Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums > Audio Measurement and Testing

Advice On Eq for TFM122M Wedges

(1/4) > >>

John Schalk:
One of my goals for this year is to learn how to use some of the audio analysis software that is out there.  I decided to start with Room Eq Wizard.  Even though its primary user base is the home theater market I've found some helpful YouTube videos that allowed me to take my first series of measurements.  I took my measurements outside in my driveway as far from any boundaries as I could get.  I decide to post the As Is measurements first and get some input before moving from the science of measuring speakers to the art of applying Eq or other filters to them.

In the screen shot, I've moved the average response up by 6dB to make it easier to compare to actual measurements.  I reduced the ground plane measurement by 3dB to match it to the three mic stand measurements, before averaging.  I used 1/24th octave smoothing.  The only processing I applied was the manufacturer's recommend HPF which is a 24 dB Butterworth @ 58Hz.

My first question is about the dip from ~200Hz to ~400Hz.  It's not present in the ground plane response, but shows up in all 3 mic stand measurements.  I believe that this is a cancellation from the ground and therefore cannot be corrected with Eq, is that right?

As far as any other changes to make, I feel like I should try to tame the low end a bit from 100Hz - 200Hz, the peak at 1.2kHz, the dip at 2.4kHz, and the peak around 6.5kHz.  Does that seem right? 

Luke Geis:
That actually looks pretty decent really as far as responses go. The dip at 250hz is likely from floor bounce and while you could EQ it out, it won't be right. I say leave that alone. It gets the mud out anyway. I prefer to high-pass my monitors pretty high, usually between 100 and 200hz depending on the instrument, the singer, and the size of the woofer. 15" usually has more low-end oomph, so I tend to roll more lows out with them. The ultimate goal is to reduce the amount of low end until the sound is natural and just full enough. I roll the HP up until the vocals sound thin and then bring it back down until it sounds right again. Usually around 140hz to 160hz.

As for the other dips, I would leave them alone. The one at 15khz is probably also another anomaly, a reflection from the horn into the grill and back of the horn or something. Or they did that on purpose to make the box have a sweeter-sounding high end? The only way to know for sure is to have a phase coherence trace. If the coherence dips at 250hz and 15khz, it is very likely reflections at the measurement mic causing the dips.

Objectively, you want to have the monitor be as linear as possible. This makes it less likely to be the cause of any issues making it easier to fix the problems at the mic channel as opposed to hacking your insert EQ to high hell. The old way of doing monitors was to turn things up until it rings, cut the offending channel on the insert EQ and repeat until you got three tones at one time. With a nice wedge, this could usually be done within 6 filters being engaged and not having much more than -6db of cut. Beyond that, you were spinning your wheels and you wouldn't get any further. With the advent of digital mixers, the game changed a little. I do not use graphic EQs inserted into my aux sends!!! If I/you can't get the monitor loud enough, stable, and sounding decent with the 6 parametric EQs most digital mixers provide, I/you am/are doing it wrong.

In your case, I would only worry about the two spikes you have at 1.2khz ( likely where the crossover starts to come into play ) and at 6khz where the dip is. There is some reason why there is a dip at 6khz with a small peak on either side of it, but what is causing that can only be determined with a coherence trace. If the trace is good and there is no other logical reason for it to be there, then it just is and all you can do is try to get the peaks to come down more linearly with the rest of the trace. Rolling the HP filter up will also get rid of the low-end build-up. I suspect by the time you roll it up to around 100hz, it will flatten out nicely.

The trace you have though looks pretty darn good though as it is. You shouldn't have too many issues getting a loud, clean, and effective monitor mix. I was looking at getting the TFM122 but was turned off by the seeming lack of availability, references, and reviews of them. I instead ponied up a little more cash and went with the RCF NX12SMA's. Very happy with that decision. They are board flat from 100hz to 16khz!!!

Luke Geis:
I took a quick peek at their online specs and the results you have are on par with what they present. The dip at 6khz is in their plot. It looks like they may have used a little more smoothing though. They do show a small dip at 15khz, but not like yours. The only thing in yours that doesn't align with theirs is the thing you have going on around 1.2khz. Which can either be because of the smoothing, their test conditions ( they say 1w/1m ), and environmental impacts. So to sum it up again, what you have looks good, you just need to get Smaart or something to really know more about what's going on.

https://mediadl.musictribe.com/media/sys_master/hc9/h38/8848079224862.pdf

John Schalk:

--- Quote from: Luke Geis on May 25, 2021, 12:15:43 AM ---I took a quick peek at their online specs and the results you have are on par with what they present. The dip at 6khz is in their plot. It looks like they may have used a little more smoothing though. They do show a small dip at 15khz, but not like yours. The only thing in yours that doesn't align with theirs is the thing you have going on around 1.2khz. Which can either be because of the smoothing, their test conditions ( they say 1w/1m ), and environmental impacts. So to sum it up again, what you have looks good, you just need to get Smaart or something to really know more about what's going on.

--- End quote ---
Thanks Luke. I have a spreadsheet that lists Turbo's Eq recommendations for the TFM series, and their PEQ settings line up pretty well to my results too.  They suggest a bump at 300Hz, another at 780Hz, a cut at 1.15Hz, a boost at 2.4kHz, and a final cut at 4kHz.  The only one that doesn't jive with what I got is the cut at 4kHz.  I've added a screen shot of the SPL & Phase for the ground plane measurement.

I looked pretty hard at the RCF wedges myself, but ended up going with the TFM series in part because I already owned the amps.  Also, our forum friend, Mr. Pyle, and some in stock that he sold me for a very nice price, so the cost per monitor "channel" was much less than if I had started over with the powered RCFs.  My amps are PSoft M50qs with DSP, so my plan is to tune them to "flat" or something close based on this forum's help, and store those settings in the amps.  That's why I'm thinking of sticking with Turbo's recommended HP settings for the amp's DSP.  That way, if I need to use these for something other than vocal monitoring, I can. 

I have a Performer 2 which has 4 bands of PEQ on the bus outputs, but it also has a 1/3 octave Eq on every aux bus, so I plan to use those for feedback suppression and voicing.

John Schalk:
Here is an SPL & Phase plot for the ground plane measurement using REW's Variable Smoothing option which is defined as follows:

Variable smoothing applies 1/48 octave below 100 Hz, 1/3 octave above 10 kHz and varies between 1/48 and 1/3 octave from 100 Hz to 10 kHz, reaching 1/6 octave at 1 kHz. Variable smoothing is recommended for responses that are to be equalised.

With variable smoothing applied, the big dip around 15kHz disappears.  I wasn't planning to apply any Eq above 10kHz anyway.

Note: I just learned how to save an image of the plot area directly in REW and it makes tiny files which is certainly handy for use on this forum!

Navigation

[0] Message Index

[#] Next page