About two and a half years ago, I wrote a post about fixing the “click of death” problem the Razer Mako 2.1 THX speaker system is prone to. My Mako worked fine until recently when it failed in another way, namely with no sound at all anymore. No clicks, nothing. While the “control puck” still kinda worked, it flickered and behaved weirdly.
I tried getting Razer support to give me schematics, but no go. Also, I can’t find any schematics on the net either. I did figure out by looking at the board that it is a dual switched power supply (two bucket converters), which implies that there are dangerous voltages on the board. Not only that, but you can’t safely ground anything on such a board, so that connecting a mains powered oscilloscope will typically result in your ground fault breaker to trip (if you’re lucky) or your oscilloscope to short out and die (if you’re less lucky), or you yourself to short out and die (I know, not funny).
To be able to work on a device like this with a scope or signal generator (anything that needs instrument ground to connect to the device under test), you must, absolutely must, power the device through an isolation transformer. But even then you need to be very careful, since even though you eliminate the ground loop problem, you can still electrocute yourself just fine by grabbing onto the less advisable parts of the device. So unless you really, really know what you’re doing, and you use an isolation transformer, I’d suggest you never work on the Mako with power connected1.
I’d suggest, if you don’t have all the gear I’m using, to simply unplug the Mako, let it rest a while, open it up, then replace the capacitor I’m mentioning below, then put it all together and test it. For disassembly, see my earlier post about the click of death.
I worked on the Mako for more than a week, an hour here and there, while thinking it through. I did conclude, and confirm, that all the power supply voltages on the board were only present if all speakers and the control puck were connected. Logical, when you think of it. (Again, only touch this thing if it’s on an isolation transformer, and even then with utmost care!)
I then fed the inputs with a sinus signal of 500 mV and a few hundred Hz and found it was present on the ADC0 or ADC1 inputs of the ADAU1701 chip (depending on if I connected to left or right channel input), so the preamps worked. But checking on the VOUT0 through VOUT3 outputs of the ADAU1701, I found nothing but a high frequency noise signal just under 60 kHz. Touching some of these outputs made the speakers click, so the power amps were probably also ok.
I then checked the power going into that chip and it looked awful, around 3.3 V but with a 60 kHz signal overlay of more than 200 mV. A horrible ripple, in other words.
That, to me, meant that a large smoothing capacitor must have failed. The huge ripple on the power line must have severely screwed up the workings of the ADAU1701, which is a ADC/DAC signal processor and which does practically everything that needs to be done to the audio before going to the power amplifier stage. No wonder I had no sound.
Back to capacitors, then. I could not see any bulging capacitors, but feeling around I found one that was so hot it almost burned my fingers. And that is really not normal. If this cap was shorting out2, it would explain everything, and sure enough, after replacing it, the system now works normally. The bad capacitor was labeled C122 on the board, a 330 µF/25V, which I replaced with a 470 µF/35V, simply because I had no 330 µF at 25V or more. Seems to do the trick, though.
(In the foreground of that last picture, you’ll see a transistor with a horrible mess around the legs. That’s normal! Razer put in a series resistor there as some last minute change, then added some gunk that turns brown. Looks like shit, but I couldn’t find any fault there.)
Moral of the story: if you have a Mako with no clicks, no sound, and a semi-working control puck, try replacing this capacitor. There is a decent chance it will work.
- Also, there are some big capacitors that may retain charge quite some time after unplugging, so give the device time to discharge before you touch it.
- It wasn’t “shorting out”; it was degrading with an increased internal resistance (ESR), which is almost the opposite. See my next post.