Driver Evaluations for a new project

[Evil Twin]

Unquestionably an Xmax of 14mm will be difficult to match... in my own case, I am more interested in BL linearity out to 7-10mm Xmax.

A clear weak spot for the Anarchy is suspension linearity.

These drivers will give me fully adequate motivation to setp two tone spectrum analysis with the Audio precision, though I expect preliminary results with conventional methods to also be revealing...

[Bear]

It will be interesting to see what kind of distortion/resonance products show up in your measurements with the big roll surround. The Anarchy is a great $60 driver, but it was/is designed to a price point with no incremental improvements expected. Not to rehash the STW-350F discussions, but extended Xmax is useless if it isn't useful.

(I'm sooooo getting Force choked over that one...)

[Evil Twin]

Ah yes, occasionally I recall the STW-350F sitting somewhere in storage in the outer rim, and ponder what it's fate will be...

At least I am not the one in carbonite in storage...

Tomorrow will start another trip back to the galactic core to retrieve one more Tie Fighter to two back to the outer rim hideout.

[JonMarsh]

New driver testing is on hold for a while- we were notified by the property management company that the owners want to sell the house we're renting, so 30 days to find a new place and move.

That's going to be the focus pretty much all of September.

It's also a difficult time- my fiancé's older daughter was diagnosed with cancer a few weeks ago and we went up to Seattle to spend time with her family and advise on some topics- which resulted in a new test, and now a full body scan yesterday.

Hopefully we'll get through the month and have a new place to live soon- I have a showing arranged this afternoon, and have already applied, but turn-over on rentals is pretty fast in this area- often just a few days.

[Scottg]

Good Luck! (..to both you and your fiance's daughter.)

[JonMarsh]

Thanks Scott- so far, things are holding.

Liubov's daughter had a full body MRI scan because genetic analysis showed it was a more aggressive type, but she's clean from cancer having spread. She'll start the follow up chemo from her surgery a couple of weeks ago, and the outlook is pretty good.

And today I signed a lease for a new place that we should be able to move into by the 10th. Their gal Friday has my kind of work ethic- the application was reviewed and approved yesterday, Saturday, and she had the lease ready for electronic signing today, the Sunday of Labor Day weekend. Of course, I'd quipped to her that myself, I never had any idea that retirement would be this much work! She's either crazy dedicated or owns the franchise!

[JonMarsh]

A peak at the new PuriFi Midrange- just a few rough measurements

It has taken even longer than I expected to get to the point where I can re-start measurements and work- as I often say, I never expected retirement would be so much work!

These measurements are not intended to be directly comparable to any other measurements of this device- think of the setup as being more like a typical Seas approach, as measurements were not done on an infinite baffle, but in a 12" cube test enclosure with a rather applicationn specific mounting, relevant to evaluation for potential purpose.





The Isiris midrange module would require some modification to accept this driver, but it is quite feasible to accomplish- but once done, there is no going back to the Accuton.

An initial check was done in near field, as this may come fairly close to 1m infinite baffle measurements. The plot below includes 2nd and 3rd harmonic distortion.



Note that the SPL plot was done with only 1/48th octave smoothing, so I think it's reasonable to say it shows a fairly smooth basic response. 3rd harmonic is quite low, also, relative to the base level, at about -70 dB above 200Hz.

Next, I took 1 meter measurements, and overlaid them with the NF for comparison. Window length was 75ms.



There are variations due both to baffle step and edge diffraction. It will be interesting to compare with an Isiris/Isis style baffle.


Last, I did a quick check at 30 degrees off axis, to see how things hold up and how the off axis response will contribute to the room power response.




My personal experience often indicates I'm happiest if the response is quite flat in the 10-40 degree off axis area, as this contributes greatly to the overall power response. Things are looking promising here, but I think I may have to bite the bullet and update the midrange-tweeter cabinet to work with the PTT6.5M04-NFA-01 and test in that configuration, and collect data for updating the crossover. But then I'll probably need to find the BlieSMa T34B that I have, as I'm considering an update on the tweeter, too.

[Juhazi]

Lars Risbo explains the fundamentals of Purifi drivers. My appetite was waked up at least!

https://audioxpress.com/article/voice-coil-spotlight-some-speaker-problems-that-needed-solving

But then this Erin's test of the 4" Purifi shows ragged response and elevated IMD at 1-2kHz... Anyway T/S measurements are very good

https://www.erinsaudiocorner.com/driveunits/purifi_4/

[JonMarsh]

Thanks for the links Juhazi! I've been so busy this year I haven't been keeping up with AudioXpress at all, and that is a very recent article to boot!

Erin's review is interesting- it confirms my impression that the 4" PuriFi's are a bit of an odd duck, yet quite high performance in some characteristics. For a 4 ohm driver of this type, the sensitivity is rather low, and the suspension nonlinearity disappointing, as well as the SPL linearity in the 1-3kHz area.

[5th element]

Yeah it's as if the research and development was done on the 6.5" driver, which admittedly has excellent performance in all areas. Then they thought, the world would like a 4" driver, so cooked one up over night and that was that. Yes the motor performance is there but that's about it. The suspension limits excursion and those issues from 1-2kHz are absolutely awful.

[augerpro]

Hificompass's results look fine: https://hificompass.com/en/speakers/...i-ptt40w04-01a

[Juhazi]

Directivity of Purifi 4" doesn't look good to me. Well, perhaps ok for a 2-way speaker and LR4 around 2kHz, but as 3-way midrange in general no... A midrange doesn't need it's low freq performance.

[JonMarsh]

Yeah, there's not much point in making an expensive 4" driver that doesn't do all that well compared with a 6-1/2" driver...

[Zvu]

Interesting little fellow is BMS 5S117

91dB/1W/1m
It's 16 ohm
Crazy good distortion figures (measured at 100W, raised 10dB to fit the graph).

Jesus Maria

[Paul Ebert]

Yeah, there's not much point in making an expensive 4" driver that doesn't do all that well compared with a 6-1/2" driver...

I've felt the same way. I'm looking forward to the 4" Purifi midrange to be released (fortunately, I'm not in any hurry). Also, the 3" Bliesma dome, if I can afford an aluminum or silk version. I've also been very curious about the Kartesian Wom120 with its very linear response, low distortion and what strikes me as an exceedingly reasonable price.

[Paul Ebert]

Interesting little fellow is BMS 5S117

91dB/1W/1m
It's 16 ohm
Crazy good distortion figures (measured at 100W, raised 10dB to fit the graph).

Jesus Maria

Was it measured at 100W, thereby raising the distortion 10dB or was it measured at 100W and raised 10dB? Either way, it's impossible for me to compare to other graphs. The distortion does appear very low.

[Zvu]

Latter seems much more plausible. Common practise for JBL also.

[tktran]

I love it that BMS turn it up to 11, and then measure the distortion.

I don't think I've ever dumped 28.3V into a 5" driver...

[5th element]

What are you people complaining about? It shows better off axis than pretty much all other 4" that hifi-compass has measured.

My issue with it is if there are consistency problems. Erin's measurements are very different from 1-2khz. Performance breaking different. Regardless of the suspension limitation on its rated xmax it's still an excellent driver. Compare the W12CY006 vs the PTT4.0. The PTT has better IMD at 9mm than the W12 at 4.5mm.

If the wonky FR 1-2kHz measured by Erin has been solved then the PTT4 is hands down the best 4" you can get for a high performance, compact, 2 way. It's sensitivity is abysmal but it's been traded for other criteria. This is not a driver you use a midrange except for maybe on an open baffle.

[5th element]

Was it measured at 100W, thereby raising the distortion 10dB or was it measured at 100W and raised 10dB? Either way, it's impossible for me to compare to other graphs. The distortion does appear very low.

Well that's the problem isn't it? You cannot make direct comparisons. It's typical practice for manufacturers to raise the distortion by 10dB to make it fit on a graph.

Its distortion does seem quite low at 100W but I want to see what it does at 1W. Some drivers distortion does not fall as much as you'd expect when the drive level is reduced. Most listening is done at less than 80dB so this area is, arguably, more important.

[Paul Ebert]

Regarding the PTT 4, correct on all counts. I just wish it made a better midrange. That was clearly not the design objective.

As for the BMS, it does look very interesting, but I'd want to see 3rd party evaluation, including 1 watt distortion. I haven't run into situations where the distortion was raised before. Is it more common for professional drivers?

[5th element]

Is it more common for professional drivers?

I would say so. Professional manufacturers are more likely to actually provide distortion data for a start and when they do it's always in combination with the FR. This usually requires raising up the level of the distortion so that it fits on the scale of a typical FR plot.

[Evil Twin]

There has been some curiosity expressed with regards to some new ongoing driver evaluations... The Chancellor forbids sharing some, but this "modest" tweeter from Beyma has been cleared for discussion. It may be used in that bane of the DIY community, another 6-1/2" two way...

[tktran]

Which Beyma model is this?

[Evil Twin]

This is for the TPL75, the “baby” one.

[DaveFred]

Many things are possible; not all things are desirable.





Do pray you can delay ordering tweeters until some investigations are complete- Injuries sustained fighting with packaging for a new testing system may delay matters a few days...


But should you be curious about the possibilities which can be examined....


[ATTACH=CONFIG]30014[/ATTACH]

Hello,

Were the Jantzen WG + T34A-4 results specifically posted?

Any pics of the rear mounting of the tweeter to the WG?

Thank you,

David.

[JonMarsh]

On holiday, I found some funnels that looked about the right size, so I bought a couple.
I pushed three 10mm felt discs in tight, 50mm acoustic foam and some loose wool stuffed in. The blue tack I used to temporarily join the funnel to the tweeter enclosure only just held when pushed hard because everything is crammed in very tight. As you can see, the impedance peak is gone.
The right hand pic is now glued together with a glue gun.

Impedance before mods -







Typed in the numbers on the label and found the funnel at CPC - https://cpc.farnell.com/whitefurze/h...nel/dp/SI18690

Ordered some silencer fibreglass wadding from ebay. 10mm thick sheet, so should be easy to cut the right sized discs to wedge in tight - https://www.ebay.co.uk/itm/133668535468

Also ordered some A-Glass Long Strand Fibre Glass Wool - https://www.ebay.co.uk/itm/111944393...r=410891458078



Not sure why there's a peak at 3khz now



Just for comparison, here's the unmodified driver response:

A very interesting study- someday when I catch up with other projects, I would like to circle back to this, and see if I can figure out why the "vintage" made in India RS52AN don't have this impedance bump issue, for the most part, and what is different in the construction (tear down of a "modern" RS52AN and one of the India manufactured versions (I have an odd number, so one can be "sacrificial"

[Evil Twin]

SB26CDC-C0004 revisited

This is a follow up on the SB Acoustics SB29CDC-C0004, which I noted in early 2019 has some interesting characteristics, and general performance at a level higher than it's price would indicate, compared with many other products from SB Acoustics or others. At that time, I was searching for candidates for waveguide use primarily, but also looking at flat baffle applications.

The web site data has been updated, including the frequency response plots. This raised the question of whether there was a useful update in performance, or if like some other vendors, things were deteriorating a bit. The web site data indicated improvements, but there is a local saying which I find appealing in cases like this- "In God We Trust, all others we verify."


Do not attempt to compare these results directly with anyone else's test- these were made on a specific test baffle related to the intended application, not on an infinite baffle. In particular, there is a "bump" in the response characteristic around 1kHz due to baffle step behavior- you can easily model this in Edge.


So, I shall share just a few of the more interesting measurements...


Polar 0-60 degrees, 10 degree increments




The polar response is promising in it's characteristic and smooth roll off out to 60 degrees... this is with the phase shield intact, as were all other measurements


For the planned/intended applications in HT, such as a center channel or small mains, a narrower off axis spread is of greater interest...

Polar 0-40 degrees




An important consideration in evaluation is the impedance curve, and whether there are artifacts present and if inductive rise is well controlled...





So far, quite promising... somethings similar to the March 2019 measurements, but the overall smoothness is better, and it's possible the phase shield design has been tweaked. It's also possible that my use of a felt rebate is superior to a routed rebate, but I question that, merely leave it open for investigation. I do not have the original test baffles anymore.


So, to a key parameter- distortion- any issues or surprises?





From 2kHz up, HD2 is at about -65dB, and HD3 is at about -75 dB. This is very satisfactory performance at 90 dB output. It is also substantially better than the last ones measured, especially in regards to HD2.



Perhaps a comparison to a previously measured more expensive model from SB Acoustics would be in order- a model exhibiting very "Scanspeak" like characteristics...

TW29BN




I think we may have a new contender, perhaps winner, in the "Under $100 tweeter that is the best $400 tweeter you can buy" category.

So, it should come as no surprise to anyone when I propose this part as the successor to the RS28a for an updated Natalie P. In fact, my interest in this part for more "budget" oriented concepts is substantial, as this picture may indicate...

[duvixan]

Is SB26CDC a worthy alternative to DA25TX?

[Evil Twin]

"Worthy" is a critical judgement left to each listener and contractor... however, this component clearly delivers higher than typical value in its price range.

Some points to note:

• The SB26CDC-C004 incorporates a phase shield to improve dispersion in the top two octaves.
• This phase shield obscures the diaphragm breakup mode, which typically occurs at 26kHz, with a peak above average levels of 10 dB.
• The lowest HD2 will be achieved with notch filter to suppress this peak- curiously, it has less effect on HD3. Overall, this breakup mode is much less of a problem than the DA25TX008 (which is typically double the amplitude), and which must address it with a notch filter.
• The linear excursion capacity is quite similar, 0.6mm Xmax for the SB26CDC, 0.7mm for the DA25TX.
• The outer baffle plate diameter at 100mm is easier to work with for specialty applications such as a CC.

[Evil Twin]

Hello,

Were the Jantzen WG + T34A-4 results specifically posted?

Any pics of the rear mounting of the tweeter to the WG?

Thank you,

David.

No, they were not. The profile of the Jantzen waveguide is not truly suitable for the T34A. Like many waveguides, the direct on axis and slightly off axis suffered from response irregularities due to reflections in the inner throat area.

I would advise examining AugerPro's waveguide design for the T34 tweeters.

[Evil Twin]

And now something in the category of "interesting and obscure vintage", difficult but not impossible to source, if you have resources out of this sector...

I am referring to the Tang Band TW 25-1983, a titanium dome tweeter with neodymium motor system, which I have tested many years ago, but lost track of my samples and the original data... but not my impressions of the experience...









The response capabilities measured by others are interesting...




Of note is the minimal breakup peak at 40kHz, and the extended off axis response at 30 and 60 degrees... one might reasonably assume a general power response will follow the 30 degree axis, and that is a most capable performance.




CLIO 10 scaling for distortion plots uses a linear scale, leaving something to be desired for the analysis of low level distortion, but this is certainly promising enough to justify the use of more capable tools in follow on tests. Particularly notable is the HD3 behavior to below 1kHz. Naturally, a modern more extended range test system should yield a clearer picture of the ultimate capabilities...


And, of course, in closing, and acknowledgment of my own personal preferences,




Black is back.

in fact, it never left.

4 samples have shipped.

[Evil Twin]

It seems that my troublesome apprentice has located the original test files for the TB 25-1983. These point to the likelihood of excessive signal level at the receiving end of the measurement given the displayed level.





Clearly new measurements may be productive, even if principally for self indulgent nostalgia, for this undervalued part.

[Evil Twin]

It appears that the Imperial investment and confidence in DHL logistics is paying off once again- this shipment, scheduled for arrival on Wednesday, is now due in today.