Need help with a TLM 102 sized PCB for DIY LDC mic
- Thread starter edenooo
- Start date
Help Support GroupDIY Audio Forum:
- Status
rogs
Well-known member
They really should make their minds up........ especially on the same data sheet!
View attachment 147527
So... based on this information, should I add a zener like Itsme recommended, or should I skip it..?
My recommendation: if you have some space left put it in as optional. You could possibly use a 6.8V Zener which can have very low noise while still having enough headroom for a decent level even with some gain. If that gives you results you are not happy with... just don't populate it.
micolas
Well-known member
I apologize.
I don't want to hijack the discussion.
But I have an observation. The Ali 102 headbasket is very small. My opinion is that it fits an 89 style microphone, a k67/87 capsule and a circuit with deemphasis. In no case is it compatible with a k47 style capsule (with linear circuit).
My question: Does the opic42 circuit allow for deemphasis?
I don't want to hijack the discussion.
But I have an observation. The Ali 102 headbasket is very small. My opinion is that it fits an 89 style microphone, a k67/87 capsule and a circuit with deemphasis. In no case is it compatible with a k47 style capsule (with linear circuit).
My question: Does the opic42 circuit allow for deemphasis?
![Soldering Iron Kit, 120W LED Digital Advanced Solder Iron Soldering Gun kit, 110V Welding Tools, Smart Temperature Control [356℉-932℉], Extra 5pcs Tips, Auto Sleep, Temp Calibration, Orange](https://m.media-amazon.com/images/I/51sFKu9SdeL._SL500_.jpg)
rogs
Well-known member
It sounds as if Itsme prefers the Alice OPA version of (virtually) the same circuit -- THIS ONE -- which has a zener, also uses a fully differential audio output, and includes resistors in the signal path which can be configured to add gain in the microphone.
I'm a great believer in simple.... I think that the impedance converter should be as simple as possible, and that additional components in the signal path at this first stage will add noise, as well as complexity. Most modern mic pre-amps will add gain with less noise, if it is required (which it almost certainly will be!).
Sadly, this being the internet, you have no way of knowing what information is important and which is not!
I suppose the sensible option would be to add provision for the zener -- and then not fit it if you discover you don't need it after all!
I've already made a working version of this microphone with the capsule inside the headbasket, is your concern just with the end result frequency response of the microphone..? If so, I'm really not too concerned with that as of now, my original plan for this microphone was to have the cleanest most transparent small size microphone I could make (for a reasonable price), I'm completely fine with processing the microphone in post as it's something that I'm always going to be doing with this microphone anyway, if anything having a "blank slate" is more preferable for me as a person.
Or maybe I'm completely misunderstanding what you mean by "deemphasis"
I think I may go with this option, I'll have multiple of these boards so I can compare 1-1 if need be, thanks!
micolas
Well-known member
1. Ok, it's absolutely fine if you don't care about the frequency response of the microphone and solve the problems in post-processing.
2. I was curious if the opic42 circuit can de-emphasize the high frequencies, to reduce the specific brightness of the k67/87 capsule. I'm really interested in this circuit, I've never built it.
rogs
Well-known member
No - the OPIC 42 - like the other OPIC projects is simply an impedance converter.
It uses the low noise, very low distortion OPA1641 op-amp in a non-inverting buffer configuration, with a (very) high input impedance and a low output impedance. No gain resistors in the signal path, only the 1G capsule bias resistor and the 47R output resistor.
I'm not a fan of applying any form of gain or frequency response correction within the mic itself. I prefer to add those things further down the signal path - at line level - where noise from additional passives in the signal path is much less of a problem.
Or, of course, the most versatle option ... do things in a DAW.
Not everyone will agree with that view of course!
I admit, I have a tendency to over-engineer...
Adding gain in would require three resistors, unfortunately...
I didn't have a symmetrical output in mind though.
The reasoning for having gain in the first stage is that the first stage dominates with the noise contribution. The following stages need less gain thus contribute less noise. Noise adds as sum of squares(root sum square).
The general rule is to apply gain as early as possible.
..sorry, physics...
Adding gain in would require three resistors, unfortunately...
I didn't have a symmetrical output in mind though.
The reasoning for having gain in the first stage is that the first stage dominates with the noise contribution. The following stages need less gain thus contribute less noise. Noise adds as sum of squares(root sum square).
The general rule is to apply gain as early as possible.
..sorry, physics...
micolas
Well-known member
Thank you for the extremely relevant answer.No - the OPIC 42 - like the other OPIC projects is simply an impedance converter.
It uses the low noise, very low distortion OPA1641 op-amp in a non-inverting buffer configuration, with a (very) high input impedance and a low output impedance. No gain resistors in the signal path, only the 1G capsule bias resistor and the 47R output resistor.
I'm not a fan of applying any form of gain or frequency response correction within the mic itself. I prefer to add those things further down the signal path - at line level - where noise from additional passives in the signal path is much less of a problem.
Or, of course, the most versatle option ... do things in a DAW.
Not everyone will agree with that view of course!
MidnightArrakis
Well-known member
[apart from one major error ..... remove R6. (Join C7 negative directly to R5)] -- While what I have shown in my schematic version above may be a -- major error -- in this circuit design, I was only following how @[SIZE=5]edenooo[/SIZE] had drawn out his EasyEDA schematic (see below). Maybe edenooo -- WANTS -- that 1M resistor in his circuit for some reason.....I don't know. All that I was doing was trying to show the OP -- HOW -- his schematic "could" look with it being drawn a bit less "convoluted" and having the component REF DES indicators labeled in a sequential -- LEFT-RIGHT / TOP-DOWN -- manner, all while using the KiCAD program that I had suggested to him previously. That's all.....
I have also provided @edenooo access to an online folder that I have containing 8GB of KiCAD training videos to help him along with learning how to use this new (to him) CAD-design program. But, so far.....I have no idea if he has downloaded my training video files or what, if any, updates he has made to his original EasyEDA schematic. I was simply trying to show the dude how his schematic could be drawn as being "the same thing, only different". That's all.....
/
Last edited:
That was just a misunderstanding of a hint i dropped (while i thought i had clearly enough mentioned the nodes those needed to go between, and those weren't it)...
I highly recomend.
(You have an idea how expert opinion it is...)
I built this version, for electrets:
https://groupdiy.com/threads/simple-quiet-mic-with-an-electret-capsule.88953/
And I like it a lot!
Low self noise is my Graal as I do field recording and I'm drawn to subtle sounds.
It is quiet.
And beautifully simple.
BTW - I really admire @edenooo's effort to design the PCB by himself.
Not long ago I tried this myself.
And let's leave it at that....
@edenooo - I looked at your PCB's drawings and I think there's something missing. Mounting holes.
I'm not familiar with the body you'll use, though. There may be some side clamps for the board, I don't know.
If not, how will you mount the board horizontally?
No sarcasm here, just friendly warning
Only recently I had to solve this problem for the round board in a modified BM-800 body.
I may have helpful ideas if you need.
I'm not familiar with the body you'll use, though. There may be some side clamps for the board, I don't know.
If not, how will you mount the board horizontally?
No sarcasm here, just friendly warning
Only recently I had to solve this problem for the round board in a modified BM-800 body.
I may have helpful ideas if you need.
Yes I'm aware! I'm have space to add those, I just didnt want to have to work around them while routing@edenooo - I looked at your PCB's drawings and I think there's something missing. Mounting holes.
I'm not familiar with the body you'll use, though. There may be some side clamps for the board, I don't know.
If not, how will you mount the board horizontally?
No sarcasm here, just friendly warning
Only recently I had to solve this problem for the round board in a modified BM-800 body.
I may have helpful ideas if you need.
Bold of you to assume someone named eden was a guy
rogs
Well-known member
The attached file is a recording of a tone applied directly to a single sided non inverting buffer OPA circuit. no capsule involved.
That is followed by the same tone fed into a differential OPA circuit, configured as per the Alice OPA project (i.e. using a second op-amp, configured as a unity gain inverter, fitted with 2k2 resistors ).
Obviously the differential audio version has 6dB more gain. .. So the single sided version has had 6dB of gain added digitally- to both the tone and the following silence - to match the signal levels.
Although both are pretty quiet, the noise figure on the differential version (the second half of the recording) is about 3 dB worse.
That is what you would expect from the additiion of a second op-amp, together with its associated resistors.
(The details were discussed in this post on the micbuilders forum: https://groups.io/g/MicBuilders/message/33002 ).
Sorry, I prefer to add the gain either digitally, or let the good mic pre-amp designers create following low noise gain stages, rather than trying to do it in the mic, where any gain stage is likely to be more basic (and probably noisier!).
View attachment opa42.noise.single.then.differential.wav
- Status
