0
\$\begingroup\$

I'm currently trying to reverse engineer a remote in my headphones (one of these three button things). The remote seems to work fine, but the mic is intermittent - at times it will work perfectly, at other times it will go completely silent.

In order to proceed further in trying to figure out what the mystery IC is doing, I need to figure out what microphone is being used so that I can identify the connections. All I know is it's an SMD mic with all pads underneath the chip so I can't trace them easily.

Mic Photo

Above is a photo of the microphone (apologies for the poor quality, I'll try for a better one later). I'm pretty sure that it must be an electret one as there are no inductors/filters anywhere that could otherwise create a supply for a MEMS microphone given that the whole controller is powered via a bias resistor embedded in the phone. (*)

The markings are as follows:

  • Top Line: 1005
  • Bottom Line: 7406 (I think. Might be 74C6)

The overall dimensions are roughly 4mm x 2.4mm x 1.1mm give or take. The black dot is the microphone membrane itself (so entry is on top).

The output of the mic is analogue not digital and seems to be on the order of a few millivolts.

A cursory Google search returned nothing, so I'm opening the floor for suggestions.

Again, the aim of this question is simply to identify the part.


(*) This is for an Android phone typically a 2.2V ish supply with a 2.2k or 5k series resistor (depends on which spec you read).


For those that want some background:

In terms of an overall picture of the circuit, as I say it's a 3-button remote in a set of headphones for controlling a mobile phone. The remote is supposedly both for Android and iPhone, but doesn't behave correctly. Initially on connection the mic works fine, but as soon as any button is pressed the mic will subsequently go silent until the whole thing is disconnected for a while.

My best guess at this is that my phone is an edge case that doesn't work properly with the way they have designed the controls - iPhones and Android use different signalling schemes so they seem to have some extra circuitry involved to switch between them. I am trying to work out the various connections so that I can start modifications to suit the Android spec.

Below is the schematic that I have thus far deduced. Again it is not possible to complete because I cannot physically see or measure what connections are beneath the microphone. In the schematic in the red box I've drawn the routing to match the physical layout as pictured in the photo.

Schematic of board

Of interesting note the voltage at the input is never more than 1.4V. When it is working correctly (i.e. mic can be heard) the voltage is about 1.06V. When the mic can't be heard it has jumped to 1.4V.

Pin 3 of the unknown 6-pin IC is sitting at 1.4V when the mic is not working. When it is working this voltage drops to near 0V. Beyond that it is hard to get an idea of what is going on from the voltages as the rest of the pins seem to be roughly the supply voltage with the exception of pin 1 which is sometimes 0 and other times floats up to 0.4V ish.

The voltage at the phone input when buttons are pressed are as expected - roughly 300mV for VOL-, 150mV for VOL+ and 0V for CNTRL. From what I can assume the unknown IC is probably some form of multiplexer, but there is no information marked on the chip.

But to be honest the aim is to simply identify the connections to the microphone after which I am planning to remove the mystery IC entirely and find a way to wire in the mic correctly.

\$\endgroup\$
  • \$\begingroup\$ A lot could be inferred from a partially reverse engineered schematic, and yes, that photo is terrible quality. My guess would be that its just the capsule. \$\endgroup\$ – PlasmaHH Sep 20 '17 at 11:47
  • \$\begingroup\$ There are a number of MEMS microphone packages that do not require external filtering, but instead have a digital interface. For example, the MP34DT01TR-M from STMicroelectronics: mouser.com/ds/2/389/mp34dt01-m-955068.pdf \$\endgroup\$ – DerStrom8 Sep 20 '17 at 11:59
  • \$\begingroup\$ Could still be a MEMS microphone. No reason they can't operate on a low voltage and be fed through the signal line like an electret mic. \$\endgroup\$ – JRE Sep 20 '17 at 11:59
  • \$\begingroup\$ @PlasmaHH schematic attached as far as I could identify it. I will get my DSLR out this evening when I get home as all I have to hand at the moment is a mobile phone (hence horrible quality). \$\endgroup\$ – Tom Carpenter Sep 20 '17 at 14:43
  • 1
    \$\begingroup\$ @DerStrom8 not a digital one it seems, very definitely an analogue output, added that to the q. \$\endgroup\$ – Tom Carpenter Sep 20 '17 at 14:44
0
\$\begingroup\$

Never mind.

Simply removed the mystery IC and shorted pins 2 and 3 of it to ground (the mic was working previously when pin 3 was 0V). As if by magic everything works.

I'm going to leave this as an unaccepted answer because it doesn't answer the main part of the original question which was to identify the microphone.

\$\endgroup\$
  • \$\begingroup\$ yes but the real issue was never the microphone, it was the switches and their effects \$\endgroup\$ – Sunnyskyguy EE75 Sep 20 '17 at 16:10
  • \$\begingroup\$ @TonyStewart.EEsince'75 I never said the issue was specifically with the microphone, if you had read the question it clear states that all I wanted was a part identification (something which is very definitely on-topic on EE.SE). Turns out by near pure guesswork my hack worked, but that's not a good way to engineer a solution, and it wasn't what the question was supposed to be about. \$\endgroup\$ – Tom Carpenter Sep 20 '17 at 16:24
  • \$\begingroup\$ I agree. but a clearer definition of the scope of the problem would have saved time wasted looking for skinny little rats in asian chip foundries. \$\endgroup\$ – Sunnyskyguy EE75 Sep 20 '17 at 16:36
-2
\$\begingroup\$

Model:TOS4737 Weifang Qinyi Electron Science & Technology Co. Limited

not same but similar

Since these companies cater to only large OEM's with 10k min MQO signing NDA's for specs. (minimum quantity order and non-disclosure agreement)

Assume they are not available and you have 2 or 3 connections. enter image description here Reverse engineering requires a scope trace, acoustic problem description and at least a plan on how you think you can improve this.

My advice

  • Compare Vbat with switch action, measure response time of analog switch. Bounce time & RC time.
  • inject open drain pulse to see if any improvement on analog output before processor, look for BT data line clues.
  • request replacement for defective product unless you prefer not to

update

If this is what I think it is...based on many web complaints of design quality failures, I have to suggest that you stay away from the V-MODA Vibrato Remote Headphones right now, and look to a different company when you want a new pair of headphones.

But by all means document the problem and request an RMA replacement ASAP.

final comment

Although I was going to suggest the solution the Op gave in answer. The question was flawed. The problem surround the transient effects of disconnecting BT and not the analog audio , its output or its specs or the lack of part number ID source.

FYI this is how some designs work to use analog DC steps to encode mic volume +/- increments in an ASIC based on an Android standard. enter image description here

\$\endgroup\$
  • \$\begingroup\$ Come on Tony, not even a datasheet? There are thousands and thousands of mics that look like this, why did you choose this particular one? \$\endgroup\$ – DerStrom8 Sep 20 '17 at 12:09
  • \$\begingroup\$ How many million parts do you need? \$\endgroup\$ – Sunnyskyguy EE75 Sep 20 '17 at 12:11
  • \$\begingroup\$ Tom's looking for a datasheet for his part, not one that roughly looks the same (I found 50 pages on Mouser in a 10 second search that looked similar but none of the ones I looked at had the correct numbering scheme). Posting a picture/part number of one that looks similar when there are probably millions of others out there is not very helpful \$\endgroup\$ – DerStrom8 Sep 20 '17 at 12:13
  • 1
    \$\begingroup\$ @TonyStewart.EEsince'75 if there is any clarification you require, that's what comments are for - we're both well versed in EE.SE to know how it works. There's a schematic to come once I get a moment, but it's not going to give much information as lack of being able to see under the mic means I can't easily determine how the mic is connected - that's specifically why I need to identify the part so I can at least figure out what connections are where. \$\endgroup\$ – Tom Carpenter Sep 20 '17 at 13:32
  • 1
    \$\begingroup\$ "the solution is not to identify the mic" - no, but the key to figuring out the solution is to identify the mic. That is the goal of this question. I'm a competent enough engineer to build a replacement that would work fine, but to do so would cost more than the headphones did. The end goal/solution is to modify the existing circuitry to suit my needs, but to do that I need to know the specific mic used so that I can find it's pinout. In any case I expanded the question with technical details to satisfy your curiosity, but I'm not convinced they will help with the question "Identify this part" \$\endgroup\$ – Tom Carpenter Sep 20 '17 at 14:41

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.