Replicating an outdoor microphone (DH-HAP120-V)

[Turlipe]

Hello,

For a project I'm using a couple of Dahua DH-HAP120-V microphones.
The specs provided are enough for my use (outdoor far-field audio recording). However the excessive price of the product and the poor integration on my device are problematics.

I'm trying to reproduce the microphone for a custom integration, but after searching for online documentation and contacting online support I didn't find any infos about its composition.

I'm have no experience in building microphones and would like to know if you have any idea about the composition of a microphone like this one, what are its key components to build a new one, or if their are cheaper and equivalent alternatives.

Thanks in advance.

 

[rogs]

I'm guessing it probably uses a waterproof electret capsule.
Those can be found quite reasonably priced. See here for example: https://www.cuidevices.com/waterproof-microphones
How you 'integrate' that into your system will depend on what kind of audio recorder input you intend to use?

 

[cyrano]

Thanks, Roger. Never found that link. My Google-Fu sure is rusty.

I've had a peek at the tech data for that DH-HAP120. It seems to include a mic preamp. It uses 12V, 60 mA. That's a bit too much for an electret only circuit. And the output is 2V.

From your link, for the OP: they have a development board with four different capsules for less than 15$. That should give you an idea and it's already soldered, so no chance of damaging the capsules if you're not used to soldering.

Mind you, these are not hydrophones. IP67 is weatherproof, only submersible for minutes.

 

[Turlipe]

Thanks rogs and cyrano for your heads-up.

If I understand correctly, the mics are probably composed of a weatherproof electret capsule and a mic preamp.
No problem if capsules aren't hydophones, it's just for outdoor use without water immersion.

I read in the DH-HAP120-V doc that the default gain is 44 dB, is this why the preamp is used for ? Or does it have other purposes ?
If it is useful just for gain, is software added gain can be an alternative to the preamp ?

For the audio recorder, I wanted to use a Scarlett 8i6 in order to retrieve signals of 6 microphones synchronously on a computer but its surely an overkill. I tried to find more appropriate synchronization solution but didn't found anything (something like the ESI Maya44 USB+ would be perfect but it only takes 4 inputs). I will probably create a new topic for this issue.

 

[rogs]

Thanks rogs and cyrano for your heads-up.

If I understand correctly, the mics are probably composed of a weatherproof electret capsule and a mic preamp.
No problem if capsules aren't hydophones, it's just for outdoor use without water immersion.

I read in the DH-HAP120-V doc that the default gain is 44 dB, is this why the preamp is used for ? Or does it have other purposes ?
If it is useful just for gain, is software added gain can be an alternative to the preamp ?

Sorry if these are trivial questions I'm an audio noob.

For the audio recorder, I wanted to use a Scarlett 8i6 in order to retrieve signals of 6 microphones synchronously on a computer but its surely an overkill. I tried to find more appropriate synchronization solution but didn't found anything (something like the ESI Maya44 USB+ would be perfect but it only takes 4 inputs). I will probably create a new topic for this issue.

The capsules I linked are just that - electret capsules. That style of device will require both a pre-amp and DC power to function.
The 2 microphone inputs on the Scarlett device you linked to could be fitted with 'Phantom power to PIP' adaptors - (one of the devices listed in this note would do the job: https://www.zachpoff.com/resources/testing-xlr-adapters-with-plug-in-power/ )
However, that is only suitable for the 2 mic inputs on the Scarlett. The remaining 4 line inputs would require mics with preamps ... and those inputs have no DC output provision to power the mics, so not really suitable.

Don;t forget, the output from 2 pin electret capsules is also unbalanced. With the application of short(ish) cable runs and screened cable this is not usually a problem, but the inclusion of a pre-amp - along with DC power - in the mic housing itself makes longer cable runs much less of a concern.

Outputs from most 2 pin electret capsules will need to interface with a 'Plug in Power' (PIP) style of input, to work without a remote preamp.
Very few mixers include that type of input - hence the need for phantom power adaptors for the mic inputs.
Most portable audio recorders with 3.5mm jack inputs will often include a 'PIP' facility - along with computer sound cards and smart phone mic inputs. ....... But not mixers!

The electronics involved are not complex, but the configuration you require is not that common..... which is why people like Dahua can charge the higher price they do.

Someone may be able to recommend a 6 channel mixer with PIP mic inputs and a USB output?.... I've never encountered one myself...

For a 'one off' installation you may be better off looking at a mixer with 6 'phantom power' capable mic inputs , and 6 XLR/PIP adaptors.

Whether that would be cheaper than a 6 input line mixer for the Dahua devices, I'm not sure?
( And don't forget you still need to provide DC power to those devices as well...... )

 

[Turlipe]

The capsules I linked are just that - electret capsules. That style of device will require both a pre-amp and DC power to function.
The 2 microphone inputs on the Scarlett device you linked to could be fitted with 'Phantom power to PIP' adaptors - (one of the devices listed in this note would do the job: https://www.zachpoff.com/resources/testing-xlr-adapters-with-plug-in-power/ )
However, that is only suitable for the 2 mic inputs on the Scarlett. The remaining 4 line inputs would require mics with preamps ... and those inputs have no DC output provision to power the mics, so not really suitable.

Don;t forget, the output from 2 pin electret capsules is also unbalanced. With the application of short(ish) cable runs and screened cable this is not usually a problem, but the inclusion of a pre-amp - along with DC power - in the mic housing itself makes longer cable runs much less of a concern.

Outputs from most 2 pin electret capsules will need to interface with a 'Plug in Power' (PIP) style of input, to work without a remote preamp.
Very few mixers include that type of input - hence the need for phantom power adaptors for the mic inputs.
Most portable audio recorders with 3.5mm jack inputs will often include a 'PIP' facility - along with computer sound cards and smart phone mic inputs. ....... But not mixers!

The electronics involved are not complex, but the configuration you require is not that common..... which is why people like Dahua can charge the higher price they do.

Someone may be able to recommend a 6 channel mixer with PIP mic inputs and a USB output?.... I've never encountered one myself...

For a 'one off' installation you may be better off looking at a mixer with 6 'phantom power' capable mic inputs , and 6 XLR/PIP adaptors.

Whether that would be cheaper than a 6 input line mixer for the Dahua devices, I'm not sure?
( And don't forget you still need to provide DC power to those devices as well...... )

Thanks a lot rogs for the detailed explanation.

I’ll try to resume it to know if I understand well:

If I use for example a Scarlett 18i20 that has 8 XLR inputs with phantom power, I will only need 6 electret capsules and 6 XLR/PIP adaptors from the link you provided, without the need of external DC power. I will not need preamps in this process because of the adaptor plug in power.

If I want to use a mixer without phantom power inputs, or plug in power inputs, like Line inputs of mixers, I will need to add a preamp to capsules signals and provide them with an external DC power. The choice of the preamp depends on the signal quality I want to have and the impedance of the capsule if I understand online documentation correctly.

If building one of these microphones requires a DC output and a preamp like those circuits, I will continue searching to build a coherent mic + preamp circuit with a line output.

 

[Turlipe]

Hello,
following the recommandations above, I'm trying to build a preamp for the CMC-9745-130t electret capsule.
I’m following the preamp guide in attachments to add a preamp to the microphone.
I have two questions about the choice of the parameters:

• in the guide the maximum output voltage goal is 1.228V but how can I choose this value ? it is used in step 7 but I don't understand why it would be a "Design goal",
• the supply goal Vcc is 5V which influences the gain at step 8. I see in the doc that the default gain is 44 dB and the mic working voltage is 12V. Should I set Vcc to 12V to mimic the mic working voltage or should I find an appropriate supply goal to match the 44 dB gain ? (with 5V the gain is -4.191 dB, is this preamp gain linked to the DH-HAP120-V “default gain” ?)

Thanks in advance.

 

[rogs]

The preamp guide you linked to is typical of the kind of presentation from a major company - like Texas. The figures are all academically accurate of course, but bear little resemblance to 'real world' applications, especially when they indicate gain figures with accuracies down to 2 or 3 decimal places.
Looking at the data sheet for the mic capsule you've posted will show - for example - that the sensitivity between capsules can vary as much as 6dB.
Not really an appropriate variable to be introduced into calculations with 2 or 3 decimal places in the answer!
Your actual gain results are likely to be much less specific than those shown.
The circuit you have listed is designed to work from a battery supply. You are presumably intending to use remote DC supplies? ...
If that is the case I think you may need to look at adding extra DC decoupling to each preamp locally... and that may include decoupling the half rail reference as well.
The choice of an actual value for Vcc will depend on the kind of headroom you are hoping for, and of course the specification limits of the op-amp.
Don't forget to adjust the value of the capsule feed resistor (R1 in the Texas circuit) to suit your chosen Vcc.
I would also look at using simpler resistor values, rather than the somewhat exotic values shown for R1, 2, 3 and 4 in that Texas schematic.
Values chosen from the E12 range should be sufficiently accurate for this task.... (220k instead of 200k, 820R insted of 787R, and 5k6 instead of 5k9, for example)
I would also personally include a small value resistor (say 100R) in series with the op amp output. I don't have any experience of the specific device being shown here, but many op-amps can oscillate if the output is connected directly to what may be a somewhat capacitive load, presented by unbalanced cabling, for example.

 

[Turlipe]

Thanks again rogs. I tried to calculate the output parameters for my circuit needs (here is a Google sheet for this preamp if someone want to reuse this preamp).

I have chosen a supply of 12V to have the same headroom as the DH-HAP120-V (? I didn't find clear documentation on how to calculate headroom). I increased the "output voltage (in Vrms) goal" to decrease R4 and stay in the same range as the guide R4 (I'm not sur if this is necessary).

For the moment I will not use battery, does the circuit need to change if I attend to use an AC adapter connected to the domestic power ?

I still have a question about gain.

I understand that the preamp final gain will have a high variability (because of the capsule sensitivity and the preamp accuracy).
But how can I match the DH-HAP120-V default gain which is equal to 44 dB? The preamp guide gain is equal to -4.191 dB.
For the gain calculation of the preamp, Imax (step 3) and Rin (step 6) values are used but will not change since they are linked to unchangeable parameters (capsule sensitivity, capsule operating voltage, capsule current consumption and preamp supply).
I came up with a gain formula where resistor R4 or R5 should be negative, and where the feedback resistance R5 have to be lower than R4, which seems not standard when I looked at what a feedback resistance should do.

Either this preamp is not appropriate for my use or matching DH-HAP120-V isn't what I should aim for, or I didn't understand how to set the gain correctly.

 

[rogs]

I have to confess, my eyes tend to glaze over when I see the details of the figures as presented in your 'Google Sheet'.
Personally, I think that number of decimal places has little relevance to the figures required for the design of a simple, single stage op-amp electret mic preamp.... other folk may well disagree with that view of course!

To get a gain of around 44dB from a single non-inverting op-amp circuit - as shown on the Texas app sheet circuit you linked to - you're going to need to change the value of R5 from 10k to 120k. ( You'll also need to reduce the value of C3 to around 10pf )
How well that particular op-amp responds to that kind of single stage gain, I'm not sure? ... It's a modern op-amp, so it probably should be OK.
(My own instinct would be to use 2 stages - but I'm a bit old fashioned!

To de-couple your AC adaptor DC supply, I would fit a small electrolytic capacitor across the DC input at each remote preamp location.
Something in the range between 10uF and 47uF @16v should suffice.

'Headroom' is defined as the difference between the level of the actual output signal being generated, and the maximum permitted by the circuitry.
With many modern preamps that is 'rail to rail' -- so 12v p-p in this case.
You'll need to confirm that parameter from the data sheet of the op-amp you'll be using.

 

[Turlipe]

Thanks rogs, here are the final metrics of the circuit and in attachments a first iteration of this circuit with your recommandations added.

Ok for the gain, I need to set R5 to get the wanted gain, but in the guide R4 depends on the R5 value. So is it acceptable to set R4 regardless of R5 ? If it is ok I picked 390k to match the gain formula in the guide.

For the opamp I have chosen the TLV9161 because it is rail to rail and it can take a supply voltage up to 16V.

If everything looks ok I will order the components and evaluate it.

 

[rogs]

Some slightly odd features of your circuit that may need to be reviewed....

• An 'AC' supply input, coupled with C5?? ------ No, that needs to be a DC input. To use C5 to de-couple that supply, you'll need to connect it between the DC input and ground.

• You might like to check your gain calculations again? With R5 as 390K and R4 as 820R I calculate the gain as 476. That's almost 54dB .. pretty high.
Not sure how stable that will be from just a single stage like this?.... You'll probably need to actually prototype the circuit physically, to see how it performs, before you decide to build anything in quantity.
I would probably try out R5 as 150K and R4 as 1K..... That should give you a gain of just under 44dB.

• I would personally drop the value of R2 and R3 to around 47K. In this configuration, you're not able to de-couple the half rail, and the input impedance is determined by the value of R2 and R3 in parallel. With 390K that will give an input impedance of almost 200K, which may make it susceptible to hum pickup, depending on the actual system layout....... Lower values for R2 and R3 might help with that problem.

• The value of R1 may have to be determined experimentally. Once you have decided on the DC voltage to be applied to the circuit, you will have to know the current the capsule is going to draw (that will vary slightly from capsule to capsule ) to ensure that R1 drops enough voltage so as not to damage the capsule.

• Connect the junction of R5 and C3 directly to the output of the op-amp - so, to the other end of R7.

• I think the selection of 1.3pF for C3 is an unnecessarily complex value -- and almost certainly too small. Use a more standard value.
The actual value required will depend on the value of R5, and where you are expecting the HF 'roll off' to occur? ... For example, fitting R5 as 150K and C3 as 47pF should give a -6dB roll off at around 20KHz. ....You can select the values of C3 and R5 to suit your specific gain and HF roll off requirements.
(In reality, anything above c.10KHz is unlikely to make any real difference to the kind of frequency response you require for this task).

 

[Turlipe]

Thanks for the review. I tried to follow the guide calculated values but I will use yours for the first iteration.
I modified the circuit in attachments, everything should be ok now ?
And yes it’s only a prototype, I will try it out before making multiple ones.

 

[rogs]

Thanks for the review. I tried to follow the guide calculated values but I will use yours for the first iteration.
I modified the circuit in attachments, everything should be ok now ?
And yes it’s only a prototype, I will try it out before making multiple ones.

The DC and the de-coupling capactior C5 are still not drawn correctly.
Your termination marked 'DC Input' should be connected directly to Vcc. C5 +ve connection is also connected to this point ( that is: VCC)
C5 -ve connection should be connected to the negative (gnd) connection. (Pin 2 of the op-amp etc....)

 

[Turlipe]

Ok I think it should be better now.