Hi Doug,
Very interesting indeed! Judging by that small dip between 7k and 8k, he likely used a single parallel notch to smooth out the entire response between 4k and 16k. I'm just guessing though, and it would be neat to see his implementation.
The good news is that the dip is in the exact same spot with my ribbons, which means there's reasonable consistency.
I'll try to model up a version with a single parallel notch, and see how close I can get it. I would be very interested in seeing a raw measurement of your NeoPros sans-crossover to see how they compare to mine.
Any chance of drawing up the crossover?
Thanks for the info!
Cheers,
Owen
Very interesting indeed! Judging by that small dip between 7k and 8k, he likely used a single parallel notch to smooth out the entire response between 4k and 16k. I'm just guessing though, and it would be neat to see his implementation.
The good news is that the dip is in the exact same spot with my ribbons, which means there's reasonable consistency.
I'll try to model up a version with a single parallel notch, and see how close I can get it. I would be very interested in seeing a raw measurement of your NeoPros sans-crossover to see how they compare to mine.
Any chance of drawing up the crossover?
Thanks for the info!
Cheers,
Owen
I could disconnect the mid-range and run measurements on the Neo5 this week for you. It will be in room though with zero system EQing.
I do not have the schematics, Its Curt's (from HtGuide) work he did custom for me so I do not think its right for me to post his work without his permission.
I do not have the schematics, Its Curt's (from HtGuide) work he did custom for me so I do not think its right for me to post his work without his permission.
Owen, make sure to take some off axis measurements as well when you're doing the tweeter xover. Some of the bumps you measure up high in the tweeter are quite prevalent right on axis but as you get off axis a little more they change. I believe due to the faceplate on the tweeter. You will want to take a look at the power response before you determine if you need to completely flatten the on axis response.
John
John
Hey Owen..did you ever give any consideration to adding a second inductor between the woofers for a .5 on the lower woofer around 400hz? Looks great BTW.
Hi Guys,
I ran across some measurements in my workspace directory the other day and it reminded me that I never posted them up, and never really concluded this project the way I should have.
As you can probably guess, I’ve abandoned this project. More importantly though, are the reasons why and the lessons learned. I’ll go over those here, and highlight them with real measurements taken in an anechoic chamber.
Overall, these drivers in this layout simply could not fulfill the initial requirements that I set out to achieve. In fact, doing an MTM this size is a very bad idea, to put it bluntly. Here’s what went wrong:
1. Unacceptable vertical response caused by the MTM arrangement of the TD10M drivers.
2. Lack of LF response from the TD10M drivers in a critically damped sealed enclosure.
3. An enclosure shape that looked interesting, but wasn’t practical
4. Unrealistic expectations for the TD10M
I’ll elaborate on the above:
1. The second post in this thread was a critical one that I chose to ignore, and shouldn’t have. The drivers ended up being spaced about 18” apart C-C to make room for the NeoPro5i in between. That 18” spacing corresponds to a wavelength of 750Hz which means the first serious null would be around 1500Hz. If you take a look at the vertical response of the two TD10M drivers in their MTM arrangement, you’ll see the severity of the problem. At just 15 degrees off axis, there’s a 20dB dip at 1500Hz which is completely unacceptable. Things only get worse from there, with a plethora of peaks and nulls as you move farther off axis.
2. It turns out that the “M” in TD10M should have been a bigger indicator than I thought. I’m guessing M stands for midrange, and rightfully so. I was hoping that with a reference efficiency of 94dB/2.83V/1m they would be good for 100dB if used as a pair. I was correct, but only above about 750Hz. If you look at the horizontal response graphs, the drivers do indeed sit right at 100dB from about 750Hz to 2kHz. By the time you get down to 100Hz though, you’re almost 12dB down. By 40Hz, you’re just over 20dB down, and that’s far too much to be used as a full range loudspeaker. Although I would only need 1W to produce 100dB above 750Hz, I would need 128W to produce the same level at 40Hz. It also means that active EQ would be required.
3. My idea of a peanut shaped enclosure seemed like a great idea when I drew it up in my CAD program. It had the benefit of minimizing diffraction for the tweeter and I thought it looked pretty unique. In the end, it was exceedingly difficult to build, extremely heavy, and couldn’t stand up on its own. It required a special stand to hold it up, and I was terrified that it would fall over. At 110lbs, it would have destroyed anything within 5 feet if it did fall. The only way to finish an enclosure that shape is to paint it, and that’s not easy with MDF, especially when the whole exposed surface is the end grain of the MDF. This was the most heartbreaking part of the whole project for me since it required an enormous amount of effort to make, and ended up being worthless in the end. A strong reminder that a making a mockup and measuring it first is well worth the effort.
4. This last one I’m a little torn on. The TD10M isn’t a bad driver by any means, in fact, a quick glance at the THD measurements show that it has extremely low harmonic distortion, and it does indeed have the rated 94dB sensitivity, but only from about 750Hz and up. The response is very smooth and well controlled right up to 2kHz, and although off axis response isn’t the best at 2kHz, at least it’s uniform and doesn’t change shape as you move off axis. I know that if the guys at AE speakers had a proper frequency response graph available, I would have taken one look and realized that the TD10M wasn’t the right driver for the job. The fact that they didn’t have one posted meant that I basically bought $1000 worth of drivers on the hope that they would do what I wanted. In the end, it was my fault for buying a driver without all the information I needed to make an informed decision. I do think this is a valuable lesson that everyone here should very much take to heart: If it doesn’t have a frequency response graph, DON’T BUY IT!
On the positive side, the NeoPro5i turned out to be very capable driver, and I’m currently looking into ways that I can make better use of it in the future. It genuinely does live up to most of the claims made on the datasheet, and is at a level of performance that most traditional dynamic loudspeakers simply cannot match.
I have two full sets of measurements for the NeoPro, one of which I will attach here, and the other I will attach in the next post.
Just before measuring the FR of the NeoPro in the anechoic chamber, we ran a quick impedance sweep on the ribbon which was accidently run from 20Hz up instead of 1000Hz up. This was run without a protection cap on the tweeter and it damaged the ribbon element (slightly stretched it). We ran the FR measurements anyhow, and the result is slightly lower sensitivity, a peak in the low end around 1kHz, and a rolloff in the high end. A few months ago I sent the tweeters in to Madisound to have new ribbons installed, and have since re-measured them. I’ll attach the first set of measurements here with the stretched ribbon, and the second set with the new ribbons after I got them back from Madisound.
Please keep the above in mind while perusing the tweeter response graphs.
In the end, I did learn a great deal from all the work I did here. There are quite a few things I’ll make sure to do differently in the future, the most important of which is more front end effort to ensure the project will work before building anything!
Cheers,
Owen
I ran across some measurements in my workspace directory the other day and it reminded me that I never posted them up, and never really concluded this project the way I should have.
As you can probably guess, I’ve abandoned this project. More importantly though, are the reasons why and the lessons learned. I’ll go over those here, and highlight them with real measurements taken in an anechoic chamber.
Overall, these drivers in this layout simply could not fulfill the initial requirements that I set out to achieve. In fact, doing an MTM this size is a very bad idea, to put it bluntly. Here’s what went wrong:
1. Unacceptable vertical response caused by the MTM arrangement of the TD10M drivers.
2. Lack of LF response from the TD10M drivers in a critically damped sealed enclosure.
3. An enclosure shape that looked interesting, but wasn’t practical
4. Unrealistic expectations for the TD10M
I’ll elaborate on the above:
1. The second post in this thread was a critical one that I chose to ignore, and shouldn’t have. The drivers ended up being spaced about 18” apart C-C to make room for the NeoPro5i in between. That 18” spacing corresponds to a wavelength of 750Hz which means the first serious null would be around 1500Hz. If you take a look at the vertical response of the two TD10M drivers in their MTM arrangement, you’ll see the severity of the problem. At just 15 degrees off axis, there’s a 20dB dip at 1500Hz which is completely unacceptable. Things only get worse from there, with a plethora of peaks and nulls as you move farther off axis.
2. It turns out that the “M” in TD10M should have been a bigger indicator than I thought. I’m guessing M stands for midrange, and rightfully so. I was hoping that with a reference efficiency of 94dB/2.83V/1m they would be good for 100dB if used as a pair. I was correct, but only above about 750Hz. If you look at the horizontal response graphs, the drivers do indeed sit right at 100dB from about 750Hz to 2kHz. By the time you get down to 100Hz though, you’re almost 12dB down. By 40Hz, you’re just over 20dB down, and that’s far too much to be used as a full range loudspeaker. Although I would only need 1W to produce 100dB above 750Hz, I would need 128W to produce the same level at 40Hz. It also means that active EQ would be required.
3. My idea of a peanut shaped enclosure seemed like a great idea when I drew it up in my CAD program. It had the benefit of minimizing diffraction for the tweeter and I thought it looked pretty unique. In the end, it was exceedingly difficult to build, extremely heavy, and couldn’t stand up on its own. It required a special stand to hold it up, and I was terrified that it would fall over. At 110lbs, it would have destroyed anything within 5 feet if it did fall. The only way to finish an enclosure that shape is to paint it, and that’s not easy with MDF, especially when the whole exposed surface is the end grain of the MDF. This was the most heartbreaking part of the whole project for me since it required an enormous amount of effort to make, and ended up being worthless in the end. A strong reminder that a making a mockup and measuring it first is well worth the effort.
4. This last one I’m a little torn on. The TD10M isn’t a bad driver by any means, in fact, a quick glance at the THD measurements show that it has extremely low harmonic distortion, and it does indeed have the rated 94dB sensitivity, but only from about 750Hz and up. The response is very smooth and well controlled right up to 2kHz, and although off axis response isn’t the best at 2kHz, at least it’s uniform and doesn’t change shape as you move off axis. I know that if the guys at AE speakers had a proper frequency response graph available, I would have taken one look and realized that the TD10M wasn’t the right driver for the job. The fact that they didn’t have one posted meant that I basically bought $1000 worth of drivers on the hope that they would do what I wanted. In the end, it was my fault for buying a driver without all the information I needed to make an informed decision. I do think this is a valuable lesson that everyone here should very much take to heart: If it doesn’t have a frequency response graph, DON’T BUY IT!
On the positive side, the NeoPro5i turned out to be very capable driver, and I’m currently looking into ways that I can make better use of it in the future. It genuinely does live up to most of the claims made on the datasheet, and is at a level of performance that most traditional dynamic loudspeakers simply cannot match.
I have two full sets of measurements for the NeoPro, one of which I will attach here, and the other I will attach in the next post.
Just before measuring the FR of the NeoPro in the anechoic chamber, we ran a quick impedance sweep on the ribbon which was accidently run from 20Hz up instead of 1000Hz up. This was run without a protection cap on the tweeter and it damaged the ribbon element (slightly stretched it). We ran the FR measurements anyhow, and the result is slightly lower sensitivity, a peak in the low end around 1kHz, and a rolloff in the high end. A few months ago I sent the tweeters in to Madisound to have new ribbons installed, and have since re-measured them. I’ll attach the first set of measurements here with the stretched ribbon, and the second set with the new ribbons after I got them back from Madisound.
Please keep the above in mind while perusing the tweeter response graphs.
In the end, I did learn a great deal from all the work I did here. There are quite a few things I’ll make sure to do differently in the future, the most important of which is more front end effort to ensure the project will work before building anything!
Cheers,
Owen
Attachments
As mentioned above, I recently re-measured the two NeoPro5i drivers after I got them back from Madisound with new ribbon elements installed.
I really need to applaud Madisound for a job well done on the ribbon replacement. I was very concerned that I would get them back and they would be poorly matched, but the results are quite incredible. Both tweeters came back matched to better than 1dB across the entire spectrum. It's exceedingly rare that a pair of drivers would ever match up this well, so my hat goes off to them! It was well worth the $90 cost to have them both redone.
I've attached several measurements I made in a medium sized anechoic chamber (good down to about 150Hz) using a calibrated 1/4" B&K microphone and an Audio Precision measurement system. The response is accurate out to 48kHz which is the limit on the AP. Measurement distance is exactly 1m on axis, and input was 2.83V. The tweeter was mounted on foam blocks and had no actual baffle.
The first graph is an unsmoothed response showing the matching between the two tweeters. I believe the ripple is either caused by the magnet structure around the ribbon, or the fact that the tweeter was not flush mounted and had no baffle.
The second graph shows the 1/6th octave smoothed response under the same conditions.
The third graph shows a close-up of the tweeter response with 1dB increments.
All the rest of the graphs show distortion referenced to 0dB (distortion ratio in dB). There's the sum of all harmonics, and then the 1st through the 7th harmonics. All are measured at 90dB.
This is a very good tweeter, and almost more impressively, the datasheet is actually accurate! You get output well beyond 40kHz which is pretty incredible, and overall the response is smooth and easily managed with either passive or active filters. This tweeter has incredible sensitivity, and the distortion performance is superb for even order and fairly good for odd order harmonics.
I'm currently considering pairing it with a single RCF MR10N301 midrange in a TM configuration. That would give you a true 99dB/2.83V/1m from about 300Hz out to 40kHz if crossed at around 1500Hz. Covering the range below that with any level of efficiency is still a bit of a mystery to me, but I'm open to suggestions.
Cheers,
Owen
I really need to applaud Madisound for a job well done on the ribbon replacement. I was very concerned that I would get them back and they would be poorly matched, but the results are quite incredible. Both tweeters came back matched to better than 1dB across the entire spectrum. It's exceedingly rare that a pair of drivers would ever match up this well, so my hat goes off to them! It was well worth the $90 cost to have them both redone.
I've attached several measurements I made in a medium sized anechoic chamber (good down to about 150Hz) using a calibrated 1/4" B&K microphone and an Audio Precision measurement system. The response is accurate out to 48kHz which is the limit on the AP. Measurement distance is exactly 1m on axis, and input was 2.83V. The tweeter was mounted on foam blocks and had no actual baffle.
The first graph is an unsmoothed response showing the matching between the two tweeters. I believe the ripple is either caused by the magnet structure around the ribbon, or the fact that the tweeter was not flush mounted and had no baffle.
The second graph shows the 1/6th octave smoothed response under the same conditions.
The third graph shows a close-up of the tweeter response with 1dB increments.
All the rest of the graphs show distortion referenced to 0dB (distortion ratio in dB). There's the sum of all harmonics, and then the 1st through the 7th harmonics. All are measured at 90dB.
This is a very good tweeter, and almost more impressively, the datasheet is actually accurate! You get output well beyond 40kHz which is pretty incredible, and overall the response is smooth and easily managed with either passive or active filters. This tweeter has incredible sensitivity, and the distortion performance is superb for even order and fairly good for odd order harmonics.
I'm currently considering pairing it with a single RCF MR10N301 midrange in a TM configuration. That would give you a true 99dB/2.83V/1m from about 300Hz out to 40kHz if crossed at around 1500Hz. Covering the range below that with any level of efficiency is still a bit of a mystery to me, but I'm open to suggestions.
Cheers,
Owen
Attachments
-
Level 1 zoom.png -
FR Zoom Out.png -
Level 1.png -
Sum of all harmonics 90db.png -
Distortion Product Ratio 2nd harmonic 90db.png -
Distortion Product Ratio 3rd harmonic 90db.png -
Distortion Product Ratio 4th harmonic 90db.png -
Distortion Product Ratio 5th harmonic 90db.png -
Distortion Product Ratio 6th harmonic 90db.png -
Distortion Product Ratio 7th harmonic 90db.png
Hello Owen,
Wanted to comment on a few things based on your recent comments here. Again thanks for taking on a project like this. Maybe I should have given you a little more direct help on this towards the beginning since it now seems that you goals were not very realistic with the drivers and enclosure type. With the original measurements and mockups you had done, I expected you were aware of the abilities of the drivers in this kind of application though.
This of course is based up on the center to center distance of the drivers. With a ribbon between you are going to have about 18-19" center to center distance which will give you problems if you aren't expecting to work within the vertical limits. The listening height limits are basically within the height of the driver centers at about 10ft away. Getting outside that point you will start to have comb filtering effects. The good thing is that the ribbon only has 5 degree vertical off axis anyway before the upper end essentially disappears. In a similar project I did in a vented cabinet with the same drivers, at about 10-15ft away I had a vertical height to work with of about 22". This was almost perfect for what I wanted. This lack of vertical response outside this window also greatly reduces the effects of floor/ceiling reflections. It's all about defining the goals. If you need more vertical response, neither the ribbons nor an MTM with big drivers is a good option. These were issues addressed in your first couple posts though.
Again this is about the goals of the system and something to be expected. In an enclosure like this the F3 is around 150hz and by 50hz the response is about 12dB down. For this reason mine used a vented enclosure about 1.5 cubic foot for the pair of drivers. Tuning to 55hz gave an F3 of 58hz and only about -1.5dB at the crossover to the subs. Still the baffle effects had to be accounted for electronically.
That was quite the project to take on without a CNC. Even in enclosures like that where I have cut all the pieces on the CNC there is still an enormous amount of work to do to sand and finish the enclosure. I give you a lot of credit for taking something like that on with standard tools.
Again as mentioned above, the tweeter has only a 5 degree vertical pattern before the top end greatly disappears so I wouldn't worry about the response of the woofers too much outside that +/-5degree window either. The goal of an MTM like this with a ribbon is specifically to get a very tight vertical pattern. We use it in recording studios where there is little ability for sound treatment to ceilings or floors and where the listening height doesn't change. The nulls simply reduce the amount of sound reflecting of the celing and floor.
Yes, the M versions are for midbass/midrange. They have significantly less excursion than the other drivers so they can't move as much air down low, but also play much higher cleanly. What you are describing are the effects of the enclosure volume on the response and the effects of the baffle step. The calculated efficiency is a value based upon T/S parameters. It will also correspond to the efficiency quite closely everywhere above where the baffle effects are no longer an issue.
If your baffle is about 12-13" wide the baffle step is going to start right around 1KHz and drop over the next octave. This is what you see in your response curves as well and is the reason for being 12dB down at 100hz or 20dB down at 40hz. If you put the TD10M's on an infinitely large baffle, you'll get rid of the baffle step and see only the rolloff of the enclosure itself. The baffle rolloff is not specific to the woofer used as the baffle step is based on the width of the baffle alone. It will apply identically to any woofer used on that baffle size. The EDGE baffle simulator can work well to give you a good idea of this prior to starting a project. As far as going to 40hz, that is a little low for a pair of 10" drivers with the 6mm Xmax in a sealed enclosure to begin with. Basic models will show you that you are limited to just over 105dB at 40hz based on the displacement of a pair of drivers. It will also take around 300w to the pair of drivers to get to that level. A vented enclosure as shown above can get much more output to a point. The vented enclosure tuned around 55hz has over 7dB more output at 60hz with the same input power. Going as low as 40hz requires much more volume though so likely still isn't practical for the pair of 10" drivers.
Nearly any time you use a sealed enclosure and want to get low extension, EQ will be required. There is just no other way to account for the rolloff of the enclosure and the rolloff from the baffle itself. Luckily units like the minidsp now are just over $100 and give some very good functionality for this.
I've been down this road too. MDF is just a pain to work with when you have layers. No matter how well you seal it, the seams come back. The only good method i found was to actually lay a layer of fiberglass matting over the surface, then do the bodywork over the top of that. All the mdf "sealers" in the world that I have tried still couldnt' fully seal the mdf to keep seams from coming back with change in humidity/temperature.
The things that you mention first are the reasons to choose the TD10M. The low distortion, consistent off axis response, etc. All the other effects are not effects of the driver, but of the enclosure chosen.
Again, there is nothing about the driver that is leading to the drop in response below 750hz. That is baffle step and enclosure rolloff. Again you measured slightly more low end with the TD10M than the B&C driver, so you would have even less bass that way. No driver will compensate for that, and the effects will be the same either way. The driver Qts of .237 tells you that it is not well suited for a sealed enclosure without EQ. A basic box model of a Qtc .5 enclosure showing 150hz F3 tells you that the response is significantly down before getting to 40hz. This is fine if you plan to use EQ and is typically what is done in many sealed subwoofers today even. Adding EQ to the enclosure should easily get the low end you want as long as you can live with the output limits based on the displacement of the drivers.
Now, had I spent the time to do response measurements in a large infinite baffle, they would not show the baffle response or enclosure rolloff. You would see a response curve that would look very similar to the nearfield response you showed right away. You would still need to model the baffle and enclosure to find those effects and then adjust the measured curve to find those. A simple enclosure model though tells you those exact same things and the measured curves just verify the models.
Best Regards,
John
Wanted to comment on a few things based on your recent comments here. Again thanks for taking on a project like this. Maybe I should have given you a little more direct help on this towards the beginning since it now seems that you goals were not very realistic with the drivers and enclosure type. With the original measurements and mockups you had done, I expected you were aware of the abilities of the drivers in this kind of application though.
This of course is based up on the center to center distance of the drivers. With a ribbon between you are going to have about 18-19" center to center distance which will give you problems if you aren't expecting to work within the vertical limits. The listening height limits are basically within the height of the driver centers at about 10ft away. Getting outside that point you will start to have comb filtering effects. The good thing is that the ribbon only has 5 degree vertical off axis anyway before the upper end essentially disappears. In a similar project I did in a vented cabinet with the same drivers, at about 10-15ft away I had a vertical height to work with of about 22". This was almost perfect for what I wanted. This lack of vertical response outside this window also greatly reduces the effects of floor/ceiling reflections. It's all about defining the goals. If you need more vertical response, neither the ribbons nor an MTM with big drivers is a good option. These were issues addressed in your first couple posts though.
Again this is about the goals of the system and something to be expected. In an enclosure like this the F3 is around 150hz and by 50hz the response is about 12dB down. For this reason mine used a vented enclosure about 1.5 cubic foot for the pair of drivers. Tuning to 55hz gave an F3 of 58hz and only about -1.5dB at the crossover to the subs. Still the baffle effects had to be accounted for electronically.
That was quite the project to take on without a CNC. Even in enclosures like that where I have cut all the pieces on the CNC there is still an enormous amount of work to do to sand and finish the enclosure. I give you a lot of credit for taking something like that on with standard tools.
Again as mentioned above, the tweeter has only a 5 degree vertical pattern before the top end greatly disappears so I wouldn't worry about the response of the woofers too much outside that +/-5degree window either. The goal of an MTM like this with a ribbon is specifically to get a very tight vertical pattern. We use it in recording studios where there is little ability for sound treatment to ceilings or floors and where the listening height doesn't change. The nulls simply reduce the amount of sound reflecting of the celing and floor.
Yes, the M versions are for midbass/midrange. They have significantly less excursion than the other drivers so they can't move as much air down low, but also play much higher cleanly. What you are describing are the effects of the enclosure volume on the response and the effects of the baffle step. The calculated efficiency is a value based upon T/S parameters. It will also correspond to the efficiency quite closely everywhere above where the baffle effects are no longer an issue.
If your baffle is about 12-13" wide the baffle step is going to start right around 1KHz and drop over the next octave. This is what you see in your response curves as well and is the reason for being 12dB down at 100hz or 20dB down at 40hz. If you put the TD10M's on an infinitely large baffle, you'll get rid of the baffle step and see only the rolloff of the enclosure itself. The baffle rolloff is not specific to the woofer used as the baffle step is based on the width of the baffle alone. It will apply identically to any woofer used on that baffle size. The EDGE baffle simulator can work well to give you a good idea of this prior to starting a project. As far as going to 40hz, that is a little low for a pair of 10" drivers with the 6mm Xmax in a sealed enclosure to begin with. Basic models will show you that you are limited to just over 105dB at 40hz based on the displacement of a pair of drivers. It will also take around 300w to the pair of drivers to get to that level. A vented enclosure as shown above can get much more output to a point. The vented enclosure tuned around 55hz has over 7dB more output at 60hz with the same input power. Going as low as 40hz requires much more volume though so likely still isn't practical for the pair of 10" drivers.
Nearly any time you use a sealed enclosure and want to get low extension, EQ will be required. There is just no other way to account for the rolloff of the enclosure and the rolloff from the baffle itself. Luckily units like the minidsp now are just over $100 and give some very good functionality for this.
I've been down this road too. MDF is just a pain to work with when you have layers. No matter how well you seal it, the seams come back. The only good method i found was to actually lay a layer of fiberglass matting over the surface, then do the bodywork over the top of that. All the mdf "sealers" in the world that I have tried still couldnt' fully seal the mdf to keep seams from coming back with change in humidity/temperature.
The things that you mention first are the reasons to choose the TD10M. The low distortion, consistent off axis response, etc. All the other effects are not effects of the driver, but of the enclosure chosen.
Again, there is nothing about the driver that is leading to the drop in response below 750hz. That is baffle step and enclosure rolloff. Again you measured slightly more low end with the TD10M than the B&C driver, so you would have even less bass that way. No driver will compensate for that, and the effects will be the same either way. The driver Qts of .237 tells you that it is not well suited for a sealed enclosure without EQ. A basic box model of a Qtc .5 enclosure showing 150hz F3 tells you that the response is significantly down before getting to 40hz. This is fine if you plan to use EQ and is typically what is done in many sealed subwoofers today even. Adding EQ to the enclosure should easily get the low end you want as long as you can live with the output limits based on the displacement of the drivers.
Now, had I spent the time to do response measurements in a large infinite baffle, they would not show the baffle response or enclosure rolloff. You would see a response curve that would look very similar to the nearfield response you showed right away. You would still need to model the baffle and enclosure to find those effects and then adjust the measured curve to find those. A simple enclosure model though tells you those exact same things and the measured curves just verify the models.
Best Regards,
John
How would a TD10S help the bottom end in this enclosure while keeping one TD10M?
I know two would be better to have a better match with the NeoPro but then it'd depend how high the TD10S could cross. Maybe the tweeter could be pad down to have a more flat response while having a lower F3.
It's always the case that is hard to find mids to keep up with high sensitive tweeters. There was talks early on of having subwoofers, was that ever completed? If so, then it's more the issue to tackle the baffle step since the subs would take over on the lower octaves. Then again, a single TD10S is not going to solve this alone. Maybe add a couple of layers (take the extra pain) to have more volume and take advantage of it?
Salvador
I know two would be better to have a better match with the NeoPro but then it'd depend how high the TD10S could cross. Maybe the tweeter could be pad down to have a more flat response while having a lower F3.
It's always the case that is hard to find mids to keep up with high sensitive tweeters. There was talks early on of having subwoofers, was that ever completed? If so, then it's more the issue to tackle the baffle step since the subs would take over on the lower octaves. Then again, a single TD10S is not going to solve this alone. Maybe add a couple of layers (take the extra pain) to have more volume and take advantage of it?
Salvador
- Status
- Not open for further replies.