The pressure from GOOD Chinese products is one of the challenges of coming up with DIY projects that make economic sense! Their presence makes justifying a DIY build sometimes a bit tricky. This is one of those.

For the cost even just using the enclosure and circuit is a great way of getting a good mic for cheap, and I expect it sounds great with a JLI 2555 or something similar.

Have you seen these OPA Alice circuits? They'd probably fit nicely.

@diyperks That's a fact Matt, no question of it.

I've bough a couple of these for testing and what I found is that older stock (I think that's the one with the chintzy gold basket) has a cheap FET and a cheap FETless condenser. 

The one with the silver basket (also NW-800) has a FETed condenser so you need a capsule with its own FET to make that work. My guess is the old (cheap) FETless electrets were old stock and have been replaced with a mass produced modern capsules - they're equally awful of course!

This is, as we discussed, where a (cardioid response) JLI or Primo capsule comes in a treat or they will lack the room rejection for a vocal mic. That's the case with my experimental Ali Express specials fall apart as they appear to be omis despite being based on the K67 design from the classic Neumans.

But just because they look like K67s, they're anything but. The K67 isn't an electret so it has to be energised (50-100v give or take). Fortunately we do have things like the JLI 2555 which is electret and well made too, so it sounds like butter.

I took a look at good quality (and I mean like high-end studio) recording capsules for the Neuman clones, especially ones that match the 2555  and that was an education into how the other half live! We're talking £100s and upward for the hand-finished ones. This is, naturally, why these things sound so amazing but the 2555 and some of the primos definitely give them a run for their money, particularly at the price!

The OPA Alice boards are made and sold on the US market and while it looks like they'll drop right in, I've been unable to locate anyone selling them in Europe. I don't know if the layout is Open Source Hardware (the circuit isn't complex enough to copyright or patent) but if it is, we could have some made for the community.

I'm still on the fence over a modern discrete FET solution which is the classic design and a self-biasing industrial standard like the OPA164x series from TI/Burr-Brown. There are pros and cons to each. I think the OPA versions are popular due to overall simplicity. I read about one guy who had built that onto stripboard [he stays sobbing softly into his cup of steaming PG Tips].

I even looked at a BiValve (not the tasty sea dwelling critters, but a valve/tube front-end with a bipolar driving stage) but I concluded that something of that nature would likely end up with us frying half of the community and getting sued by the others. 🙂 So maybe not that then.

But what's really interesting to me is the self-noise. One of the reasons (not the only one) instrumental musicians love condenser is their low self-noise. The capsule itself produces very little noise with most of it coming from the biasing resistors and impedance matching circuits. Even then, by their nature, condensers produce more "raw" signal voltage than a dynamic which has all the mass of a coil to drive and that swamps the noise produced at the front end.

You can see this when comparing capsule self-noise for a FETted condenser to a dynamic capsule and then looking at the relative output voltages at a standardised sound pressure level (SPL). 

But putting all of that aside - I wonder if you would consider doing a PZM (pressure zone) microphone with your usual panache? Mine has a 1mm acceptance slot on a piece of oak, but I don't think anyone has done one with a condenser like the JLI255. It would probably be better to use one of Primo's larger FETted SDCs for this though but it would still outperform the Crown by a significant margin. They're fantastic for conferencing and musical instruments including (but not limited to) grand pianos since you can mic up the entire instrument with the lid SHUT.

The other one I'd love to see you you lend your brass and metal design skills to would be a boundary mic. These are a little more specialised but would still be amazingly useful. The existing ones look ... well, like microphones rather than pieces of Art Deco!

The ultimate for builders is Ambisonic but that's probably a bit specialised. There are carriers designed to take Primo capsules but you need a set of four matched transducers and that's gonna run you about £100 plus you also need a four channel ampifier and recorder and then the software to decode the signals to ... everything from stereo to true surround sound.

https://en.wikipedia.org/wiki/Ambisonics

Rode, I think, lead the field in popular commercial ones. The NT-Sf1 won't leave you much change from £1000!

Given the matched capsules are about £100 that makes a DIY one sound like a very attractive proposition indeed. You'd just need to 3D print the head and you're off!

@marcdraco Jules from Sound Sleuth designed the circuit I believe. I've had an email exchange with him a while back and he's very much doing this for the passion of sharing mic builds so I'm sure he'd be up for making the design more widely available for people to construct. He's done a couple of videos about it on his YouTube page and it's been fascinating and educational to watch. I actually have one of PCBs with me here for testing, but (and I may be doing something wrong) it appears to have more noise than pure FET attempts, which is something I've mentioned before.

He's even got a PZM build!

Honestly I hadn't really considered such a build before, but now you mention it it might be really useful for presentation without having to wear a lav mic as it could just sit on the table in front of me...

Pretty neat aren't they? I've got a full design now working at 5V for the desk mic. I'll post it in the main thread later. I want to build it and then make a PCB to go on the "head".

Oh wow the perfect thread for me. I bought NW800 as a starter mic, one that included a boom but excluded the USB interface. I used a Bengoo 7.1 channel audio interface to go from the 3.5mm to USB. It seems to be pretty nice, and has an adjustment knob for a physical volume adjusting.

Though the boom did break after maybe 6 months of use, the thin stamped steel was weaker than the springs tension. So it broke just above the round part at the bottom. I'll have to break out the TIG welder and do some tacks, then reinforce with with epoxy or JB Weld. Right now I have some round steel filling the free space to mostly support it. Another thing to note for future buyers, reinforce that area for long term durability. Even a wooden dowel to fill the space would increase durability. And a drop of grease on the outside of course! 

Overall it works great for the money but seeing DIYperks video makes me want to upgrade it. And that OPA Alice video is perfect! I posted in the mega thread but this is actually what I'd be doing. Though Im not sure if I want to try just upgrading the sensor first and seeing how that works, but I should probably upgrade the circuitry as well. I just opened the mic and the included one does look fairly decent, better than I thought it would. But clearly these enthusiast made upgrades are better just looking at the filtering done at the XLR connector alone.  

The two MAJOR issues with the BME range are (a) the microphone capsule which is horrible (some come with FETs though, so be aware of that) and the on-board FET, where fitted. That's right, some of them use traditional electret microphones ...

Ironically, they are probably the cheapest ones to upgrade well since you can fit a more performant cardioid JLI electret, albeit for about twice the money for the part, which has a low-noise FET on board. 

Most Chinese mics use the Schoeps circuit downstream of the FET and that's not bad no matter what you do with it since it's just a current amplifier (the noise source is mostly voltage or pickup mixed with the signal due to the lack of current).

The on-board FETs are passable - those that have them as FETs tend to be on the quiet side anyway (at least for high-impedance work). Discrete FETs work better because they physical device is much larger than normally fits on the an Op Amp die. InterFET make an absolute huge device which is, I believe the quietest in the world right now, but it costs £60/$70 in BULK and they'll only sell you a bundle at once. I forget the number now but I've seen them and they look rather industrial/agricultural. They're actually meant for military and high-end medical gear of course.

My designs (in consultation with Matt) are later than I'd hoped (cough, cough) but the last lot are due first week of June - many of which solder directly to the back of a JLI mic per Matt's design without needing any extra parts. Everything is on the board but we don't know until they get here how well they perform. The best one, which to be fair, does use a a costly discrete FET, is an absolute beast but it's really too good for that capsule since it extends well below the mics own noise floor of about -60dB.

I've done a "version" of the OPA Alice that mounts direct on the rear of a one of these capsules - you know because, why not... but it doesn't have a regulator so you can't upgrade a BME800 with it unless you took supplies from the board and that's fiddling.

I've part designed an adapter board to fit into this body so you can fit one of our head units - I'll have it ready in a day or two but I've not been myself lately.