A while ago I saw this post by DIY perks.
video here
and resources here https://diyperks.com/project_31_high-quality-usb-c-microphone/
I was fairly intrigued so I bought the mic capsules. well they've been sitting on my shelf for a while and I was going to build the rest of it, but looking at his amp circuit i'm not really happy with what I see. it's ac coupled in 3 places. I also see that it doesn't have any negative feedback.
Then I did a bit of digging and found the "alice" electret amp circuit
https://www.instructables.com/OPA-Based-Alice-Microphones-a-Cardioid-and-a-Figur/
Now from a bit of digging I've heard this one is supposed to sound a bit better, but I see it's still AC coupled on the outputs, and it's for XLR while I want to use my sound blaster's 3.5mm microphone input. for power I'll probably use usb, and if I need to boost levels / get negative voltages I can use the NMA0515Sc from the first schematic. I would probably add some LC filters after the power regulator though.
Now onto my questions. Does anyone have a better circuit, maybe something dc coupled that would work for this situation. I'm also really unclear about what voltage levels the microphone input on my soundcard uses and I'm afraid to blow it out.
Also would bootstrapping the opamp be applicable at all?
Thanks
video here
I was fairly intrigued so I bought the mic capsules. well they've been sitting on my shelf for a while and I was going to build the rest of it, but looking at his amp circuit i'm not really happy with what I see. it's ac coupled in 3 places. I also see that it doesn't have any negative feedback.
Then I did a bit of digging and found the "alice" electret amp circuit
https://www.instructables.com/OPA-Based-Alice-Microphones-a-Cardioid-and-a-Figur/
Now from a bit of digging I've heard this one is supposed to sound a bit better, but I see it's still AC coupled on the outputs, and it's for XLR while I want to use my sound blaster's 3.5mm microphone input. for power I'll probably use usb, and if I need to boost levels / get negative voltages I can use the NMA0515Sc from the first schematic. I would probably add some LC filters after the power regulator though.
Now onto my questions. Does anyone have a better circuit, maybe something dc coupled that would work for this situation. I'm also really unclear about what voltage levels the microphone input on my soundcard uses and I'm afraid to blow it out.
Also would bootstrapping the opamp be applicable at all?
Thanks
We don't hear DC. Cap coupling may be most important at the microphone end of the chain, where signals may be smaller than DC offsets and distortion is unlikely. The split-load JFET is heavy NFB and the THAT1512 is a heavy NFB chip (look it up).
Anyway I think the best feature of the one in the video is the way-cool brass batwing. Brass makes any $5 capsule sound better. As I learned on my early mikes. As you can still see on reruns of WoodSongs TV.
@TNT once tried to make a digital field recorder with microphone that had no AC coupling in the direct signal path, only in optional distortion-reducing bootstrapping circuits. When you have a very large dynamic range ADC, you can digitize the signal plus offset and remove the offset digitally, either with a digital high-pass or an autocalibration algorithm.
I think TNT gave up hope after blowing up the expensive ADC evaluation module, at least I haven't heard anything about it in months.
https://www.diyaudio.com/community/threads/fixed-gain-field-recorder.373352/
I think TNT gave up hope after blowing up the expensive ADC evaluation module, at least I haven't heard anything about it in months.
https://www.diyaudio.com/community/threads/fixed-gain-field-recorder.373352/
There is an interesting Linear Audio article by Scott Wurcer with lots of microphone circuits. He does use AC coupling, but with small capacitors with a good dielectric.
https://linearaudio.net/article-detail/2095
https://linearaudio.net/article-detail/2117
https://linearaudio.net/article-detail/2095
https://linearaudio.net/article-detail/2117
Absolutely not giving up - just regrouping for new attack ;-D
It will be the perfect autumn project - my yearly speaker project came in between
Marcel, I hope for your continuing invaluable suport.
//
Here it is btw... https://www.diyaudio.com/community/threads/fixed-gain-field-recorder.373352/
But less FET, is that possible?
//
But less FET, is that possible?
//
EDIT: Never mind, that capsule is FETless. I like to use capsules that have the FET built in, there are so much easier to use. The "Alice" circuit is well known, for sure.
yeah fetless is in reference to there being no fet on the electret mic itself. I just want to make sure I'm getting as good of sound quality as I can. I'll be sure to check out all the content posted and hopefully make a decision on which way to go with the circuit. I really appreciate all the input.
just a little update on what I decided to go with. mostly the first circuit. I've added a 5G ohm resistor to ground per the EEVBlog video I linked. I've also added provisions for negative feedback. if any of those makes it worse I can just desolder them so no harm done.
I've decided to replace the 100ohm resistors on the power rails with an LC filter ala a ferrite bead and some 22uf caps. Now I don't have the power rail completely figured out because I'm not sure if I want it on the same board or if I want to keep the power down stream and away from the capsule. also that's why I don't have any real bulk caps on here yet. when I figure that out I'll update it.
The datasheet for the that1512 had those 470pf caps. again if they cause any issue I always have the option of removing them.
Why? What? Have you read and digested the THAT1512 datasheet/app-notes? It HAS negative feedback built in. Great heaps of it.
http://www.thatcorp.com/datashts/THAT_1510-1512_Datasheet.pdf
I suspect that your one NFB resistor will not only unbalance the action but howl supersonically.
Your circular board around the microphone will significantly alter frequency response and polar pattern.
Good luck.
good info. I'll probably leave the negative feedback pads in there, but just not populate them. I like being able to play around with values.
I'm not so worried about the effect of the circular pcb. I'm thinking of making the pcb a little larger / farther from the mic and then using bungie to shock mount it to the pcb. so there will be some space between the capsule and pcb.
I'm not so worried about the effect of the circular pcb. I'm thinking of making the pcb a little larger / farther from the mic and then using bungie to shock mount it to the pcb. so there will be some space between the capsule and pcb.
I found some more good design info on the that1510/2 http://www.thatcorp.com/datashts/dn138.pdf
so I've fixed more issues with the schematic, and I was about to order and then I noticed something. the THAT1510/12 datasheet references +48v and ground when I had been building off of -24v and +24v. so I started changing everything to make -24v the gnd in the datasheet and adding another voltage supply for the +15v and -15v, but then I realized that if they call the -24v a ground and the -15v is referenced off of that, then that means the circuit expects a -15v, 0v, +15v +48v with the 0v being 24 volts lower than the gnd coming in on the USB, which would be a maximum difference of 63v between the highest and lowest voltages?
that seems so strange. the voltage regulators I have are isolated so I have a real ground and 5v coming in from USB. then the first isolating regulator makes -24v, +24v, and a midpoint off of that.
I would rename those ref_GND, 48v, and mid_GND.
Then I would have to drop the 48v down to 5v with another regulator, then I would use another isolating regulator to make a -15v, +15v, and second midpoint ground mid2_gnd
that seems so strange. the voltage regulators I have are isolated so I have a real ground and 5v coming in from USB. then the first isolating regulator makes -24v, +24v, and a midpoint off of that.
I would rename those ref_GND, 48v, and mid_GND.
Then I would have to drop the 48v down to 5v with another regulator, then I would use another isolating regulator to make a -15v, +15v, and second midpoint ground mid2_gnd