7
\$\begingroup\$

On electronic devices in the '80's, electronic malfunctions could sometimes be temporarily fixed by physically "hitting" the device.

For example, I once had a colour TV - "ECTV" (from Electronics Corporation of India ECIL) which would lose image. It was enclosed in a MDF/particleboard case, and as a kid I discovered that whacking it on both sides together would sometimes temporarily make it function correctly (for a few minutes to a few hours).

enter image description here

I think it wasn't purely physical (like an unseated component or dry solder) because it would stop working again after a while, and eventually it was fixed by a repairman replacing a capacitor in the power supply if I remember correctly (this happened around 1996 and my recollection is hazy).

We also see this happen in various movies, where hitting an old piece of electronic equipment makes it start up. There's even a term for it, "percussive maintenance".

What's the science behind this?

\$\endgroup\$
  • 4
    \$\begingroup\$ It's not science - it is an art knowing where to hit the offending device, and how hard to hit it. \$\endgroup\$ – Peter Bennett Jan 2 '16 at 4:52
  • 1
    \$\begingroup\$ As an aside, back in the days many tube TVs burned because they got draped with a cloth, and vents got blocked. \$\endgroup\$ – Nick Alexeev Jan 2 '16 at 6:02
  • \$\begingroup\$ Yes, I often wondered why folks would drape electronics with a cloth or knit cover when they had obvious vents. I thought it was an Indian thing :) Also, this is not a pic of my TV, which is long gone. I found this image on a local classifieds portal. \$\endgroup\$ – Pranab Jan 2 '16 at 7:17
  • \$\begingroup\$ People doing certain things in movies, especially of a technical nature, rarely has any basis in reality. \$\endgroup\$ – whatsisname Jan 2 '16 at 8:40
  • \$\begingroup\$ I independently discovered this in real life, as have many people. Anecdotally this seems to have been common with CRTs, both monitors and TVs, perhaps because of the form factor, popularity of these objects, and something in the electronics that made this method work - hypothetically like unseated chips, soldering issues etc. My question is specifically about if someone knows what could have made this work. \$\endgroup\$ – Pranab Jan 2 '16 at 8:52
5
\$\begingroup\$

Re: how this happens in the first place, or how it happens again soon after "fixing" it- Heat expansion. Power on your device, and everything heats up, and expands slightly. Power it off again, the components cool down and contract. Do that enough times, and the different materials expanding and contracting at different rates leads to connectors "walking" loose.

Three reasons why this is less likely to happen with modern electronics- first is directly related to power/heat cycles. New equipment tends to be more energy efficient, i.e. it doesn't get as hot under normal operating conditions as old stuff did, thus less mechanical stress from thermal cycling.

Two- chips in sockets are rarely used anymore. What some people lament as making it difficult to fix anything is actually a result of the much higher reliability of components, it's unlikely to ever need to replace anything, so connections are more permanent. Also, with flash memory, firmware can be replaced "in circuit" rather than needing to swap out an EPROM.

Three- lighter weight. Don't laugh, really. Consider the difference between a CRT and a LCD panel- a CRT is not only heavy, but oddly shaped, and its case must not only support that weight, but also present controls and connectors to the user that end up being physically far apart. If the case isn't strong enough, it can flex, moving those parts relative to each other. I have a tube HDTV that weighs a ton, and the case has so much flex I can rock the top back and forth an inch or two with the base on a flat surface. Every so often the picture goes dim and blurry, and smacking it near the top on one side (not the middle) of the screen will return it to normal for some time. I one day took it apart and found it had 4 different boards across the bottom, in two pairs- the front to back were connected by soldered ribbon cables, but the left to right were pins and sockets. There were no connectors near the top of the screen, the only reason that was the "magic" spot was leverage- it imparted maximum flex to the case, causing those board connections to move somewhat. After taking it apart and putting it back together, I didn't need to smack it again for a couple years. It started doing it again a couple months ago, and I can't be arsed to fix it again. It's now sitting on my patio with a "free if you can lift it" sign, and a LCD has taken its place in the living room.

\$\endgroup\$
4
\$\begingroup\$

I think this has become much less common as modern processes have made products tremendously more reliable than they once were, but I do have a real example and a real explanation:

A few years ago, I noticed that a friend was constantly tapping the front of his home theater audio receiver with his remote control before being able to do things like adjust volume and change the input. I asked him about it, but he had no idea why this worked. Years later, still doing the same thing, I asked him about it and someone out on the Internet had figured out that the IR receiver diode for that model (a through hole part on a nearly completely surface mount circuit board) had not been soldered well due to some process problem or contamination. After a few more months of going through his ritual of tapping the receiver, I finally convinced him to open it up and let me touch the solder joints with an iron. This successfully corrected the bad, intermittent solder joint and he uses it without a problem to this day (still more years later)!

The tapping was almost certainly re-adjusting the bad solder joint, possibly knocking off a small layer of oxidation that was preventing the low voltage from conducting. Once the incomplete joint was melted together, it was no longer a problem.

\$\endgroup\$
  • \$\begingroup\$ Tapping something like that would not loosen oxidation. Ever. \$\endgroup\$ – Passerby Jan 2 '16 at 5:12
  • 1
    \$\begingroup\$ @Passerby It's just a theory, but I can't imagine it takes much mechanical force to break an oxidation layer that is only a few atoms deep \$\endgroup\$ – Daniel Jan 2 '16 at 5:42
  • \$\begingroup\$ The offending oxide builds up again over time, leading to later failure, until the joint is soldered properly. \$\endgroup\$ – Neil_UK Jan 2 '16 at 8:19
  • 1
    \$\begingroup\$ I have a TV connected to a DVD player via SCART and every time I dust the TV, the next time I turn it on the screen is black. The "Millennium Falcon Manouver" always works and restores the image. My idea is that some contact on the scart is oxided and the tapping (not on the scart, on the the top of the TV) is helping in temporarily reducing the gap between conducting surfaces and starting a tunnelling through the oxide. Now, how it is sustained afterwards, that I still have to work out...;-) \$\endgroup\$ – Sredni Vashtar Jan 2 '16 at 12:17
  • \$\begingroup\$ FWIW I had a modern (less than 5 years old) USB DVD drive that used to respond to percussive maintenance until it stopped working entirely last year. It was the cheapest model available. This probably was a mechanical thing though, as the issue was that it would make strange crunching sounds and fail to read the DVD until I tapped it on the top. \$\endgroup\$ – Nathaniel Jan 2 '16 at 14:01
2
\$\begingroup\$

Calling it 'science' is probably a stretch but it generally makes the hitter feel better :-) The only valid explanation of why it works is mechanical - loose connection or dry joint. Modern electronics seems less amenable to percussive maintenance :-9

\$\endgroup\$
  • \$\begingroup\$ Agreed it must be something mechanical if a mechanical impact resolves it, but I don't understand what would then cause it to stop working after a random interval, from minutes to hours (without any causal physical whack). This leads me to doubt if it was a loose connection (which wouldn't go loose again by itself perhaps) or even something that was expanding due to operational heat (because then the time observed for it to go off again would be similar). \$\endgroup\$ – Pranab Jan 2 '16 at 7:22
  • \$\begingroup\$ It may not work with modern electronics. But I sure have seen my dad try to use it quite often when he is not entirely satisfied with the way some smartphone software responds to accelerometer inputs. \$\endgroup\$ – kasperd Jan 2 '16 at 13:10
2
\$\begingroup\$

I've been playing the electronics game for many decades. Percussive Maintenance [tm] has had its place along the way, and sometimes even works.

The obvious reasons seem to be some sort of intermittent contact, as various people have said. In devices with sockets for eg valves, relays, various plugs etc it is conceivable that a contact pair whose mechanical contact force was provided by a spring that had lost "temper" with age and heat and oxidation may be restored briefly. Less likely, but conceivable is that where a capacitor was dying, opening a contact elsewhere may cause a change in level which re biased a point till the capacitor recharged or transferred a pulse that pushed a circuit into a working mode. That's getting desperate - but so is PMing your equipment, and that does sometimes work.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.