I have another question, this time about REW measuring this enclosure.
I have set-up a basic measurement system with a Scarlet 2i2, Dayton EMM6 and REW. The set-up is pretty common: microphone 1m away, no possible reflections except floor and ceiling. For what it's worth, I'm using the Dayton Mic calibration file.
I've done both near and farfield measuring, and the nearfield looks as expected. My issue is with the farfield, especially from 500Hz to 2500Hz.
Dayton official specs. The SPL from 500Hz to 2.5Khz is let's say from +- 86.5dB to 90dB.
Now my REW measurement. I repeated the measurement several times, also turned my test set-up 90 degrees, exactly the same results. Note: I changed the SPL to match the official specs, my Scarlet 2i2 is not calibrated.
It this to be expected? The difference is 7,5dB instead of 3,5dB. Am I measuring wrong? The highest peak at +- 5,5Khz is 95dB in REW, just like the official specs. So I'm doing something right. But my frequence response is much more erratic.
I don't know if it's related or not, but when I'm measuring my tweeter I get this in Impulse. Ceiling is 2,5m high so this is exactly as it should be.
Microphone is between tweeter and woofer vertically.
Now the woofer. No clear dying out of the initial sound wave, no clear reflection.
Thanks again, I might be a bit out of my element here, although I read all I could about REW.
Edit, might be unimportant, but I don't have a speaker stand and used this box. Could this have something to do with it.
Vincent
I have set-up a basic measurement system with a Scarlet 2i2, Dayton EMM6 and REW. The set-up is pretty common: microphone 1m away, no possible reflections except floor and ceiling. For what it's worth, I'm using the Dayton Mic calibration file.
I've done both near and farfield measuring, and the nearfield looks as expected. My issue is with the farfield, especially from 500Hz to 2500Hz.
Dayton official specs. The SPL from 500Hz to 2.5Khz is let's say from +- 86.5dB to 90dB.
Now my REW measurement. I repeated the measurement several times, also turned my test set-up 90 degrees, exactly the same results. Note: I changed the SPL to match the official specs, my Scarlet 2i2 is not calibrated.
It this to be expected? The difference is 7,5dB instead of 3,5dB. Am I measuring wrong? The highest peak at +- 5,5Khz is 95dB in REW, just like the official specs. So I'm doing something right. But my frequence response is much more erratic.
I don't know if it's related or not, but when I'm measuring my tweeter I get this in Impulse. Ceiling is 2,5m high so this is exactly as it should be.
Microphone is between tweeter and woofer vertically.
Now the woofer. No clear dying out of the initial sound wave, no clear reflection.
Thanks again, I might be a bit out of my element here, although I read all I could about REW.
Edit, might be unimportant, but I don't have a speaker stand and used this box. Could this have something to do with it.
Vincent
Last edited:
If you are measuring the woofer and not the system, measure on the woofer axis. This will show the IEC baffle response from Dayton as adapted to YOUR baffle, and is the main reason it will change. This is called the combination of Baffle Step, diffraction, and position on a baffle. Measure the tweeter on its axis.
If you are measuring with a time-lock and want all of the measurements to account for offset inherently, then you measure ALL sweeps at 1m on the tweeter axis.
Measuring between is only preferable if that is the listening axis intended. However, the tweeter axis is the best or preferred listening axis to avoid upper treble rolloff.
If you are measuring with a time-lock and want all of the measurements to account for offset inherently, then you measure ALL sweeps at 1m on the tweeter axis.
Measuring between is only preferable if that is the listening axis intended. However, the tweeter axis is the best or preferred listening axis to avoid upper treble rolloff.
The idea is to get the offset more or less in my measurements so I can do the XO design in VituixCAD with 0 on the Y axis.
It's only a 7 inch woofer and there's less than 10mm of space between woofer and tweeter (my current reference height), so were I to measure on the woofer axis then my mic would be only 3.5inch or so lower. I can measure this too if it's useful, but I'm afraid it won't make up for the decibels of difference, not by a longshot.
Is there anybody else that has this specific woofer Dayton Audio RS180P-8 with some self-made measurements?
Do respectable vendors like Dayton Audio publish SPL graphs that are too optimistic or clean?
It's only a 7 inch woofer and there's less than 10mm of space between woofer and tweeter (my current reference height), so were I to measure on the woofer axis then my mic would be only 3.5inch or so lower. I can measure this too if it's useful, but I'm afraid it won't make up for the decibels of difference, not by a longshot.
Is there anybody else that has this specific woofer Dayton Audio RS180P-8 with some self-made measurements?
Do respectable vendors like Dayton Audio publish SPL graphs that are too optimistic or clean?
ROTFLMAO! Thanks! I needed that! 
Regardless, they did provide a note as to how it was measured, so I'll give them that, though as you proved it's the next best thing to useless to the average DIYer.
Don't know about today, but historically, nearfield meant < 1/4"/6.35 mm (> ~18 kHz) from the dustcap (whatever is nearest to the VC) and its 1/24th octave smoothing means that any 'ripple' below 450 Hz becomes ~smooth as a baby's bottom.
Fh = Fl*2^n
Fl = Fh/2^n
n = ln[Fh/Fl]/ln[2]
where:
Fh = upper frequency
Fl = lower frequency, or the XO point in this case
n = octave spread
ln[2] = 0.6931
Regardless, they did provide a note as to how it was measured, so I'll give them that, though as you proved it's the next best thing to useless to the average DIYer.
Don't know about today, but historically, nearfield meant < 1/4"/6.35 mm (> ~18 kHz) from the dustcap (whatever is nearest to the VC) and its 1/24th octave smoothing means that any 'ripple' below 450 Hz becomes ~smooth as a baby's bottom.
Fh = Fl*2^n
Fl = Fh/2^n
n = ln[Fh/Fl]/ln[2]
where:
Fh = upper frequency
Fl = lower frequency, or the XO point in this case
n = octave spread
ln[2] = 0.6931
I don't know anything about measurements or other tech things, but we have two RS180Ps in a MLTL enclosure and the designers' F3 is around 40Hz: I usually leave the sub off as we find them more that adequate bass-wise for our listening. Volume is about 30 litres and a Precision Port supplies the venting.
Here's a well documented project which uses one RS180P:http://projectgallery.parts-express.com/speaker-projects/supernova-minimus/
HTH
Geoff
Here's a well documented project which uses one RS180P:http://projectgallery.parts-express.com/speaker-projects/supernova-minimus/
HTH
Geoff
A couple of things:
- Have you done nearfield port and driver measurements and applied the appropriate sum for setting the port measurements at the right level. Can you publish these, please?
- Your test set up with the baffle running flush with the Thomann box will give a measurement showing increased baffle size like you are measuring a 'floor standing' speaker not a 'book shelf'. Try a structure that has thin legs or better yet one thin leg
- Have you done nearfield port and driver measurements and applied the appropriate sum for setting the port measurements at the right level. Can you publish these, please?
- Your test set up with the baffle running flush with the Thomann box will give a measurement showing increased baffle size like you are measuring a 'floor standing' speaker not a 'book shelf'. Try a structure that has thin legs or better yet one thin leg
No respectable vendors like Dayton usually don't publish optimistic graphs, and they didn't this time. You didn't measure the speaker in the same test baffle used by Dayton, so you will not get the same results. If you build a large flat standard test baffle, you would get those results. As was mentioned before, your measurement includes cabinet diffraction, baffle step, box resonances, floor and ceiling bounce and likely other room artifacts unless you were very careful and gated the impulse response, That's the great thing about speaker design, it's really complex and interesting. You can load the Dayton measured response into software like the freeware Visaton Boxsim and enter all you box and baffle design and likely reproduce the response including baffle step and diffraction effects that you measured not including the room response.
- I bought an adjustable speaker stand with a singular round support pole. So the cardboard box is gone.
- I got rid of a hum in my set-up by using a laptop, and my impulse responses look normal now. First reflection around 4.6ms.
- I also built a new closed enclosure with the same speaker. It has different dimensions.
- I am using a gated response for my farfield measurements.
- I'm changing the port output with the correct formula to account for port dimensions.
- I combine farfield and nearfield at 300hz so farfield isn't of interest here but it's included.
My measurements result in a driver that's very hard to work into a good XO. And to be honest, these measurements are bad, and don't correspond to the official frequency response.
Anything else I can try?
I found this review of the DA driver: https://audioxpress.com/article/tes...-8-7-reference-series-midwoofer#&gid=1&pid=12
While it doesn't have as big a dB difference between 500 and 700Hz, it also shows that the frequency response isn't as smooth as it should be. I don't know why my two RS180P-8 are even worse though.
While it doesn't have as big a dB difference between 500 and 700Hz, it also shows that the frequency response isn't as smooth as it should be. I don't know why my two RS180P-8 are even worse though.
It's been a while, more like 4 yrs, since I last worked with this RS180P-8.
Below is my FR of the RS180P-8 in a 13L BR with a baffle width of 8.5". Taken on-axis with woofer at 1 meter.
As you can see, there's not much bass. There's a +5dB rise from 500Hz due to baffle step. That's expected.
At 4kHz, the cone starts to breakup.
Actually, your FR is very similar to mine.
Regards
Mike
Below is my FR of the RS180P-8 in a 13L BR with a baffle width of 8.5". Taken on-axis with woofer at 1 meter.
As you can see, there's not much bass. There's a +5dB rise from 500Hz due to baffle step. That's expected.
At 4kHz, the cone starts to breakup.
Actually, your FR is very similar to mine.
Regards
Mike
Can you post the individual measurements for near field port, near field driver and far field driver in .FRD, please? What are the baffle dimensions and mic distance in these instances?
It would be interesting to sim the baffle and make a simulation of output and see where they differ.
Is there any stuffing in the cabs, by the way?
PS. I agree with Mike, the blue far field measurement does show some consistencies with other measurements.
It would be interesting to sim the baffle and make a simulation of output and see where they differ.
Is there any stuffing in the cabs, by the way?
PS. I agree with Mike, the blue far field measurement does show some consistencies with other measurements.
Very nice Workmanship. Is that the aluminum reference driver? Supposed to have good bass? Since you are so good at cabinet construction consider building a Voight transmission design. Usually a full rage driver but who's to say? Built one and first put the driver in a small ported design according to parameters and was BAD. Dropped into an easy to build Voight cabinet and the bass was way better and much more over all airyness and open. Boxed in sound was gone. Might try that. Michael
That 5 k might be corrected with a notch filter? Bottom looks funky. Try it in a Voigt design since it's darn good to10 k. Just supplement with the same tweeter. Bass in these designs address resonance and usually have solid and more important natural sounding Michael