Hello
I'm quite new in the compression driver world and still try to understand most theory behind them.
Here a few question than bug me. Did search but did not came with answer yet...
I own a pair of Eminence N151M (1" annular ring CD)
as per datasheet:
http://www.eminence.com/pdf/N151M-8.pdf
it has 2 impedance peak
one at 3200hz and a broad at 1200hz
Why is the 1200hz so wide?
Is 3200hz realistic the lowest point to aim? Eminence recommend above 1.8khz...
A BMS 4545ND
As per Datasheet
http://www.bmsspeakers.com/fileadmin/bms-data/curves_compression/neodymium/bms_4545nd_curves.jpg
It has 3 impedance peak, not 2 like the eminence
(750hz, 1.5khz and 2.4khz)
Why is there a 3rd one?
Yet a Celestion CDX1-1445
as per datasheet
Celestion CDX1-1445 Compression Driver - Ferrite
it only has 1 impedance peak
at 1600hz
What lesson to take of 3 different 1" exit compression driver showing different impedance peak to the source?
I'm running in active so passive crossover impedance match do not matter much for me.
Can impedance plot be useful for spotting other parameter in selecting proper CD?
Is the impedance important at all given you operate active AND above FS?
Any comments welcome.
Thanks
I'm quite new in the compression driver world and still try to understand most theory behind them.
Here a few question than bug me. Did search but did not came with answer yet...
I own a pair of Eminence N151M (1" annular ring CD)
as per datasheet:
http://www.eminence.com/pdf/N151M-8.pdf
it has 2 impedance peak
one at 3200hz and a broad at 1200hz
Why is the 1200hz so wide?
Is 3200hz realistic the lowest point to aim? Eminence recommend above 1.8khz...
A BMS 4545ND
As per Datasheet
http://www.bmsspeakers.com/fileadmin/bms-data/curves_compression/neodymium/bms_4545nd_curves.jpg
It has 3 impedance peak, not 2 like the eminence
(750hz, 1.5khz and 2.4khz)
Why is there a 3rd one?
Yet a Celestion CDX1-1445
as per datasheet
Celestion CDX1-1445 Compression Driver - Ferrite
it only has 1 impedance peak
at 1600hz
What lesson to take of 3 different 1" exit compression driver showing different impedance peak to the source?
I'm running in active so passive crossover impedance match do not matter much for me.
Can impedance plot be useful for spotting other parameter in selecting proper CD?
Is the impedance important at all given you operate active AND above FS?
Any comments welcome.
Thanks
The 1200Hz is so wide because it resonates lively there. Parts of it are from the horn used for the measurement. Important are there just the fs of the diaphragm. XO'd steep you're probably go a little lower than 1800Hz.
That's a horn resonance or a resonance of the back chamber.
That's an exceptionally resonance-free compression driver. That's a unique trait of the plastic diaphragm. It got a extremely narrow resonance at ~18kHz which can be seen on a decay measurement. It's not audible. It sounds excellent btw. but not that 'in your face' a lot of ppl want.
The lession would be 'don't read too much measurement graphs'
..especally if they are made under completely different circumstances - they aren't comparable.Partly with experience. Select them from other informations.
Well, with that information you can determine how much power exactly is consumed. For that you have to measure them yourself in your speaker and with your horn. Besides that it can be a hint of resonances but for that a THD measurement is much more important.
Thanks ICG
This clarify a lot for me.
I'm experimenting with Radian 760NEO PB mounted on a JMLC horn (340hz azura)
It cover from 600-7khz and so far, I really like it
I'm trying to find the best tweeter to mate with it as treble seem my weak point right now.
Had the TPL150 and didn't had enough resolution above 10K and directivity wasn't the best. I sold it back, maybe too quick.
Had Raal 70-10 but directivity and subjective dynamic were lacking so I sold it too...
Currently use a Beryllium dome as it seem as the best compromise so far for sound quality/dispersion but want to get higher efficiency if possible.
Await to horn load it (horn salvaged from Scan speak H2606)
Not sure what to expect with this but it shall be fun.
The dome sounded the best with my previous cone midrange but not sure if it is still true with my horn...
Tried the Fostex FT96H and BC DE35, they didn't pass my subjective assessment at all.
BC has no definition and fostex had a annoying sizzle to it's sound. (all fostex bullet horn I heard seemed to have this attribute I think, heard the ft17h and t500mkii on different systems)
The Eminence N151 seem best compromise so far but it ain't a cost no object tweeter either. I'm yet to direct compare it to my beryllium dome on my main system.
I'm tempted to buy a pair of Faital HF108 on a tractrix horn, Visaton TL16H, Beyma cp21 and faital FD371.
Any comment on those 4?
I think the HF108 is the safest option because, if I revert to my previous cone midrange, the 108 will have no issue crossing at 3khz...
This clarify a lot for me.
I'm experimenting with Radian 760NEO PB mounted on a JMLC horn (340hz azura)
It cover from 600-7khz and so far, I really like it
I'm trying to find the best tweeter to mate with it as treble seem my weak point right now.
Had the TPL150 and didn't had enough resolution above 10K and directivity wasn't the best. I sold it back, maybe too quick.
Had Raal 70-10 but directivity and subjective dynamic were lacking so I sold it too...
Currently use a Beryllium dome as it seem as the best compromise so far for sound quality/dispersion but want to get higher efficiency if possible.
Await to horn load it (horn salvaged from Scan speak H2606)
Not sure what to expect with this but it shall be fun.
The dome sounded the best with my previous cone midrange but not sure if it is still true with my horn...
Tried the Fostex FT96H and BC DE35, they didn't pass my subjective assessment at all.
BC has no definition and fostex had a annoying sizzle to it's sound. (all fostex bullet horn I heard seemed to have this attribute I think, heard the ft17h and t500mkii on different systems)
The Eminence N151 seem best compromise so far but it ain't a cost no object tweeter either. I'm yet to direct compare it to my beryllium dome on my main system.
I'm tempted to buy a pair of Faital HF108 on a tractrix horn, Visaton TL16H, Beyma cp21 and faital FD371.
Any comment on those 4?
I think the HF108 is the safest option because, if I revert to my previous cone midrange, the 108 will have no issue crossing at 3khz...
I am not surprised. You are starting it from the wrong side. To get a homogenous sound, you have to match the directivity at least roughly and get something with the same sound character. And there's your problem, your JMLC is extremely wide at the bottom (120°) and disperses very narrowly on the top (just ~45°). I know, that's the measurement of the 350C but I couldn't find it for the 340 and it will behave almost identical.
An externally hosted image should be here but it was not working when we last tested it.
That means, you're using it a lot too far up because it's so narrow there and the tweeters you bought got a lot broader dispersion (and aren't supertweeters anyway and need EQin too). The energy frequency range in the room does not fit at all with these drivers. The big distance between the mid horn and the tweeter is another issue, especally at very high xo frequencies.
That won't be fun. That's not a horn, that's a wave guide and it hardly gives you higher efficiency that far up, it 'starts to end' loading where you want to cross it over.
No, it won't.
The Fostex FT96H is hopelessly overpriced, IMHO it's not worth a third, other, older Fostex horns were better but I wouldn't buy them either. I haven't heard the Fostex T500A but I strongly doubt they are worth the money.
The N151 is not a supertweeter, it's for a 2 way PA top, best in the range 1,8-10k. IF you want an Eminence tweeter and xo them at 7k, the APT 80 would be the one to go with but I wouldn't go for that either. You should start your search from the other end. You need either something that matches the dispersion of your mid horn that far up (45°) OR you cross over deeper and use a wider horn there, that would be at 3-4k and 60-80° radiation pattern. For the former you should look for a small diaphragm supertweeter (complete with horn preferably), for the latter a compression driver with a small to medium diaphragm (25-34mm) and a matching horn. Decide for either solution and then start looking for a driver.
Well, the TL16H is good, the FD371 maybe too, haven't heard it yet. I would go for a Beyma, but the CP22, that's a ringradiator, identical to the CP21 except for the radiation pattern, with a big coil but still very light membrane and very good motor/mass ratio and it got a reasonable price. If you find that's the way to go, you could try the TL16H but mounting that one is a hassle, you have to manufacture a flange or something, you should be sure that's what you want before you buy that one.
Er, no. Do not buy a compression driver with an unnecessarily big diaphragm (44mm), the Celestion CDX1-1445 is a better fit there or, if you want to stay at Faital, the HF100 is very nice. The HF100 can be used either as supertweeter aswell at a lower xo frequency. For 3k the 18s XT120 horn would be a good fit for the HF100, nice horn, wide and constant radiation pattern, looks good, very good haptic, strong built and very cheap.
I look at LeCleach as a more correct and complete family of horns in the spirit of tractrix, so I ignore tractrix as an option. These horns are excellent loaders, coming at the expense of directivity (but with the benefit of excellent termination). This means once you build one you are more or less stuck with your choice of crossover frequency.
'More correct' in which regard? Linearity, dispersion, directivity, gain, ..look?
There are a lot different horn types which work great - if used right. That includes you can get a worse result if driver and horn don't play along well. And that also means there are examples where the LeCleach horns sound bad and others do better. Does that make the LeCleach bad? No. But the other way around it's the same, the LeCleachs are not 'more correct' than others. They probably sound better with driver x than driver y but the same goes for other horns. And, not to forget, the setup and crossover, EQing has all to fit aswell. So no, LeCleach is not 'more correct' in general and they did not discover the holy grail of horns.
ICG
Thanks so much for all the info
Ordered a pair of Beyma CP22
Will buy the Celestion cdx1-1445 as well but finding the horn for it will be harder.
the 18s you recommend, I know no distributor in Canada
and the JMLC option (looking at the autotech 2000hz) is months for delivery, will probably have to wait on this one...
Despite your lack on enthusiasm, I'll still try the waveguide on my dome tweeter.
It's A Transducer lab 28BER
It is very true that the scan speak waveguide can load only in the 1500-4000range but I think it will look cool and already bought the h2606...
Will report back
Thanks so much for all the info
Ordered a pair of Beyma CP22
Will buy the Celestion cdx1-1445 as well but finding the horn for it will be harder.
the 18s you recommend, I know no distributor in Canada
and the JMLC option (looking at the autotech 2000hz) is months for delivery, will probably have to wait on this one...
Despite your lack on enthusiasm, I'll still try the waveguide on my dome tweeter.
It's A Transducer lab 28BER
It is very true that the scan speak waveguide can load only in the 1500-4000range but I think it will look cool and already bought the h2606...
Will report back
I'm very sorry, I was not aware 18sound is not available in NA/CA.
The FaitalPRO LTH102 would be an option and is sold by Parts Express. Downside is the price, with 60 bucks noticable more expensive (but surely worth it) but I'll search for another one which is closer to the XT120.
Well, it's a very good driver but it lacks the spl and dynamics. Sure, you've already got it so go and try it, why not? I'd just suggest to not modify the driver irreversible, that would be a real loss.
Too late, plastic waveguide horn is out and I sanded it down to fit the Transducer lab tweeter
Irreversible mods and I threw the brand spanking new SS in the garbage... (2 scan speak 2606 and 2 peerless H26 to be exact)
I also have a set of transducer lab with aluminum dome (Ti vc)and I want to compare it to the beryllium with the same waveguide.
Is there a reason you recommend the elliptical tractrix over the JMLC for the cdx1-1445?
with Ebay, I can buy both mentioned horn but will pay brooker/custom fees.
Well, yes, it got the matching radiation pattern I can't find the JMLC type now but I think I remember the dispersion wasn't matching. The Tractrix wasn't my first choice but the probably best fit what Parts Express has in their program.
Edit: Ah, well, I think I remember now, the JMLC continues to beam narrower further up which the Tractrix does not.
Edit2: Could you please provide a link to the JMLC?
Last edited:
Thank you. Yes, it's as I remember for the 1400, 120° at 3k and less than 80° at 8-15k but especally narrows down at 16k, just 50° is not exactly harmonic. Radiation pattern and interferences are the things no DSP can correct. The 2000 is exactly the same horn contour so it will behave the same, just shifted to ~1/3-1/2 octave higher.
This mean, I shall go with your elliptical instead of the JMLC?
to follow your 2nd recommendation earlier and use the 340hz JMLC from 600-4k (may hit lower and go to 550hz)
the cdx1-1445 will kick in at ~4khz, what horn shall I get, the elliptical tractrix the 1400 or 2000?
you basically control my wallet at this point
I shall receive the Beyma CP22 next week and will try them quick.
Martin
to follow your 2nd recommendation earlier and use the 340hz JMLC from 600-4k (may hit lower and go to 550hz)
the cdx1-1445 will kick in at ~4khz, what horn shall I get, the elliptical tractrix the 1400 or 2000?
you basically control my wallet at this point
I shall receive the Beyma CP22 next week and will try them quick.
Martin
It's not mine, I don't sell it and I don't own it.
The Faital Pro STH 102 is a good match for XO at 4kHz and up in your case because that matches the JMLC in that range, if you want 2-3k XO the Faital Pro STH 100 would be a better choice (and it's also cheaper)
you basically control my wallet at this point
Well, I feel rather uncomfortable with that kind of liability.
I think you'll like them. They have a good resolution/detail, they are dynamic and they are capable of a very high spl if not xo'd too low (>5k).
Ordered STH100
Ordered Celestion Cdx1-1445
not sure if I shall order a JMLC 2000...
Love DIYaudio, ask for a question on impedance on CD and then order CAD$500+ worth of stuff...
I mean, I'm more than happy to try as I'm soon to be called a "horn convert".
I will try the beyma first at ~7khz (48db/oct)
then, once I receive the rest of the stuff, I'll try to cross at 4khz with the 1445 on the sth100
Will see which configuration win
Have you any prediction?
Ordered Celestion Cdx1-1445
not sure if I shall order a JMLC 2000...
Love DIYaudio, ask for a question on impedance on CD and then order CAD$500+ worth of stuff...
I mean, I'm more than happy to try as I'm soon to be called a "horn convert".
I will try the beyma first at ~7khz (48db/oct)
then, once I receive the rest of the stuff, I'll try to cross at 4khz with the 1445 on the sth100
Will see which configuration win
Have you any prediction?
This is quite a wide band for this kind of horn, especially when you want to cross both above and below it. Some use JMCL as a tweeter all the way up and like the sound because it has graceful higher frequency properties, except that the power drops with frequency.
When going elliptical, keeping the expansion will retain the loading but when you force the profile in one dimension for dispersion, you pay for it in the other dimension.
What is your reason for wanting this variety of horn, is it because you liked the sound? I don't know whether you'd find it easy once you have to cross in the highs.
Lynn Olson was using the Azura but doesn't anymore. I think you'll find his comments in the JMMCL thread.
I would not (yet). Just listen to what you've got so far and then decide which way to go.
That's too steep, the phase won't add cleanly. And it's not homogenous, it will sound like two separate parts. You probably even have to xo both partners asymmetrically (electrically) to achieve a good cross over range. You should start with 12dB and then up to 24dB max. 18dB is maybe the best choice. 70% of the sound impression depends on the settings of the DSP, not the drivers. And a flat response is not guaranteed to sound great, also dependent on what response you're flattening (direct sound, 0° or different angle, short distance, room, etc).
You'll probably like both. And then you can't decide..
I personally would likely prefer the CP22, it's crisp and detailed, the Celestion is detailed too but a bit less ..eager.. tad tamer. But you can always change the character by the setup.
Done
here my blog about it
Audio by Martin
will have to listen to them, just measured so far.
Been busy this weekend.
_________________________________________
Received the Celestion CDX1-1445 and the Faital STH100 horn.
here some quick line about them.
Audio by Martin
I really need to burn them in (and the new amp driving them too)
right now, it is very promising at ~4khz crossover point.
way better than my heavy magnesium dome tweeter.
More to follow.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.