Mark Audio Alpair 10M gen 3 in a sealed box

[waxx]

I did build last weekend a new box for this driver i had in an old ported box. As i made a better version of that ported box a while ago, i was not listening to it anymore.

I wanted to explore the BBC/Spendor/Harbeth way of building cabinets with thiner plywood box but damped with bitumen, and a thick front and backplated screwed (not glued) to it. So i made this sealed 11.5L box quick, simulated it in winISD and build it in 2 sunday afternoons. The box is stuffed almost fully on the inside and has 5mm thick bitumen pads on the sidewalls and the top and bottom

I did not take many pictures during the build (as it was ment as a test) but i'm so pleased with the result that my other ported box is in storage now and this took it's place in my waw setup in my living. I wonder why nobody else tried this before?

The driver is an old (as a few years and many hours playing) Mark Audio 10M gen 3. I did not measure it yet, but the few hours playing showed me already that this box is way more accurate in the reproduction, and got bass lower than i expected when put close to a wall. And with the sub it works like magic...

 

[planet10]

The A10.3 does not go as low as A10.2, which we did sealed and mated with a subwoofer, works well.

With 11.5 litre heavily stuffed i expect you will see F6 near /F10 in the low 40s and a Q of less than Bessel (0.58). Will have very good transient response and i’d start with an XO at >170 Hz (an octave above F3, or if you use the natural 2nd order rolloff of the sealed box at about 85 Hz.

dave

 

[waxx]

Well, it works on it's own better than most bookshelf 2 or 3 ways i know, as long as you put it against or close to a wall. in the middle of a room it lacks a bit bass (wich was to be expected).

But i use it now in a WAW config with my big ported scanspeak sub (tuned to 25hz), with the old crossover (180hz serial first order with bsc) for the ported box i had original. I still need to measure it (will only have time next weekend) to rework the crossover. But even with this not ideal crossover, it sounds a lot better. Indeed very good tight transient response and a more balanced open sound (and i already loved this driver in the ported box). I even have the impression that the bump in the high is slightly less pronounced than in the ported version.

And i'm actually considering to change the ported sub to a sealed one also. That scanspeak 26w/8534G000 drivers models rather good in a sealed 80L sub. But i don't have some spare wood availeble like i did with this box so i'll have to order and spend money. This box was made from unused parts of 12 and 18mm birchply sheets of other projects, recycled drivers and other parts of obsolete projects of the past...

 

[JoeyGS]

Wow this is nice ....

Will the same strategy work with the Alpair 10p Gen3? I have a spare pair stashed away

 

[waxx]

It is the 10M Gen3 i use, Dave uses the Gen2. Both can work in a sealed config of that volume.


And i think a part of the great sound mine makes is due to that i use the BBC/Harbeth/Spendor way of making cabinets, wich is thin (12mm) ply damped with bitument, and a screwed on (not glued) front and back from 18mm ply.I did put some windows sealer rubber strip between the box and the front and back plate to make it airtight.

The theory behind it is a study of the BBC done by H.D. Harwood (Harbeth) and R. Matthews for the BBC and it was the prestudy that delivered the infamous LS3/5A and BC1 BBC speakers. You also need to brace is (i made a frame on the inside of the box to do that) as the panels are rather thin.

http://downloads.bbc.co.uk/rd/pubs/reports/1977-03.pdf

 

[waxx]

Wow this is nice ....

Will the same strategy work with the Alpair 10p Gen3? I have a spare pair stashed away

THe 10P (i red it wrong) should work also. It's more or less the same specs, Modelling gives it a slightly bigger box (17L in stead of 11.5) for more or less the same curve and qtc. It will give a little bit extra volume says winisd

 

[planet10]

And i think a part of the great sound mine makes is due to that i use the BBC/Harbeth/Spendor way of making cabinets

This technique is not my favorourate, but in this case, where little bass is reproduced, moving the resonances down in frequency is a viable strategy.

dave

 

[planet10]

A10p and A10.2 sims for completeness.

Worth noting, that at the same Q, A10.2 hits under 80 F3, A10.3 about 90 and the A1-p 5 hz higher. F3 choosen as the figure of merit for XO functions.

dave

 

[waxx]

This technique is not my favorourate, but in this case, where little bass is reproduced, moving the resonances down in frequency is a viable strategy.

dave

I would not use this technique for a sub, that is true, there 18mm or thicker MDF, or even better, good (birch) multiply is a better id. But for tops or low power fullrange speakers this can work great. And those Harbeth speakers are among the best i've ever heared and are all build like that and going rather low.

But when i change my subs to sealed like mentioned before, i'm going the classical way with 18mm ply, and heavy braced internally.

 

[planet10]

No MDF here, particularily a sub, where it is totally wrong. Good plywood is considerably better and your back will thank you.

dave