Measuring my AST2560 in 'the Bordeaux'

[krizork]

Hello everyone,
First time post, although I've been scrolling this forum for quite some time.

My question concerns the production change in the AST2560 (https://www.htguide.com/forum/showth...lease-chime-in).

In januari 2018 I build my own set of Bordeaux speakers and I've been very happy with them overal but can't help wondering if my build has been effected by the 2k-ish dip.



I've bought a measuring microphone, the MiniDSP UMIK-1, and I've done some measurements of the full range of the speaker.
1. at about 4cm from the front of the baffle, microphone positioned between the tweeter/mid woofer
2. at the about 60 cm (2 feet), same height.



During the first (near field) measurement I was quite sure my build wasn't effected, but the further away I measure the less sure I get.

Do you think my measurements conclude to anything? Or do I need to do additional tests, and if so, what is the most effective way of figuring this out (without me damaging the tweeter 8))?

Many thanks in advance!

[Efalegalo]

Krizork,

Try taking another measurement from about 1 meter away. Point the mic approximately between the tweeter and the mid-woofer.

Also, bring the speaker into the middle of the room, away from the corner currently pictured as well as other near boundaries. Make sure the nearest boundary is at least a 1 meter away.

Next - gate the measurements. Go to "Tools" --> click "IR Windows", then under "Right Window (ms):" ----enter say 3.5

Gating will kill measurement resolution below say 500hz, but you should still enough resolution to test whether you have an issue in the 2Kz - 5Khz region.

Lastly - increase the steps for your measurements. A 10db step is too much to pick up any oddities. Go for at most 5db per step.

If you find that you have a dip in the 2Khz - 5Khz region, try moving the mic up and down by few centimeters (e.g 5 or 6 cm) and remeasure. If the dip persists, you may have a similar issue as I did.

Goodluck.

[Dave Bullet]

What you have there @ 60cm shows a great speaker IMHO. Most designers pick a listening axis. This is usually the tweeter. So measuring at short distance and not on the design axis will a) introduce errors associated with nearfield measurements (not taking into account the baffle step which you can see rising in your 4cm distance measurement - but allowed for in the design at your 60cm measurement) and b) introduce a lobe off vertical axis that may emphasize (peak) or recess (dip) a particular frequency band which won't be apparent on the listening axis

The rule of thumb is to measure at a distance at the lowest XO point / 3 or 3 * the CTC spacing between widest drivers (crude and from memory). In the case of most 3 ways - this is at least 1 metre.

As Efalegalo stated - the longer the wavelength (lower the frequency) the bigger the listening space you need (further from first reflection surfaces) to get an accurate measurement. In most normal listening environments this is from ~ 300 to 500Hz and up. Anything below that will include room effects. This isn't a bad thing (i.e. you can use Room EQ wizard to calculate an equalisation curve to compensate for your room) - but it doesn't indicate the actual speaker low end performance.

[krizork]

@Efalegalo and @Dave
Hi guys, many thanks for the replies! I'm refinishing the front baffle right now so I'll be able to make the measurements in a couple of days.
Thanks to your tips I now know how to get an accurate graph of my response!

To be continued...

[Curt C]

Krizork,
Jim and I designed and voiced many of our designs with drivers centered on the baffle. This driver layout has two advantages. One is that there is no left or right side baffles to cut and simplifies the build on the enclosure. You will never build a pair of Jim’s speakers only to find out you made two lefts or two rights, or cut one of the rebates for driver flush mounting on the wrong side. With an eye to the DIY community, Jim has always striven to make his enclosure designs simple to construct, but has not sacrificed anything to quality of construction. This choice also required another choice to be made: That is enclosure alignment in the room.

As all of you know, there are diffraction aberrations that occur when the acoustic energy travels along the surface of the baffle until it hits the acoustic discontinuity that we perceive as the edge of the baffle. The baffle edges in essence become secondary acoustic radiators, and some of the energy travels to the listener, but delayed in time. At frequencies where this delayed energy reaches the listener in phase or out of phase with the direct sound, they sum constructively or destructively respectively. If the drivers are centered on the baffle and the listening axis is perpendicular to the speaker face, the delayed sound from both sides of the baffle arrive delayed by the same amount and the aberrations in response are twice as significant. However if the speaker is not toed into the listener the time of flight is not the same from each baffle edge and it responds in much the same way as offsetting the drivers on the baffle.

The Bordeaux optimally need to be placed in the room with no toe-in. Consequently my measurements were done approximately 10 degrees off the horizontal axis with the mic 1 meter away and the mic height set approximately between the mid and tweeter. When you do your measurements, it might be useful to try it both on axis, and also the intended listening axis.

C

[krizork]

Hi Curt, thank you for the extra info and tips.

I've done the measurements as described by you guys, approx 1meter away, right window at 3.5.



Here are two measurements with the mic height just between the mid-woofer and sweeter, one on-axis and the other is 10' off axis:



I've also altered the height at my off-axis measurements:



To me it looks like there is a dip around the 2k2 region, would you guys agree?

Many thanks again, also to you Curt for sharing your wisdom on this matter.

[Zvu]

You can measure it from 75cm distance and you (or anyone else) can hold the mic in hand. It would be much cleaner measurement, without that much combing, and easier to see the problem.

But from what i see on your measurements, i'd say you have at least 6dB dip between 2kHz and 3kHz.

[Dave Bullet]

Have you got the old or new tweeter? there was a post on here a while ago where the tweeter FR had changed.

Next step for me is to take single driver raw measurements (without crossovers) as a starting point. If that is a ribbon - you may need an inline capacitor of some sort to prevent a short to the amp due to the very low DCR ribbons present (and protect the ribbon!). If you feel unsafe about raw measurements (tweeter only impacted) - then leave the XO in place for each leg and measure each driver independently. Compare these to the designers in box measurements (if these are published).

The above may sound like a lot of work - but it is sometimes less effort overall to systematically eliminate and "step through" a design confirming each component.

[Curt C]

Krizork,

Did you try the crossover posted by Jim here:

https://www.htguide.com/forum/showthread.php?44323-Updated-Bordeaux-crossover-now-available&highlight=bordeaux

C

[Dave Bullet]

I should have added a system impedance sweep. This is a good way of quickly checking XO wiring (although you'll need to manually check driver polarity).

[krizork]

Have you got the old or new tweeter? there was a post on here a while ago where the tweeter FR had changed.

Next step for me is to take single driver raw measurements (without crossovers) as a starting point. If that is a ribbon - you may need an inline capacitor of some sort to prevent a short to the amp due to the very low DCR ribbons present (and protect the ribbon!). If you feel unsafe about raw measurements (tweeter only impacted) - then leave the XO in place for each leg and measure each driver independently. Compare these to the designers in box measurements (if these are published).

The above may sound like a lot of work - but it is sometimes less effort overall to systematically eliminate and "step through" a design confirming each component.

I'm actually trying to figure out if I have the old or the new tweeter, because if it's the new tweeter I need to build me that x-over mentioned by Curt.
I was hoping the full range measurement of the speaker would be enough to figure out which tweeter version I have.

@Efalegalo, as you have ran into the same issue, does my freq-response match yours?
@Dave, what kind of inline capacitor are we talking about here? I'm sorry if it's a noob question but I have zero crossover building experience.

[krizork]

Krizork,

Did you try the crossover posted by Jim here:

https://www.htguide.com/forum/showthread.php?44323-Updated-Bordeaux-crossover-now-available&highlight=bordeaux

C

Hi Curt, looking at my response graph, do you think I have the newer AST2560?
If yes, I need to go and buy me some crossover parts. It costs about 200$ here in Europe so I'd like to be sure.

[5th element]

You can always buy cheap non polar electrolytic capacitors, some cored inductors and basic wire wound resistors as a trial xover. That shouldn't be expensive at all if bought from the right place.

Like from here

https://www.intertechnik.de/shop/weichenbauteile/kondensatoren/elko-rauhe-folie/1768,de,606

I would assume that the xover redesign an would reuse a lot of the original components anyway.

[Dave Bullet]

Hi Curt, looking at my response graph, do you think I have the newer AST2560?
If yes, I need to go and buy me some crossover parts. It costs about 200$ here in Europe so I'd like to be sure.

If it isn't obvious from labels printed on the actual driver (such as MMYY / period of manufacture or batch numbers), I suggest you do a raw measurement WITH protective series capacitor to compare which version you have. .... I believe a "raw" response was posted in the thread.

[Efalegalo]

Krizork,

Looks like you too are experiencing the broad dip around 2Khz or so as I experienced. The new parts are minimal and have low values and will be rather inexpensive. I would do as 5th Element suggested, but really inexpensive parts to test which crossover works best and order premium parts once confirmed.

Thanks


Sent from my iPhone using Tapatalk

[krizork]

Hi all,
5th Element, Dave and Efalegalo thank you for the help and tips!
I'm emailing Mark from Meniscusaudio if he can hook me up with replacement parts (I originally also bought the crossovers from them). It looks like the 100uF capacitor is reusable but the rest of the parts have all different values.
I'm not much of a tweaker so I'd rather just get the crossover Efalegalo came up and use that, with premium parts from the get-go.

I'll get back to you guys with an update when I've made some progress!
Thanks again!

[sfdoddsy]

It may be a bit heretical, but I’m a believer in doing listening position measurements as well. After all, that’s where your ears are.

Certainly in your case the tweeter measures worse the farther away you get.

Nice looking build though. I hope you can sort it out.

[krizork]

The new crossovers are in! Mark from Meniscusaudio send me new x-overs for the tweeter and mid-woofer.
I could re-use the 100uF cap from the original ones.

The difference is very noticeable!






With the old x-overs I always found the voices a bit low in volume compared to the rest of the instruments (at first I thought maybe this is just how these speakers sound) but now with the correct x-overs it is a lot better.
I could get my receiver's Audyssey to correct this somewhat but this has more focus and more articulate. The right frequencies coming out of the correct speakers apparently helps (who'd a thought that?!).
Stereo imaging has improved as well, where with the old x-overs I had to sit exactly dead-center for the voices to find it's spot in the center, now there isn't that much difference in the stereo-image when I move to the left or right.

I also took my chance to refinish the speakers a bit, instead of rounded baffle edges I now have a 45' angle which I like a lot more. And instead of acrylic polyurethane I now have a hard-wax oil (Osmo PolyX) finish which looks great.





Thanks everyone for the help and your thoughts!

[Jim Holtz]

Excellent! I'm very glad to hear you had a successful outcome!

Jim

[fish fingers]

Loving that v natural wood finish..