VictoriaGuy said:
Unfortunately, this reduces Novak's and Pi Align's boxes by ~ a cubic foot, so just made them worse performing at a lower efficiency
. The basic lumped parameters T/S box formula is Vb = 20*Vas*Qts^3.3, so Qts is the big deal here.
I mean if Freddy's specs are accurate/typical though, then it's like a car dealer advertising whatever its econo-car model is with a relatively powerful motor for 'x' $$$, but at that price yours arrives with an asthmatic little four cylinder. This is plain and simple 'bait n' switch' marketing IF TRUE.
Without more folks measuring theirs in a similar way such as with a WT3 used per its test procedure instructions, then all Freddy's measurements do is raise a Caveat Emptor (buyer beware) 'flag', but other data has been posted to imply his are well within the acceptable 10% measurement accuracy 'window', so for now I'm not going to waste any more of my quality time helping folks with AN products without measured specs.
GM
GM said:
Interesting point.
Perhaps Freddi and some of the other folks could measure some Fostex drivers with the same setup and see how close they are to the factory specs as well? (Fair is fair.) Lots of design time seems to have been put into enclosures for the Fostex drivers, so I assume the factory specs are 'trusted'?
I just don't have a lot of confidence in results from amateur basement labs without some replication- especially when they may be used to question the ethics of a driver designer/seller.
Anyhow, I'm quite happy with my AN10s- 'as is' (certainly a huge improvement over my fostex167s in Brines 1600s). I was just (like most of us?) looking for something 'a bit better'.
GM- thanks for your time on this little project.
Cheers
John
GM said:
Thanks for pointing out the appropriate formula
(..scribbling on the back of the envelope...)
Even with 'acceptable' 10% errors in the Vas and Qts parameters, the Vb could be off by 50%. It certainly gives lots of room for the 'art' in speaker enclosure design as well as the 'science'.
John
Scottmoose said:
Thanks for the reminder- I HAD read that but forgot that you had mentioned Fostex as well.
I know you've done a lot of sims and designs- do you test the drivers yourself, or are there 'reliable' spots on the web where actual specs are listed, or...?
I'm also curious to know your thoughts on why manufacturers would publish inaccurate specs- it seems strange to me. Surely they have the technical facilities to test drivers properly, if they wanted to?
Cheers
John
VictoriaGuy said:
I used to be able to measure units. Most of my gear was sacrificed on the alter of my PhD though, so I try to get reliable average specs. from other sources (owners, distributers & the like). Lowther have been very kind to me in the past too I must say, when I've had direct dealings with them on behalf of other people.
They do. Or most of them do at any rate. There won't be any one reason -if you want a small handful, there are marketing considerations; sample variability; QC; changing driver / component specs., & so on & so forth.
wakibaki said:
If you want to optimise something correctly, that's exactly what you need to do. Sometimes factory specs will get you in the ballpark, often they won't / don't. So always look at them through your best cynical eyeball.
Well I can say thay my WT3 comes very close to specs published by Eminence for their drivers. The guys at Eminence do know how to measure. (tho they are starting to cheat on the sensitivity).
Others with the WT3 have found the same with Eminence.
Ditto for Selenium drivers.
We have to remember, tho, that the specs at the begining of a production run may drift over time. QC says they shouldn't, but they will. And who updates the specs for the current run? Not many, I would guess.
If you want to see specs that are waaaaay out, take a look at the cheaper "pro" drivers. Pure fiction, there.
Others with the WT3 have found the same with Eminence.
Ditto for Selenium drivers.
We have to remember, tho, that the specs at the begining of a production run may drift over time. QC says they shouldn't, but they will. And who updates the specs for the current run? Not many, I would guess.
If you want to see specs that are waaaaay out, take a look at the cheaper "pro" drivers. Pure fiction, there.
VictoriaGuy said:
You're welcome!
I believe you're missing my initial point........ I'm well aware that many manufacturer's specs are off enough that using a textbook T/S max flat alignment will be anything but flat, but if the specs fall within the ~ accepted 10% manufacturing tolerance, then when I do a typical (i.e. oversize) MLTL and especially a ~ Vb = Vas/Fb = Fs alignment that I can be secure in the knowledge that with a bit of vent fine tuning it will perform well. This is NOT the case with the Super 10 based on the current available information which requires a considerably larger Vb than Vas.
WRT the Fostex and most other popular brands, mine, Scott's, etc. 'success' rates using published specs for MLTLs, ML-horns, TQWTs has been near/at 100% AFAIK, so as far as I'm concerned they are close enough. The same can't be said for Lowthers, the late Moth Audio Cicada and a few others though.
Oh really?! Please elaborate WRT actual box performance differences, not those in the mids/HF.
GM
Hi John - - what box and port do you have for An10?
I'd have to take one outside to see if there's any evidence of LF peaking in a 2.4 cubic foot box. For me that little box and 41Hz tuning were not good. I don't see a point of a really large box.
with most recorded material (not Earthworks drum demo) the mids and highs swamp the "bass" - - AN10's bass is pretty solid in K15 and cone excursion is reduced vs reflex.
FWIW here's a crude inroom comparison of a little Yorkville C190 to Behinger "Truth" to AN10 in 2.4CF/41Hz reflex.
btw - -I've seen that 3k peak more than once - anyone here notch their AN10 around that spot?
I'd have to take one outside to see if there's any evidence of LF peaking in a 2.4 cubic foot box. For me that little box and 41Hz tuning were not good. I don't see a point of a really large box.
with most recorded material (not Earthworks drum demo) the mids and highs swamp the "bass" - - AN10's bass is pretty solid in K15 and cone excursion is reduced vs reflex.
FWIW here's a crude inroom comparison of a little Yorkville C190 to Behinger "Truth" to AN10 in 2.4CF/41Hz reflex.
btw - -I've seen that 3k peak more than once - anyone here notch their AN10 around that spot?
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
for anybody interested, on another thread Gainphile is helping me figuring out an Openbaffle - very narrow baffle - solution to try with AN super 8", seems like they can work down to 450hz on a 25cm baffle...
you can found it here http://www.diyaudio.com/forums/showthread.php?s=&postid=1729451#post1729451
in the next week i'll be making up the 2.8mkII and reporting on the performance,
is there any thread on this forum to make a simple cheap reliable measurement tool?... ok i know, i know, i'll be searching first
you can found it here http://www.diyaudio.com/forums/showthread.php?s=&postid=1729451#post1729451
in the next week i'll be making up the 2.8mkII and reporting on the performance,
is there any thread on this forum to make a simple cheap reliable measurement tool?... ok i know, i know, i'll be searching first
For the sake of interest, here are some measured specs on the AN Super12, found on http://diyaudioprojects.com/Forum/viewtopic.php?f=11&t=21&start=50 -see 4 posts up from the bottom. However, for simplicity,
Measured / Factory published
Re 6.971 / 7.2
Fs 43.74 / 33.48
Qts .6965 / .487
Qes .8252 / .528
Qms 4.467 / 6.347
Le .2126mh / n/a
Mms n/a / 28.195
Vas 8.966 /10.29
Bl 6.432 /8.995
So, not a whole lot of motor power. Needs a hell of a large box, although a more modestly sized sealed cabinet would be a decent enough option.
Measured / Factory published
Re 6.971 / 7.2
Fs 43.74 / 33.48
Qts .6965 / .487
Qes .8252 / .528
Qms 4.467 / 6.347
Le .2126mh / n/a
Mms n/a / 28.195
Vas 8.966 /10.29
Bl 6.432 /8.995
So, not a whole lot of motor power. Needs a hell of a large box, although a more modestly sized sealed cabinet would be a decent enough option.
I caught that too - assuming that's the trend, qts could come in ~40% higher than spec - thats not a lot of motor for a 12 but HF should be extended and with good sensitivity - an ancient SP15 sounds more muscular imo than AN10 on plucked string bass - -- seems like sealed box AN12 would be very un-dynamic (?) . I've got an old Semprini 12 with qt~0.7 in a Karlson 12 playing so I'm one to talk :^) (it sounds pretty good in places with a helper tweeter using 0.75uF highpass LOL)
lets see some measured T-S and measured BLH, etc to get an idea how close sims come and maybe to get a grip on measurement technique and where it fails subjectively . (I'd imagine someone with real pretty cabinets would not want to drag them outdoors for testing - if outdoors data can be extrapolated to get indoors performance.
how sensitive is the average listener to a "bad" alignment? what happens if they use high Z-out amp which wrecks the alignment? can "hot" midrange and HF tonal balance plus baffle step dropoff mask-bad alignment? can a weak distorting amp mess with perception of "bass"? do certain chord clusters sound deeper than their real LF content?
lets see some measured T-S and measured BLH, etc to get an idea how close sims come and maybe to get a grip on measurement technique and where it fails subjectively . (I'd imagine someone with real pretty cabinets would not want to drag them outdoors for testing - if outdoors data can be extrapolated to get indoors performance.
how sensitive is the average listener to a "bad" alignment? what happens if they use high Z-out amp which wrecks the alignment? can "hot" midrange and HF tonal balance plus baffle step dropoff mask-bad alignment? can a weak distorting amp mess with perception of "bass"? do certain chord clusters sound deeper than their real LF content?
All good philosophical points Freddi. Who knows? We're back to what the individual is happy with again. And most of the time, we're assuming a 'perfect' amp in the mix that isn't altering the behaviour of the speaker system, let alone reflecting (no pun intended) on what the room is going to do.
Yeah, there'd probably be a dynamic trade-off with the sealed box -that said, when I said 'modestly sized' I was thinking in the order of about 5ft^3 (these things being relative), which shouldn't completely cripple it.
One thing that occured to me -with the interest in the AN12in growing (and why not), perhaps it's also time to revisit the Hammer Dynamics unit. It's a 12in unit after all, albeit a wide-band, rather than ~FR. Just wish they'd measured the drivers as the specs are incomplete. Interesting situation though, because people have had major reservations about that enclosure too, so I suppose there's an analogy to be drawn somewhere. I suppose we could add the Eminence Beta12LTA that formed the basis of the Hammer driver to that too -they were quite popular a while back, but seem to have been forgotten of late.
Who knows? We're back to what the individual is happy with again. And most of the time, we're assuming a 'perfect' amp in the mix that isn't altering the behaviour of the speaker system, let alone reflecting (no pun intended) on what the room is going to do. Yeah, there'd probably be a dynamic trade-off with the sealed box -that said, when I said 'modestly sized' I was thinking in the order of about 5ft^3 (these things being relative), which shouldn't completely cripple it.
One thing that occured to me -with the interest in the AN12in growing (and why not), perhaps it's also time to revisit the Hammer Dynamics unit. It's a 12in unit after all, albeit a wide-band, rather than ~FR. Just wish they'd measured the drivers as the specs are incomplete. Interesting situation though, because people have had major reservations about that enclosure too, so I suppose there's an analogy to be drawn somewhere. I suppose we could add the Eminence Beta12LTA that formed the basis of the Hammer driver to that too -they were quite popular a while back, but seem to have been forgotten of late.
wakibaki said:
Scottmoose said:
The problem is, Scott, that this isn't much of a basis on which to build a buying strategy for most people.
With Lowther and Fostex out of the running the choice is becoming somewhat limited anyway.
Some of us are not tuned in to the folk grapevine, and anyway, a great deal of folk knowledge is superstition.
I have no inclination to indulge in a round of experimental buying and testing. I certainly wouldn't be thinking about measuring them until they had been run for some considerable time. Open baffle is too fragile for my circumstances, so it's a box.
Therefore my first port of call and my recommendation to any reader when seeking information about any piece of equipment remains the manufacturers published specification.
I am often the first to castigate correspondents for their failure to apply rigorous analysis, but there is such a thing as over-dependence. There is an important component of engineering practice which is beimg ignored here. This is empiricism.
Mathematical models vary in their accuracy.
A loudspeaker is a complicated system. Many of the interactiog components, despite the best efforts of manufacturers, are possessed of ill defined or poorly understood qualities and behaviours such as e.g. the paper in the cone or the suspension.
Anyone comparing a simulated plot of a loudspeaker with a real one will instantly notice that the real one has a lot more departures from a smooth response than does the simulated one.
Cut-and-try is a perfectly respectable engineering technique, particularly if it gets better results than a sophisticated but necessarily incomplete model.
w
All my eyeballs are cynical.
freddi said:
The on-axis response of my pair jumps up at 2KHz and then goes pretty flat
up to 20KHz. If you are eq'ing for flat response probably you will want a
baffle step filter.
From actual listening, you may want a broad band notch centered around
4KHz with about a 6 dB depth. Depends on the angle and your taste -
Your Mileage WILL Vary.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.