The modder’s old standby from MXL has gone to surface mount technology. This makes replacing components more difficult for the upgrader.

Yellow = SMT 990     Cyan = previous THT 990     Green = Interface noise floor

1KHz output at 500mV p/p (a very loud signal) injected along with a bit of hum from sig gen.


So the best version of the MXL 990 so far is the old model with Micrphone Parts’ Advanced upgrade kit. Unfortunately, home upgrades of surface mount boards are very difficult. Removing and replacing such tiny parts is a skill few people are willing to learn, and sourcing small quantities of suitable upgrade parts is just about impossible. As it happens, I have been designing SMT products for the last 20 years or so, and have design kits of resistors and capacitors, but I don’t have transistors, zener diodes, etc. and I don’t like working with tweezers under a microscope. So the new version really isn’t a modder’s platform unless you can get a replacement printed circuit board. I’m thinking of making one, and I’d bet others are doing the same.


That said, I did modify my pair as well as I could with a pair of large capsules, a pattern selector switch, simplified input circuit, and higher bias voltage. The capsules I had on hand are Alctron 35 mm dia. K-67 types with 6 mil diaphragms. I like them with a high bias voltage, around 80 V, so I replaced the stock 6.7 V Zener, D3, with a 12 V part. The replacement diode’s leads are tacked to the pads where the SM diode used to be.


One of the denizens of the Yahoo! micbuilders forum, Johnny Zhivago, came up with a neat way of mounting a switch in one of his mics, and I copied it here. The switch is mounted in the plate which divides the capsule compartment from the electronics compartment, with the lever facing downward. The capsule is wired to the switch inside the headbasket, and a lead goes from the switch down to the FET.

Cool, huh?


But there’s an OOPS! when you try to put the mic together.

The rim of the headbasket casting is in the way. First I tried grinding it down with a Dremel. The soft metal just loaded up the stone and was going nowhere fast. I tried a diamond cutter in the Dremel. It was also very slow going. What I wound up doing is making a bunch of kerfs with a saber saw and then breaking out the metal between the cuts. Pretty crude hack, but what the hey...

Nobody’s going to see it.

The advantage is it allows using a high quality switch with SPDT on-off-on contacts so in future with a suitable printed circuit, I can have omni, cardioid, and fig.8 patterns selectable by spinning the base off and flipping the switch.


This time I pulled both inner layers of screen out of the headbasket. With this capsule and the flat response electronics, this is a crisp sounding mic. I find the boosted bias voltage tames the 8 KHz screech common to this category of mic, making it useful, though not my favorite everyday mic.


Sample recording here.