advertisement


Mike's speaker thread

Mike P

Trade: Pickwell Audio
Here we go folks, this could be a long one!

For the past 4 or 5 years I've been a very satisfied owner of a pair of A.O.S Studio 100 ATC XL loudspeakers. Some of you may have seen them in the classifieds recently as I had plans to sell them to fund a sideways step to a pair of big 15" Tannoys.

To cut along story short, my wife wasn't best pleased with the idea of some massive Arden sized enclosures in the lounge and so it was decided that I would keep the Studio 100's. My apologies to Toby who was interested in buying them!

I expect most of you won't be familiar with AOS but they are a German high end kit speaker company. Here's a link to their website:

http://www.aos-lautsprecher.de/

The Studio 100 can be found in the 'Classic' speaker kit section.

So my plan now is to spend some time, money and effort making my pair of Studio 100's as nice as possible and I have just embarked on a major refurbishment project.

But before I get into what I want to do to them here are some photos of the speakers in the state that I bought them in.

P1100656 by Michael Pickwell, on Flickr

P1100653 by Michael Pickwell, on Flickr


As you can see from the photos they are a large floorstanding transmission line design. The drive units are:

Tweeters: ScanSpeak D2904/980007 (AOS bespoke version)
Mid: ScanSpeak 18W16545 K
Woofers: ATC 24W (AOS bespoke version)

The cabinets are primarily made from 21mm MDF. The tweeters and mid-range drivers are housed in the upper chamber which has a sloped floor and a rear facing small port.

The woofers are housed in the lower chamber, which is of course the transmission line labyrinth.

P1100655 by Michael Pickwell, on Flickr

Time for a tea break, more to follow!
 
Tea break over, let's continue....

Crossovers:

The crossovers are made with very high quality components and are housed in the cabinets.

The low pass filter for the woofer consists of a large Mundorf transformer and big ClariryCap SA film cap which are fixed to the cabinet back wall in the lower chamber.

The crossovers for the mid-range and tweeters are on separate PCB's which are mounted on the floor of the upper chamber.

The speakers are set up for bi-wiring with separate speaker cable binding posts for the bass and mid/treble (assuming you don't use the links).

This is the only circuit diagram I have for the crossover but it isn't completely correct for my speakers. In my speakers the capacitor in series with the tweeter is actually 8.2uF and the resistor is apparently 1R8 from the colour code (I'll check it later).

Also the 2R7 resistor in series with the mid-range driver unit -ive terminal is not fitted.

I assume that the differences in the crossovers are because my speakers are fitted with the ScanSpeak D2904/980007 dome tweeters rather than the R 2904/7000 ring radiators shown on the schematic.

20171014_132336[1] by Michael Pickwell, on Flickr

Here are some photos of the crossovers removed from the cabinets.

20171012_114120 by Michael Pickwell, on Flickr

20171012_114127 by Michael Pickwell, on Flickr

20171012_114109 by Michael Pickwell, on Flickr

The 68uF Clarity Caps are absolutely huge!
 
Over the last couple of days I have disassembled the speakers, removing the drive units, crossovers and (almost) all of the acoustic damping material.

I found that the original builder of the speakers had gone a little 'freestyle' with the damping material and hadn't followed the build instructions supplied with the kit. This is one aspect I hope to be able to improve upon in the rebuild.

I also want to smarten up the cabinets with some new wood veneer. The original veneer looked pretty good from distance but on closer inspection it was bubbling in places. I want these speakers to have a high class aesthetic so I'm going to take the opportunity to re-veneer them while they are in pieces.

Yesterday I stripped off all the old veneer. Here's what they look like now!

20171014_121326[1] by Michael Pickwell, on Flickr

The white stuff on the outside of the cabinets is Isopon filler which I'm using to fill some minor surface imperfections.
 
I can't see what you would change about them?

Yeah, I haven't gotten around to explaining that very well yet.

So what am I hoping to achieve?

  • Smarten them up with some nice new wood veneer.
  • Improve the execution of the internal acoustic damping and also the damping cabinet resonances.
  • Improve the crossover components where applicable.
  • The grilles have been attached with Velcro strip which looks unsightly when the grilles are off. I want to do something much neater, perhaps using concealed small neodymium magnets.
  • Repaint the fronts after removing the Velcro tabs and sorting out a better grille mounting system.
  • The Studio 100 has gone through various upgrades over the years and one of the major upgrades has been the progression to ever better tweeters. My pair has the ScanSpeak D2904 metal dome tweeters. This was later upgraded to the R2904 ring radiator (also from ScanSpeak). The ultimate incarnation uses the D3004 Beryllium dome tweeter. While I have the speakers apart I'm also considering an upgrade to better tweeters.
  • Another possibility is a change to the Studio 100 MKII crossover specification. My German language skills are almost non-existent but using Google translate it seems that this basically involves changing the mid-range from a 12dB (per octave) roll-off to a 6dB roll-off. This is done by modifying the existing mid-range crossover. If any Fishies here speak German I'd appreciate your help. Similarly, I could use some advice from anyone fully versed in loudspeaker crossover theory.
  • I'm also looking at making some sort of better plinth/base for the cabinets to improve their stability and add some protection from the vacuum cleaner.
  • Finally, I'm musing over the costs/benefits of going down the external crossover route.
 
I look forward to following the project. Do you know if AOS is now owned by Scan-Speak or perhaps the company that owns Scan-Speak? I note they list a price for the ATC midrange which Wilmslow no longer sell. Do you know if they are still able to sell this driver to DIYers?

The correct amount of relatively light damping in a transmission line is crucial to it's performance. It needs to let through relatively unimpeded the lowest pipe resonance and impeded as much as possible the higher pipe resonances. Do you know the amount and type of damping to achieve this for your speaker?

I would have some doubts about 6 dB/octave slopes and "better" crossover components. Are you in a position to thoroughly measure what the current speaker is doing and hence to identify where improvements may lie based on something perhaps a bit closer to evidence?

An external active DSP crossover offers both almost limitless flexibility to try things and an optimum performance. It may not be a large improvement if the current implementation is good but it will be safe to assume an improvement. Is this what you mean by an external crossover?
 
I look forward to following the project. Do you know if AOS is now owned by Scan-Speak or perhaps the company that owns Scan-Speak? I note they list a price for the ATC midrange which Wilmslow no longer sell. Do you know if they are still able to sell this driver to DIYers?

Sorry, I've no idea. AOS use a lot of Scan-Speak but they also use ATC and Volt too. I've a vague idea that they are now using Volt soft dome mids in place of the no longer available ATC mids.

The correct amount of relatively light damping in a transmission line is crucial to it's performance. It needs to let through relatively unimpeded the lowest pipe resonance and impeded as much as possible the higher pipe resonances. Do you know the amount and type of damping to achieve this for your speaker?

Fortunately I don't have to guess, it's in the build sheet for the speakers which call for bitumen sheet followed by felt mat on the side walls and profiled acoustic foam in the labyrinth. The chap who built mine left out the bitumen sheet in both the upper and lower chambers. Also, the upper chamber he stuffed full of some sort of fibre wadding putting in way more than the build instructions call for.

I would have some doubts about 6 dB/octave slopes and "better" crossover components. Are you in a position to thoroughly measure what the current speaker is doing and hence to identify where improvements may lie based on something perhaps a bit closer to evidence?

I didn't measure the speakers before taking them apart and it's obviously not possible now. It's not me who is saying that the 6dB crossover slope sounds better, it is AOS themselves that advocate the modification backed up by whatever testing they've done and positive testimonials from many happy owners who've made the retrospective modification. I too am pretty sceptical though as they sound great the way they are.

An external active DSP crossover offers both almost limitless flexibility to try things and an optimum performance. It may not be a large improvement if the current implementation is good but it will be safe to assume an improvement. Is this what you mean by an external crossover?

No, I was thinking about at external passive crossover to start with using what I have. Of course I could easily change this at a later date to an active crossover if I wanted when funds allow. The first decision I need to make is whether to rebuild the speakers with the crossovers inside or not.
 
Before I can start the rebuild I need to make a decision as to whether I'm going to put the crossovers back inside the cabinets or if I'm going to leave them outside as an external passive crossover.

Reading up on the subject the benefits of an external crossover are purported to be:

  • Crossover components are much less affected by vibration than they would be inside the cabinets. Vibration creates distortion.
  • It is much easier to tweak the crossover and change components. This seems like a real benefit to me, particularly with regard to getting the series resistor for the tweeters correct if I decide to upgrade the tweeters.
  • It is easier to get the inductors a good distance apart and in correct physical orientation to each other.

And the downsides:

  • Considerably more expensive by the time you've built boxes with all the connectors etc and the extra cables.
  • Less neat, more space taken up in the lounge.
  • Someone could accidentally power the speakers without crossovers and damage the drive units.

I really need to make a decision on this as it is holding up the rest of the rebuild.
 
Fortunately I don't have to guess, it's in the build sheet for the speakers which call for bitumen sheet followed by felt mat on the side walls and profiled acoustic foam in the labyrinth.
As I mentioned earlier the type, amount and location of damping material in the pipe needs to be right with little wriggle room. If it sounds good to you and you have not performed measurements to check the performance then there doesn't seem to be much of a case for changing anything.

The chap who built mine left out the bitumen sheet in both the upper and lower chambers.
The cabinet is a single structure with the midrange driver directly bolted to it and so the lowest resonances in the side walls are going to be driven relatively strongly. The labyrinth partitioning will likely raise the frequency of these resonances into a range where the ear is more sensitive. Although extensional damping is not particularly efficient a little damping will reduce high Q resonances a lot. Since MDF has little internal damping the bitumen pads are likely to significantly reduce the radiated sound from the resonant peaks in those side walls with their very large radiating area. Adding damping pads looks wise to me.

Also, the upper chamber he stuffed full of some sort of fibre wadding putting in way more than the build instructions call for.
Yes this looks wise to me. Did he also block the port? Not that this should matter too much if there is a decent amount of damping material used.

Ports tuned to a high frequency that use resonance to boost the output of midrange drivers is not a particularly sensible thing to do. It degrades the transient response and as the frequency rises squirts sound at the front wall that is progressively not in phase with the sound from the front of the midrange driver. It should be easy enough to perform an experiment to determine which you prefer.

It should be noted that removing the resonance will change how the midrange rolls-off and depending on where this roll-off lies relative to the high pass filter in the crossover it may change the output in the crossover region between woofer and midrange.

I didn't measure the speakers before taking them apart and it's obviously not possible now. It's not me who is saying that the 6dB crossover slope sounds better, it is AOS themselves that advocate the modification backed up by whatever testing they've done and positive testimonials from many happy owners who've made the retrospective modification. I too am pretty sceptical though as they sound great the way they are.
There is little doubt that some audiophiles derive significant pleasure from changing this and that and having a listen guided by the enthusiasms of others with absolutely no interest in getting on top of the technical side in order to use it to guide what to do. There's nothing wrong with either approach as a hobby interest but many audiophile enthusiasms can be somewhat in conflict with a view based on conventional acoustical engineering.

No, I was thinking about at external passive crossover to start with using what I have. Of course I could easily change this at a later date to an active crossover if I wanted when funds allow. The first decision I need to make is whether to rebuild the speakers with the crossovers inside or not.
An external crossover is more flexible if you know you are going to fiddle about making changes but it is also messy and likely to devalue the speaker a bit if you sell it on. Perhaps hanging it off the back in a box?
 


advertisement


Back
Top