Originally Posted by jimboy
Just got back from CC. The replacement unit seems stable on all receivable channels but it stills fails the whap test. It must suffer from a case of Microphonics.
Microphonics describes the phenomenon where certain components in electronic devices transform mechanical vibrations into an undesired electrical signal (noise). http://en.wikipedia.org/wiki/Microphonics
Some ceramic capacitors, such as high density multilayer ceramics which are generally better suited for decoupling rather than being in the signal path (they tend to also have high variation in capacitance value from part to part and with temperature), exhibit a very significant piezoelectric effect. This means they act like microphones - or speakers, though that tends not to be noticeable. These parts are smaller and cheaper than other capacitors of the same value. This is one of the things you take into account when designing circuits, especially ones that will be used in high vibration applications.
Traces on PCBs (Printed Circuit Boards) couple capacitively to the metal case and if there is relative motion that produces a change in voltage. Rigid mounting of PCBs is a requirement in high vibration applications for many reasons.
Manufacturers of consumer electronics often cut corners on capacitors; audiophiles and others often rebuild them and capacitors are often replaced with expensive precision caps.
The crystal used for the master clock for the processor or other digital electronics is inherently piezolectric - that is the property which makes it work in the first place. I haven't seen adverse effects from microphonics, since they produce a digital signal out in this type of circuit, but sometimes when they are incorporated in poorly designed oscillator circuits they will fail to start oscillating and you can flick them with a finger and get them to start.
I hadn't noticed it before but my DTT901 is extremely sensitive to being tapped. Every sharp tap produced dropouts. This is an AC powered device which makes two reasons it would not be well suited for mobile applications (TV reception in vehicle on the go can be tricky, anyway).
My guess would be the coupling capacitor between the tuner and the demodulator chip. After the demodulator, it would have digital signals. It is possible it could be inside the tuner instead but digital designers are more likely to overlook this effect. The Tivax converter box has a differential signal between the two. Each of the two lines has a 0.1uF capacitor (and 100 ohm resistor) in series and there is an inductor and a 47pF capacitor across the lines. 47pF is a low value that is less likely to use a piezoelectric ceramic. But 0.1 is a typical value of decoupling capacitor and those often are piezoelectric. Further, there is incentive to use the same model of capacitor used in other places where 0.1uF is used. On the tivax, these two capacitors are in an 0805 package but 0.1uF decoupling caps are in the much smaller 0402 package, so they have used a lower density cap for coupling the tuner and demodulator, as well as on the video output circuit. See the tivax schematic
. Because this is a differential signal, if the microphonics are in phase on the two capacitors they will add and if they are out of phase they will cancel to some extent. This could depend on the position of the two caps on the board relative to the vibrational modes or the orientation of the capacitors at each site.
In most cases, this won't be a problem, but if it is, the two signals in question are on pins 20 and 21 of the DTT76850/DTT76851 tuner module (these are the two pins furthest from the TV_OUT connector). Those with suitable electronics skills can look for capacitors in series with those two pins to replace.
Bad solder joints, hairline fractures on PCBs, and poor connections often respond to mechanical stress which is why giving electronics a good whack often helps temporarily "fix" it. I have seen a number of undersized resistors operating at high power essentially unsolder themselves on TVs due to what is probably thermal cycling stresses shearing the joint. These resistors are often levitated above the circuit board for better airflow.