Using REW to "tune" an external EQ

[jakex]

Hi everyone,

I'm new to REW. I used it a couple of years ago to check my room's acoustics (I run a small project recording studio) I have a UMIK mic with calibration file.

I have "corrected" my room using "other" room correction software and I'm reasonably happy with it.

I like building music gear and I've recently finished a clone of the famous Night EQ (6 band stereo mastering EQ with "air band")

It has a 10k trim pot for each band, and a 50k trim pot for the gain.

The calibration "manual" says to pre-set each trim pot to 5.6k then put all gain knobs to centre or zero gain and "adjust all trimmers to come as close as possible to the original flat curve" and gives me a multi curve REW file (attached) to "match" to.

My problem: I have absolutely no idea how to do this!

How do I set it up? What tone do I use? Do I use a tone?

I'm sorry if this is a big topic, and I know I should really have a full grasp of how REW does what it does and how to leverage that, but I don't, so please go easy on me!

Any pointers as to how to do this would be really helpful. I obviously have a multi-input/output interface so can easily put the EQ in/out through REW which is where I'm guessing I need to start.

Thanks!

jake

 

[Wayne A. Pflughaupt]

It’s really not clear what you’re trying to accomplish. If you’re trying determine what the setting for each filter’s gain knob is that results in flat response, then you’re on the right track connecting the EQ between interface’s input and output. You want to run the Soundcard Calibration routine for this (“Preferences” / Soundcard tab). It will be a tedious “trial and error” thing – make an adjustment on the equalizer, run the routine, compare the graph to the previous one to see what changed, rinse and repeat.

Regards,
Wayne

 

[jakex]

Hi Wayne,

Thanks for your reply. I'm not sure that is what I'm trying to do.

If you open the attached REW file I have been given, there is a curve for each knob at full + and - settings, and I have to use the small trim pots on the PCB to "match" that particular knob's curve as seen by REW with each curve in the REW file.

It's that bit I have no clue how to do. It may well be exactly as you say, but do I need to run a tone or music through it or does REW run a series of tones through it? So, when tuning the 150hz knob am I running a 150hz tone through the unit or what?

Apologies for not having a clue what I'm on about, and thanks for taking the time to give me a pointer. I wish the guy who makes the kits was a bit more helpful!

cheers,

jake

 

[Matthew J Poes]

If I understand correctly, it looks like these trim pots are setting the center frequency of the EQ? Is that right? It appears the Knight mastering eq is a graphic EQ, not a parametric EQ?

The best way to do that is to run sound from the output of the computer into the input of the EQ and then from the output of the EQ to the line input on the computer. If you run a small studio, you probably are using something like a USB audio interface. Then what I would do is adjust each EQ band to its max individually (one at a time) and run a test tone. You could even use the RTA if you want, that should be accurate enough (but you can run a sweep too). Then adjust to match the curve you were given. To match, I would just load the file for now and compare both together.

One thing that might help would be to know what it is you hope to accomplish with the EQ you have built. I think these devices are fun for DIY projects and even to add to a mastering chain to achieve a specific vintage sound. However, if the goal is accurate mastering and to EQ your mastering system to a flat and accurate response, I would not use such a device. This is not an accurate or precise way to EQ your mastering speakers for a neutral response. Instead, I would use traditional PEQ. If your mastering includes both Analogue and Digital and your speakers have a pure analogue path, then EQ using an analogue PEQ. If your system would always involve an AD step and convert to digital before going to the speakers, then a much cheaper and easier option is to use a software EQ. There are even free options. Even better would be to use something like DIRAC as that allows full frequency response correction of amplitude and phase and would give the most neutral correction. This ensures that when mastering, you are not in fact eqing out flaws in your own reproduction path.

Also keep in mind that some vintage mastering equipment had been tested using modern testing equipment and found to be very flawed. Floyd Toole would probably refer to this as part of the circle of confusion. If you master something using a flawed product (say vintage speakers) then people using modern and more neutral speakers will not hear what you heard. They would have to EQ in the same flaw to their system to get accurate sound. EQ's like you are using don't concern me as much as vintage studio monitors. Especially something like the vintage Urei's, which Harman tested on the Spin-O-Rama and were given the lowest sound quality score under blind conditions of any speaker they ever tested (They were discontinued for a reason). It's measured and subjective flaws were so great that Toole concluded the love for these in the studio was totally bonkers (My phrase not his).

 

[John Mulcahy]

Easiest way to observe the response as you make adjustments is to play the Pink PN test signal from REW's signal generator through the system and look at the response on the RTA using these settings:

 

[jakex]

If I understand correctly, it looks like these trim pots are setting the center frequency of the EQ? Is that right? It appears the Knight mastering eq is a graphic EQ, not a parametric EQ?

The best way to do that is to run sound from the output of the computer into the input of the EQ and then from the output of the EQ to the line input on the computer. If you run a small studio, you probably are using something like a USB audio interface. Then what I would do is adjust each EQ band to its max individually (one at a time) and run a test tone. You could even use the RTA if you want, that should be accurate enough (but you can run a sweep too). Then adjust to match the curve you were given. To match, I would just load the file for now and compare both together.

One thing that might help would be to know what it is you hope to accomplish with the EQ you have built. I think these devices are fun for DIY projects and even to add to a mastering chain to achieve a specific vintage sound. However, if the goal is accurate mastering and to EQ your mastering system to a flat and accurate response, I would not use such a device. This is not an accurate or precise way to EQ your mastering speakers for a neutral response. Instead, I would use traditional PEQ. If your mastering includes both Analogue and Digital and your speakers have a pure analogue path, then EQ using an analogue PEQ. If your system would always involve an AD step and convert to digital before going to the speakers, then a much cheaper and easier option is to use a software EQ. There are even free options. Even better would be to use something like DIRAC as that allows full frequency response correction of amplitude and phase and would give the most neutral correction. This ensures that when mastering, you are not in fact eqing out flaws in your own reproduction path.

Also keep in mind that some vintage mastering equipment had been tested using modern testing equipment and found to be very flawed. Floyd Toole would probably refer to this as part of the circle of confusion. If you master something using a flawed product (say vintage speakers) then people using modern and more neutral speakers will not hear what you heard. They would have to EQ in the same flaw to their system to get accurate sound. EQ's like you are using don't concern me as much as vintage studio monitors. Especially something like the vintage Urei's, which Harman tested on the Spin-O-Rama and were given the lowest sound quality score under blind conditions of any speaker they ever tested (They were discontinued for a reason). It's measured and subjective flaws were so great that Toole concluded the love for these in the studio was totally bonkers (My phrase not his).

Thanks for your detailed response, Matthew.

Firstly, let me clear one thing up: I am neither a mastering engineer nor am I going to use this EQ as a mastering EQ. I'm using it as part of a signal chain on my master bus. It's more of a "finishing" EQ. I want to use its lovely gentle +/-5dB Baxendale-style curves to put a bit of sheen on my tracks.

I run a mostly analogue studio, with lots of nice outboard (that wasn't built by me) and this was a relatively inexpensive way of indulging my electronics bug with something I can use. I recently built two Pultec-style 500-series PEQs and this was a logical next step. The Pultec's didn't require "tuning" and all of this is new to me. RTFM as I would tell my customers!

My room has been tuned using a dedicated microphone system with 14 measurement points. It's not perfect, but the difference between the uncorrected and corrected sound is dramatic, so I'm guessing I'll need to tune this with my correction profile turned off (as i do when i print to tape - yes, I have a reel to reel tape deck too - sad!)

I'll give it a go and probably fail, so please bear with me if I need to ask a couple more questions. I'll plug away and see how I get on. I'm going to email the creator again.

cheers,

jake

 

[jakex]

Easiest way to observe the response as you make adjustments is to play the Pink PN test signal from REW's signal generator through the system and look at the response on the RTA using these settings:

View attachment 6845

Thanks John. I am starting to get a grasp of what I'm supposed to be doing. I'll report back

Cheers,

jake