The best tweeter I've tested yet...

[Reet]

...came from AliExpress. "Best" is a subjective of course, as I have others that do better <2kHz. But >2kHz this is of the lowest 3rd harmonic I've seen, and the lowest over a wide frequency range of anything on my shelf. Tested at 2.8V, I did test again at +5dB (~5V), 2nd order went up ever so slightly, 3rd harmonic didn't budge. Of course, it's not made of magic, there are other tweeters that reach this level of performance or better, but at multiples of the price of this one.



Filtered for 2kHz LR4 target, 3rd harmonic is all but non-existent.

[duvixan]

Which tweeter is this, please?

[Reet]

Only 1 person wants to know?

It's Melodavid Be28. Is it really Be? I don't think so, but I don't know for certain. I think its Al or Ti with a nice grey matte ceramic coating. Dome resonance is 26-27kHz, a bit low for Be, comparable with many Al tweeters. But, given the performance, I don't think I care about that fact. It looks nice inside and out.



Behind the face plate is a nice flat surface, great tweeter for some waveguide experiments. The dome however doesn't really look very Be, but rather a grey ceramic matte type of coating. I would not be surprised if it's Al or Ti underneath.


The dome assembly also includes a couple alignment pins, making disassembly and reassembly easy without messing up the coil alignment. Underneath there's a nice looking motor and voice coil assembly. Maybe a little more felt behind the dome would be nice, but I'm not complaining.


Impedance between the pair is a good enough match, as good as any other "name brand" drivers I've tested. Fs is low enough at 580Hz. Dome resonance shows at 26-27kHz, a bit low for what I would think for Be, any other Be model I've seen has resonance >30kHz. 26-27kHz is comparable with a lot of aluminum domes. Otherwise, the inductance is low, looks good.


I haven't done a full FR measurement set yet at 1m, but the initial measurement at short 315mm distance doesn't show any cause for concern, it looks like a perfectly well usable tweeter.


So here's the thing. I don't think this is a Be tweeter, that much is possibly a scam to charge double for a different colour variation of the CM25Nd "ceramic alloy" model. There is also a Titanium Ti25Nd as well, both these models share the same motor structure, with just a slightly different rear chamber, and unfortunately aren't available in a 28mm version. With that said though, at only CAD$110 for a pair shipped to your door, if they perform anything like this one, they're a steal!

CM25Nd

https://www.aliexpress.com/item/32329525755.html

Ti25Nd

https://www.aliexpress.com/item/1005003234048593.html

Be28 (this one)

https://www.aliexpress.com/item/1005006256231999.html

[Reet]

Motor image taken from the Ti model page. Full copper sleeve and lots of neo power.

[duvixan]

Woud you recommend the Be28 without reservations?
Any subjective listening impressions?

[duvixan]

Found a thread about them on Russian forum

https://forum.vegalab.ru/showthread.php?t=93941

[theSven]

I'm always interested in reading about new speakers. At first glance I thought it looked similar to the peerless tweeter that I have in 2 builds.

Thanks for sharing the data Reet. It's great to have as many options on the table as possible for DIY designs and building.

[duvixan]

There are many reviews from Russians for all options on aliexpress, them seem quite happy with them.
Looks like also they sell high quality Dynaudio knock-offs

[Reet]

I'm always interested in reading about new speakers. At first glance I thought it looked similar to the peerless tweeter that I have in 2 builds.

Thanks for sharing the data Reet. It's great to have as many options on the table as possible for DIY designs and building.

I've a bunch more drivers that I tested over the holidays that I will post soon, just collating the data and deciding if I should bother posting about a couple high end drivers that tested fairly poorly. This one was particularly exiting though, true Be or not, it's a top drawer performer.

[Reet]

Woud you recommend the Be28 without reservations?
Any subjective listening impressions?

Not listened to yet, just raw data. After testing many drivers over the holidays, it was surprising to be able to easily detect the clarity in the sine sweep, and I can confirm, this one was clean. I've also tested a few that have been done on Hificompass, and with fairly good agreement in my own measurements, I have confidence in the results shown above.

There are many reviews from Russians for all options on aliexpress, them seem quite happy with them.
Looks like also they sell high quality Dynaudio knock-offs

I've a pair of the cheaper CB25 "copper beryllium alloy" with ferrite motor, it also tested pretty darn good for distortion, but frequency response left a bit to be desired on the top end. I wouldn't hesitate on the CM or Ti models mentioned above. I don't really want to support knock-offs, but there is copies of Dynaudio Esotar, as well as Illuminator tweeter and 18WU woofer. They're still fairly pricey, but looks like about half the cost of the real thing. This tweeter doesn't appear to be a direct knock off, but it does appear that Melodavid has dissected some high end drivers and made something quite comparable at a low cost.

[JonMarsh]

Thanks for sharing and posting these measurements. I have only a few thoughts or comments to offer-

The question of whether the Be28 actually is or isn't Beryllium is interesting in a theoretical sense, regarding business ethics, I think, but in a practical sense, it comes down to a question of just how does the part perform and can you work with it successfully in a given application and target value proposition...

On the topic of Be domes, I offer the observation that geometry and fabrication of the dome is just as important as the material... one need look no further than the efforts an variation of a single company, ScanSpeak, to appreciate that. Look at the differences between the D3004/6640 (used in my Wavecor Ardent design from 2013, and in a OEM version in a number of recent Rockport models) and the D2908/7140, selected for ET's Kurosawa tower design with Accuton woofers.

The 6640 has what I would call a near perfect dome design, and virtually no discernible breakup mode at the top end in either my measurements or ScanSpeak's; the 7140 has behavior more like what I would expect from a "typical" Be tweeter, with a roughly 6 dB peak of middling Q, at 32 kHz.



The 7140 shows the typical SS vice of rather high asymmetric distortion manifesting in HD2, possibly due to the mechanics of the dome suspension, though the basic motor linearity is evident in the low level of HD3 above 1kHz.

Now, just relating how other dome materials may perform with careful design, this is the "venerable" (as in rather old design) Tang Band TB 25-1983, an inverted titanium dome tweeter with a high flux neodymium motor (inverted titanium domes being a favored configuration of mine in the 90's):



Sample measured in 2016:




Off axis shows "hints" of an intrinsic dome resonance at ~32 kHz, there is little evidence of it in the 0-30 degree plots (10 degree increments here), and interestingly, distortion plots of this part show HD2 lower than HD3, and both at fairly nice levels.


What's the old saying? "Takes all kinds..."

[Reet]

Of course, there is a lot more to a dome than just the material it is made from. My comment was purely at whether the advertisement is accurate or not. I don't think this one is a "pure Be" diaphragm like what is available from Scan-Speak or SB. Possibly some Be is in the coating layer, on top of an Al or Ti diaphragm. Unlike what you see from other Be tweeter models, this one rings like a bell up top, in a similar fashion to many Al domes. I'd really like to compare the CM25Nd model at half the price, as well as others from Melodavid like the soft domes.

[PANDINUS]

My BE28 bought a year ago has resonance at 32kHz. The distortion is really very low, that's right.

[duvixan]

But is it better than SB26CDC or Seas 27TAC ?

[Reet]

My BE28 bought a year ago has resonance at 32kHz. The distortion is really very low, that's right.

I got 27kHz, that’s al I can say. Maybe Be25 will be at 32kHz.

[Reet]

But is it better than SB26CDC or Seas 27TAC ?

You'll have to try them all and report back

[tktran]

I’ve tried the 27TAC/GB. But I wasn’t that impressed with the dip at 20KHz and the peak at 27Khz. The harmonic distortion is excellent, if it wasn’t marred by the mentioned dip and peak

[duvixan]

Looks like an ultrasonic notch would do well

[Reet]

Ain't nobody but your dog gonna hear that dip or that peak. Put the 27TAC into service and enjoy

[technodanvan]

Coming from a guy with two dogs, I might consider reducing that peak a hair just so we don't go irritating those good bois.

[tktran]

What??! We don’t need diamond to solve this?

https://hificompass.com/en/speakers/measurements/bliesma/bliesma-t34d-4

[Reet]

Need? No, but there is a market for it. I'm sure Bliesma has put considerable engineering effort to this product, but In this price range, performance is secondary to being pretentious.

[bobev]

Hi Reet, could you predict the sensitivity of the Melodavid Be28 compared to other drivers you have tested under the same conditions?

[Reet]

Hi bobev, I think it's actually a bit on the lower side, maybe 90dB/2.8V/1m, but let me check with Omnimic to confirm this evening.

Couldn't help but notice, you joined this forum 15 years ago and this is your first post, welcome!

[bobev]

Hi bobev, I think it's actually a bit on the lower side, maybe 90dB/2.8V/1m, but let me check with Omnimic to confirm this evening.

Couldn't help but notice, you joined this forum 15 years ago and this is your first post, welcome!

Thanks!

I would be very grateful, sensitivity is important to me.
I have been here for a long time and I read with interest, but I don't write mainly because of my bad English.

Greetings from Bulgaria!

[technodanvan]

You English is at least as readable as mine, and I grew up with it! Welcome!

[tktran]

bobev
Здравейте! Как си? Приятно ми е да чуя от теб

Has anyone seen an off-the-shelf 25-26mm hard dome tweeter with a neo magnet, no faceplate and minimum 92dB/2.83V?

[Reet]

Thanks!

I would be very grateful, sensitivity is important to me.
I have been here for a long time and I read with interest, but I don't write mainly because of my bad English.

Greetings from Bulgaria!

Here you go, measured with Omnimic, at 315mm, adjusted by 10dB to represent SPL at 1m. I used 27uF series cap, since Omnimic doesn't allow frequency range limits, so SPL is only accurate in the high end, but good enough for a sensitivity check.

Black trace is Scan Speak R2604/8330, rated at 92dB, I got ~92.5-93dB.
Red trace is Be28, 90.5-91.

[duvixan]

off-the-shelf 25-26mm hard dome tweeter with a neo magnet, no faceplate and minimum 92dB/2.83V?

Audax?

[bobev]

Thanks Reet

I ordered a pair of 28Be-4ohm, I also have an old 25Be about 6-7 years / on the other one I broke the dome / from the same seller.
When they arrive I will share impressions and measurements of the new and the old.

@ tktran
Поздрави на Австралия, чел съм ваши публикаций тук и в diyAudio, ентусиазмът ви е впечатляващ!
Бъдете здрав!

[tktran]

Благодаря, оценявам го. като цяло се опитвам да споделя това, което съм научил, и да привлека хората, като демонстрирам. В ранните дни на Интернет всичко, което имахме, беше монохромен текст, така че това, което можех да комуникирам, беше ограничено от владеенето ми на английски. След това се появиха снимки, а наскоро и видеоклипове, които могат допълнително да ни помогнат да обясним защо преживяхме това, което направихме. В много отношения да говорим/пишем за звук е като да говорим за вкус или друго сетиво. Става въпрос много за превода.
Очаквам с нетърпение да прочета повече за вашите аудио преживявания.


A new/old comparison would be great!

[Reet]

I guess I have the 8 ohm model, but it's closer to 6 ohm. I don't actually recall what I ordered, and when I checked it just says "110mm panel". I don't even recall if impedance was an option when I ordered, the product page only gives information for the 8 ohm model , but there is clearly a selection now for 8 or 4 ohm coils. Hmm.

[tktran]

Audax?

Which model?

looking for one with a cutout diameter of around 2”.
the idea is to mount it into a cylindrical enclosure, and have roundovers starting from the edge of the dome.

lets see if all baffle-less tweeters give controlled directivity:

https://www.htguide.com/forum/forum/primetime-a-v/mission-possible-diy/driver-testing-discussion/950681-a-study-on-directivity-ribbons-domes-wg-horns-baffled-or-naked?p=950858#post950858

[bobev]

Hello, I am uploading several measurements of the 4 ohm version of the 28BE, for comparison I will also upload measurements of the SB SATORI TW29TXN-B under the same conditions.

impedance

frequency response



harmonic distortion at 90db





harmonic distortion at 94db





For comparison, SATORI TW29TXN-B



harmonic distortion at 90db



harmonic distortion at 94db



I also managed to listen to them in one of my projects, they sound wonderful and I think they are quite similar in character to the mentioned SATORI

[bobev]

And measurements of the old Be25, the resonance is at 21khz and it is not any beryllium but some brittle alloy, it breaks like ceramics or glass.

frequency response


harmonic distortion at 90db

[Reet]

What are you using for a mic? Flat line 1% 2nd order is not right, so focus will have to be on 3rd order and up.

I think the Melodavids are excellent tweeters, they just need a small bit of QC on the chamber damping and ferrofluid.

[bobev]

Hi, the second harmonic is inaccurate, but I don't really care, it's the microphone Dayton emm-6 calibrated by https://www.cross-spectrum.com/measu...ed_dayton.html

Yes, they are great, I have an idea to modify them without ferrofluid and some more magnets in the motor to saturate the iron completely.

[duvixan]

I wish he was truthful about the actual diaphragm material, but that's secondary, I suppose..

[bobev]

https://vi.aliexpress.com/item/1005006725847389.html?spm=a2g0o.detail.pcDetailTopMoreOtherSeller.1.6320XjSsXjSsoE&gps-id=pcDetailTopMoreOtherSeller&scm=1007.40050.354490.0&scm_id=1007.40050.354490.0&scm-url=1007.40050.354490.0&pvid=cb5a903b-b22d-4893-8279-aa5546b677e8&_t=gps-id:pcDetailTopMoreOtherSeller,scm-url:1007.40050.354490.0,pvid:cb5a903b-b22d-4893-8279-aa5546b677e8,tpp_buckets:668%232846%238112%231997&pdp_npi=4%40dis%21BGN%21365.33%21328.80%21%21%211440.00%211296.00%21%4021032dcb17114621544876951e5c51%2112000038102947728%21rec%21BG%214790202002%21&utparam-url=scene%3ApcDetailTopMoreOtherSeller%7Cquery_from%3A

Small waveguide version, will also have 4 ohms

[rhalf]

Hi, I was drawn to this forum by this topic. A youtuber Raw-Cat (is he here?) posted some helpful comparisons with these drivers.

We can draw some conclusions from them. The part that caught my attention was the comparison with a "ceramic" version of this tweeter. They're very similar, but the ceramic one has more sensitivity and similar, but more pronounced features in FR. Could it be, that the Be is simply deposited on the ceramic diaphragm? That's how Chinese make headphone drivers. They take regular off the shelf drivers and metallise the plastic film with some titanium or berillium for a slightly brighter effect.
This rises some questions. Wouldn't it be better to just get a ceramic driver for quite a bit less and then use EQ on it? I can't find decay data for them.