Although I would use the Axi2050 with a larger horn here is the fr for this driver attached to TH4001 with a questionable adapter that was delivered with the horn:
To avoid comb filter effect I placed driver/horn on the floor pointing to the ceiling. The mic was about 25cm away from the mouth.
To avoid comb filter effect I placed driver/horn on the floor pointing to the ceiling. The mic was about 25cm away from the mouth.
I bought them from this German distributor: https://www.behringer-electric.de/truextent-beryllium/beryllium-zum-nachrüsten-bex4008-4016/
The webpage has contact details to ask for pricing, so I assume they're still available.
Marco
Nice 4 inch Be diaphragm driver indeed. Thank you for sharing.
Please could you show a photo of your throat adaptor? And please if you could PM me how you got it. Thanks!
The HA14-2 adaptor helps the Radian quite a bit. Here is the measured response of this 3 inch Be diaphragm driver on the same TH4001 at throat mouth. (SPL is arbitrary).
Comparing these curves I note:
The Radian with the composite throat adaptor extends flat to 500 Hz and is down 5 dB at 400 Hz. Better response than I got using the "stubby" throat adaptor but not quite as nice as the results posted by docali where one can see the 18s ND4015Be extend flat to 400 Hz.
The 18s ND4015Be shows a notch at 10 kHz. This is clearly seen also on Vance Dickason's measurements. Would this be the larger diaphragm breaking up? The smaller Radian 745Be is smooth at 10kHz, exhibits a small notch at 15kHz and a more severe one at 20kHz (comparable to the ND4015Be's). For this reason, when crossed at 700 Hz or higher, I continue to like the 3 inch diaphragm Radian in a two way design. I would love to hear the larger drivers though...
This is definitely no break up. These occur around 26kHz. HF measurements are a bit arbitrary from my experience. The mic a little bit at a different position and you get a slightly different fr curve. Also I found that it is more difficult to measure large mouth horns in the near field. Here you can see how the ND4015Be performs in the LTH142 horn:
Just show me another large format cd that performs like this ;-)
Btw, the notch between 500Hz and 600Hz is because of the mouth area of this tractrix like horn is too small. This is the case for most tractrix horns.
Just show me another large format cd that performs like this ;-)
Btw, the notch between 500Hz and 600Hz is because of the mouth area of this tractrix like horn is too small. This is the case for most tractrix horns.
I believe you and appreciate the engineering behind such a driver. Still, trying to understand what the curves show. Your measured curve using the TH4001 and the one (from VD) below using a different horn show a similar dip at 11-12 kHz. Would you attribute this to the driver? the matching between the driver and the horn? Or (as you imply) interference due to microphone placement within a large mouth horn? (I could see this btw)
edit: I just recalled that Vance measured the version of the driver that has a 2 inch throat. Docali I believe your version has the 1.5 inch throat correct? This might account for some measured differences.
edit: I just recalled that Vance measured the version of the driver that has a 2 inch throat. Docali I believe your version has the 1.5 inch throat correct? This might account for some measured differences.
Last edited:
What was done in VC magazine is imo one of the worst scenarios you could do for a driver as it was measured in an 2" XR2064 horn and with an 1.5" to 2" conical adapter which opens up very fast.
To be clear, there is only one version of this driver - it is a 1.5" driver! There is no 2" driver! I bought it together with two adapters: 1.5" to 1.4" and 1.5" to 2". So the version that Vance used must have had the adapter attached which is quite short and conical and this goes into the diffraction horn. No wonder about the strange curves. I have measured the ND4015Be in several different horns and it always performs way better compared to the VC measurements. I would even call it one of the very best large format drivers ever made.
I wonder that 18sound accepted this article and I would have never bought this driver without receiving different measurements from the German distributor in a real setup and another horn.
To be clear, there is only one version of this driver - it is a 1.5" driver! There is no 2" driver! I bought it together with two adapters: 1.5" to 1.4" and 1.5" to 2". So the version that Vance used must have had the adapter attached which is quite short and conical and this goes into the diffraction horn. No wonder about the strange curves. I have measured the ND4015Be in several different horns and it always performs way better compared to the VC measurements. I would even call it one of the very best large format drivers ever made.
I wonder that 18sound accepted this article and I would have never bought this driver without receiving different measurements from the German distributor in a real setup and another horn.
Filtering and EQ for TH4001
Caveat: Output from the Radian driver is limited above 15 kHz. I personally find that a response covering 30 Hz to 15 kHz satisfies me, but this is personal taste.
Shaping the response of the Radian 745Be + HA14-2 + TH4001 horn combination in a two-way design is straightforward. The efficiency of the Radian driver in the 10 to 15 kHz range is sufficiently high to match that of a good woofer, allowing the use of passive EQ. This can be done many ways (the TAD filter is out there somewhere), but as a starting point below is an example circuit that requires only 5 components. This is for the 16 ohm version of the diaphragm-coil.
In series:
Using the REW alignment tool, I could achieved a smooth transition at 700 Hz by reversing the phase on one of the drivers and positioning the horn mouth 3 inches (75 mm) in front of the baffle.
Below shows the response measured at 20 inch in front of the speaker, mid-height between woofer and horn motors, after time windowing to reduce reflections. The black curve is horn+woofer, the dashed curve the other speaker (to show matching) and the blue curve the CD-horn alone. The depression above 1 kHz is, I believe, an artifact of measuring close and below the axis of the horn. The response can be tilted, moved up and down by adjusting Ra, Req and C.
Caveat: Output from the Radian driver is limited above 15 kHz. I personally find that a response covering 30 Hz to 15 kHz satisfies me, but this is personal taste.
Shaping the response of the Radian 745Be + HA14-2 + TH4001 horn combination in a two-way design is straightforward. The efficiency of the Radian driver in the 10 to 15 kHz range is sufficiently high to match that of a good woofer, allowing the use of passive EQ. This can be done many ways (the TAD filter is out there somewhere), but as a starting point below is an example circuit that requires only 5 components. This is for the 16 ohm version of the diaphragm-coil.
In series:
- An attenuation resistor on the amplifier side, Ra = 4.7 ohm for my woofer-enclosure. Adjust to taste.
- An RC EQ filter, Req = 50 ohm and Ceq = 0.77 uF in parallel. This attentuates the midrange below 10kHz or thereabouts for flatness.
- Filter Capacitor on the driver side, C = 1.2 uF.
- Filter Inductor L = 10.9 mH.
Using the REW alignment tool, I could achieved a smooth transition at 700 Hz by reversing the phase on one of the drivers and positioning the horn mouth 3 inches (75 mm) in front of the baffle.
Below shows the response measured at 20 inch in front of the speaker, mid-height between woofer and horn motors, after time windowing to reduce reflections. The black curve is horn+woofer, the dashed curve the other speaker (to show matching) and the blue curve the CD-horn alone. The depression above 1 kHz is, I believe, an artifact of measuring close and below the axis of the horn. The response can be tilted, moved up and down by adjusting Ra, Req and C.
Last edited:
@AllenB
To be honest, I'm a little tired of responding to your postings, which don't really have a direct relation to my statements and whose overall relation doesn't fit either in the context of this thread.
First, do you want to teach us that the example I presented with the driver impedance free-air with significant peaks and with horn loading almost flat does not have any effect on the driver/horn performance neither for controlled diaphragm motion nor for spl or frequency response?
Second, if you have a driver attached to a waveguide-only device (less LF loading) with a certain spl especially in the lower frequency region and the same driver attached to a good loading horn which gives much higher spl in the LF region and both are normalized to the same level of spl then the second loaded example has much lower excursion. This is the key point: you have to look at normalized results.
To be honest, I'm a little tired of responding to your postings, which don't really have a direct relation to my statements and whose overall relation doesn't fit either in the context of this thread.
First, do you want to teach us that the example I presented with the driver impedance free-air with significant peaks and with horn loading almost flat does not have any effect on the driver/horn performance neither for controlled diaphragm motion nor for spl or frequency response?
Second, if you have a driver attached to a waveguide-only device (less LF loading) with a certain spl especially in the lower frequency region and the same driver attached to a good loading horn which gives much higher spl in the LF region and both are normalized to the same level of spl then the second loaded example has much lower excursion. This is the key point: you have to look at normalized results.
Yes this may have been what you meant and it's good, but it's a completely different thing and the amounts may be unrelated. Others in the thread may want to know. I can see you sometimes don't need to drive the horn as hard so you can turn down the signal, but the horn doesn't load down the driver in doing so, it still moves as freely.
It also doesn't help in this area if the narrow horn has lost directivity by being narrow.
It also doesn't help in this area if the narrow horn has lost directivity by being narrow.
I had planned to experiment with TH4001 and a 300 Hz JMLC horn. Waiting for the athosaudio TH4001 at the moment. The approach and drivers will be a little different but I am curious about your subjective impressions on the two horns types. If you had to pick one which one would it be?
The last time similar topics were discussed people ended up on the sin bin:
https://www.diyaudio.com/community/threads/compression-drivers-in-undersized-horns.376472/
I am out here in this thread...
I welcome technical comments and find that they are more useful if participants elaborate and explain "how it works" so that non experts (like me) can understand.
AllenB - Your statement that the horn or throat adaptor does not load the driver is intriguing. (Or am I misquoting you?) I can visualize that most of the compression (damping) may come from the chamber side of the diaphragm. If this is the case, then I am genuinely curious to learn what causes the changes in driver impedance response that we observe in our measurements? Nothing else has changed but resistance of air movement on the exit side.
Diaphragm excursions is not a concern to me, as I don't even dare to measure distortion at the level used by Vance Dickason (too loud in my room).
Flatness of response and maintaining low distortion are more important to me as I prefer passive filtering and aim for accuracy.
AllenB - Your statement that the horn or throat adaptor does not load the driver is intriguing. (Or am I misquoting you?) I can visualize that most of the compression (damping) may come from the chamber side of the diaphragm. If this is the case, then I am genuinely curious to learn what causes the changes in driver impedance response that we observe in our measurements? Nothing else has changed but resistance of air movement on the exit side.
Diaphragm excursions is not a concern to me, as I don't even dare to measure distortion at the level used by Vance Dickason (too loud in my room).
Flatness of response and maintaining low distortion are more important to me as I prefer passive filtering and aim for accuracy.
Thank you for asking. It is too easy to tell (novelty bias), however I will be happy to share subjective impressions in due time. For instance I like to rotate electronics and sources to get "a feeling" and not just a left brain analysis. This takes time.
For now, I can say that I enjoy both horns (JMLC and TH4001) in a medium size room and find that they achieve correct tonality in their sweet spot when properly integrated in a two way speaker. They throw a deep soundstage and let one hear nuances deep down into the dynamic range. Yet there are many differences, for instance in what I would call "event presentation", or "imaging" if you prefer. I will be happy to explore and describe these in time. The room is a big factor here, which will not surprise any of you.
Incidentally, the three-way configuration that I briefly tried using the stubby throat adaptor and small compression driver tweeter (EV T350) on top of the TH4001 sounded uneven: ok on voices, good "air", but tiring on plucked guitar and percussions. Not my cup of tea. I've tried integrating a ribbon without success before. Perhaps I need to vertically align the motors of the two compression drivers? This should be more practical to try on the TH4001 than on the JMLC (because the tweeter would end up physically very high). Before any of this I intend to spend quality time with the two way.
Last edited:
There are few open questions I might be able to close.
The throat adapter needs to match the flare rate of the horn to maintain loading and reduce any impedance change that could cause a reflection. Alot of BEM simulations were done to see how sensitive particular horns were to the adapters.
Loading does change the behaviour of the driver as shown in @docali graphs and for a good reason. I've attached an explanation from the Kolbrek-Dunker horn book High Quality Horn Loudspeaker Systems . Its a fantastic book and covers theory behind the horns discussed in this thread. The theory agrees with the BEM simulations and the measurements of prototypes.
The throat adapter needs to match the flare rate of the horn to maintain loading and reduce any impedance change that could cause a reflection. Alot of BEM simulations were done to see how sensitive particular horns were to the adapters.
Loading does change the behaviour of the driver as shown in @docali graphs and for a good reason. I've attached an explanation from the Kolbrek-Dunker horn book High Quality Horn Loudspeaker Systems . Its a fantastic book and covers theory behind the horns discussed in this thread. The theory agrees with the BEM simulations and the measurements of prototypes.
Attachments
This is the kind of explanation I was hoping to find. Thank you so much for sharing!
A couple days ago I looked into ordering the Kolbrek-Dunker horn book as it appears to be well reviewed (your support increases my confidence further) and I could not find another textbook on the topic! Alas shipping from the UK to Canada is expensive, so I am standing by to order it from the US when they receive a new shipment.
Also, I have enquired about having a custom throat adaptor made that precisely matches the exit angle of the Radian driver (4 degrees). If anybody knows suppliers or someone willing to make this, please PM me.
A couple days ago I looked into ordering the Kolbrek-Dunker horn book as it appears to be well reviewed (your support increases my confidence further) and I could not find another textbook on the topic! Alas shipping from the UK to Canada is expensive, so I am standing by to order it from the US when they receive a new shipment.
Also, I have enquired about having a custom throat adaptor made that precisely matches the exit angle of the Radian driver (4 degrees). If anybody knows suppliers or someone willing to make this, please PM me.