3
\$\begingroup\$

I'm a safe engineer and have little to no idea regarding electrical engineering, however I'm more than capable of soldering components and very willing to learn but also begrudge spending £300 for a sound amplification system (again!!)

I've seen a couple systems that people have made themselfs but not inquired further as to ask for a component list etc. I understand the basics (very basic) I realise I'll need mic's to pick up the sound and that the circuit will need to amplify the sound into a pair of earphones in order for me to deduce what is going on through a couple inches of steel!!

Advice on components required etc etc. I have looked into it in some detail but as yet I'm at a loss. What I can say is they seem to be similar in construction to covert listening devices ie the ones that listen through walls etc but still would not know where to begin So I thought I would come to the experts.

\$\endgroup\$
  • 3
    \$\begingroup\$ You need to define what exactly you are trying to listen too. You mention offhandedly a couple inches of steel. The full problem needs to be clearly stated up front. \$\endgroup\$ – Olin Lathrop Sep 25 '12 at 21:56
  • 1
    \$\begingroup\$ Thanks for the reply, I'll try to explain the best I can so it depends what your listening to in each different safe as the locks tend to differ but in all cases your detecting the sound of metal on metal these noises are very faint. For example in a key lock you may have 7-9 levers each lever has a true gate(the height the lever needs to be lifted in order for the boltstump to pass through the lever) it also has false gates or notches. These notches produce a different sound against the bolt stump than the true gate this is what I'm listening for whilst picking the lock. Hope this will help \$\endgroup\$ – user13553 Sep 25 '12 at 22:14
  • \$\begingroup\$ My old amp I assumed had a contact mic as I used to fix the mic on the safe door with a magnet, it simply just used to attach itself and I would move it around the door for best affect. My old amp also had noise cancelling capabilities but I realise this could be a long shot!! \$\endgroup\$ – user13553 Sep 25 '12 at 22:18
  • 2
    \$\begingroup\$ Ah, so you're trying to build something to aid in safe-cracking. You should have made this clear up front. While there are legitimate uses for such things, there are also many illegitimate ones. No, I'm not going to help you break into other people's safes. Perhaps the Federal Burglar's Institute (FBI) can help. \$\endgroup\$ – Olin Lathrop Sep 26 '12 at 13:52
  • 1
    \$\begingroup\$ This sounds like a classic make/buy decision that is, for whatever reason, being resolved the wrong way. How many hours do you expect to spend designing, building, and tweaking your gadget, how much is your time worth to you per hour, and how does the product of those two numbers compare to £300 (plus credit card charges, presumably)? \$\endgroup\$ – John R. Strohm Sep 27 '12 at 15:20
4
\$\begingroup\$

I'll advise you to not build this yourself, and instead spend the money on something designed for this task. If you value your hearing as much as I value mine, it is money well spent.

Simply taking the signal from a mic and amplifying it and sending it to a headphone is relatively easy. On a scale of 1 to 10, it is a 1 or 2. Doing it with high audio quality and low noise is about a 2 or 3. Doing it with safety is about a 5 or 6. I would argue that anything 2 or beyond is not something a beginner EE is able to do as their first or second project.

So I've mentioned safety a couple of times, so here is what I mean: The amount of amplification required for this is huge. Not unreasonably huge, but still huge. Let's say that you're at a safe with this device in your ear and your cell phone rings, or something falls on the safe, or someone is behind you and talks loudly. That sound will be amplified a huge amount and sent into your ears. Without some safety circuits, you will receive ear damage. Maybe temporary, or maybe permanent, but still damage.

A proper device will have something called a "limiter", which will detect the loud noise and reduce the volume before it gets into your ear. It turns out that designing a limiter that both sounds good and protects your ears is a difficult thing. You can roughly divide up the limiter types into two categories: Analog and Digital.

One component of a good limiter is an audio delay. A short delay (less than 100 mS) is required to properly detect the audio peak and reduce the level BEFORE it goes to your ear.

If you look up limiter circuits on the 'Net you will likely find analog limiters. The problem with analog limiters is that it is hard to do an analog audio delay. It can be done, but it is difficult and often sounds terrible. So almost all analog limiters do not have a audio delay. This is often OK for limiters for recording music, but is NOT ok for protecting your hearing. The part of the circuit that turns the volume down does not respond fast enough to protect your ear (the audio delay compensates for this).

Digital limiter circuits get around this by doing things digitally. Normally this is a small DSP plus ADC and DAC (sometimes all integrated into the same chip). Then some software is written for the DSP to do the limiting function.

For an application like this, the DSP would be used for other things too. It might enhance some frequencies while removing others-- all to make safe cracking easier.

For someone who doesn't know electrical engineering at all, and probably doesn't know the math behind digital signal processing, this project is probably way beyond your ability. This isn't an insult, just reality. We can't all be proficient at everything. You could learn it and become proficient, but that can take a long time. Maybe years.

Disclaimer: It might be possible to make an analog limiter that does protect your hearing and doesn't require years of training to do. But I can guarantee that it'll sound terrible and not be near as good as if it was done correctly. I also would not want to be the guy who is the beta tester.

\$\endgroup\$
  • 1
    \$\begingroup\$ Don't assume that an amplifier will have the same gain when presented with a signal 100,000 times larger than it was designed for. It'll clip, becoming non-linear and sounding terrible, but not hitting your ear with a dangerous level of power. No delay circuit needed either. And the speakers themselves won't handle a huge signal without distortion/clipping, either. So power-limiting the amplifier is a viable approach, and it shouldn't particularly affect small-signal linearity, so you won't have terrible sounds except when the input goes out of range. \$\endgroup\$ – Ben Voigt Sep 26 '12 at 17:57
  • 1
    \$\begingroup\$ @BenVoigt A "normal" output level will be far below the power rails of the output amplifier. If the input is high enough to clip the output, the output could still be higher than what is required to damage your hearing. Yes, it's not 100,000x the input, but it could be 100x or even just 10x if the input noise is loud enough. And the output is highly distorted which could make things worse. \$\endgroup\$ – user3624 Sep 26 '12 at 19:00
  • \$\begingroup\$ I'd be surprised if the amplifier for headphones has more than 30dB of headroom above your normal level (if it does, pick an amplifier with less power). And that may certainly be startling and uncomfortable, but still below damaging. Remember that audio perception works on a logarithmic scale... you need a huge amount of power increase to make an output go from "pleasant" to "really loud". You need equipment designed for high power in order to get dangerously loud, whether with or without distortion. \$\endgroup\$ – Ben Voigt Sep 26 '12 at 19:20
  • \$\begingroup\$ @BenVoigt There are lots of factors to consider, including peak vs RMS (I.E., Crest Factor). An amplifier that can accurately reproduce the fast impulses required for reproduction of many sounds (without excessive RMS power) can frequently deliver severe levels of sound when that amplifier is clipping. Add to that the wide variation of headphone efficiency can in fact lead to being able to deliver audio above the 140 dB peak sound pressure limit for impulse sounds that OSHA imposes. \$\endgroup\$ – user3624 Sep 26 '12 at 19:43
  • \$\begingroup\$ @user3624 The fact that you can devise a scenario which is unsafe without clipping (and I don't dispute that such scenarios exist) doesn't mean it is a relevant scenario for the OP's task. \$\endgroup\$ – alephzero Jun 23 '18 at 15:20
3
\$\begingroup\$

I am not an expert but I have made some circuits to listen to sounds through pipes and solid objects and I always run into the noise issue, so I would suggest using some kind of ICs that are designed for this job like TDA7297, or you can even find something ready on ebay link or this link2.

You should be able to use the microphone from your old amplifier or use a condenser microphone like the WM-61A which has a nice frequency response, just make sure that your circuit provides phantom power for the mic.

\$\endgroup\$
  • 3
    \$\begingroup\$ And make sure to use a limiter or compressor so that when someone taps on the safe you're cracking you don't go completely deaf. \$\endgroup\$ – travisbartley Sep 26 '12 at 1:24
0
\$\begingroup\$

Get a mechanic's stethoscope. There are acoustic ones out there as well as electronic ones.

\$\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.