Getting a "thump" from external speakers.

[Greg Miller]

Very interesting, Chris, including the trivia!

I like your idea of using a switched-mode power supply to test for high impedance mains wiring. I hadn't thought of that but it makes good sense. Maybe the OP's TV/monitor has such a supply (quite likely unless it's ancient) in which case that might explain the entire scenario.

I haven't seen a fused power plug since the 1950s when my grandfather had an old floor-model AM/MW/SW radio with cartridge fuses in both sides of the line plug. (Nothing here was polarized then, so, depending on chance, either side of the plug might have been connected to the neutral side of the mains.) Of course if the fuse in the neutral side had blown, the entire radio would have been "hot" with respect to ground. (And back then, it was not uncommon to have a pair of disc capacitors for "noise suppression," one tied to "hot" side of AC, one tied to "neutral" side of AC, and the remaining lead of each cap tied to the metal chassis, making a capacitive voltage divider. So even in the best of times, the chassis was floating at 1/2 of the mains voltage. I got "surprised"on more than one such set (the surprise ranging from a tingle to a strong zap)... the current was limited only by the value of the capacitor (i.e. the relative impedance at our 60Hz mains frequency).

Yes, the metal creep can happen with any conductor, it's just more pronounced with Aluminum. Older terminals (on switches, outlets, breakers, etc.) were made from brass, which corroded rapidly in contact with Aluminum; newer terminals are made from an alloy that is much less prone to corrosion. That, in conjunction with anti-corrosive paste (NoAlOx, DeOx, Penetrox, etc.), makes the problem much less serious than it was when Aluminum wiring first went into widespread use.

BTW, vacuum tube (valve to you) rectifiers were the best, in terms of inrush current. They conducted not at all when cold, so the current just gradually ramped up as the filaments (or cathodes) got warmer. But on the downside, all those filaments were a very low resistance when cold, so turning on the filament supply caused a pretty significant -- though brief -- downward spike in the mains voltage.

But about your "dead PCs" scenario: don't you folks have slow-blow fuses? A lot of equipment has same in the mains circuit, although "fast blow" in the downstream DC buses.

[Denis Danatzko]

Greg and Chris,
THANKS SO MUCH for your patience and perseverance.

As for the noise, Item # 1 in Greg's response, it's definitely a "thump", and not a "pop" or "click". (I've been unusually busy of late, but I promise to record a brief audio segment so you can hear it, it just may not be until the weekend).

As for # 2 in your reply, I do have a meter, but, again, I probably won't get to that until the weekend.

However, I did notice yet another anomaly that has me totally perplexed. I'm not trying to play "stump the experts" here, but today I noticed something totally unexpected. Here's the tale:

I've recently recorded and edited some footage for a customer. During editing, and what prompted me to pose my dilemma here, I was getting the "thump", it seemed during editing whenever the audio would peak/red-line, i.e. exceed 0 dB. I thought it was my speakers "going bad". Well, I did some rather simple noise removal in Audition CS6, saved that audio file, finished editing, saved my project in PPro CC, exported the sequence via AME to one of the vimeo presets, and posted the results to vimeo for the customer to approve. No thump in vimeo.

Today, after having moved the wall wart for the speakers to another outlet and circuit, and not getting the thump since doing so, I re-opened that project to write the finished product to an SDHC card for the customer, and played the final cut from the sequence in the PPro timeline just to refresh my memory of it. Partway through, I start hearing the thump again, as if it had been "recorded" as part of the finished product. This is the exact same file I uploaded to vimeo. Additionally, I don't hear the thump when I play the footage from another computer.

I don't have a separate audio card in any of my computers. In the computer where I hear the thump, I have an nvidia Geforce GTX 670 and both nVidia Control Panel and Realtek HD Audio Manager. However, in the computer which first produced the thump, I have an nvidia Quadro FX 1500. That has, or I should say, had Nvidia Control Panel, but no equivalent I know of to Realtek Audio Manager. (That computer began acting-up months ago, which, along with PPro x64, spurred me to buy my current one).

Is it possible, if a sudden drain in current is causing the thump, that it would also be recorded in my edited audio file and/or my project file? Because I edited the audio in Adobe Audition, I have an audio-only version after my "cleanup", and that, too produces the thump when I play it on the same computer on which it was edited. When just that wav file is played on another computer, I don't hear the thump, and I've played it with VLC player and Windows Media Player.

I'm wondering if this is a bug in either Audition, or PPro and however/whatever it uses to process audio, or some bug in nVidia Control Panel or Realtek Audio Manager. (It may also be one of the perils of using a video card not "certified" by nVidia for CS6 or CC .In any case, I was totally unprepared for this quirk.

I have another editing project coming my way within the next day or 2, and will see if plugging the speaker wart into a different circuit has any effect on what gets stored while editing that one, and will report back here with results. (Though I just realized that I can see the thump in the waveform on the timeline, so it MUST be getting recorded. I'm surprised; I wouldn't expect that to happen).

Thanks again for your help and patience.

[Chris Soucy]

But about your "dead PCs" scenario: don't you folks have slow-blow fuses? A lot of equipment has same in the mains circuit, although "fast blow" in the downstream DC buses.

Yeah, well we, er, they, do.

The problem is that UK law states that every single plug on every single piece of kit that "plugs" into a standard UK 13 amp socket has to have it's own inbuilt fuse, rated for the device it's connected to (and I mean EVERYTHING - fridges, tv's, vacuums, bedside lights - you name it, if it plugs in, it has it's own fuse in the plug, live side).

Great idea in theory, don't work so well with the older "hit the button and pray" switch mode power supplies. As I mentioned, the only plug fuses (from memory) available were/are 3, 5 and 13 amps, no "slow blow" option on offer.

The PC terminals themselves would have had double protection from their own internal fuses and the fact that switch mode power supplies (of the day) when they decided to do the flash/ brown smoke thing tended to fuse themselves anyway.

But, the manufacturers quickly worked out that the only way to stop them destroying sensibly rated plug fuses was to fit them standard with 13 amp ones, which could withstand the humungous inrush current without karking it in very short order.

Then the "expert" (ha!) dipsticks come along, go anal and the whole thing goes straight down the toilet.

Speaking of the "Flash, brown smoke" thing, another snippet to keep the viewers entertained.

I was the guy in charge of the Operational DP function back in the bad old mainframe days, at a company which shall be nameless (I will not mention Sotheby's, OK?).

One morning, after the scheduled preventive maintenance by the Burroughs engineers that happened every Wednesday from 8 AM till 10 AM, I just happened to be standing behind the head computer operator when he asked me "good to go?".

I say "yep", he hits the green "GO" button, there's a blinding flash from the left side of the mainframe stack immediately in front of us and three of the removable panels are blown clear off the side of the stack frame. At the same time there's a bang like a grenade going off in a dustbin, right next to you.

The scorch mark(s) were still on the wall about 15 feet away when I left the company some years later.

The main computer power supply had utterly shat itself, nothing to do with the engineers, they hadn't been near it.

Cue: sea of engineers trying to figure out what the hell happened. The days pass, components are tested, replacement flown in from the States and the entire power supply is rebuilt from scratch, the (our) company is facing bankruptcy as it can't get the bills out and I'm getting piled on from everyone from the Chairman on down.

The engineers finally say "It's good to go".

Same head OP hits the green button, almighty flash, 3 panels blow off the side of the stack, grenade goes off in dustbin and the wall gets another scorch mark.

I pack up entire site and decamp (with crew) to the Burroughs "failsafe" backup site in the City (of London), only to find most of their tape drives won't read our tapes.

I might only look about 60, I sometimes feel about 135.

PS: The head OP in question was never again seen to hit the green "ON" button, always delegating to a junior member of staff. Don't blame him.

A few years later I was the poor unfortunate who was the sucker the day we took down the entire system, network and all, lock stock and barrel, for over a week. Man, did I age a few years that week (and no, I didn't "DO" it, I just hit the "GO" button which let the systems self destruct, ooooh, Bugger!. You've never heard Anglo Saxon adjectives like it in your entire life, starting right at the top and working their way all the way to the bottom)


CS

[Greg Miller]

Great story, Chris! You certainly got more than a "thump." I'll bet Mr. Danatzko doesn't feel so bad now.

[Greg Miller]

Originally Posted by Denis Danatzko

Is it possible, if a sudden drain in current is causing the thump, that it would also be recorded in my edited audio file and/or my project file?

(I just realized that I can see the thump in the waveform on the timeline, so it MUST be getting recorded. I'm surprised; I wouldn't expect that to happen).

• First, it seems an extreme coincidence that you would suddenly develop "thump" problems in two different pieces of gear at the same time.

• Second, your computer's sound system is running from its own highly regulated supply. If the internal DC voltage is dropping low enough to get a thump into the sound, I'd expect your video to be flickering, or the whole computer to be re-booting, BSODing, or failing in some significant way.

• Third, a very serious analog problem might superimpose a thump on the audio while it is being recorded in analog form. In other words, if you were using the computer's mic or line input, and recording in analog mode, there's a slight chance a "thump" might get recorded as part of the track. But there is no way that an analog noise will be added to the file once it has been digitized. When you open the file to edit or process it, that's all happening as digital mathematics, there is no analog circuitry involved (except for the playback to your monitors). So once the file has been captured, it's immune to any further analog problems.

• Fourth, what do you mean by,

I was getting the "thump", it seemed during editing whenever the audio would peak/red-line, i.e. exceed 0 dB.

How can the audio exceed 100%? By definition that's impossible. It might peak at 100%, but beyond that, you will certainly generate digital clipping. That will sound terrible, but nothing that I would describe as a "thump"... it will have a lot of higher frequency harmonics and will sound distorted and raspy. I make a point to be sure my files never exceed -0.25 dB. And if an earlier generation file is clipped, it's time to go back and replace that with clean audio before proceeding.

Q.: How did the audio get into the computer in the first place? Please explain in detail.

Can you post just an audio file(s) -- preferably WAV -- of some of the parts of the file where you hear and see the "thump"? It's probably best if you just post the bad part, with a few seconds of good audio before and after. At this point, I would think there's no point in posting the entire file.