X,
A couple of points. Could you post non-smoothed responses? For example, if you are setting the gate to 200 Hz, smooth only to 1/48th, or at the worst 1/24th. Any more smoothing and the native smoothness of the driver's response will be covered up by the software smoothing. Smoothness is important, according to Olive, together with flatness.
Second, it's got to be a 50 db scale with 5db increments. That, IMO, gives the right resolution to see details. With a larger scale, everything looks smooth.
Thank you for doing these tests, BTW. I have collected quite a few full range drivers over the years and I planned to do something similar but never got around to it.
Another question, are the drivers being flush mounted?
Thanks! And great work!
Edit: I would make a new thread for full range driver measurements. You can keep editing the first post of the thread and link to posts where new measurements are posted.
A couple of points. Could you post non-smoothed responses? For example, if you are setting the gate to 200 Hz, smooth only to 1/48th, or at the worst 1/24th. Any more smoothing and the native smoothness of the driver's response will be covered up by the software smoothing. Smoothness is important, according to Olive, together with flatness.
Second, it's got to be a 50 db scale with 5db increments. That, IMO, gives the right resolution to see details. With a larger scale, everything looks smooth.
Thank you for doing these tests, BTW. I have collected quite a few full range drivers over the years and I planned to do something similar but never got around to it.
Another question, are the drivers being flush mounted?
Thanks! And great work!
Edit: I would make a new thread for full range driver measurements. You can keep editing the first post of the thread and link to posts where new measurements are posted.
Last edited:
Ra7,
The 5dB scale data and no smoothing will have to come in a later installment. This will give us an idea for relative differences from a listening standpoint (I recall human hearing is similar to 1/12th octave smoothing. You will see why I need such course resolution and wide range in a bit. But sure I can present gated 200Hz and up with 1/48th octave smoothing from 50 to 100dB in 5dB major increments. Another reason is that you physically need a larger monitor to show that in REW as these were done on a tiny netbook screen.
No flush mounting was used - all surface mounted, but most drivers have a pretty thin bezel except for Faitals which have an upturned lip for the screw mount ears.
It's not that I am trying to build suspense with the CHN70's. I am just taking extra time and effort to make sure I am doing this right because I am seeing rather puzzling results which I will show shortly now that I am convinced they are real and inherent in the driver and not the enclosure or the way they were measured. I made an open baffle to compare with the trapezoidal supra baffle and Nautaloss rear chamber to ensure that the baseline setup was not the source of the anomaly. I also measured the impedance of the CHN70 in the Nautaloss and free air, and measured the T/S params to make sure the unit is matching published specs.
The 5dB scale data and no smoothing will have to come in a later installment. This will give us an idea for relative differences from a listening standpoint (I recall human hearing is similar to 1/12th octave smoothing. You will see why I need such course resolution and wide range in a bit. But sure I can present gated 200Hz and up with 1/48th octave smoothing from 50 to 100dB in 5dB major increments. Another reason is that you physically need a larger monitor to show that in REW as these were done on a tiny netbook screen.
No flush mounting was used - all surface mounted, but most drivers have a pretty thin bezel except for Faitals which have an upturned lip for the screw mount ears.
It's not that I am trying to build suspense with the CHN70's. I am just taking extra time and effort to make sure I am doing this right because I am seeing rather puzzling results which I will show shortly now that I am convinced they are real and inherent in the driver and not the enclosure or the way they were measured. I made an open baffle to compare with the trapezoidal supra baffle and Nautaloss rear chamber to ensure that the baseline setup was not the source of the anomaly. I also measured the impedance of the CHN70 in the Nautaloss and free air, and measured the T/S params to make sure the unit is matching published specs.
Last edited:
Maybe not trying, but doing so none the less😛
It's great stuff that you are doing - thanks for the effort.
It doesn't show up as ringing, it is there, period, as ringing, that's why there is a peak in the frequency response. We don't hear cones ringing, like a bell rings, all we hear is the uneven frequency response.
Sibilance is also not an artefact associated much with 15kHz, sibilants are usually around 1 octave lower at around 6-8kHz. A peak at 15kHz will add in an artificial sense of air to the sound and most wont even be able to hear it.
CHN-70 Measurements
The CHN-70, is a paper cone driver with published specs that appear suitable for many vented alignments, especially MLTL's and bass reflex. The $27 price puts it in the same class as the other drivers in this lineup - especially the RS100/P which has a similar Qts, xmax, and sensitivity.
CHN-70 mounted in trapezoidal supra-baffle Nautaloss:
Upon measuring the response in the trapezoidal supra-baffle Nautaloss, I obtained the following response and harmonic distortion plot which was very puzzling: a big dip near 670Hz accompanied by high harmonic distortion there, and several other dips on high with a notable 12dB dip at 7kHz, and a rapid falloff above 12kHz. Initially, I thought my setup might be to blame, but none of the other drivers showed anything like this kind of response. In any event, I had to look deeper to make sure the measurement and enclosure were not the source of the dips and falloff.
CHN-70 mounted in OB:
I thought that the Nautaloss chamber may be the source of this so I then proceeded to put it in a 24in wide x 36in tall open baffle and repeated the measurement. The OB data matches the Nautaloss data almost perfectly, which indicates that the dips are not caused by the enclosure. This test also shows that the trapezoidal supra baffle Nautaloss is very similar to an infinite baffle in that the rear chamber act's like a reflection free rear chamber.
Comparison of Nautaloss and OB response:
I then did an impedance sweep of the driver in the Nautaloss and in open air and see that the Nautaloss TL helps to smooth the impedance peak but there is a 700Hz peak corresponding to the dip and high HD, even in free air. I also measured the T/S parameters and the values are within family for the fs and Qts.
From DATS v2:
Based on this info, it appears that the dips and response seen are intrinsic to the driver and not the enclosure or the measurement method.
Here is a comparison of the CHN-70 and TC9FD:
I also used a 670Hz sine wave generator and looked at the RTA to see what the distortion components look like. This is for 2.83v drive with mic at 1m away - most of the distortion is coming from the 2nd order.
Given the very uneven response and its lack of HF extension, I think that the TC9FD is the clear winner. The CHN-70, will have better bass performance in an appropriate enclosure as it has a lower fs and more xmax, as well as a moderate Qts very suitable for MLTL's and BLH's. However, the very high distortion in the midrange and dips and peaks will color the sound from how it was intended to be heard. I really was hoping to find a good driver for use in full range applications here, unfortunately, that was not the finding. I want to make it clear that I really did try to repeat the measurement using other baffles, looked at the free air impedance sweep and the anomaly at 670Hz is there regardless. If anyone else has a measurement of a CHN-70, I would like your feedback to see if what I have is a fluke or defective unit.
The CHN-70, is a paper cone driver with published specs that appear suitable for many vented alignments, especially MLTL's and bass reflex. The $27 price puts it in the same class as the other drivers in this lineup - especially the RS100/P which has a similar Qts, xmax, and sensitivity.
CHN-70 mounted in trapezoidal supra-baffle Nautaloss:
Upon measuring the response in the trapezoidal supra-baffle Nautaloss, I obtained the following response and harmonic distortion plot which was very puzzling: a big dip near 670Hz accompanied by high harmonic distortion there, and several other dips on high with a notable 12dB dip at 7kHz, and a rapid falloff above 12kHz. Initially, I thought my setup might be to blame, but none of the other drivers showed anything like this kind of response. In any event, I had to look deeper to make sure the measurement and enclosure were not the source of the dips and falloff.
CHN-70 mounted in OB:
I thought that the Nautaloss chamber may be the source of this so I then proceeded to put it in a 24in wide x 36in tall open baffle and repeated the measurement. The OB data matches the Nautaloss data almost perfectly, which indicates that the dips are not caused by the enclosure. This test also shows that the trapezoidal supra baffle Nautaloss is very similar to an infinite baffle in that the rear chamber act's like a reflection free rear chamber.
Comparison of Nautaloss and OB response:
I then did an impedance sweep of the driver in the Nautaloss and in open air and see that the Nautaloss TL helps to smooth the impedance peak but there is a 700Hz peak corresponding to the dip and high HD, even in free air. I also measured the T/S parameters and the values are within family for the fs and Qts.
From DATS v2:
Based on this info, it appears that the dips and response seen are intrinsic to the driver and not the enclosure or the measurement method.
Here is a comparison of the CHN-70 and TC9FD:
I also used a 670Hz sine wave generator and looked at the RTA to see what the distortion components look like. This is for 2.83v drive with mic at 1m away - most of the distortion is coming from the 2nd order.
Given the very uneven response and its lack of HF extension, I think that the TC9FD is the clear winner. The CHN-70, will have better bass performance in an appropriate enclosure as it has a lower fs and more xmax, as well as a moderate Qts very suitable for MLTL's and BLH's. However, the very high distortion in the midrange and dips and peaks will color the sound from how it was intended to be heard. I really was hoping to find a good driver for use in full range applications here, unfortunately, that was not the finding. I want to make it clear that I really did try to repeat the measurement using other baffles, looked at the free air impedance sweep and the anomaly at 670Hz is there regardless. If anyone else has a measurement of a CHN-70, I would like your feedback to see if what I have is a fluke or defective unit.
Attachments
Last edited:
Good work.
The impedance peak around 670 Hz is in the published measurements also, so, you are not seeing any anomalous behavior. But that big dip and peak smack in the midrange has got to be disconcerting. And the response up top is quite ragged also. Your 80 db scale is actually making it look better; it is much worse in reality.
I would suggest you move all your posts with the measurements to a new thread and link to individual posts from the first post. Make it a sticky. This is helpful information to all newcomers in the full range forum.
The impedance peak around 670 Hz is in the published measurements also, so, you are not seeing any anomalous behavior. But that big dip and peak smack in the midrange has got to be disconcerting. And the response up top is quite ragged also. Your 80 db scale is actually making it look better; it is much worse in reality.
I would suggest you move all your posts with the measurements to a new thread and link to individual posts from the first post. Make it a sticky. This is helpful information to all newcomers in the full range forum.
thank you for the good work XRK971. Subjectively, how might you characterize and correlate these measurements towards the way we perceive instruments, male and female vocals vs the Vifa? I realize the responses are quite different.
the soft top end should have effect - some listeners might like that for some program.
since the midrange is elevated, could a K-slot shave some of that off and improve off axis response? (it might be difficult to add a helper tweeter to a little K if one wanted a normal top octave)
(ra7 has a good suggestion) - oh - is that little dip a spider artifact? - surround reflection?
the soft top end should have effect - some listeners might like that for some program.
since the midrange is elevated, could a K-slot shave some of that off and improve off axis response? (it might be difficult to add a helper tweeter to a little K if one wanted a normal top octave)
(ra7 has a good suggestion) - oh - is that little dip a spider artifact? - surround reflection?
Last edited:
Here is a fantastic chart for interpreting how too much or too little of a frequency region will be perceived:
Interactive Frequency Chart - Independent Recording Network
Scroll over the chart and the relevant info will show up in the right side boxes.
Interactive Frequency Chart - Independent Recording Network
Scroll over the chart and the relevant info will show up in the right side boxes.
xrk971, thanks sharing great fair documented data this diy hobby.
One gets greedy 😱 look forward to PS95 and if we lucky Godzilla throw in some spare drivers.
One gets greedy 😱 look forward to PS95 and if we lucky Godzilla throw in some spare drivers.
I would love to see the response of the CHR70 in the FH3 box. Not that I could tell necessarily but I wonder if it has a similar dip? I think the HF extension is better than X's graph shows. My ears are not the greatest and I am not for wanting in the highs. I am thinking the CHN70 is similar to the EL70.
Attachments
Cal,
I think the CHN70 is a completely different animal. The inverted rubber surround, the shape of the cone and dustcap may look similar but it is a stamped steel basket and paper cone vs the CHR70's plastic basket and bezel and aluminum cone.
I asked over in the FH thread if anyone over the years has made a measurement of the A7 and FH3 and not a single one. Instead I got all sorts of reasons why measurments are worthless.
Send me your FH3 and I will gladly measure for you 🙂
X
I think the CHN70 is a completely different animal. The inverted rubber surround, the shape of the cone and dustcap may look similar but it is a stamped steel basket and paper cone vs the CHR70's plastic basket and bezel and aluminum cone.
I asked over in the FH thread if anyone over the years has made a measurement of the A7 and FH3 and not a single one. Instead I got all sorts of reasons why measurments are worthless.
Send me your FH3 and I will gladly measure for you 🙂
X
Having a Tang Band bamboo and a Radio Shack FR will round it all out. As well as the Dayton PS95. And maybe even a Visaton B80 in the future.
The Nautaloss is turning out to be a very versatile and accommodating cabinet that "gets out of the way" to allow the driver's character to come through. I have used it for something as big as the PRV 5MR450NDY and it works there too.
I think I need to pair this setup with my XKi sub in a fast XO at 200Hz and listen to music - really get a sense of what the differences are with my ears.
Stretching the truth Mr X? You got reasons why LF measurements were not very useful and potentially misleading. HF measures will be substantially the same as those made elsewhere in different boxes.
dave
Yes that's something even the unique Hypercube had difficult hang on when both had an TC9FD installed.
Last edited:
Awesome measurements! Thanks xrk!
Folks, I will be away on a business trip starting tomorrow and unable to ship drivers until I return sometime after NEXT WEEKEND. I'm sorry! I messaged xrk and will send ASAP when I arrive home.
Otherwise, the Alpair measurement was a disappointment to me. I was considering it as an inexpensive way to hear the Alpair sound. Other sites have reported poor and strange measurements for other Alpair's so I'm really not sure what to conclude. I don't want to purchase drivers that measure this way. I wonder if despite their measurements they sound good much like Lowther's have their followers.
Folks, I will be away on a business trip starting tomorrow and unable to ship drivers until I return sometime after NEXT WEEKEND. I'm sorry! I messaged xrk and will send ASAP when I arrive home.
Otherwise, the Alpair measurement was a disappointment to me. I was considering it as an inexpensive way to hear the Alpair sound. Other sites have reported poor and strange measurements for other Alpair's so I'm really not sure what to conclude. I don't want to purchase drivers that measure this way. I wonder if despite their measurements they sound good much like Lowther's have their followers.
It was a CHN-70, not an Alpair 7 - those are much more expensive and have plastic baskets.
I set up the CHN-70 in a FAST with my XKi dual W5-876SE sub with a 200Hz XO. I recorded sound clips with it raw (no EQ) and with EQ to level the response as best I could. You can listen for yourself and see what you think.
Here is XO & response (with and without EQ). Sound clips correspond to magenta (no EQ) and blue (w/ EQ) curves below :
Here is HD:
Here is IR:
Attachments
Last edited: