Simple and boring…

Another completed project courtesy of holidays and so-so weather 🙂

I know it doesn’t look like much, but this is yet another part of my attempts to make loudspeaker testing easier (and therefore something I am more likely to do). It is basically a box with a two-channel speaker input and output, and then a capacitor in series with one of the positive lines. This means I can put a capacitor in series with a tweeter for testing purposes without having to open up the loudspeaker cabinet, screw/glue the cap into the case and/or interfere with the speakers existing wiring. Of course bi-wiring terminals on everything would solve the problem as well, but this is way more flexible overall.

If you are a bit cynical this is just three (ok, four) capacitors in a box, but I have added a little to improve the basic concept. Not only is it built on a simple PCB to keep everything in place, I have also fitted switches to add extra capacitors in parallel meaning I can get the same low-ish cutoff frequency with 8, 6 and 4-ohm tweeters. Input and output are via Neutrik Speakon connectors to make hooking up the box easy and safe. I also already have the needed adapter cables because the pinout is the same as I normally use elsewhere (including on my Hypex test amps)

I have already used it, and it works well (although I should probably label the switches…). In fact, it works so well that I was able to break out two old projects that I haven’t managed to build passive filters for yet, hook them up to an active filter and confirm that they work. At least one of them definitely sounded so promising that I’ll put the passive filter nearer to the top of the to-do list, so even simple and boring projects can still be incredibly useful 🙂

Proof of concept?

So, summer is here and a slightly late break for me compared to the norm where I live. That means two weeks off and while I did spend a few days away from home last week – clearing my brain in anticipation of a very busy autumn/winter season at work to be honest – this week I have time to catch up on various projects around the house plus of course some DIY audio stuff as well.

Right now the weather is playing ball (by only being suitable for indoor activities…), so I started by finishing up my “Open Baffle” experiment speakers (shown here). They consist as described of a Tang Band full range driver (W5-2143) on an open baffle, coupled to a 10 inch Seas L26RFX bass driver in a sealed box.

First impressions? Well, physically I have to say that I think they turned out better than expected. Even if they are a little rough, the finish is not 100% and some details could be improved, I think the basic “form factor” works well. Despite the bigger/wider baffle I could actually see something like this work in my living room on a permanent basis, which I am not sure I really expected.

Soundwise they of course pretty raw right now. I suspect some basic EQ will fix most of the issues and I believe I can already hear that the Tang Band full range has quite a lot of potential. It has the same openness I heard from the Seas FA22 full range drivers and while they are not anywhere near as detailed or neutral as the Mundorf AMT tweeters I have in my main speakers at the moment, I can definitely understand while som people speak highly of full-range drivers.

On one hand the speakers are finished and playing, but on the other hand I think the term “proof of concept” is accurate. They were at least never meant as anything other than an experiment to satisfy my curiosity of OB speakers. Right now I don’t think I’ll be building a “version 2” of the design, but I’ll let these live for a bit longer and do some more listening because as a “PoC” they seem to work very well 🙂

Fake it till you make it…

Not really the way I like to do things, but sometimes it does make sense…
In my ongoing quest to make my loudspeaker prototyping endeavours as simple as possible, I have come up with what is probably my simplest PCB-layout ever.

What it does is “fake” a Hypex FA-series plateamp by breaking out the connectors used for connecting the Hypex amps onto FAST-ON connectors and holes suitable for soldering wires to them (yes, that is all it does, sorry!).

Not only does that allow for prototyping using my prototyping rig, it also allows wiring a speaker with connectors suitable for a plateamp but without having to fit one from the outset. Given the number of speaker designs I would like to try at some point buying separate plateamps for all of them doesn’t make a ton of sense, so this should help with switching amps around between boxes.

Apologies also for the extra-long gap between posts, which was partly due to a planned holiday. Just before heading off I managed to send a PCB-order and so some more advanced designs will hopefully follow this one in the not-too-distant future 😀

Distant cousins?

I’ve showed off my fascination with (small) coaxial loudspeaker drivers before and recently I scratched the itch again with a pair of SB12PACR25-4-COAX.

This is a new driver from the “cheap” SB PFC/PAC lines. Like the Seas driver it has an alu cone and a softdome tweeter, but the price is around 1/3 of the Seas driver, courtesy of cheaper SB manufacturing and a cheaper plastic basket compared to the die-cast aluminium on the Seas. That said, it doesn’t feel that cheap to be honest, because even though the basket is plastic it seems well-made and solid. I have several others of these PFC/PAC SB drivers and they as well-engineered as the rest of the SB lineup IMO and excellent value for money. In case you doubt their “audiophile credentials”, there is even a Troels Gravesen design based solely on the PFC-drivers.

Now I could have just bought the paper-coned version of the coax-driver that has been on the market for years, but I have the 16cm version of that already, the price was about the same and I thought it would be fun to compare the alu-version to the Seas L12 – oh, and I think the alu-version looks cooler as well :).

This is one of the drivers that I have bought with a specific project in mind, namely to try and build a small active all-in-one “desktop” speaker. The original plan for that was a three-way design with a small built-in subwoofer, but the enclosure ended up being very large and impractical (i.e. desk-sized instead of for desk use…). Two of these SB12s plus some electronics is an altogether more reasonable size, so I have repurposed the other drivers for something else instead.

I’ve already cut some wood for the basic design and with a holiday coming up soon, I hope in a week or so I can get these drivers into cabinets and do some basic listening tests. In parallel I am also looking for suitable electronics, but that’s a story for another time (mostly because it’s not done yet 😉 )

Promising sounds…

It’s been the Easter break here, which I once again took full advantage of. “Full advantage” in Denmark means 10 days off work and just because I felt like I needed a break, but also because I’ll be starting a new job in the beginning of May so it felt like a good time to relax a little.

After a (predictably) slow start, I did actually manage to do quite a lot of stuff during my days off. Since the weather was nice most of the time it was time to start working outside and that meant working on some loudspeaker projects. To cut a long story short I managed to finish a few things, most notably the full range Seas FA22s that have been waiting for better weather for a few months.

I’ve stuck with the standard Seas design of a 60L closed box for the FA22, but instead of the short and deep enclosure design proposed for the FA22 I have gone with the more traditional floorstanding form factor suggested for the more expensive Seas X1 full range (with a simplified internal bracing setup).

As the title alludes to I’ve only just managed to connect them up and listen to them, but I have to say I am impressed already. As there is no filter to correct the frequency response some kinds of music including most male vocals sounds a bit anemic, but I think the filter might help with that. On the other hand the soundstage is very good, the resolution and timbre on most female vocals is excellent and there is obviously some bass as well.

First order of business now is to play a little with the recommended filter values and also some additional EQ just to see how much sound quality it’s possible to extract from these drivers. If they live up to the initial promises I’ll have to do something finish-wise to the cabinets, but for now I can live with the raw MDF 🙂

Mass production….

No, I am not starting a business, but right now you could be forgiven for thinking that my house looks like a loudspeaker cabinet factory… 🙂

After taking my sweet time with these projects, I’ve finally gotten some decent progress going. One of the main challenges with building speakers for me is that I only have a small room in the basement to use as a workshop when it’s too cold to work outside. The problem with being space-constrained like that is that it requires you to work in a relatively structured, focused and sequential way – something I am notoriously bad at in my hobbies 🙂

However, after spending far too much money on tools and workshop accessories I have now got a setup that works for most tasks, gives sufficient accuracy to the work and is able to manage dust and noise to a level where inside work is possible (if not entirely optimal). The only thing I can’t really do is proper painting etc. so I’ll have to live with raw MDF cabinets for a while, but that is manageable.

I’m still missing a few details to be able to start putting the speakers together, but among the first things I’ll hopefully be able to have up and running is a permanent home for a pair of Seas FA22RCZ full-range drivers that I look forward to testing.

New Year’s resolutions…

Well I don’t really do New Year’s resolutions as such (and not just because I can never stick to them anyway…), but starting a new year is still a good time to reflect on what to prioritise, change and improve for the coming year.

One of my goals for this year is definitely to get some traction under a couple of my speaker building projects. I’ve already started the woodworking and things are progressing so there is still hope on that front. However, once I get the mechanics worked out will come the hard (and interesting part) of experimenting with crossovers and tuning the sound.

As I have already decided to start out with active crossovers, I spent parts of the Christmas break making some temporary homes for my prototyping amplifiers. These boxes are just made from scraps of plywood I had on hand and they are definitely not pretty. They should however be functional, even in a semi-permanent setting as the projects (inevitably) drag on 😀

The amplifiers are 3-way Hypex Fusion FA123s and some readers might think “Hang on, we’ve never seen those presented before?” – and that is absolutely correct 🙂 I have two-way amplifiers from both Hypex and MiniDSP available for the projects I have in mind, but at some point I want to get to a real three-way speaker project as well.

Since much of the prototyping is is going to be “two-way and sub” configurations which is nice to be able to do in one setup if required, I looked at these. They come up occasionally as used pairs, but Hypex have been raising their prices regularly and so a Black Friday offer actually looked tempting enough to make me take the plunge and buy a new pair.

Next step: Getting some drivers into some cabinets and diving into the Hypex filter designer software – wish me luck!

Small wonders?

Every year for at least the last 20 years I have made a point of buying myself an X-mas present (OK, the last few years it’s been more like a handful of presents, but nevermind that…).

Picking something out has usually been done with the help of three guiding criteria, namely “shiny” (= something that is a little out of the ordinary or feels special), “expensive” (= not necessarily in absolute terms, but a step above what I would typically spend), and “unnecessary” (= something I didn’t need, but still wanted 😊 ). Admittedly these goals were more relevant back when I was a student and the budget was tighter, but consciously buying stuff that is a little out of the ordinary in this way still feels worthwhile.

This year I have bought quite a lot of speaker drivers and so it was sort of appropriate that I ended the year with buying one more pair. It’s a pair of small Seas L12RE/XFC coaxials that I came across at a good price. I’ve wanted to try these more or less since they were launched but the price was a bit too steep, however at very nearly 50% off list price for this pair I had to take the plunge.

I don’t really know why but I have had a keen interest in coaxial drivers for a long time. It probably started when I originally began working in a hifi-store that sold KEF speakers equipped with KEF Uni-Q coaxials, but it might also have something to with the engineering fascination. Coaxial speakers are at the same time both a brilliant and a deeply compromised idea, but they often achieve a perceived sound reproduction that many other speakers struggle to replicate.

Apart from these coax-drivers, I am also regularly eyeing the various PA-coaxials available to make something with a pair of those as well. The problem there is that PA-coaxials are normally very bass-shy (because they are meant to be used with subwoofers), or very expensive (or both!). Who knows, maybe that’s next year’s present? 😉

The first stop for this pair of drivers is probably a small vented box like in the Seas application note but depending on how that turns out I might also (eventually) experiment with a 3-way stand mounted design where these are the top and middle section together with a woofer.

Fully activated?

My speaker projects are hibernating a little at the moment – mostly because woodworking outside isn’t really possible in the current sub-zero temparatures – but once the weather improves I am definitely starting up work again.

However, the one thing I dread a little about speakerbuilding is to have to design crossovers (more or less) from scratch. Not because it’s complicated as such (I am quite happy experimenting and doing “trial and error”), but because there are tons of variables and takes time before you get through all the different variations and find something that’s really listenable. All the more reason to keep an eye out for good offers on something that makes life simpler then…

As I think I mentioned in the past I actually bought a pair of MiniDSP PWR-ICE plateamps but stupidly sold them on again because I didn’t think I would ever get around to using them. I still regret that, so when a pair of lightly used Hypex Fusion FA122’s popped up at a decent price on my local classifieds page last week I jumped on them. My previous experiences with the NCore amps show that they are very capable, and this time I am sure I will eventually use them 🙂

The somewhat improvised casing and mounting system (yep, thin plywood and electrical tape!) is not mine, but comes courtesy of the previous owner who used the amps in a test/prototyping setup. While it’s a bit of an eyesore it works well, so at least for now it is not going anywhere. If the amps are eventually going to mounted on the back of a pair of my DIY speakers a few scratches and tapemarks aren’t going to matter and it’s actually a very flexible solution for prototyping.

A more practical problem though: As a committed Mac-user I don’t actually own a Windows-PC which is required to run the Hypex software, so while my woodworking activities are on hold I need to get started on fixing that problem – preferably cheaply 🙂

• 

• 

Soulmates?

A while ago I showed one of the results of my decision to restart buying speaker parts – a pair of Scan-Speak Illuminator midwoofers – and a few weeks ago I managed to get my hands on a pair of tweeters to match.

My original plan with this project was to build a version of Troels Gravesen’s Illumina-66 design with the Scan-Speak R3004 ringradiator tweeters. I had actually designed an enclosure plan with the right 5 degree tilt and started cutting up the MDF for it, but mid-project I sort of changed my mind. Fundamentally the reason I want to build another pair of speakers is not because I need them, but because I want to experiment with them and see how good I can make them sound (possibly even with DSP/active filtering at some point). So, instead of just ordering the ring radiators I decided to wait a little and see if I could find something else that was more “interesting”…

After some weeks of waiting I spotted a pair of very slightly used Mundorf AMTs (the closed-back 25 model) on a forum and decided to pick them up instead. They were about the same price as a new pair of R3004s and I have heard good things about the Mundorf AMTs. Also from past experience with my AMT-based Elac speakers I tend to like the sound “character” so I thought it was worth a shot. I was a little unsure about achieving good matching between the larger AMT and the woofer, but it looks doable to get a good result (even without DSP).

As my original enclosure plan had to be scrapped I have yet more wood to cut up before I can proceed with the build. Woodworking outdoors is still much easier than indoors so I will need to take advantage of gaps in the rain over the next few weeks to get that done 🙂 I’m going to keep the basic reflex tuning of the Illumina-66 because Troels probably knows more about this than I do, but I can change the overall cabinet design a little bit to suit my need since I’ll need to redo the crossover from scratch anyway.

Timeline for the project? Don’t bother asking… 😀