Any microphone experts out there?
in [Design]
|
Prev: Steve Jobs taking a "Poo Poo" :)
Next: Last Call for Papers Reminder (extended): World Congress on Engineering and Computer Science WCECS 2010
From: Grant on 18 Jul 2010 18:17 On Sun, 18 Jul 2010 13:36:42 -0400, Phil Hobbs <pcdhSpamMeSenseless(a)electrooptical.net> wrote: >Jim Thompson wrote: >> On Sun, 18 Jul 2010 08:37:46 +0100, "Midge" <midge(a)somewhere.com> >> wrote: >> >>> Have a small microphone mixer http://www.audiomate.co.uk/PDF/G105C.pdf >>> >>> I've been using a couple of dynamic microphones on it, but I wondered if it >>> was possible to either get a tie-clip microphone (these seem to be electret >>> condenser) which can be plugged straight in OR match the different >>> technologies somehow. >>> >>> Thanks, Midge. >>> >> >> If the electret has a built-in FET buffer, all you'll need is power >> for the FET. >> >> Otherwise the Audiomate input impedance is too low. >> >> ...Jim Thompson > >As an aside, does the standard explanation of how electret mics work >bother anyone but me? It's usually stated that the 'frozen in' >electric polarization in the material works just the same as the applied >voltage in a real capacitive mic. Why is frozen electric field in plastic any more difficult than frozen magnetic field in a whatsit to make a magnet? > This is of course nonsense--charges >such as air ions migrate around and eventually neutralize any E field >external to the electret material, so that waving it around does nothing >whatever. (Leakage--that's why you have to keep the capacitance mic's >supply connected for it to work, duh.) Yeah but can you easily neutralise the charge trapped in the plastic? > >The reality appears to be that they work by a combination of the >piezoelectric effect (like a quartz crystal) and the variation of >capacitance with stress (like coaxial cable with a DC voltage on it). That too. Grant. > >Any comments from audio folks? > >Cheers > >Phil Hobbs
From: Grant on 18 Jul 2010 18:21 On Sun, 18 Jul 2010 12:49:54 -0700, Tim Wescott <tim(a)seemywebsite.com> wrote: >On 07/18/2010 10:36 AM, Phil Hobbs wrote: >> Jim Thompson wrote: >>> On Sun, 18 Jul 2010 08:37:46 +0100, "Midge" <midge(a)somewhere.com> >>> wrote: >>> >>>> Have a small microphone mixer http://www.audiomate.co.uk/PDF/G105C.pdf >>>> >>>> I've been using a couple of dynamic microphones on it, but I wondered >>>> if it was possible to either get a tie-clip microphone (these seem to >>>> be electret condenser) which can be plugged straight in OR match the >>>> different technologies somehow. >>>> >>>> Thanks, Midge. >>> >>> If the electret has a built-in FET buffer, all you'll need is power >>> for the FET. >>> >>> Otherwise the Audiomate input impedance is too low. >>> >>> ...Jim Thompson >> >> As an aside, does the standard explanation of how electret mics work >> bother anyone but me? It's usually stated that the 'frozen in' electric >> polarization in the material works just the same as the applied voltage >> in a real capacitive mic. This is of course nonsense--charges such as >> air ions migrate around and eventually neutralize any E field external >> to the electret material, so that waving it around does nothing >> whatever. (Leakage--that's why you have to keep the capacitance mic's >> supply connected for it to work, duh.) >> >> The reality appears to be that they work by a combination of the >> piezoelectric effect (like a quartz crystal) and the variation of >> capacitance with stress (like coaxial cable with a DC voltage on it). >> >> Any comments from audio folks? > >Some piezoelectric materials do need to be polarized to work. > >I suspect that the "charge frozen in" explanation is valid to the extent >that anything else could be without going into the details of the >physics, and it has the desirable property that it confuses people into >shutting up while still using the device correctly. > >Do you accept the "trapped charges on a floating gate" model of EEPROM >behavior? How is that different from the "frozen-in charges" of the >piezo microphone? That's the one I wanted to mention too :) What gets me is the current 3bit per cell technology, reliably trapping one of eight voltage levels in the memory cell gate, and remembering that level for years... Grant.
From: Phil Hobbs on 18 Jul 2010 19:39 Tim Wescott wrote: > On 07/18/2010 10:36 AM, Phil Hobbs wrote: >> Jim Thompson wrote: >>> On Sun, 18 Jul 2010 08:37:46 +0100, "Midge" <midge(a)somewhere.com> >>> wrote: >>> >>>> Have a small microphone mixer http://www.audiomate.co.uk/PDF/G105C.pdf >>>> >>>> I've been using a couple of dynamic microphones on it, but I wondered >>>> if it was possible to either get a tie-clip microphone (these seem to >>>> be electret condenser) which can be plugged straight in OR match the >>>> different technologies somehow. >>>> >>>> Thanks, Midge. >>> >>> If the electret has a built-in FET buffer, all you'll need is power >>> for the FET. >>> >>> Otherwise the Audiomate input impedance is too low. >>> >>> ...Jim Thompson >> >> As an aside, does the standard explanation of how electret mics work >> bother anyone but me? It's usually stated that the 'frozen in' electric >> polarization in the material works just the same as the applied voltage >> in a real capacitive mic. This is of course nonsense--charges such as >> air ions migrate around and eventually neutralize any E field external >> to the electret material, so that waving it around does nothing >> whatever. (Leakage--that's why you have to keep the capacitance mic's >> supply connected for it to work, duh.) >> >> The reality appears to be that they work by a combination of the >> piezoelectric effect (like a quartz crystal) and the variation of >> capacitance with stress (like coaxial cable with a DC voltage on it). >> >> Any comments from audio folks? > > Some piezoelectric materials do need to be polarized to work. > > I suspect that the "charge frozen in" explanation is valid to the extent > that anything else could be without going into the details of the > physics, and it has the desirable property that it confuses people into > shutting up while still using the device correctly. > > Do you accept the "trapped charges on a floating gate" model of EEPROM > behavior? How is that different from the "frozen-in charges" of the > piezo microphone? > I'm not denying the existence of electrets, either the PVDF plastic ones or the floating-gate CMOS ones. Inside the material, charges aren't free to migrate around, so there's no worry. It's outside, where air ions and charged dust can accumulate, where you can't hold a static charge for too long. (Rubbing a toy balloon on your hair can make it stick to the ceiling, but eventually it unsticks itself as the charge leaks away--or air ions arrive to neutralize it.) The same is true of magnets, except that there is no such thing as a magnetic monopole--i.e. the magnetic equivalent of free charge doesn't exist. That's why magnets continue to produce a field external to themselves, but electrets don't. I think you're probably right that it's a "Mr Science" type explanation, designed to shut people up rather than to actually answer the question. Cheers Phil Hobbs -- Dr Philip C D Hobbs Principal ElectroOptical Innovations 55 Orchard Rd Briarcliff Manor NY 10510 845-480-2058 hobbs at electrooptical dot net http://electrooptical.net
From: Robert Baer on 20 Jul 2010 01:29 Phil Hobbs wrote: > Jim Thompson wrote: >> On Sun, 18 Jul 2010 08:37:46 +0100, "Midge" <midge(a)somewhere.com> >> wrote: >> >>> Have a small microphone mixer http://www.audiomate.co.uk/PDF/G105C.pdf >>> >>> I've been using a couple of dynamic microphones on it, but I wondered >>> if it was possible to either get a tie-clip microphone (these seem to >>> be electret condenser) which can be plugged straight in OR match the >>> different technologies somehow. >>> >>> Thanks, Midge. >> >> If the electret has a built-in FET buffer, all you'll need is power >> for the FET. >> >> Otherwise the Audiomate input impedance is too low. >> >> ...Jim Thompson > > As an aside, does the standard explanation of how electret mics work > bother anyone but me? It's usually stated that the 'frozen in' > electric polarization in the material works just the same as the applied > voltage in a real capacitive mic. This is of course nonsense--charges > such as air ions migrate around and eventually neutralize any E field > external to the electret material, so that waving it around does nothing > whatever. (Leakage--that's why you have to keep the capacitance mic's > supply connected for it to work, duh.) > > The reality appears to be that they work by a combination of the > piezoelectric effect (like a quartz crystal) and the variation of > capacitance with stress (like coaxial cable with a DC voltage on it). > > Any comments from audio folks? > > Cheers > > Phil Hobbs > Regarding presumed "frozen in" charge.. About 50 years ago, in high school, there was this "true" rumor of a related experiment: Take a sheet of glass and place a metal plate on each side to make a capacitor, then charge it (2KV to 20KV if i remember correctly). Disconnect completely and then remove the glass sheet. Measure the voltage (virtually zero). Remove volt meter, replace the glass sheet and measure again. According to the story, one reads near the original charging voltage. But it gets better.. Short this cap and verify zero voltage. Wait..(minutes, i think an hour max i think) then re-measure and find a rather shocking (literally) voltage present. ** Naturally, that second aspect is well known - most especially to TV repairmen dealing with those CRTs.
From: Fred Bartoli on 20 Jul 2010 05:01
Jim Thompson a �crit : > On Sun, 18 Jul 2010 14:33:20 -0400, Phil Hobbs > <pcdhSpamMeSenseless(a)electrooptical.net> wrote: > >> Jim Thompson wrote: >>> On Sun, 18 Jul 2010 13:36:42 -0400, Phil Hobbs >>> <pcdhSpamMeSenseless(a)electrooptical.net> wrote: >>> >>>> Jim Thompson wrote: >>>>> On Sun, 18 Jul 2010 08:37:46 +0100, "Midge" <midge(a)somewhere.com> >>>>> wrote: >>>>> >>>>>> Have a small microphone mixer http://www.audiomate.co.uk/PDF/G105C.pdf >>>>>> >>>>>> I've been using a couple of dynamic microphones on it, but I wondered if it >>>>>> was possible to either get a tie-clip microphone (these seem to be electret >>>>>> condenser) which can be plugged straight in OR match the different >>>>>> technologies somehow. >>>>>> >>>>>> Thanks, Midge. >>>>>> >>>>> If the electret has a built-in FET buffer, all you'll need is power >>>>> for the FET. >>>>> >>>>> Otherwise the Audiomate input impedance is too low. >>>>> >>>>> ...Jim Thompson >>>> As an aside, does the standard explanation of how electret mics work >>>> bother anyone but me? It's usually stated that the 'frozen in' >>>> electric polarization in the material works just the same as the applied >>>> voltage in a real capacitive mic. This is of course nonsense--charges >>>> such as air ions migrate around and eventually neutralize any E field >>>> external to the electret material, so that waving it around does nothing >>>> whatever. (Leakage--that's why you have to keep the capacitance mic's >>>> supply connected for it to work, duh.) >>>> >>>> The reality appears to be that they work by a combination of the >>>> piezoelectric effect (like a quartz crystal) and the variation of >>>> capacitance with stress (like coaxial cable with a DC voltage on it). >>>> >>>> Any comments from audio folks? >>>> >>>> Cheers >>>> >>>> Phil Hobbs >>> I've used them many times in hearing-aid chip designs, but I have no >>> idea of how they actually work. The leader, in hearing-aid usage is >>> Knowles... >>> >>> http://www.knowles.com/search/technologies/tech_condenser.jsp >>> >>> But I've also used a MIMS equivalent (Akustica) which was simply a >>> capacitor biased with a _very_high_valued_ "resistor", powered from a >>> 4X charge-pump (my part of the design). >>> >>> ...Jim Thompson >> I just did a design for a startup making radiation sensors from >> conductive polymers--it used a $5 CCFL inverter with a bridge rectifier >> and a rail splitter to form half of the readout bridge. It makes 600V >>from a 5V supply. There are a lot of things you can do with those, for >> very little dough. (Maybe even make capacitive mics.) >> >> Cheers >> >> Phil Hobbs > > I don't the luxury of a wall socket for my hearing-aid designs ;-) > > ...Jim Thompson How about an ear plug? -- Thanks, Fred. -- Thanks, Fred. |