Denon DHT T100 DESIGNED TO FAIL : bad caps (ceramic caps this time)

A friend gave me a Denon DHT-T100 to look at.
Do not spend much time on it.. ok.


It had a problem, it started clicking and ticking after power on, and after some time.
(a so called intermittent problem)

Sometimes it did not tick or click, but it was basically not usable.
The clicking had a sharp click in one channel and repeated after about a second, then sound recovered and it would click and drop out again.

He told me, do not spend any time on it, yeah, right. ;-) like I would give up after 15 minutes.
I wanted to know what was going on with this thing, I found some schematic online and started measuring the usual things.

The power supply, 24V did it drop down when a tick occurred ? Difficult to find out because sometimes the thing would play for hours without a glitch.

I eventually found out the 24 V PSU, the step down SMPS on board and the LDO's were all ok, all power rails remained within spec, but it still glitched sometimes. (while power was ok)

I investigated methodically, did it occur just with the blue tooth or also with the cable? Both ..
Was it digital or analog? Was the DSP restarting? All these things were ruled out eventually.

I finally found the a signal changing during the glitch.. it was the FAULTZ signal on both the power amplifiers (class D, TPA3118)

The signal is bi-directional, you can pull it low to disable the power amplifier, and it gets pulled down by the amplifier when a short, over-temp, over-voltage, under-voltage etc occurs. It is also connected to a reset circuit.



http://www.ti.com/lit/ds/symlink/tpa3118d2.pdf

The 3 pin reset IC was finally ruled out as the culprit. (U28 was not bad)

To make a long story shorter, one power amp was detecting a failure and taking the other one down too.

I ordered a new TPA3118 and spent some hours soldering it, it has a heat-sink on the bottom and access to the pins is hindered by big caps, it was a struggle, but it got done.
The de-soldering of the thermal pad on the underside  was not easy, but with a 40 Watt big iron it finally worked.

I powered up the thing again and expected success, but was greeted by the same clicks as before.

I really had a WTF moment, you can be sure. ;-(

I had to change my plan, I knew the problem was probably thermally triggered so I took my soldering-heat-gun and turned it all the way down to 100 deg C. A very small nozzle on it, to point at potential problem components.

I pointed at specific points on the PCB to find the location of the fault and after some trial and error found 2 small capacitors that resulted in ticking/glitching very quickly.

These were the DC blocking capacitors between the TL074 analog filter and the input of the TPA3118.

I measured the capacitance, 10 uF all fine, the voltage over the caps (DC was about 9.2 Volt) and looked this all up in the schematic.
It seemed to make some sense, if the capacitors would briefly short and recover the power amp would trip into FAULTZ and recover.

Looking at the schematic I noticed the caps are specified as 6.3 Volt caps, but there is a 9.2 Volt DC on them, no wonder these things go bad! If interested, please google 'failure modes of MLCC capacitors' to learn more than you ever wanted to know about the problems with these caps.
The datasheet says the DC level at the inputs is biased to 3 Volt from ground.
I checked the schematic, the 'virtual ground' of the TL074 chips is at about 12 Volt, so 12 -3 is about 9 volt DC over the blocking caps.

My conclusion : bad specification, the blocking caps should be 25 volt or higher.
This is a design FAIL FAIL FAIL.
During startup the spikes on these caps will be about the full 24 Volt of the power supply. 

If the capacitors are really just 6.3 Volt as the schematic suggests, then all these DHT T100 sound bars will break down very quickly.

What you need to do is replace caps C126, C138, C154, C160, C171, C184, C199, C210 with higher voltage ones.

Also, the caps are 10 uF and the reference design by TI specifies 1 uF, the 10 uF brings the high pass filter's corner frequency down to 1.6 Hz, which is just plain nonsense. Better use a 1 uF cap of 50 Volt than a 10 uF cap of 6.3 Volt. It sounds just the same but does not break down within 2 years.

I replaced the caps with some 25 volt ones of 4.7 uF I had in stock.
Powered up. Problem is gone now.
I can heat it up, cool it down whatever, it is really gone.

So if you ever have no-sound or clicks/ticks/glitches take a look at the DC blocking caps.

I spent a lot of time to find this problem, but I get a good feeling that I cracked this one!
Hope this helps you in repairing your system.
If you have no experience with electronics then this is not something you can fix, please ask someone with soldering experience. I also take no responsibility for things you do or accidents you make.

(This is the broadest possible legal disclaimer you can think of)

Good luck and happy tinkering.
Edwin (Oetel-x)