Driver Evaluations for a new project
Collapse
X
-
As-168-9-251
One more test basics, for the Accuton AS168-9-251...
This was also mounted on an application relevant test baffle, the same size as for the PuriFi woofer.
Polar response 0-30 degrees
Polar response 0-60 degrees
The AS series Cell drivers clearly do have some relatively unique properties due to the overall engineering effort. They have been engineered to present an acoustic origin which appears at the baffle, and also with a response curve engineered for specific range of baffle dimensions- there is no EQ or BSC applied to the driver for this measurement. In this regard it behaves identically to the AS190 and AS250 I have measured.
Also, there is no conventional exposed surround to contribute to forward radiation, and the hemispherical aluminum sandwich dome does appear to convey some benefits with regards to consistency of off axis behavior.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Some basic zT25 BlieSMa tests
For Mr. Manning's plans for CC and a possible tower TMWW using a Duelund crossover, evaluating small form factor tweeters is a necessity to minimize the CTC for the midrange to tweeter.
While there are many small form factor tweeters on the market, many have undesirable response irregularities that makes the crossover development more difficult. The BliesMA T25A is one of the more promising ones available at a reasonable price...
With this small form factor, it seems the best method for rebating for the driver frame and motor is the use of lager Forstner bits... and an additional rear backing plate.
This technique may also facilitate clean bores for the connector wiring...
The finished mounting met expectations for fit...
DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
But this is just the foreplay... data is what is needed.
The tweeter was not tested on an infinite baffle setup, but a defined baffle with round over in a width range that may track well with potential applications...
The key data is a polar set from 0 degrees to 60 degrees.
There is not as much uniformity of the response characteristic as there is for the Peerless DA25, but the observed behavior is similar to that measured by others. I speculate that off axis behavior is not just a function of the absolute diameter of the radiator, but also the dome profile and the flatness of the front plate of the tweeter.
From a design viewpoint, it may be most useful to compare the 30 degree response and the on axis. However, the on axis is not very representative of the total power response, in comparison, because it is a small point, as opposed to a donut covering a much larger area, and of similar characteristic.
From a design viewpoint, especially for an HT type speaker, the 30 degree response is probably the most useful design axis to use...
So for the purpose of crossover design, I will likely use the 30 degree response as a primary design axis.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
For Mr. Manning's plans for CC and a possible tower TMWW using a Duelund crossover, evaluating small form factor tweeters is a necessity to minimize the CTC for the midrange to tweeter.
While there are many small form factor tweeters on the market, many have undesirable response irregularities that makes the crossover development more difficult. The BliesMA T25A is one of the more promising ones available at a reasonable price...
With this small form factor, it seems the best method for rebating for the driver frame and motor is the use of lager Forstner bits... and an additional rear backing plate.
This rebel scum is curious about the width of the test board. It seems .... inadequate .... to the grandeur of your lordship, and also to the typical 20"+ width of such a weapon as this. For testing purposes, is the baffle step diffraction irrelevant for testing purposes?Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
Indeed, with a 2-3/4" Forstner for the flange rebate, and 2" for the magnet bore, this was very straightforward and quick to fabricate- anyone doing this at home must remember to set their drill press for the lowest speed and highest torque. This is not a job for a hand held drill...
In the frequency range the tweeter operates in, above 1kHz, BSC per se is not of great interest... ultimately is it wise to consider using Edge or a similar tool if there are unusual baffle conditions to to consider. But nothing beats physical testing.
It should be noted that this particular test enclosure and baffle size (Denovo 0.56 cu ft, 8.5" wide (or if using additional side panels and a replacement center baffle, as for the 6" MT test enclosure, to achieve 10" baffle width) correlates quite well with the Ardent D baffle for these frequency ranges. Data taken on that baffle used for crossover designs in VituixCAD has generated very good agreement overall.
For the CC, a test enclosure has been built 10" high and 32" wide. This will be for a more definitive design than a basic POC effort, of course. That one may end up being configured as a four woofer 3.5 way, using Anarchy 554, while the basic design to test will use two 558 for now.
Of course, a dual 708 configuration is possible, and parts are on hand... timetable and time resources are the key question there.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
current production RS52 not recommended for new designs
2020 is teaching the value of using strict design processes for design assessment and development. And having NOS parts on hand may not be a valid choice for initiating designs for parts still in production.
Keeping to the concept that pictures are worth many words, I will share a few... just a few, which tell the story. This first post will focus on data for the 2008 versions purchased and evaluated by Jonmarsh for the NeoD CC. Some of these have still been on hand, in long term carbonite storage.
First, an Excel plot of the factory FRD file...
Next, a plot of the locally measured response in the test cabinet target:
Next, a plot of the locally measured impedance:
And a plot of the RC2 (Release Candidate 2) version of the NeoD CC system response with crossover:
The results at this point were most satisfactory- using a modified three way crossover topology derived from the Duelund but not a pure Duelund in order to accommodate the time offset of the HiVi D6.8 woofers.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
And now, the 2020 data
For consistency checks, a number of RS52 samples were purchased this year. They are all consistent with each other, but not with the 2008 version.
Impedance plot:
This is from a "late model" RS52 mounted in the Ardent test cabinet in the living room. Note the difference in the shape and height of the peak, as well as the different frequency. My Force intuition suggests a surround resonance, and I have disassembled units and investigated remedial damping efforts, but with no success. The height of the impedance curve suggests a high energy EMF/mechanical resonance.
And the response curve of the 2020 parts confirms it...
Nearfield checks confirm this is not at all baffle related.
It appears that below the resonant point at 1800Hz, the radiation is anti-phase, and above it in phase.
So much for using a surround as a radiating element.
This is the part where I say, as in the movies, there is another... that may be strong in the Force, and suitable for the current project. They should arrive mid week...DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Quite similar, considering different equipment for measurement.
Thank you for sharing this here.
I will forward the data I have taken to Dayton Audio.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
The MD60N might be an interesting option, but it's not yet available. I'll be curious which bounty hunter catches the assignment.Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
I am also looking forward to the MD60N too.
Though it seems it can only flat down to about 800Hz not low as RS52
I know the SPL on RS52 in the low frequency is limited due to XMAX but it still making RS52 more applicable.
One of my planning project is a pair of something like cello stradivari grandmaster to use 8 RS52s per side crossing about 500-600Hz.- Bottom
Comment
-
I am also looking forward to the MD60N too.
Though it seems it can only flat down to about 800Hz not low as RS52
I know the SPL on RS52 in the low frequency is limited due to XMAX but it still making RS52 more applicable.
One of my planning project is a pair of something like cello stradivari grandmaster to use 8 RS52s per side crossing about 500-600Hz.
Look at the relative SPL at any given drive level, not just the shape of the curve. The RS52 is ~90dB at 2.83v while the MD60N is ~95dB. Based upon factory curves, both can do about 90dB flat down to 500HZ, but you would need a nominal shelving filter to get the MD60 to be "flat". Of course, as a mid-range, flat isn't necessarily what you really want anyway (nice inverted parabola when I've been playing with three-ways....). Depending upon slopes and SPL, they should fill a similar niche, but the MD60 may need more parts for any given application (or fewer if the natural curve and output fits the need).
It does look like the spec sheet has gone missing from most places, but Intertechnik still has it published:
Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
What you screamin' for, every five minutes there's a bomb or something. I'm leavin' Bzzzzzzz!
5th Element, otherwise known as Matt.
Now with website. www.5een.co.uk Still under construction.- Bottom
Comment
-
Detailed examinations were made.
Three new drivers from a shipping lot of 4 and 1 legacy driver were examined in detail, including a fairly complete tear down. Visibly, they do not appear different, and apart from the difference in the 1-2kHz impedance curve and response, behavior is quite similar. This included checks of the impedance curve with each step of teardown and verifying that the rear chamber seal for each driver was intact when testing was done, without or without a faceplate.
Both near field (6") and farfield tests were done, and results agreed closely.
The driver construction is not meant to be disassembled and repaired. The motor plate with Neodymium ring magnet floats between the plastic front and the plastic rear chamber. It is sealed/clamped to the front by the screws mounting the metal front plate. Those screws can be re-attached after removing the front plate, as shown above. At this point, the impedance and near field response curves before and after removing the front plate are essentially identical to each other for any specific driver, legacy or new.
If the dome and surround assembly are cut away, one can see the copper gap rings, and the screening used venting to the rear chamber and containing the chamber synthetic acoustical stuffing.
This stuffing is substantial and packed into the rear chamber carefully. The rear chamber is normally completely sealed from the outside. If the mounting screws for the top plate are loosened, then there is an air leak which shows up in the impedance curve as a collapse of the primary resonance and a double peak resonance above and below replacing it.
These visible aspects of construction did not vary between the NOS legacy driver and the new purchases.
A driver modification investigation was pursued, using damping compounds in several experiments on the surround edges- it seems possible from that to impact the level of the anti phase radiation, but not the in phase peak. This allows a usable response curve by implementing a series LCR network to tune down the peak, but this is an extra expense, and still leaves a slight dip in the response.
This information along with a selection of the measurements made were forwarded to Dayton Audio. No response so far.Last edited by Evil Twin; 20 August 2020, 17:59 Thursday.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
- Bottom
Comment
-
There is a relationship between the driver diaphragm diameter and the usable near field measurement distance and frequency - you will get high frequency cancellation above 2500Hz with this close a position. but this technique can be used well at a distance of about 2" for this driver. A general if inexact method I use is the measuring distance the same as the driver diaphragm diameter- it will get you in the ball park.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
It is also interesting to check HiFi Compass measurements- he measured a set of RS52AN which is what I would call "legacy configuration". He also measured the RS52FN, which has the surround resonance and other issues, and, well, there's no charitable way to put it, it is a hot mess.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
No response from Dayton Audio. I did not contact PE- they are just a distributor.
With selective doping of the surround and adjustments to the crossover, I can get some reasonable results. Later today I expect to be testing an alternative driver.
I was not as productive yesterday as I should have been, because I was fitting a new head unit to the entertainment system in my cargo shuttle. That effort proved to be quite successful in the end, and I was able to integrate and test advanced communications protocol in this older utility vehicle never intended to support them... (Apple CarPlay). There are now effective alternatives for extensive music library playback also... between USB memory systems and the CarPlay functionality.
DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Very nice...surprised to see it has a stand alone clock! I had to bite a similar bullet a month ago when my factory cd player started acting up. The in dash cd changer had died a couple years ago, and the volume control was temperamental. A Kenwood Excelon with a USB connection went in scarily easy, and I removed the CD changer, leaving its faceplate mounted so it looks factory and weighs a few pounds less to boot- Bottom
Comment
-
It still may be worth contacting parts express because of the relationship they have with Dayton. They seem to have done a great job digging into the issues reported on the rs28 tweeters when they went though unexpected changes.- Bottom
Comment
-
Is Dayton audio a subsidiary of parts express?
https://trademark.trademarkia.com/da...-76165629.htmlWelcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
An alternative has passed some basic preliminary tests, and a system evaluation should occur soon...
The primary region polar response from 0-30 degrees appears quite usable for the intended application...
There are no diagnostic issues to be seen in the impedance curve...
Basic harmonic distortion results are within the desired parameter range...
It even appears to be reasonably mechanically compatible with the RS52, facilitating use in this current project...DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
So, did my electronic communication reach your realm with unscathed and usable information as to a replacement?
If so- this will be good!
Wolf- Bottom
Comment
-
I've heard both the EM1308 and the MDM55, and really like both of them. I've also heard both of the RS52 variants, and think they sound like #$%^.
(I see it is the EM1308 I recommended now, but don't know if I was the one whom tipped off the OP or not.)
Later,
Wolf- Bottom
Comment
-
If you heard versions of the RS52AN that measure like the recent ones I have purchased, I can certainly understand the reason for your opinion. Troels and others point out that the kind of resonance and abrupt phase shift described above is quite audible, apart from the frequency response issues, and I concur with that judgement.
The part I have been evaluating is the EM1308, and I have five on hand currently, just to be sure that I did not get a "lucky" golden sample pair.
Now, just recently I did hear back from a Parts Express technical advisor- note, I did not write Parts Express, but Dayton Audio.
He attempted to send a picture showing the response of a golden sample and four recent production drivers, but the picture did not come through in the email- possibly it was on a networked drive in their facility.
He proposed that based on the impedance and response anomaly that there is some sort of resonance affecting my results, and questioned if I had the drivers securely fastened to a baffle, and that the baffle it self might be resonating...
I provided him with a picture of the test baffle,
Which is fabricated from two layers of 1-1/2" pin block maple ply, which I doubted had a resonance at 1800-1900 Hz of that magnitude. Also, the resonance in the impedance curve is quite visible and not changed in level for testing the drivers on a baffle or free air.
I also offered further background information-
Impedance and polar measurements from HiFi Compass of the original RS52AN, which correlate well with my measurements on NOS drivers...
And measurements by HiFi compass of the newer RS52FN, which exhibits similar impedance irregularities as the RS52 I have purchased recently.
As well as completely unacceptable polar response
I informed him I was not expecting action or refunds from Parts Express, but wanted the management at Dayton to be informed that their appears to be a process change in the drivers since their introduction, and that the current behavior of what they are shipping may not be what they expect, or their customers prefer.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Center Channel
What is the "Duelund concept"?Welcome to Rivendell, Mr. Anderson.- Bottom
Comment
-
5 more new unopened RS52AN were taken from the nearby off site storage and tested. All exhibited the signature impedance problem. All have been discarded, along with the ones in process here for testing.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
- Bottom
Comment
-
Relatively so... but unlike a Tymphany DA25, or the original version of the RS52AN, the value proposition in the currently shipping version is not there.
Altogether, a dozen were discarded today.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Here are the normalized equations for the total transfer function (which sums to unity), the low frequency transfer function, the midrange, and the high frequency transfer function. There is only one gain coefficient aleph which is used as a main component of the polynomial equations throughout.
Here, I plot the combined Low, mid, and high frequency transfer functions for different values of aleph, with the center frequency set to 1600Hz, a reasonable choice, for example, for dome midrange drivers.
The summed phase of alignment calculated above is calculated here:
NOTE that the midrange driver is connected out of phase, for the driver phase alignments to match.
The target being used for a current design study is shown below- note the level marker lines (this work was all done in MathCAD, but Excel can be used for this kind of calculation also)
A necessary requirement for this configuration is that the drivers all have the same Z axis origin- usually the front baffle. This is a difficult requirement to fulfill with standard drivers. Conventional cone drivers have their acoustic origin behind the baffle mounting plane.
As bear has noted, if properly calculated and implemented, the drivers will all be in the same phase at any frequency- up to a point, usually being the response limitations of the driver. What I have found in practice is that it is necessary to assure the response and phase tracking down to about the -18 to -20dB level. As you can see from the last chart, the woofer and tweeter still contribute to the output at the center frequency of the midrange in the chosen alignment.
Using conventional terminology, one would describe the crossover frequencies as being at 650Hz and about 4kHz. But the tweeter needs to be in control to the transfer function target down to about 1600Hz (not too difficult with better tweeters) and the woofer needs to be tracking to the same frequency, in my opinion. I have found subjectively that the midrange response needs to track to about the -12 dB level accurately. And the drivers should not have any spurious resonances or other nonconforming output down to about -24dB.
it is a demanding concept to fulfill. However, it minimizes the total phase shift and group delay for a three way system, and unlike first order butterworth crossovers, does not have vertical lobing.
Electrically, it appears as an approximately 2nd order bandpass for the midrange, and initially a 2nd order function for the woofer and tweeter, moving towards a fourth order function as the filter frequency is farther from the center frequency. This results in some unusual requirements for component sizing.
As you can see for the examples plotted in the first graph, different aleph coefficients allow tuning the spread and levels for the midrange and tweeter- but there are consequences as well as advantages to some of these alternate tunings.
For example, with an Aleph of 4, the demands are somewhat lowered on the woofer and tweeter, as the lower nominal crossover point is moved down to ~425Hz and the upper crossover to 6kHz. But for reasonable level integration, the midrange must cover a range down to -15 db from 200Hz to 12kHz. This is a challenging proposition.
As a three way configuration, this crossover type has a fairly specific step response signature, and there are commercial systems such as those from Tidal and Vimberg which appear to be using this approach, sometimes with diamond midrange drivers in order to cover the wide frequency range of a specific alignment. The impulse response is much different from conventional multi-way crossover approaches with or without elliptic elements, as employed, for example, by Magico or Gauder Akustik.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Another pair of SB Acoustics SB26CDC-C004 are on the way; this set not to be evaluated for a waveguide application, but for possible use in the current three way. The original data was quite promising, but that was taken 18 months ago... a refresher is due, especially since SB Acoustics has updated web site information on this part.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
-
Another pair of SB Acoustics SB26CDC-C004 are on the way; this set not to be evaluated for a waveguide application, but for possible use in the current three way. The original data was quite promising, but that was taken 18 months ago... a refresher is due, especially since SB Acoustics has updated web site information on this part.- Bottom
Comment
-
Yes. There were updates, and a new SPL response graph. This is a relatively high value tweeter, and after researching many tens of parts again recently, it was selected for the next evaluations.DFAL
Dark Force Acoustic Labs
A wholly owned subsidiary of Palpatine Heavy Industries- Bottom
Comment
Comment