Need help with a TLM 102 sized PCB for DIY LDC mic

[Екатерина]

You must have a really nicely treated room, if you can afford to be that far away (especially when singing / speaking softly) without "room sound" becoming an issue.



Is that important?

This is useful for increasing the level. If the output impedance is relatively low, a transformer can be used to amplify the level signal

 

[Екатерина]

Singing is okay, but I usually prefer to keep it away from me

 

[rogs]

... For what? A soft voice a foot away? A snare drum from a couple inches away?

I should have qualified that of course! ...
Using an Arienne Audio flat 47 capsule - with 65v polarisation (as I described in post #14) that is the sort of gain I needed to apply for spoken word at around 6".

For a 'snare drum at a couple of inches' -- I'm not sure the transformer output version would be appropriate for that task?
As kingkorg demonstrated in his tests a while back, those transients can be in the 10s of volts -- (hence the logic of trying to maximise headroom)..... Probably overdrive the transformer, I would think?
Now whether that overdrive would sound 'nice' is another matter..... (The use of the word 'nice' there shows I'm starting to learn some of the non-scientific 'fluffy' microphone description words ! )

 

[rogs]

At present, the simplest way to increase gain is to add two resistors and turn the follower into a in-phase amplifier. The level signal will increase by 6dB, and of course, the noise will also double with the amplification factor. Have you measured the output impedance of this circuit? I made a diagram and I feel sorry for altering your pictureView attachment 147665

Dont forget to AC couple the '1K' resistor to ground. The op-amp output is reference to half rail, not ground.

I have no idea whether this would produce good results. As I've mentioned before, I'm not a fan of applying gain inside the microphone.
Applying gain by using a low noise pre-amp will almost certainly get you better results. Even cheap audio inerfaces - like the Behringer UMC series - use quiet preamps these days.
Fitting extra gain resistors into the mic will almost certainly add noise -- and the higher value the resistor, the more the noise.
But this is a DIY mic forum, so experimenting is good... Give it a try and see how it goes!...

 

[Екатерина]

Dont forget to AC couple the '1K' resistor to ground. The op-amp output is reference to half rail, not ground.

I have no idea whether this would produce good results. As I've mentioned before, I'm not a fan of applying gain inside the microphone.
Applying gain by using a low noise pre-amp will almost certainly get you better results. Even cheap audio inerfaces - like the Behringer UMC series - use quiet preamps these days.
Fitting extra gain resistors into the mic will almost certainly add noise -- and the higher value the resistor, the more the noise.
But this is a DIY mic forum, so experimenting is good... Give it a try and see how it goes!...

Oh yes, 1K to ground should be connected in series with a capacitor, maybe 47 μ F is a good choice

 

[Itsme]

At present, the simplest way to increase gain is to add two resistors and turn the follower into a in-phase amplifier. The level signal will increase by 6dB, and of co

Only two resistors won't work in this case: The non-inverting input is biased to Vsupply/2. Hooking the feedback divider to V- would saturate the output at V+ . Splitting one of the resistor into two would make it work. Note that R1 and R2 are virtually in parallel meaning R3 needs to be R1IIR2 for 6dB of gain.

 

[Khron]

Only two resistors won't work in this case: The non-inverting input is biased to Vsupply/2.

Hence rogs' tip of

Dont forget to AC couple the '1K' resistor to ground. The op-amp output is reference to half rail, not ground.

Which was received and understood (as far as i can tell):

Oh yes, 1K to ground should be connected in series with a capacitor, maybe 47 μ F is a good choice

Splitting one of the resistor into two would make it work. Note that R1 and R2 are virtually in parallel meaning R3 needs to be R1IIR2 for 6dB of gain.

I've yet to see THAT particular arrangement for feedback anywhere else. Probably because a mid-supply is already available for biasing the non-inverting input, rendering this convoluted thing sort of redundant (imho)... And it's nothing that can't be remedied by adding one capacitor, anyway.

 

[rogs]

Oh yes, 1K to ground should be connected in series with a capacitor, maybe 47 μ F is a good choice

Or maybe select a smaller value to adjust the LF frequency response a bit? ...... The thing to do is experiment!
(One reason why I do my experiments with stripboard, rather than PCBs -- it's cheaper and easier to try out different things quickly. )

 

[Itsme]

A capacitor would work, of course. Some downsides though:
A 47uF cap is larger and more expensive than a resistor.
A cap influences the frequency curve.
A cap would possibly cause a large spike when the mic gets plugged in. The resistive feedback would settle immediately.

 

[rogs]

A capacitor would work, of course. Some downsides though:
A 47uF cap is larger and more expensive than a resistor.
A cap influences the frequency curve.
A cap would possibly cause a large spike when the mic gets plugged in. The resistive feedback would settle immediately.

I remain intrigued as to why folk seem to want to add gain in the mic with this project? ..... It will create noise. Just use a pre-amp to add the gain.

If you want an impedance converter with gain, there are many hobby project circuits out there, using discrete FETs and bipolar transistors, which offer many more 'no compromise' gain and frequency resposnse control adjustments.

The OPIC project really is just a simple impedance converter...... It doesn't lend it self to 'tidy' gain additions.

 

[Екатерина]

I remain intrigued as to why folk seem to want to add gain in the mic with this project? ..... It will create noise. Just use a pre-amp to add the gain.

If you want an impedance converter with gain, there are many hobby project circuits out there, using discrete FETs and bipolar transistors, which offer many more 'no compromise' gain and frequency resposnse control adjustments.

The OPIC project really is just a simple impedance converter...... It doesn't lend it self to 'tidy' gain additions.

OPIC is a project with great potential. With the development of electronic components, there will be better operational amplifiers that can reduce the use of many components. Currently, it is because it is easier to manufacture and increase the gain to use older sound cards

 

[Екатерина]

It has to be said that laziness can make people more curious, making it easier to produce fewer parts. If it is more widely applicable, more people will like it

 

[rogs]

OPIC is a project with great potential.....

Very kind of you to say so. It is really just a simplified version of Jules Ryckebusch's 'OPA Alice' concept.

I'm sure there will be increasing numbers of mic projects using op-amp front ends as the devices improve.
(So much easier than individually biasing discrete FETS!)

I think taking the op-amp 'front end' concept, and devising new circuits - which include gain and frequency response options at the outset - might be better than trying to tinker with the simple (and not that easily adapted ) OPIC project?

 

[Екатерина]

But at present, there is no such design, and when you are the first to use this method, sometimes the first one used is often the most important

 

[Itsme]

The OPIC project really is just a simple impedance converter

Please don't get me wrong. I really appreciate your approach of a simple but well performing circuit!
I could think of a few additions however, that don't really break your concept but offers optional improvements, if one doesn't mind three additional resistors...

 

[rogs]

..... I could think of a few additions however, that don't really break your concept but offers optional improvements, if one doesn't mind three additional resistors...

I look forward to seeing - and perhaps hearing? - your results.
As I have mentioned before, the OPIC project only really came about becasue JLI's shipping costs ouside the USA for the ''Alice OPA' project were very expensive. I wanted to know how the OPA164* op-amps performed, so I designed the OPIC project to find out!

Looks as if there may now be further 'spin off designs', using new circuit variations...

 

[edenooo]

I look forward to seeing - and perhaps hearing? - your results.
As I have mentioned before, the OPIC project only really came about becasue JLI's shipping costs ouside the USA for the ''Alice OPA' project were very expensive. I wanted to know how the OPA164* op-amps performed, so I designed the OPIC project to find out!

Looks as if there may now be further 'spin off designs', using new circuit variations...

I really do think this project is incredibly versatile, it's just SO wildly simple that it makes adding anything extra just a matter of preference, anyone can (relatively) easily make their own version that's catered to their tastes while still keeping a very small number of components, which allow it to fit in basically any shape or size that you so desire.

 

[micolas]

I look forward to seeing - and perhaps hearing? - your results.
As I have mentioned before, the OPIC project only really came about becasue JLI's shipping costs ouside the USA for the ''Alice OPA' project were very expensive. I wanted to know how the OPA164* op-amps performed, so I designed the OPIC project to find out!

Looks as if there may now be further 'spin off designs', using new circuit variations...

I haven't built any microphones with OpAmp so I have no experience.
That's why I recently asked you about the linearity of the circuit.
But I was thinking about an NFB adapted to OPIC that would do deemphasis for the bright k87 capsules.

 

[Itsme]

Opamps can be very linear thus creating very little distortion (harmonics).
With the addition of a feedback network it is fairly easy to add arbitrary emphasis or de-emphasis on both ends of the frequency range.
If there is interest, I could post some simulations to demonstrate some variations of the circuit and explain how that works.
It would probably a good idea to start a dedicated thread for that.
Please let me know if you are interested...

Andreas

 

[micolas]

Opamps can be very linear thus creating very little distortion (harmonics).
With the addition of a feedback network it is fairly easy to add arbitrary emphasis or de-emphasis on both ends of the frequency range.
If there is interest, I could post some simulations to demonstrate some variations of the circuit and explain how that works.
It would probably a good idea to start a dedicated thread for that.
Please let me know if you are interested...

Andreas

Andreas, I'm sure many agree that a feedback network that simulates, for example, U87-style NFB would be of great interest.

Edit:
In order not to deviate from the TLM102 thread, we will address this topic here:

https://groupdiy.com/threads/microphones-with-opamp-and-de-emphasis-feedback-network.90278/