Hi all This is the frequency response of my creation
It was taken with an uncalibrated microphone in the garden (noise one) at approx. 1 meter
Smoothing 1/48 no gate applied
I am an hobbyist and i have to say they sound really good to me however. What about that deep at around 3000Hz? it should not be there the crossover frequency is around 1800Hz so out of that range and the tweeter is suppose to be linear there.
Could it be caused by baffle refraction?
What could i try to reduce it?
Again i am a new bee so any help is appreciated
Thank you
It was taken with an uncalibrated microphone in the garden (noise one) at approx. 1 meter
Smoothing 1/48 no gate applied
I am an hobbyist and i have to say they sound really good to me however. What about that deep at around 3000Hz? it should not be there the crossover frequency is around 1800Hz so out of that range and the tweeter is suppose to be linear there.
Could it be caused by baffle refraction?
What could i try to reduce it?
Again i am a new bee so any help is appreciated
Thank you
Could you post more information and maybe photos, baffle size, tweeter position, driver models, crossover, response at angles ...
Last edited:
I always ask, if you like what you hear, why do you want to mess with it. You might tune out the dip but then cause something else. It is like having super glue, to try find things to fix.
Hi all
First Tks for replay. This is the picture of the speaker
it is a two way some six liters bass reflex. component are Monacor SPH135c and DT254
Crossover frequency is 1800Hz baffle size are 21X32Cm
As said i like the sound of it but
I like to fix it not because i am looking for perfection but just because as i am learning i would like understand what is causing it -)
Is 3000Hz a tollerable place for a deep?
First Tks for replay. This is the picture of the speaker
it is a two way some six liters bass reflex. component are Monacor SPH135c and DT254
Crossover frequency is 1800Hz baffle size are 21X32Cm
As said i like the sound of it but
I like to fix it not because i am looking for perfection but just because as i am learning i would like understand what is causing it -)
Is 3000Hz a tollerable place for a deep?
I suppose so. In addition it will probably disappear once you listen to the speakers at some angle.
Have a look at the second image in my post here.
A tweeter mounted centrally with the same (or similar) distance from the top edge will have some very severe baffle edge diffraction effect that will lead to dips and peaks.
You should measure the speaker at angles before modifying it. Maybe listening at 30° is perfectly ok and you could either not toe the speakers in at all or direct them to a point in front of the listening spot.
If you need to change the speakers eventually consider cutting chamfers around the tweeters.
Have a look here.
Cross over slope? Have you tried reversing the tweeter phase? What do impulse response and phase look like? Mic position? Have you tried measuring at several other points?
I think looking for root cause and mitigations will pay dividends. Once you know the cause and possible remedies you can decide whether the subjective performance is what you prefer. (I usually like a bit of a bit a dip between 1 - 4kHz)
They look very good. 😀
I think looking for root cause and mitigations will pay dividends. Once you know the cause and possible remedies you can decide whether the subjective performance is what you prefer. (I usually like a bit of a bit a dip between 1 - 4kHz)
They look very good. 😀
Try to measure the response some 30 degrees above and below tweeter axis (same distance as on the original response plot) This will give more info about the polarity, the "rightness" of slope and possible x-over frequency separation. The dip at 2-3.5 kHz is rather deep (6 dB), and this might degrade -"rob"- the high frequency "slam" and attack of certain instruments - drum rim shots, cymbals, plucked intruments etc.
tks Kevinkr
Unfortunately I lost all other measures so I need to do again!!
From what i remember
All measures were with no major problems phase and impulse were okishh
Crossover slope is the best I could achiv
Inverting polarity create a even big deep at crossover point but did not affect the One at 3000hz I shall make new measurements at different angles.
I really suspect that position of tweter on the baffle maybe the root cause
Need to experiment more
Ref the look thank you it took a lot of work!!!
The box is tick MDF and the side are solid wood strongly glued on
The box is rock solid
Thank you again
Unfortunately I lost all other measures so I need to do again!!
From what i remember
All measures were with no major problems phase and impulse were okishh
Crossover slope is the best I could achiv
Inverting polarity create a even big deep at crossover point but did not affect the One at 3000hz I shall make new measurements at different angles.
I really suspect that position of tweter on the baffle maybe the root cause
Need to experiment more
Ref the look thank you it took a lot of work!!!
The box is tick MDF and the side are solid wood strongly glued on
The box is rock solid
Thank you again
I'm pretty sure its the baffle width and tweeter symmetry on the baffle causing this specific dip. A quick way to confirm this is measuring the vertical axis tipped up and down. You'll see a shift in the dip across various angles across the axis.
A ground reflection path length is longer than the direct path, which can cause a dip in the response.
What was the height of the speaker and mic from the ground?
A Umik-1 or Umik-2 from MiniDSP is a good reasonably priced USB choice for use with REW. I have some fancy mics as well, but mostly still use the Umik mics. (I have several) You download the calibration files from MiniDSP based on mic serial #.
Sometimes there is no great substitute for a measurement outdoors on a tall stand if you can manage it. My speakers weight about 100kg and are not Kevin friendly when it comes to moving them. 🤣 It would make my life simpler if that was not the case.
If your speakers look nice, which they are, then listen to them until you grow tired of them and tackle a new project. On the other hand get an equaliser and tailor it to your liking and be done. I don't care how good or adequate your speakers are, you will at some time or other want new ones because it is like a girlfriend, after some time you need a change and hanging on makes matters worse.
So what you are saying is cancel the ground reflection. I don't care what speakers you have there will be reflections and there is no way to avoid it. Your brain will do your signal processing.
I'd pursue correcting the FR just for the sake of getting the experience dealing with a design problem like you're encountering. It's good practice and makes you more knowledgeable, improving your next project outcome. Otherwise you end up with something that collects dust and didn't end up really being finished. You'll learn the most from fixing your mistakes.