo outputs , which is feeding my main bookshelf speakers via the 'balanced' out to an amplifier, and also sending (the same) signal to a subwoofer via a pair of stereo RCA cables.

The subwoofer has an in-built amp, with an adjustable low-pass filter. The amplifier for the bookshelves has has not Eq / filtering capability of its own (Audiophonics s125nc, or somesuch).

What I'm trying to understand is what the 'correct' procedure would be, for using REW to measure the setup, and generate a set of 'filters'/eq-paramters that I could put to EqAPO to get the best in-room response for this 2.1 system. I have a Umik-1 to do the measurements with, and have worked through a few 'tutorials' for how to Eq the bookshelves on their own.

From most of the tutorials I've read / seen, they seem to either assume that REW is used to generate filters for a 'stereo setup' (i.e. one pair of speakers), or for integrating a pair of stereo speakers, with a subwoofer fed via a different set of outputs (e.g. via a MiniDSP). there are also some tutorials that look feeding a fully-active 2.1 system studio monitor system via a stereo system, but those assume that both the mains and subwoofer have additional low/high-pass adjustments of their onw. In my case - whatever gets sent to the bookshelves will be full-range.

So, since I don't have an access to a 3 or 4 channel DAC - what would be the standard/best approach towards room correction of a 2.1 system, with a 2.0 feed from a DAC?

Any advice is warmly welcome !

 

[boxerfan88]

Over a year ago, I had a similar setup, using SMSL DO100, XLR to mains, RCA to 1 sub.

What I did then was to use REW to sweep L & R normally. Basically when sweeping L, you’re recording L+Sub; the same with R (R+Sub). Then use REW to analyse & generate the room correction PEQ. After that transfer the generated PEQ to EQ-APO.

Hope the above helps…

 

[staticV3]

Hi @bjamk! Welcome to ASR.

Here's what I'd do:

First, iteratively perform a set of REW measurement sweep+Moving Microphone Method frequency response measurement for different low-pass filter, phase, and level settings on the subwoofer (sweep for phase response, group delay, step response, MMM for frequency response), until you find settings that result in optimal subwoofer integration.

Then use MMM only to measure the in-room frequency response and use the EQ feature to knock down any peaks below the Schröder frequency of your room. Export the filter parameters as .txt.

If there is spin data of your Main speakers on https://www.spinorama.org/, then go to the "Data & EQ" tab and copy the AutomaticEQ (IIR) parameters straight into the .txt file.

If your speakers are not on Spinorama, then you can perform ultra nearfield, gated response measurements of the tweeters to measure their anechoic response, then use the EQ function again to correct any nonlinearities there.

Loudspeaker measurements

miniDSP is a leading manufacturer of Digital Audio Signal Processors for the HomeTheater, Hifi, headphone and Automotive market. Join our large community of Audiophiles, Engineers and DIYers using our innovative products.

www.minidsp.com

Copy the resulting filter parameters into your correction file, which you can load into Equalizer APO like so: https://youtu.be/jkgcsVNTJvc

In EQApo, above the Include block, add a Preamp to avoid digital clipping:


This is a pretty minimalistic approach to IIR filtering that avoids many pitfalls.

If you'd like to give advanced FIR filtering a try, then have a look at Obsessive Compulsive Audiophile on YouTube.

 

[bjamk]

Thank you both, @boxerfan88 and @staticV3 , that's really helpful, and will help me on my way

The L+sub then R+sub approach is sort of what I stared with, but used those mesurements to follow the OCA's 'recipe' for generating filters/IR's. this seems to generate some latency, and also, it may not be working particularly well since I may not have correctly integrated the sub first.

Will follow your suggestions for using PEQ next, and see how it goes.

Of course, very happy for anyone else to chime in with additional tips. Seems like there's a lot to learn!

 

[staticV3]

Thank you both, @boxerfan88 and @staticV3 , that's really helpful, and will help me on my way

The L+sub then R+sub approach is sort of what I stared with, but used those mesurements to follow the OCA's 'recipe' for generating filters/IR's. this seems to generate some latency, and also, it may not be working particularly well since I may not have correctly integrated the sub first.

Will follow your suggestions for using PEQ next, and see how it goes.

Of course, very happy for anyone else to chime in with additional tips. Seems like there's a lot to learn!

For sub integration and response correction below Schröder, it's best to measure the combined response of L+R+Sub.