I am stuck! Looking for the perfect midrange, I ask for help

Æ · « **Reply #20 on:** 22 Oct 2011, 06:45 pm »

Quote from: THWO on 22 Oct 2011, 06:33 pm

Meanwhile I have found a woofer with figures and response so good I barely dare to beleave: Precision Device DP.1550
I made a scan of the datasheet =>

Looks like this could be a nice one for a system with only one woofer instead of two per side? And EUR 200 seem well OK...
What do you wonderful people think about it and does anyone enjoy any own experience with it or other 15" PDs?

Regards
Till

Specs usually look good, but the reality is that too often the received product measures nowhere near what the spec sheet claims. I had to tilt my head to the left to read the numbers, the "Qts" of the woofer is very low. Were you planning on using this woofer in an open baffle configuration?

THWO · « **Reply #21 on:** 22 Oct 2011, 07:06 pm »

Yes, I plan OB. Do you think QTS is too low to be helped out by going active plus DSP?
I was looking for high efficiency, low QES and low Mms : B/L ratio as well as high Xmax.
I have heard several times that very high QTS meant reduced dynamics and precision in upper bass / low mids... Am I wrong with this or too naive?

Regards
Till

Æ · « **Reply #22 on:** 22 Oct 2011, 07:31 pm »

Quote from: THWO on 22 Oct 2011, 07:06 pm

Yes, I plan OB. Do you think QTS is too low to be helped out by going active plus DSP?
I was looking for high efficiency, low QES and low Mms : B/L ratio as well as high Xmax.
I have heard several times that very high QTS meant reduced dynamics and precision in upper bass / low mids... Am I wrong with this or too naive?

Regards
Till

The quoted sensitivity will only holdup through the midband, you'll get nowhere near that much in an open baffle. That woofer was designed for use with an enclosure. I have no facts about it's performance in an open baffle without modeling it first.

"I have heard several times that very high QTS meant reduced dynamics and precision in upper bass / low mids." That is an unsubstantiated statement. It depends on the woofer and application.

Here was one very successful application of high Q woofers.
http://www.nytimes.com/2011/10/18/business/edgar-m-villchur-hi-fi-innovator-dies-at-94.html?_r=1

THWO · « **Reply #23 on:** 23 Oct 2011, 11:40 am »

I love creating and handcrafting speakers (cabinets and open baffles, recently) but admid that I am not very experienced with developing complete speaker systems from scratch, i.e. crossovers etc. I own a measurement system now (ATP pro) and have started playing and experimenting around. My last OB-project was more succesful than prevoiusly expected, not to forget by the support of some forum members, too, and this very positive result is it what made me interested in building up a top quality system in near future. I do know that I appear a bit naive in this regard... But that´s exactly why I have been approaching you wonderful people here in this forum. Thanks for your patience with me.

Well, my latest project did include a high-QTS-woofer, indeed, by using two Eminence Alpha 15A per side, but in the overall setup these are the weak part as well.

Let me modify my questions: Should I look fo a woofer with QTS of at least 0.5 only - or: How low may I go with QTS considering going active incl. DSP? Would 0.3 be still OK? Is it at all possible to give an overall valid answer? I can well imagine that an answer like "depends..." is inviting - but will that hel me? How should I preselect just from available technical data? It will not be extremely realistic to listen to all kind of woofers in open baffles before deciding...

If I went for two BG Neo 10 coming in at, say, about 300-400 Hz and if I decided for just one 15" woofer instead of two, I would need an adequate companion being fast and accurate / dynamical enough to come up with the speed and resolution of the Neos. I don´t need subbass (i.e. 20 Hz), but look for precision and top resolution.

Till

Æ · « **Reply #24 on:** 23 Oct 2011, 06:45 pm »

Your best bet is to take a serious look at the woofers other people are using successfully.

"Fast Bass" is a bogus term. Several technical articles on the subject.

http://stereointegrity.com/docs/WooferSpeed.pdf

scorpion · « **Reply #25 on:** 23 Oct 2011, 09:16 pm »

Till,

We have had an intense discussion in our Swedish Hi-FI Forum advicing a guy who wanted to build a 3 unit H-baffle for bass to supplement his ownconstructed ribbons.
Like you he started out with Eminence Alpha15s (on my advice among others) but realized that the result wasn't optimal. Now he wanted the best.

To make a long story short the decision at last stood between AE elements and PD elements. He choose AE TD15X units for his project.
But Thomann still seem to have PD18BR40 at about 220 Euros. This is an outgoing model but with better OB parameters than your choice.
For the Swedish guy we were discussing PD15BR40 but this is evidently outsold at Thomann now.

I think that one 18" per side would be sufficient and should produce 100 dB at 30 Hz within X-max limits, two would be great.
The rest is EQ. Theese are very good units.

/Erling

studiotech · « **Reply #26 on:** 24 Oct 2011, 03:35 am »

A pair of GR Research servo 12" open baffle provide world class bass reproduction as well. This is another item I almost used after hearing them several years ago at the RMAF. I am not sure you can take them up high enough to meet the Neo10 though...

More reviews this year that mention best bass at the show. Danny gets these type of remarks year after year.

Greg

Toaster · « **Reply #27 on:** 24 Oct 2011, 01:19 pm »

There is a thread on the GR Research forum on here that shows the 12" servo OBs running up to 300Hz- which appears to be as far as they go. It should therefore work fine with one or two Neo 10s. I've been considering that sort of setup myself, but can't really afford it all in one shot.

scorpion · « **Reply #28 on:** 28 Oct 2011, 10:12 pm »

No shade over the GR servos. They obviously are good. But I should perhaps show a bit more about my PD proposition.
First compared to yor choice, Till, the 18" basically has more X-max and more cone surface other things more or less equal and that would certainly be in favour of the 18".
I have made some simulations for the PD18BR40 driver. First I put it in a 50 x 50 x 40 cm (BxHXL) H-baffle just to illustrate what it will take to EQ. MJK uses his H-baffles for EQ-purpose and that is fine with high Qts units. Maybe also Alpha15 is better housed in an H-frame than on a bare baffle. However a lower Qts unit like the PD would sound more clear where the Alpha is muffled.

Now for the simus. First let's see the H-output unequalized when 40 Hz is 100 dB, thats how it always will look, regarding if EQed in H-baffle or on a flat baffle.
In H-baffle it will probably be less sensitive by a margin but more flat responding, so not so much EQ is needed. But in the end all will converge.

So first the unEQed H-baffle response:

Next let's look upon the cone displacement:

Now have a look at cone displacement. From MJK simus it is suggested to multiply with 1.414 to get the real cone amplitude. Now it pretty well clear that X-max requirement is more than fullfilled to produce this kind of SPL output.

So what more to wish ? Now 30 Hz - 100 dB will stress this units almost to it's max but it will do this beautifully, and just remember there is a +- 20 mm mechanical limit.

/Erling

zmyrna · « **Reply #29 on:** 31 Oct 2011, 03:57 pm »

I am at the driver selection/purchase stage of my OB built as well.
I am going in complete opposite direction of modern mainstream speakers.
So no good are boxes, heavy cones, rubber surrounds, high excursion/long throw.
I have already bought a pair of vintage Isophon alnico 8" as mid drivers.
Modern alternatives would have been Visaton B200, Supravox 8", Tone Tubby/Hemptone/Hempcone 8".
Line source type mid drivers can also be considered but are harder to implement.
Good vintage bass drivers are hard to source and expensive.
Alternative is pro audio driver.
Lampizator has massive experience with rolling drivers on OBs. His conclusion is that the optimum Qts for a 15" is about 0.35 (0.7 for a 8" and anything in between should be fine around 0.4/0.5).
At Qts 0.35, a big baffle or U/H baffle is required. And for a 15", separate baffle for mids is always better.
I am thinking two 12" per side. This has several benefits: the cone is lighter (better dynamics, less muddy mids), similar surface to a single 15", 0.4/0.5 Qts should be fine (smaller baffle/wings), upper 12" will be placed closer to the mid thus can be crossed higher. I will probably test the cheapo 12" from Soundlab first.
If the bass is not deep enough in my small room, I can add subs later.
Those Qts 0.2 PD drivers will need equalisation or TBass etc to give any bass on OB. Warning 2: I read on another forum that PD woofers have noisy engines and are not suitable for OB.

THWO · « **Reply #30 on:** 1 Nov 2011, 01:56 pm »

Thanks once again for your interest and support. Great forum.

Erling, when shopping around the recent days, one pair of the PD15BR40 has still been available at the Netherland´s distributor:
HW Sound vof
Maasdijk 142
4264 St Veen
The Netherlands

T: +31 65 3390308
F: +31 41 6696031
e-mail: hennie@hwsound.nl
web: www.hwsound.nl

Just in case someone here is interested in obtaining this model.

For myself, it was kind of internal struggle to finally come to the decision not to buy this item. However, thanks again for your recommendation, Erling, but I felt it was a bit premature to buy woofers before the entire concept was ready to lift off. In case I should come up with an entire different idea one day, I would be limited already.

I admit that I feel a bit reluctand going for an 18" as this means real big fronts. Yeeaaah, those silly design considerations... But your simu is nice work! I asume, when setting a depth of 40 cm, you meant driver in the middle and, therefore, 20 cm each in front and back. My own experiments with the Alphas in a H-baffle made them audibly mulmy. Less precise. They don´t seem to like any "load" on their membranes. My extreme experiment was setting them into a Ripol-Sub concept including measurement instructions and correction filter by its inventor, Axel Ridthaler, but that resulted in horror. Yes, extremely low with full 20 Hz, but extremly mulmy, "slow", and a-l-w-a-y-s moving and giving subsonic preassure even when no bass was envolved in the music. In short, unacceptable for me.

How would a low-QTS driver (say, around 0.3) perform in an open baffle without H-frame? I expect early drop in low freq. Is there a realistic chance to come up with this by equalisation to get reasonable 35-40 Hz or will that lead to too muchg stress / excursion and unprecise performance? I have been hoping not to need a H-frame... (yes, the design...)

Zmyrna, considering two 12" looks like it could be a rewarding choice. Would going for one 15" and one 12" be a combination that could work as well? The 12" could run up to 400 Hz, and the 15" would be used as a fill up in the low and mid-bass frequencies. Or will I buy tonal disharmonies when mixing sizes? Advantage would be that placing them upon each other would not eat too much height of the baffle and left enough space for the rest of the setup. Or will you guys show me "your red tongue" on this, as we would say in Germany, (for: no go).

If it worked, I could imagine a 15" plus 12" on top, followed by a Neo 10 (would a Manger be an option?) from 400-4000 Hz plus supertweeter, i.e. Harwood AM 25 (that looks nice!) or even a RAAL. Possibly quite an elegant, slim pyramid (at least at the top) and yet not exceeding my envisioned maximum of 49 inches in height.

Till

fsimms · « **Reply #31 on:** 1 Nov 2011, 02:33 pm »

If you were thinking of a 15" and a 12" you could change the 12" to an 8" to make it easier to crossover into the neo10. I have some EV Sixes that have an 18" woofer and a 8" lower midrange and it works well.

Bob

PS With the 8" you could probably get by with only one Neo 10.

Æ · « **Reply #32 on:** 1 Nov 2011, 07:53 pm »

Quote from: THWO on 1 Nov 2011, 01:56 pm

How would a low-QTS driver (say, around 0.3) perform in an open baffle without H-frame? I expect early drop in low freq. Is there a realistic chance to come up with this by equalisation to get reasonable 35-40 Hz or will that lead to too muchg stress / excursion and unprecise performance? I have been hoping not to need a H-frame... (yes, the design...)

It's not just the "Q" of the woofer you have to take into consideration, but the baffle width too.

scorpion · « **Reply #33 on:** 1 Nov 2011, 10:19 pm »

OK Till,
let's get serious. If you are going for 15" and 12" I see no reason costwise not to go with the best. So this is Acoustic Elegance TD15H on a 100 x 45 cm baffle, 25 cm up from floor in bafflemiddle. It can be crossed beautifully to Neo10 at 300 Hz. This simu has a Butterworth 4 order crossover at 280 Hz wich would meet Neo10 at 300.

The simus show SPL at 40 Hz 100 dB, 30 at about 96 dB and the SPL-curve will show what kind of EQ that will be required for flat response.
Here is SPL-response at 128 watts:

Next is cone amplitude in this situation, at 40 Hz 100 dB cone movement will be about +- 8 mm (5.6 x 1.414 mm) well within its 14 mm x-max.

So going active on a narrow baffle with this speaker would be a joy I am certain.

/Erling

scorpion · « **Reply #34 on:** 2 Nov 2011, 01:36 pm »

Let me illustrate my points a little more with some images.
First a frequency measurement of my Blindstone speaker with the IB15 properly EQed on its 45 x 55 cm baffle. I have built a new Panasonic WA-61 mic with simplest possible driving from the Sound Card's 5 volt output. Also a new Laptop ensures 50 Hz free measurements, which plagued my stationary PC. Here we go:

Then I want to show augerpro's distortion measurements at 100 dB SPL of AE IB15 and AE TD15X - a sister element to TD15H. I had to convert these images to jpg format to get them here properly, the main link to augerpros 'drivervault' is: http://sites.google.com/site/drivervault/driver-measurements/15 .

First the IB15:

Then TD15X:

Now, these measurements talk for themselves.

/Erling

THWO · « **Reply #35 on:** 3 Nov 2011, 09:51 pm »

Thanks for all that info.

As I have learnt from your blindstone thread, the IB15 should not be driven higher than up to 250 Hz, too low for the Neo 10. Going for an additional 12" or (Hi, Bob) an 8" would mean that this one should go until takeover by the Neo 10 at, say, 400 Hz. Nice idea...

Or would it be a better idea to reduce to three way and take a 15" to 250 Hz and then another cone driver up to the 2-3000 Hz range in order to pass the ball on to an overwhelming AMT Tweeter?

I think we are narrowing in. Once the general configuration is decided upon, it should be easier to decide on the very items to be set.

Regards,
Till

zmyrna · « **Reply #36 on:** 4 Nov 2011, 11:24 am »

Well, the main reason for my choices is that I am pursuing the passive crossover route.
You may have good results by using long throw drivers with heavy membranes, low efficiency and high Qts, and crossing them active with steep slopes.
However I still doubt how those drivers sound on open baffle. Added lack of dampening should have negative effects on dynamics and clarity. Historically, open baffle speakers had light cones and stiffer surrounds. With box designs came the pp cones and foam surrounds.
Compare the moving mass of 20g on Supravox GMF 285 vs 220g on IB15. Just as a reality check, IB15 is no more available for sale on AE website. That leaves Dipole 15, with higher Qts, and lighter cone (90g). Much more expensive though.
I am going towards high eff, light papercone, passive route. I will try to keep the main baffle 3 way. I am hoping to match the LF extension of a single 15" by way of using two 12". This will allow a 1st or 2nd order CO. If needed I can add subs (active or passive, box or OB).

scorpion · « **Reply #37 on:** 4 Nov 2011, 12:13 pm »

I have to back a bit regarding the IB15. Today I would not hesitate to play it much higher than I do, at present 277 Hz 48 dB Butterworth LP crossover. My earlier judgment was probably too much influenced of a very new driver. Today when it has had its duty of beeing played in. I think that AE's argument that the midrange is smooth is valid, at least it definitly sounds so when I test. So crossing an AE driver in the 300-400 Hz range would be no problem at all, I think and know. The low induction argument is valid. The reason to show my measurment was really to show the powerful 30 Hz output that I enjoy with my IBs. Do try to engage for a Group Buy from AE !

/Erling

THWO · « **Reply #38 on:** 8 Nov 2011, 05:36 pm »

Hi Zmyrna,

your preference of highly efficient drivers and hint to Supravox made me look around a bit. Two Supravox 400 GMF together with one 215-2000 plus a nice suiting supertweeter could be a real treat - and my banker´s suicide, probably. But a truly inviting idea...

Before deciding, I would have to find a spot to listen to such drivers, of course.

Heard that there was a serious and audible dip somwhere around 1500 Hz with the 215-2000 bringing strong coloration? If that was true, active equilisation would then be necessary, right? If the drivers were flat throughout its range, a passive configuration could be feasable as well.

Earling, I am really convinced that you do love your IB 15. I admit that I am a bit reluctant just because of its very low effiviency. But nothing decided yet.

Thanks all. I do apreciate your considerations. What a wonderful hobby.

Regards,
Till

Remlab · « **Reply #39 on:** 10 Nov 2011, 12:16 am »

Why Does everyone think the neo 10's can't go lower than 300hz? Zaph's measurements show very low distortion at 150hz and lower. The neo 10's designer told me that this driver has inherently low distortion down to 40hz relative to the signal, due to the diametrically opposed magnet arrangement(Admittedly, there's hardly any output at 40hz, but still...)
The neo 10 also measures identically on both sides. Perfect for a dipole arrangement. I cant imagine why anyone would think that a big magnet attached to the back of a midrange driver could be ideal for a dipole speaker.