I normally rarely buy any speakers, I'd rather build them myself. Yet, there was a particular occasion where I quickly needed some decent portable speakers. I'll save you the details, but just assume I wanted a set of speakers that could produce a reasonable sound volume at reasonable quality, at a reasonable cost. The things should be cheap enough to store in a cupboard in an office that might be visited by people with less than good intentions. So, I looked on eBay, and dug through the enormous amounts of crap. Then a particular speaker set caught my eye, and I ended up buying it. Here goes…
The eBay page didn't contain any brand name and the photos didn't show any hint of a brand either, but once it arrived, the box bore the brand name “Dr. Hank”. (Nope, no relation to ‘Dr. Lex’ whatsoever.) The speakers bear the model number “SP-M600”. Unlike the image in the eBay ad, the speakers also sport the “Dr. Hank” logo on their fronts. The packaging mentions a website, but it appears to be dead at the time of this writing. I won't be surprised if these things pop up again under a different name.
So, as promised on the eBay page, these things are actually made of wood. From the outside, the wood appears to be some kind of tropical hardwood but don't be fooled: it's just regular MDF with fake veneer on top. Not that MDF is a poor material to build speakers, on the contrary: it is often used in high-end cabinets. I guess these must be the smallest commercially built MDF speakers available. Some may argue that MDF is not wood, but for all practical purposes it is and it is a better material to build speakers from than many types of ‘real’ wood.
The eBay page didn't show any photos of the rear of the ‘amplifier’ speaker, so I was unsure if it had any controls like volume/tone or power on/off. Well, it doesn't. There are only three cables: one to a USB plug for power, one to a 3.5mm for audio, and one to the other speaker. I don't mind the lack of a power switch or a tone control. The single tone control on typical PC speakers is utterly useless anyway: when turned to the left one has too little trebles and to the right, too little bass. I always end up setting it in the middle which has the same effect as no tone control at all. But the lack of a volume control is a bit annoying, it's not always possible or easy to change the volume on the audio source.
Note 2012/03/08: while taking these speakers apart, I discovered a web address inside the plastic cover that points to the roots of these speakers. They are made by a company called Enkor. On their webpage they are shown with a slight rounding at the front and some Engrish descriptions like “For latatop Design”.
The eBay page stated two power ratings. One is the regular RMS rating, claimed to be 5W per speaker. This rating is also mentioned on the box. RMS stands for “Root Mean Square”, and means the speakers should be able to handle a continuous sine wave of 1KHz, dissipating 5 Watts of power. This is a reliable, well-defined method of specifying the power rating of a speaker or amplifier. Now, the other rating mentioned on eBay was a dreaded ‘PMPO’ rating, claimed to be a whopping 800 Watts.
You see, the problem with the PMPO rating is that it is completely, utterly and totally nonsensical (feel free to use the shorthand for “bovine excrement” as well). ‘PMPO’ stands for “Peak Music Power Output.” It's supposed to be the absolute instantaneous power output that could theoretically be handled by as many parts of the entire circuit summed together as the manufacturer thinks is relevant, during a time interval that is as short as the manufacturer wants it to be. Even the speaker from a telephone handset could handle an instantaneous power of 1000W if the time interval is short enough. There is no standard for the time interval. This makes the PMPO rating a scam, and I believe it should be forbidden to mention it anywhere unless it is standardised. Some manufacturers multiply the RMS rating by 10 to obtain their PMPO rating, others multiply it by 400. There is no way to compare PMPO ratings between manufacturers, so don't even try. When you see a PMPO number, pretend it is not there, you'll be better off. Look for an RMS rating instead.
Now, after this bit of education, the good news: there is no PMPO rating mentioned on the actual box, only the 2×5W RMS. This earns the manufacturer some karma points. Unfortunately there are two problems with that RMS figure too: first, the speakers are powered through USB, which by design can deliver at most 500mA at 5V. This amounts to a maximum of 2.5Watts, or 1.25W per speaker if the amplifier would be ideal. In reality, the amplifier will have an efficiency of about 60%, so there will be less than 1W going to each speaker. Sure enough, the speakers themselves are marked “5W 4Ω”, so that's where the rating comes from. You may think that by hooking up an external power source which delivers more juice than USB, the speakers could reach their full potential. Alas, the second problem is that the built-in amplifier is based on a TEA2025 chip. For a 4Ω load, the maximum power supply voltage is 9V. Under those conditions it can reach its maximum power output of 2×2.3W, but only if the IC is fitted with an adequate heatsink. There is none on the built-in amplifier, so the IC's thermal protection will limit the power to something below 2×2.3W. In other words, if you want to squeeze the promised 2×5W RMS out of the speakers, you'll need to swap the built-in amplifier with something more capable.
The packaging also touts a frequency range of 20Hz up to 20kHz. I have built speakers myself, and believe me: coming even near to 20Hz is not an easy task and becomes harder as the dimensions of the speaker become smaller. That is something that can be understood through a basic knowledge of acoustics, and was formalised by Thiele and Small in the 60's and 70's. The fact that it is impossible to get loud bass sounds out of tiny speakers is also formulated in ‘Hoffman's Iron Law’. For speakers of this size, reaching 20Hz at any perceivable amplitude is as good as impossible. My guess is that their lower 3dB point is at about 120Hz.
Of course, as with all my purchases these things were doomed to be taken apart shortly after exiting the safe environment of their packaging box. But for a while I resisted my urge to hack and just plugged them in like a normal person would. An immediate disappointment was that when I plugged the USB power cable into the same PC as the 3.5mm cable, a noticeable hum emanated from the speakers. When trying my laptop instead, the hum was mostly replaced by noise that was clearly correlated with CPU activity. The cause of this may be some kind of ground loop, so the amplifier doesn't seem to be designed very well. But the biggest problem seems to be the flimsy audio input cable. It looks fancy with its silver lining and transparent plastic, but it obviously is not shielded at all. The silver lining is not connected to anything, so if it is conductive it will only worsen interference problems. The hum and noise is mostly gone when plugging the USB cable into another device than the audio source, but for many people that won't be an option. A wall wart with USB power port works well but kind of defeats the idea of getting power through USB.
With the hum/noise problem solved, the speakers do perform pretty well. They have a clear and balanced sound and the lack of bass can be partially compensated with some equalising. Many ordinary PC speaker sets are only good at reproducing midtones, but not these ones. The bass response is probably as good as one can get with speakers of this size, and the trebles are excellent. Combined with a subwoofer they could provide a very satisfactory full-range sound.
Overall, this looks like yet another Chinese product that was well designed but shoddily constructed in a factory, cutting corners to save time and money. For instance, in my specimen of this product, the magnets on the loudspeakers are not even properly centered and one of the loudspeakers is mounted somewhat slanted and with one of the four screws of a different type. But this does not affect the sound and it is not visible.
If you want to get the most out of these speakers, there are a few things that you could do. First, because powering the speakers through USB is troublesome anyway and limits the output power, you could swap the USB power cable with a 9V mains adaptor. There is no use in going above 9V, because that's the maximum rated voltage for the TEA2025 IC when driving 4Ω speakers.
Second, to get the maximum power out of the IC, it should be fitted with some kind of heatsink. There is some room around it to glue a small heatsink with thermal glue. Maybe not enough to allow the maximum continuous 2×2.3W power, but better than nothing.
Third, a volume potmeter is relatively easy to install. The input impedance is around 30kΩ, so a logarithmic potmeter of about 10kΩ or less is suitable, especially given the fact that you're likely going to drive these things from a low-impedance source like an iPod. Put the potmeter between the 3.5mm input wire and the amplifier board. There's room directly under the hole where the wires enter the cabinet.
To reduce the noise, replacing the flimsy audio cable (with the 3.5mm jack) is highly recommended. Use a cable that is shielded, i.e. whose ground conductor entirely envelops the signal conductors. You may want to use this occasion to make the rather short cable longer while you're at it.
If you really want to get the most out of these speakers, you should replace the TEA2025-based amplifier by something that can deliver 2×5 Watts. I can't tell if the loudspeakers themselves are really 5W RMS or 5W “music power” which generally means 2.5W RMS. But they look as if they're capable of handling 5W RMS. I have built my own 2×5W TDA1516-based amplifier with a PCB that is even smaller than the original one (see photo), even with a pretty large heatsink attached. It works very well at 12V and certainly delivers more oomph than the original amplifier without sending the speakers into overdrive. Alternatively, you could throw everything out of the cabinets and leave the drivers with wires connected directly to them, allowing to use any external amplifier.
The drivers themselves may or may not be mounted in a semi-sloppy way. It's the name of the game given the “Made in China” label. This may improve with time, remember how people laughed at the “Made in Japan” label long ago. Even on the eBay page of the seller, the photo showed one of the drivers mounted slightly rotated, and in my case, there was a visible gap between the driver and the cabinet on one of the speakers. My standard way of eliminating any loss of sound quality due to air leaks, is to apply silicone kit between the driver and the cabinet. To make it feasible to take the speakers apart afterwards, apply grease to the driver before mounting it with the silicone. Silicone kit hates grease and will easily detach while still providing an airtight seal.
These are some pretty unique speakers: you won't easily find anything made out of actual wood at this size and price. The finishing is pretty good, the cabinets really look like miniature Hi-fi speakers, and the materials they're made from are OK. The result is that there's no noticeable ‘tiny plastic speakers’ sound.
If you're like me and you tend to hack everything you buy, these speakers offer a good base to create something pretty good. Add a volume control, replace the worthless amplifier with e.g. a TDA1516 and hook it up to a 12V 1A power supply, seal any gaps around the speakers, and you have some tiny but powerful and good-sounding speakers that look awesome too. If you just want to buy and use these things instead, I can't really recommend them unless you really like the looks, don't mind some hiss and noise, and don't want to play your music loud. Given their shoddy finishing and rather poor electronics, you should not pay more than €10 for them. If they would have been properly constructed, they could be worth a lot more.