[Sticky] USB-C Microphone (official topic)

Yeah, the Alpha needs to be powered from a stable 5V - the easiest way to do that is to get a little breakout - but you can just snip the end from an old USB lead like a phone charger and use that. There are four cables usually, black (0V or GND); red which is your 5V and the other two are data (white and green usually but there's no actual colour code that's enforced). Alternatively you could just power it from a 5V wall wart. It can handle a little bit of ripple and noise - Matt designed that into his original.

Sorry, yes a "meter" is just shorthand for a multimeter (sometimes called an AVO from the old Brit manufacturers which in turn came from Amps, Volts and Ohms.) 

An oscilloscope is very useful to have on hand but not essential.

I think Matt is waiting on me to demo the final capsule adaptor and the P24 power (which is a recent addition thanks in part to @bromalite). I often get my best ideas from interacting with people on the forum. Just because someone doesn't understand how stuff works, doesn't stop them from suggesting or discussing their ideas. I wish I had Matt's imagination (he's probably blushing) but I'm a big fan.

I can run some of the book by you and Mus - as you're about at the level it's aimed at. I have an unusual way of looking at stuff - forensic detail but from angle where practical results matter. You can get a perfectly functional schematic but set the board up incorrectly and it won't work well (or at all in some cases). So the book teaches the stuff that matters - to give you people the intellectual tools they need to get cracking. 

It's got a crash course in the best of KiCAD (not everything of course) and practical solutions to common problems - mostly by example of the development of Michelle and the associated designs.

this sounds great, i look forward to seeing more about this book!

It's about 1/3 done so far and called (speculatively) "Making Michelle: the fibs your school physics teacher told you." 

I have a dummy cover but it's not very good since I only did it for early (sponsor) promos. https://www.codeweavers.com/ has already donated a software licence so it deserves a mention. Most people here will be using Windows but for those of you on Linux or other Un*x-like systems, it's a very good way to run a lot of Windows software without having to fool around with dual-booting or run Windows in a container. While some of that software runs under Wine, Crossover is even better and it's gotten me out of a pinch running stuff like Saturn PCB Toolkit.

Why "fibs"? Well we're taught that voltage is a thing when it's actually the ability to do a thing. While current and resistance are very real, voltage indicates how much pressure is available to do that - so it's a virtual variable that we get by implication.

Sure, we can measure it with a multimeter but meters only measure current through a known resistance. In much the same way as we measure current through a known resistance and resistance by measuring the current through a resistor. Voltage measured this way is showing the work being done. The difference is very subtle, to the point that we usually just ignore it.

But voltage is more than that - it's better to think of voltage as the presence of the two fields, something we're taught as part of electro-magnetism. The voltage is an indicator of how powerful those fields are - and it's the fields that cause current to flow. Better yet, we're taught that when we connect up a source, the current flows from the positive to the negative (conventional current flow) or from negative to positive - electron flow.

Only that's wrong too - at least, it's inaccurate.

When we connect up a power source the fields begin to appear and race along the conductors at roughly 2/3s the speed of light  - which means that the current begins flowing in just a few nano-seconds (billionths of a second). It's the fields that move the free electrons through the conductor but more bizarrely, the fields are not in the wire, rather they are moving in the substrate or even the air. All of the actual energy is in those two fields - the electric and the magnetic field.

Hertz was the first person to prove this in 1886 and that work is what gave us everything from television to GPS and radio telescopes. Even visible light is an electro-magnetic field (or, at least, the presence of one). This gets into particle-wave duality because the energy acts mostly like a wave until it hits something and then it behaves more like a particle. The more energetic waves like ultra-violet and x-rays are so powerful that they can cause burns and even cancers.

While I'm leaving the quantum stuff to people who shoot sub-atomic particles around 17 miles very high powered magnets, it's useful to understand how electricity works.

All transistors but most notably JFETs and MOSFETs work by changing the field in a capacitive region between the gate (control terminal) and channel (the bit between drain and source) and for that reason they only "leak" almost undetectable amounts of current. Knowing stuff like this prevents us from getting  tripped up by capacitive loading (more of that in the book) and indicates why some devices, power MOSFETS being a good example, require special driver ICs to switch them on and off a very high speeds - we're literally charging and discharging a tiny capacitor; and when they're "empty" capacitors look like a instantaneous short.

The difference between JFETs and MOSFETs is that the capacitor in a MOSFET is a real capacitor (albeit a minute one) and in the JFET it's caused by a region of differently charged particles causing a barrier at the junction. Many of these things are sidelined in teaching texts, until you get into the higher levels of a degree course and beyond, but they don't need to be that complicated as long as you know they're there, you are less likely to find yourself scratching your head.

You can do this experiment yourself by putting your meter into continuity (with the buzzer). Find a small capacitor - 1uF is good, and make sure it's discharged by shorting the two leads. Now when you measure continuity - remember that capacitor blocks DC - you'll hear your meter "blip" as there's a very short surge of current.

I'll show you how this happens and why in the book but current is a function of charge vs. time - and that matters in some cases. For example, a very large capacitor (1000s of uF) will exhibit huge inrush currents until the device starts to reach full charge (with the energy locked in place by the electric field). Transformers/inductors exhibit similar behaviour as the magnetic field is built up, with the energy store in the magnetic field.

The other gotcha (and one you won't be taught in school) is that every component, no matter what it is, has some amount of capacitance, resistance and inductance and, particularly in high-speed digital circuits, these tiny values can really ruin your day. Similarly, the fields don't hug the wires so much as they appear around them. This means that if you put two conductors too close to one another, fields from one (called the aggressor) will cause currents to flow in the surrounding wires (called victims). 

Once you understand fields, electronics makes a LOT more sense and by the end of this book, you will.

Hello, and thank you for the fun project.

I have had fun working on my version of the V1 microphone over the past month. I was excited to get to use it as I just finished it a few days ago. However, when I try to use it, I get nothing. Absolutely no sound, not even a buzz or hum from interference. I opened it up to see if I could try and troubleshoot with the goal of getting it to work and noticed I had swapped the 'mic +' and 'mic -' around on the Jfet transistor at some point and never noticed it. After I fixed this I still get nothing, complete silence. As far as I can tell, everything is connected where it is supposed to go, and the gain has ~10 ohms of resistance for testing.

The only things I can think of that are causing issues are either I fried something with the swapped wires going to the Jfet, or my use of a few components that differ from what Matt used in his video are to blame. The two things that I used that do not match with Matt's microphone is both of the capacitors. For the four 2200uf 16v capacitors I instead used 35v rated ones as it was something I already had available and I figured that the higher rating wouldn't hurt. I feel is if I may have been a bit silly/foolish when ordering the 22uf nonpolarized capacitors that I ended up going with and got hyperfixated on them being polyester film and not electrolytic as I read somewhere that they were 'better' for some reason. However I am now left wondering if I shot myself in the foot by using the wrong component here. I currently have TKD B32523R0226K000 Polyester Film Capacitors in my microphone and I would really appreciate if anyone with good knowledge on how these circuits work would be able to tell me if that is what's causing my issues.

I tested the audio capture card by playing audio from an mp3 player into its input and it worked fine, so I suspect that is not the issue. As that was my first thought when trying to figure out what's wrong, so I checked it first. The NMA0515SC also appears to be okay(?), I measure 35.2 volts from the v+ and v- pins, and 5v going into it, which I would mention here as the 35v seems abnormal. I know/could not find any way to test the THAT1512 or Jfet, any help with that would be great as well.

I feel as if I have exhausted my limited electrical knowledge trying to troubleshoot and I am hoping someone here has some good ideas that I can try to get me one step closer to a working microphone. I would like to avoid throwing parts at it, but I could order another THAT1512 if anyone here thinks that is the problem, and just see what happens. I ordered the 22uf capacitors, THAT1512, and NMA0515SC directly from mouser, so it should be safe to rule out counterfeit components from being the cause. I can send pictures of the microphone if there is anything you want to see, just ask.

Any help is hugely appreciated, I am getting desperate lol.

Thanks!

OK, well it sounds like you've done a pretty good job so far (despite things not working).

Let me put your mind at rest for a couple of things.

Strictly speaking Matt's use of the 16V capacitors on the high-side of the NMA0515 inverter was riding a bit close to the edge of the cliff (for my liking). It's one of those things we can sorta get away with because the loaded (operating) voltage from the inverter never really gets to even 15V when the circuit is operating and the current limiting resistors slow their roll (charge time) and reduce wear on the part. A 35V part is a more costly but safer design choice as the inverter isn't regulated and, when not loaded, produces well over 40 V typically. The combination of the 2N4416 FET and THAT1512 serve to draw sufficient current as to cause its output to buckle (well within tolerance) but keeping it below the "magic smoke" level of those parts.

Poly film caps are far superior to electrolytics in any application  (ANY classic capacitor is superior in fact). We don't tend to use them in places like this because the blighters are expensive and very bulky. Electrolytics are nasty things to be avoided in all but the most desperate of circumstances (LOL) like bulk local storage and for smoothing out the worst of the "wobble" in a supply. 

The 2N4416 (like every JFET I've ever used) are tough little fellas and despite what the doom-merchants from some parts of the web might say, they don't give a fig which way around you put them in circuit provided that you only mix up the Source and Drain terminals.

One way to look at a FET is like a hose with a kink in it. The hose represents the channel from Drain to Source and water can flow either way. Control of the flow comes from how much the middle is squished: with the gate being the thing that does the squishing.

To really stretch the analogy to breaking point, you can think of the hose passing through the hinged part on a wooden gate and someone moving the gate closed so the hose gets "pinched" ever tighter until the flow stops entirely. The hose is very strong because it's made of a very tough plastic, the gate on the other hand, that isn't. Pulling the gate closed as hard as you can and the water stops flowing but force it OPEN too hard and it will snap right off its hinges.

In other words practically the only way you can destroy a JFET (cf. MOSFET) is to reverse the gate terminal with respect to Source or Drain. Do that and it's bye-bye baby. You won't see it go up in a cloud of Magic Smoke, that ignominious end is most often reserved for small resistors and electrolytic capacitors.

Unlike MOSFETs which can be destroyed simply by touching the gate (static electricity enters the terminal and blows a hole in it!) at the electrical level, a JFET’s gate is a diode junction so it’s very robust indeed when handled.

But here’s the problem – if you happen to wire the Gate to either supply by mistake you will “forward bias” the diode and overload it. If there’s enough current available (and there is here) the diode will silently just blow and you won’t even notice.

Testing a JFET

I’ve gotten this wrong more than once myself because I’m forever forgetting which type of FET I’m talking about… (cough, cough) but ALL JFETs are depletion mode. I know this because teacher made me write it out 100 times on the blackboard recently…

Depletion mode? Vs. Enhancement mode…

Yeah. Right, because that makes everything clear: as clear as a house brick.

All it means is that when the JFET is sat on your desk out of circuit, the “channel” – that’s the bit between the Source and Drain terminals looks a little like a modestly sized resistor (as far as a multimeter is concerned).

This makes them relatively easy to test because the channel can only fail open circuit (completely blown) or closed circuit where the internal terminals are welded together. Open is far more likely though, I’ve never heard of someone shorting a JFET but I assume it could happen.

Testing the other connections is “simply” a matter of putting your meter into DIODE mode (usually found on the resistance range) and checking you still have a functional gate.

You should read a diode voltage drop (around 0,7 V) from Source OR Drain and Gate with the red (positive) lead to the Source/Drain and negative to the Gate; and 0 (or high impedance) in the other direction.

This test should work OK with the FET wired up but not connected to the inputs although it might be a tad tricky to get at if you’ve sealed up the capsule end as the Gate is connected to the centre terminal of the JLI capsule.

The other quick test you can do is to remove the capsule connection from the board and (with everything running and digitiser in place) test to see if anything happens if you probe the inputs with your finger. We’re meaty radio aerials and mains “hum” can easily pass through us into the circuit so it should roar and grumble like my ex. wife prior to her morning gallon of IV espresso.

Have a go at that and we’ll see where we go from here.

Also, welcome to DIY Perks community.

Thanks for the quick response marcdraco!

Posted by: @marcdraco

↑

Testing the other connections is “simply” a matter of putting your meter into DIODE mode (usually found on the resistance range) and checking you still have a functional gate.

You should read a diode voltage drop (around 0,7 V) from Source OR Drain and Gate with the red (positive) lead to the Source/Drain and negative to the Gate; and 0 (or high impedance) in the other direction.

This test should work OK with the FET wired up but not connected to the inputs although it might be a tad tricky to get at if you’ve sealed up the capsule end as the Gate is connected to the centre terminal of the JLI capsule.

I tried this test and I think it looks good, I get .7v measuring from gate to either source or drain. As you also predicted, it measures 0 from gate to drain the other way. I cannot get a measurement from the source pin doing it this way.

Posted by: @marcdraco

↑

The other quick test you can do is to remove the capsule connection from the board and (with everything running and digitiser in place) test to see if anything happens if you probe the inputs with your finger. We’re meaty radio aerials and mains “hum” can easily pass through us into the circuit so it should roar and grumble like my ex. wife prior to her morning gallon of IV espresso.

I gave this a try, and I get absolutely nothing for an output when touching the pins. Still complete silence.

I also looked over how everything is soldered up, and even though I moved a few things around on the board to make them fit everything seems to be going to the correct spot. I will also note that I moved the 22uf capacitors out of where they would normally go due to size issues, and used some wire to 'remotely' connect them to the circuit.

I am looking forward to what you think to test next.

Hi Guinea, happy to help.

I hit an issue like this caused by capacitors oddly enough but not for a reason I'd have expected. We'll get to that if it proves to be an issue for you.

OK, so to the task at hand. So remove the THAT1512 from it's socket first so we can get a picture of what's going on. We need to know the voltage with respect to GROUND (0V) on each pin of the THAT.

First check for continuity from the 0V on the power input to the ground reference terminal on the socket - Pin 5. If you've used the little USB-C breakout the shell is a good place to get the 0V. The resistance here should be less than an Ohm (a lot less actually but that's not easy to measure with basic kit). 

Pin 5 on the THAT sets the output reference to 0V in Matt's design but it can actually be used to move it to somewhere more convenient - that's what happens in the Alpha version of this design. 

If you've got 0V at that point, you're good to use that as a 0V (ground) reference to measure the other voltages. You can use anywhere that's convenient but this is a good place to start.

Now check the voltages on each pin and list them out. Pin 0 and 8 should be 0 because nothing is connected to them - they make a circuit inside the IC. Pin 7 is your positive supply and pin 4 is the negative, so you should see around 13 V on Pin 7 and around -13 V on Pin 4. 

If you're good there that means the THAT is powered correctly but leave it out for now.

Moving on you probably have some odd voltages on 2 and 3 (ignore that for now) but check to make sure you still have around 13 V and -13 V on the output to the capsule. It might be a little higher but the exact value isn't too critical.

Next up, test for voltage on Pin 6 - this is the output terminal and should have 0V on it.

OK if that all works as it should, next job is to remove power, replace the THAT in its socket and repeat the test again.

The voltages on Pins 2, 3 and 6 matter most here. Pin 6 will probably have a few 100 mV on it - this is normal. A small voltage should appear between 1 and 8 although it's going to be quite small if you're using a sub 100 ohm resistor in there.

Have a crack at those and let's see where we are.

Man I wish I'd finished Michelle at this point because it works out of the box but ... good things come to those who wait and I want it perfect and the last bit isn't quite done yet, Totally understand the problem with those film capacitors. Ironically, I know I'm gonna catch it from purists for using MLCC and carbon comp resistors in my design (they're not as good as film caps and metal film resistors - but these folks don't have to pay for the R&D and most of them earn a lot more than I do. 🙂

Hello, I am having trouble getting mine to work properly. I get some popping ground buzz sound when I touch the potentiometer or audio interface board, but nothing from the mic. Also, the voltage converter gets quite hot. Not sure how to go about checking if everything is hooked up correctly. I have not found an updated schematic of the electronics beyond the printed-out layout sheet. I am still fairly new to electronics, would love some help.

Hot? How hot... if you can't hold it between your thumb and forefinger for more than a few seconds it's WAY too hot. Modern electronics can run for years at temperatures well above the boiling point of water but just because they can... besides that's the temperature of the part and those parts are mounted in a solid epoxy resin so the temperature we feel at the surface is well below what the parts are experiencing.

If you look back a couple of posts there are some diagnostics that Guinea is doing, that's the best place to start.

(If I ever finish the book I'm current writing, I think you'll find it invaluable, since it attacks electronics from a hobbyist perspective and I've kept the maths as simple as possible. Even to the point of showing you how to enter the values into the amazing Qalculate! which is a scientist/engineers dream and it's free too.)

My apologies for the hastily written response. I got logged out or something(?) and my nice one got deleted. (I do not feel like typing the whole thing again)

I did test the socket until I found some issues.

From pin 5 to ground I measure .3 ohms.

I checked if I measure anything from pin 5 to pin 8 and I measure ~2000 ohms and .35 volts when powered on. I will be checking tomorrow that the traces are cut correctly and what on earth is bridging the gap there. I am only able to measure .35 volts anywhere on the socket so I think I have a short somewhere bringing my voltage down, which I will be looking for and reporting back tomorrow with whatever I find.

Pin 7 & 4 are also the .35v.

Hopefully the THAT didn't blow up with .35v on pin 8, we will see I guess.

That's Veroboard unfortunately. I can vividly recall many years back spending hours trying to figure out why a little AM receiver I was making didn't fire up.

The cause was a tiny sliver of copper (narrower than a flea's eyelash) shorting the "cut" track. I hate the stuff and I really do admire anyone who can work reliably with it.

@marcdraco I fixed some messy solder joins, which seems to fixed the no audio signal from the mic and the hot dc to dc converter. But now I only get the static and some popping if I tap the wires coming from the microphone/transistor. So you think this is the microphone side, or could it be on the board?

Can you give us some sample audio please?

Also if you get a chance, run those voltages for me if you don't mind. 

ssm2019bnz (This is my amp IC) - pins readings (ground pin (5), each of the other pins): all 0.0003-0.0001 V. powered:1=1v, 2=11.6v, 3=12.4, 4=0.2v, 6=0.0002v, 7=0.03v, 8=0.2v. I also had it hard-mounted to the board, so I'm not sure if that affects anything. Maybe it got too hot, I do not know. By the way, I am using 2n5485 jfet on the mic module.

You don't have any power (to speak of) at your amp.

Pin 4 should be -12 -> -15V

Pin 7 should be +12 -> +15V

First thing to do is check each track for a short circuit and see if everything appears to be connected correctly. That's reading as if the NMA0515 is shorted, perhaps only the 15 or -15 rail but it's not a good look.

Also check those voltages with the IC removed if you are able, that can be useful and it's also why it's smart to use sockets, esp. for expensive parts.

I did not have a ic cradle yet, so I made one with the cheap black plastic gold color pined board mont pins connector. Also, I think I got myself mixed up with what orientation the pins were in, so hopefully these are more of a correct representation of the circuit.

Powered in the circuit (v): 1:0.6v, 2:0.009v, 3:0.01v, 4:-13v, 6:3.3v, 7:13v, 8:0.6v. Pin 4 rail is 13v, pin 7 rail is also 13v

No Power in circuit (ohms): 1:~15kohms, 2:3.8kohms, 3:, 4:climbs to 13ish kohms and falls to point kohms and back up, 6:30kohms, 7: same as pin4 but up to 16kohms, 8:15.17kohms.

Solo testing off board (ohms): 1:8.89mohms, 2:8mohms, 3:6.39mohms, 4:6mohms, 6:9mohms, 7:o.l, 8: 15.18mohms (volt test was 0.0001-3 on all, same for on the board with no power on)

Tested the NMA0515: volts in 5v, volts out is 29v

Still get electrical interference/popping sounds when touching components. The first and last parts of the recording are me touching the mic module components.

Those voltages are good except Pin 6 (which I'll get to).

The energy spectra of your recording:

Shows a nice bunch of stuff around or deriving from the mains so that suggests something is working, the fact that you can make it "pop" is another sign that the amp is working (so it didn't end up in Silicon heaven when you had it wired backwards. 😉

It's dropping off quite quickly too but that could easily be R/F leaking into the inputs. Matt's design relies (mostly) on the Faraday cage around the capsule and the interconnects to keep passing EM fields out of the electronics. Without that (and your digitiser is design to NOT render this)  you're leaving yourself open to amplifying everything from local radio stations to the CMBR (afterglow of creation).

It's highly likely that's what's causing what appears to be a DC offset at the output although that can be caused by an imbalanced in the input capacitors. I've found (and added to my designs) that a small resistor say 3K9 - > 4k7) across "shorting" Pin 2 and 3 together causes any DC overshoot to balanced out so the inputs "see" the same voltage on each pin. If they don't and you turn the gain up to hear something, what you end up doing is swamping the amplifier with DC.

I think you said you were using a SSM2017? (I could go back and look but it keeps my memory if I make myself remember)... now what was I cooking for dinner?

Sorry just my warped sense of humour, I've spent a fair few hours over the last few days trying to make quantum mechanics, more specifically, quantum electrodynamics, understandable to the cat. He's smarter than I am so if he gets it, everyone should have a fair crack. It's part of my next book (shameless plug) which teaches electronics in a way that doesn't require anything more complicated that a very basic scientific calculator.

For what appears to be a very simple schematic, there's a lot that can go wrong here (some of it due to the perils of low-cost strip-board) and some of it due to the ridiculously high impedance of electret capsules, requiring the use of a JFET.

If you've got "RF" getting into your inputs (via the capsule and JFET) that will guzzle up most of the power and you can't hear it because it's way outside the range of human hearing. You can isolate that by removing the capsule or/and (ideally) shorting Pin 2 and Pin 3 together AND to Pin 5, which is Ground.
 
When you do this the output (pin 6) should drop to a no more than a 100 mV DC (+ or -, doesn't matter) although it's likely to be a LOT less.

One other note of caution - less important but still worth considering is the connection to the gain control. That isn't a high-impedance node like the JFET but it can still pick up stray radio and other passing rubbish that you don't want getting into your carefully built mic.

So if you can try those out and get back to me, I'll try to reply by the next day. I'm a bit slow at the moment.

Pin 2+3+5 with some thin small wires readings: as low as -0.006v to -0.003v (the mic capsule wires are not plugged into the circuit). Not sure fully what you mean by removing the capsule part.

Noise when the mic is plugged in: popping when the capsule cable is moved too much, and a whining electrical sound when touching the gain switch/potentiometer.

Sorry, yes. (Ideally) remove the microphone "head" assembly so we can just get the signal from the amplifier with nothing going in.

Nothing in = nothing out.  

The two input pins are 2 and 3 - so we short BOTH to ground. It won't hurt the device.

You might get a small DC offset but that's expected (at Pin 6).

If you record the whine in Audacity (or similar) it might be the wretched noise I spent nearly a year trying to find. That what I get for looking in the wrong place! It turned out in my case the power from the digitiser's USB port was interfering with the main 5V power putting a 1KHz pulse on there. This is the USB communication and it plagued me for ages thinking I'd done something wrong.

The NMA0515 can remove some of that but it's still going to be in the "air" around you so you're probably coupling it when you touch the potentiometer pins.

Okay, I shorted out (connected) pin 2 and pin 3 to the ground pin 5, tested voltage, and got -0.001v or 0.000v from pin 5 (ground) to pin 6. This is without the microphone module assembly connected to the board. *I get the same voltage with or without the mic assembly connected to the board.* I also recorded the audio; the first part is the interference of the switch when touched. And the last is the popping sound of the mic wires being touched when I plug the wires back in. 

With the microphone assembly, I ensured that the mic module was grounded to the brass mesh. The only thing that could be something is that my wires for the mic are not in the desoldering wire; I am using it as the ground connection wire. The shielding for my wires is heat shrink. I was unable to find tubed desoldering wick. Not sure what else to do/change.

Excellent. 0V is perfect. I need to listen to the sound now and see if that offers any clues.

Digging into the signal, you can see the peak at 998Hz which is the USB signalling. So that looks like a win for you pre-amp section from the input to the computer.

Doesn't leave a lot to be "wrong", except the FET and capsule assembly.

It's unlikely the FET is busted if you have it the right way around.

Can you double check this please?

Okay, I checked and replaced the JFET. Everything looks right to me. Do you need any more photos of up close of the board or anything? Readings?

The FET is wired correctly but I am concerned that it looks like the brass shield could be shorting the connections.

If you didn't toss your original FET, we could perhaps use that to test things are working by making up a simple connection without a capsule.

At this point, I have been cobbling together materials and working on this project for about a month, and I am just ready to get this thing off my desk. I feel this was a bit over my head of understanding. From how Matt made it sound, it was going to be easy to make, drop some components in, solder here and there, and have a working studio-grade microphone. I would love to get it to work, but I am losing faith in my abilities to make it work. So, whatever I must do to try to get something out of this, I am fair game.

Writing "War & Peace, the Microphone Years" (Ok, "Making Michelle") about my own quest to "create a studio mic with toothpicks, sticky-back plastic, some sellotape and white wood glue" I know exactly how you feel.

I've literally got an embarrassingly large box of failures (might even auction them off for charity) because I rarely got close to the quality Matt did. There are so many blasted variable that have to be "just right" that experience can be crushing. Here's a painting of me contemplating how much weed beer I could have bought with that lot. 

There is a sweet spot (isn't there always) but the road can be a bit rocky.

First thing I noted when I started mucking about with these mics is the number that attract huge electro-magnetic fields and others don't. I'm not entirely sure why this is and it doesn't appear to be the construction but it might have to do with the materials. One of the things about Farady cages is that the metal we use makes quite a bit of difference. There are other features too, but that seems to be a major one.

I haven't taken all of mine apart to find out what uses what but one would have thought that the JLI-2555 would have the best. In fact it appears to be the worst! This HAS to be a design choice given how beautiful the design is. I can't think of any other explanation. Why? I don't know. Most of the cheap alternatives work at least partially out a cage at all.

The ones that DO work without a cage generally also have a JFET and this might be what you need to pick your chin up. 

A classic electret capsule (certainly the vast majority of them) come with an internal FET and those are orders of magnitude easier to work with but the best bit is this - you only need TWO wires!

Matt's design should work just fine wired up to the positive and ground of a classic electret (you'd ground the other pre-amp input).

Best of all these things only cost "a few bob" (less than a $1 a piece and often a lot less than that) because we make so many of them. I even considered using them as a supply of cheap JFETs at one point, although digging the part out proved fiddly and the gate isn't easily accessible (being part of the mic).

I suspect your pre-amp is working a bomb but there's something iffy at the capsule end. I have developed some boards to fit at the capsule - with a "final" one coming later this year, along with the other improvements I've been adding as the book progressed. When the piggy bank is sufficiently recovered (or I spend a few months down the local docks, in a mini - with my legs!) but anyway, I'm doing a version with a REAL tube!

Let me know if that helps or you need more assistance.

So you think if I buy a better mic modual with jfet I could just plug into the existing amp, with little to no change, depending on quality of audio. If so which once do you think would work best? Or put it on the back burner and wait for the new boards and book you are making? By the way been loving the humor.

I'm so sorry, I missed your post. That's entirely my fault.

Better ... start with one of those el-cheapo ones. Cheapest you can find. That way we can make sure you have the setup right.

Does  that make sense? The cheapest capsule you can find - probably $1 or less. Larger is better and pre-wired is going to make it easier to goof around and not worry to much.

I have a couple 3.5mm headset mics and a audio jack to type c, along with the type c breakout board. Could this work. Plug the mic into the adapter, then into the c breakout board (as a c to wire adapter), then however many wires out to the amp.

Not likely. Those little USB-C plugs can carry a whole ADC, etc. that they need to convert audio from a condenser into the USB.

BUT

If you mean the type with a little 3.5 mm jack at the end, that could help! Most have four rings, a few have three (mono ones) so long as they follow the standard layout which is highly likely that would work. The question then is do you have a 3.5 mm socket? Can you salvage one from old parts?

Alternatively you could chop the plug off but that can be a fruitless exercise now so much is done by machine and the cables are tiny.