One Man’s Quest to Build His Own Speakers

Why build your own stereo speakers? Some people like to work on cars in their garage. Some people build fast computers. Others seek the perfect audio setup. The problem for a newcomer is the signal to noise ratio among audiophile experts. Forums are generally filled with a vocal group of extremists obsessing on that last tiny improvement in some spec.  It can be hard for a beginner to jump in and learn the ropes.

[Ynze] had this problem. He’d finished a custom amplifier and decided to build his own speakers. He found a lot of spirited debates about what was important for good speakers. He tried to wade through the discussions and determine which things had real practical value. The results and his speaker build are documented in a post that you’ll want to check out if you would like to design and build your own speakers.

Some of the topics ranged from solder type to capacitor construction and 700 Euro capacitors. [Ynze’s] goal was to build something that sounded good while keeping costs in line. He claims he spent about 250 Euro and wound up with speakers equivalent to 750 Euro store-bought speakers.

We love posts like this that are more than just a glimpse at what was built and plans for how to reproduce it. [Ynze] shares a lot of links and the logic behind the different decisions made. One decision was to buy reasonable speakers, tweak an off-the-shelf crossover, and then spend time experimenting with different cabinet designs made out of chip wood. In the end, he created six different cabinets using a combination of computer-based modeling and trial and error.

There’s also a lot of practical tips like getting pillow filling from Ikea for sound isolation material. We haven’t heard the speakers, but we imagine they sound pretty good. Good old-fashioned carpentry is a great counterpoint to the 3D printed speakers we’ve seen. If you want something more exotic, check out an electrostatic speaker.

48 thoughts on “One Man’s Quest to Build His Own Speakers

    1. That has to be a typo. I recondition/restore audio equipment for a living, including vintage McIntosh gear, and I’ve never encountered capacitors that expensive. Massive 120v 15,000uF caps barely break $30… 700 euro for small filter caps? There’s no way.

      1. 1: I accidentally clicked the report button because I thought it was a reply button, so hopefully the Mod(s) read this comment and then don’t do something to yours.

        2: I completely agree, I’ve never seen caps that expensive. I have seen some expensive hand wound inductors, but those are usually used in large amplifiers (which the OP already completed).

      2. Sounds a bit high, but crossover components can get crazy expensive. Part of it is the audio premium, but also they are of a different construction. Just taking a guess but I’m guessing the capacitors mentioned above are Mundorf, they have a 47 uF model that sells for $762. If you are just filtering the power stage then yeah you can get oodles of capacity for a cheap rate.

        https://www.madisoundspeakerstore.com/mundorf-supreme-evo-silver-gold-oil/mundorf-evo-sgo-47-silver-gold-oil/

        1. Also just as a heads up, if someone wanted to really go nuts with DIY speakers it is actually cheaper to shoot for the moon and go with full active amplification. Use a DSP or discrete components to alter the signal before it hits your amplifier stage, then you can just run wires to the individual drivers. Way cheaper when you are trying to compete with the boutique manufacturers (think the Pagani or Koenigsegg of audio).

          1. Yeah but that’s interrupting the purity of the analogue path or some other audiophile bollocks.

            I can’t take anything audiophile seriously anymore, it’s only slightly less dumb than the flat-earthers or anti-vaxxers.

        2. To my ears, there would be absolutely no point in using those Mundorf caps in loudspeaker designs with – let’s be honest – less-than-stellar drivers. Seems like it was just thrown into the article for oohs and ahhs.

          1. Oh I would definitely agree that they are referenced exactly for that reason. Really that is just one of the priciest caps around for audio, I really doubt anyone would use it in a professional speaker.

        1. No they need to be made of 99.9999999999999% silver and quenched in the blood of a unicorn.

          But in all seriousness, best commercial audio cables have to be Blue Jeans Cable. They are a solid company that doesn’t talk up pseudo science mumbo jumbo. Just quality conductors and terminations, they offer ultrasonic welding and have a very reasonable approach to things.

          1. I love the stuff from Blue Jeans. I typically make my own cables, but I’ve used the Blue Jeans stuff on a few setups, typically for other people. Great stuff without all the snake oil.

          2. I like blue jean because they stood up to the bogus BS from monster cable. Monster wanted to shut them down over patent issues but did it was trumped up anti competitive baseless scare tactic cease and desist patent infringement claim BS.

        2. When I was DJing.
          I just used 250′ Long #12 2 conductor cabtire. With the connectors on the ends. Worked great.
          If I was to buy the cables, It would of costed me over $500 a cable. NO WAY.
          Those were the days. Getting payed to party. Loved it.

      3. > 700 euro for small filter caps? There’s no way.

        Decent film caps for filters are not that small, with some of the values for quality XO designs.
        http://www.humblehomemadehifi.com/Cap.html

        Rather than the $200+ caps, get decent sub-$200 film caps, remove the metal cases, mark the +/- leads & cover with teflon heatshrink, wrap cap in shelac-soaked cotton, wrap in shelac-soaked hemp twine, let dry, put in an oversized wood container, fill with beeswax chips/shavings, flood with melted beeswax. Instead of wood container, build a wood XO case, cork floor tiles cut to line and make a separate chamber for each XO component, same beeswax treatment.

        Removes the interference of the metal case. Contrains and damps mechanical movement.

      4. You can buy power transformers with solid silver windings for tens of thousand dollars. There are also IEC mains cables for thousands of bucks. Even fuses with golden comductors.

  1. Why does hackaday always make me login to wordpress to post a comment in the process loosing everything I’ve typed?

    Summary of lost post:
    Speakers are cool, and building them is fun.

  2. Years ago I was part of a study to gather 50 so called audiophiles. We lured them into the group by saying we need the best audio ears for our new audio gear. So these people poured in all taking a short survey first to get in to see if they felt they were true audiophiles. Behind the scenes we had 7 different amps and 12 different sets of speakers. Only one setup used the highest quality speakers, amp and cabling. We played music through each combination and they all had a clicker to rate which combo was the best to their so called audiophile ears .

    When the test was over I laughed because all 49 picked the setup that you can get from best buy and Walmart. So ya, i’d have to say most people that call themselves this it’s just all in their head.

    1. This is exactly why i prefer to call myself a fanatic instead of an audiophile.

      Fanatics do things for self-justified reasons.
      Audiophiles need to convince others of their self-justified reasons.

    2. That’s why it’s necessary to qualify the particupants to those who can tell the difference. Like an IEEE speaker comparasion some years back.
      Twisted the other way, a chemistry prof needing funding and wanting to make violin varnish from his work, identified those who couldn’t tell the difference between his violins/varnish and Strads, then used them to say his was just as good as a Strad…

    3. Thing I’ve learned with audio gear, and this applies to a lot of other things too, is buying the middle of the road stuff is about where it’s at, if you buy the absolute cheapest you can find, you will hear it. Buy the stuff that’s more middle of the road, and it is going to be orders of magnitude better for not really that much more money, but the ridiculously expensive stuff isn’t the same amount better than the mid grade stuff is from the cheap shit. Like my $99 Grado headphones sound way better than the $10 pair I bought once because I was traveling and forgot headphones, but I was at Best Buy once and listened to a $400 pair of Bowers & Wilkins headphones, and they did not sound all that much better than the Grados, or the $100 Sennheiser headphones that were two aisles over.

      1. Like the TV-show “wheeler dealers”. Calculating the profit of a car restauration without the cost of labor by their professional mechanic.
        Nice show, but the “bill” at the end was totally ridiculous.

        1. That was because the show had hosts with different ideas of what the show is about.

          Edd China wanted to make it a DIY show about repairing your own car, but the producers wanted it to be more like the Gas Monkeys.

          In the end they did a lot of dubious “wheeler dealer” repairs, like putting a nylock nut inside a gearbox to fix the shifter, which will dissolve in the oil over time and cause the gearbox to fail again.

      2. >” ignore the hundreds if not thousand+ hours it took to do it.”

        The problem with equating time with money is that the exchange rate is variable according to supply and demand. Even hundreds of hours are cheap, since you couldn’t use those hours to make money anyhow, but you can use them to make something worth of money.

        1. Depends on your situation. I personally would be able to work for money as many hours as I want. So I do DIY-stuff only because I want to and for fun. For me it would be, in almost every case, totally stupid to build things myself just to save money.
          Don’t get me wrong, i DIY quite a lot. I also do speakers once in a while.

  3. Please don’t use OSB (that wood in the speakers on the left in the pic) to make speakers. MDF (good damping) or plywood (more rigid) are far better and don’t cost much more.

    also diyaudio.com is an excellent forum, with a notably low level of BS. More relevant to anyone building speakers, amplifiers, etc than something like instructables.

      1. Poor rigidity (it’s flappy) and poor damping-factor (it rings), plus it’s prone to delaminating under heavy loads like you see in a subwoofer cabinet. You get box panels that radiate out of phase with the cone, and not in a way that’s very predictable – so you get a bunch of nasty peaks and dips in your in-room frequency response. It’s literally orders of magnitude worse for the sound quality than any fiddly difference arising from capacitor/opamp/whatever choices and nearly any not-completely-stupid amplifier design decisions.

        There are a bunch of “MDF vs Plywood” threads on diyaudio that make for good reading, as far as a holy war can make for good reading, and occasionally someone will jump in with “yeah I tried OSB too… don’t make the same mistake as me”.

  4. That looks like a nice tweeter for the price, but there’s no data on off-axis response. Too bad DynAudio stopped selling drivers, their tweeters were close to perfection and only $50 each when I bought them in 1990.
    My own tests with crossover networks showed me not to use electrolytics, the high ESR meant a loss of 1 dB. The use of power resistors in the article’s crossover was anathema to me, I couldn’t stand the thought of throwing away power.
    Nice article, and I like the symmetrical design.

    1. I’ve sometimes put poly caps in parallel with electrolytics in the belief that it improved their HF performance. It’s a fact of life that driver efficiencies will differ, so resistors become a necessary evil.

      I do believe that active crossovers and separate amplification is the best possible, because it’s more efficient, easier to make precision filters at line level, the amps can be sized and optimized for the specific driver, and it’s easier to tweak the crossovers once it’s all built.

    2. Sometimes you need to burn power, otherwise you get it where you don’t want it. Peaks and things in your frequency response, or a Q that’s wrong and therefore crossover curves that don’t match.

      The answer, of course, is not to build low impedance crossovers that go after the power amp. Power amps are super cheap now (a couple of dollars for an integrated chip amp), so you use electronic active crossovers instead.

      But “waaaah chip amplifiers” I hear you say… active crossovers into cheap chip amps into simple loads – single drivers, will beat the pants off the fanciest gold-plated amplifier driving a passive crossover and 3 drivers interacting with 3 different boxes, all presenting a wonky and complex impedance curve, possibly with resonances, back to the amplifier.

      Amplifiers are not the limiting factor in nearly any real system – it’s speaker physics and passive crossover parasitics.

      You can have steeper filters, don’t lose Q on low-frequency filters due to series R in your L or have non-linear nastiness from iron-core L, tweak your crossover frequencies, gains and slopes at will to match the room, adjust phase or time delay, etc.

      Join the 21st century already!

  5. I’m actually a bit sad this was posted on Instructables. It’s a great and inspiring article for sure. But it is not an instructable. It is pretty useless for someone who knows how the speakers are built, but a first-time builder, if inspired to build a pair, will not know where to start and how to proceed. It’s basically “I cluelessly poked around on the Internet and tried to guess a speaker design by building a bunch and succeeded”, which leaves nothing to a newcomer other than repeating the same process.
    Providing all those source links was helpful though.

    1. Indeed. Much better to link to a few excellent projects on diyaudio forums than someone who flailed around a bit and thinks they got an OK result, but can we really tell?

      Some of the projects on diyaudio are *spectacular* by HAD standards, including home-made DACs, digital crossover systems, amplifiers, crazy bandpass subwoofers, etc.

  6. Vented design (a.k.a. bass reflex) isn’t “a forgiving design”. If you do it wrong, get an annoying hum you can’t suppress instead of a true bass. In case of closed boxes improperly designed boxes bass is suppressed, but you can work it around by boosting it in the amp. All this happens because, vented boxes are higher order systems.

    Otherwise, I think building your own loudspeakers is a great way to save a lot of money.

    1. That really depends on what you have to work with. It is like sewing our own clothing to save money. If you have some of the material and all of the tools, yea. If you have to go out and buy drivers and wood, and tools. not so much.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.