EQ-ing HIFI speakers for a flatter response


New member
I've had a pair of Polk Monitor 5A speakers for nearly 10 years, and I've found them to be great listening speakers - detailed highs and nice, fast low end, but suffer from the ubiquitous mid-range "suck" that so many HiFi speakers do. So, I had the idea to analyze their response and boost frequencies above 1k accordingly. I measured the speaker by placing it on a chair in a very dead room (rockwool panels/moving blankets on the ceiling and walls, rug covering the floor) to minimize HF reflections, and placed a Behringer ECM8000 about 16' from the speaker, right in between the midbass and tweeter drivers. I used REW to collect the data. No calibration file for the mic, but I figured it would be good enough for this experiment. Below is the curve before applying EQ.

Screen Shot 2021-10-31 at 12.36.15 PM.png

As I imagined, there was a pretty significant cut between 1k and 10k, with the biggest dips around 2k5 and 5-6k. So, I used a Behringer Ultra-Curve (just a stereo rack EQ) to bring everything beyond 1k up to the 60db mark. A 2.5db increase at 1k, 5.5db increase at 2k, and so on.

So, I put the speakers back in position, and the results were positive. The midrange is more present (obviously), and details in the upper-mids that were previously inaudible can be heard clearly. The Behringer unit has an EQ in/out button that allows you to quickly A/B the speakers with and without EQ, and the difference is significant. The speakers are currently in my mixing position, and I plan to listen and mix on these for a few weeks. Curious to see how this experiment pans out!

I'd love to hear yall's thoughts on this. I don't fully understand the consequence of the phase aberrations that might occur by boosting so much of the spectrum using non-linear EQ, and I don't think that EQ is a substitute for a speaker with a well designed crossover/driver selection. I'd also like to add that there is no LF EQ. The room is treated, and I think most folks know that using EQ to correct room modes is a fool's errand. This was just an attempt to boost the mids and HF of the speaker itself, not in relation to the room it is used in.
I've never had any luck at all with anything other than tiny adjustments. In the PA world we always had 32 band graphics for making the system flat(ish) but I think it really just helped remove some room anomalies that would cause feedback and exaggerate resonances - but we did it almost as a 'norm'. However in a studio, it never worked for me. If we had nasty nodes producing weird peaks and troughs - trying to fix these always created extra problems somewhere else. My studio monitors are 40 years old. They were not flat when new, and are a little less 'crisp' than they were, but they've aged gradually, alongside my own hearing. A few months back I was fiddling with some 20-20K sweeps and discovered a nasty honk in my room (from the curved green screen I installed for a project at one end. I tried to EQ it out but while I tamed the big hump - anything around the B on the E string on my bass sounds dull. The Bflat and the C are both less clear, but it's gone above or below. I've left the notch in a bit, so I get a bit of extra volume at the B, but it's not horrid. In your plot - you have a peak and a notch down the bottom. I suspect these are similar to mine - so treating them needs care.

I think my view now is to learn your room and your speakers and work with them. I know that if I hear a little too much bass end, it will be fine on other systems. I think in mid to small rooms, flat is rarely going to happen with ANY speakers.
EQing monitors with an eq, 31 band or anything else does not work. Making the room sound right is the only way to do it.