Keeping the Family Business Going

Among the more lamented departures from the world of electronics was RCA's microphone business. A few years ago I visited the college radio station where I'd worked in the 80s, and one of my former professors was proudly showing off an RCA 44A ribbon microphone the school had acquired. Though it had a few dents and scratches, it was a work of art. Soon after I saw it, the university's music department borrowed the mic for recording sessions and never gave it back.

A typical RCA 44A ribbon microphone(Source: Wikipedia)

A typical RCA 44A ribbon microphone
(Source: Wikipedia)

Like original McIntosh MC275 amps, old ribbon microphones are collected because, when used correctly, they produce sounds hard to get elsewhere. But eventually even the vintage electronics section of eBay runs dry.

One of the companies helping to provide new ribbon microphones and preamps, Cloud Microphones, has been able to continue the work RCA did through collaboration between a musician/businessman, Rodger Cloud, and Stephen Sank, son of RCA's ribbon microphone designer Jon R. Sank.

Jon Sank had done design work at RCA that included the BK-10A and the BK-11. When he passed away in 1998, he left Stephen with the knowledge and other tools to continue the ribbon microphone work that RCA had left behind.

Cloud came to Tucson as a singer/songwriter and started getting involved with bands and recording music. Sank moved into an office suite next door to Cloud's studio, and in 2006 they started working together.

“So we collaborated on the initial designs of some of the mics, and I worked with him first refurbishing the vintage mics and so forth, and I really fell in love with the technology as it was in the 1930s and 1940s. It was just pretty darn great,” Cloud told me. “However, there were certainly challenges with the technology, most notably those types of microphones had a really weak sensitivity and are really difficult to deal with in terms of amplifying them.”

Cloud said that even with exceptional amplifiers, it was very difficult to use ribbon microphones to record some softer sounds because of the gain requirements. This led Sank and Cloud to develop the JRS-34 ribbon microphone, named in honor of Sank's father and noting the year he was born.

Rodger Cloud with his version of the RCA mic

Rodger Cloud with his version of the RCA mic

“As we worked on it, I really felt the need for an active microphone,” Cloud said. “And we really pushed ourselves to come up with something. We weren't just coming up with something that was OK. It had to be absolutely audiophile quality.”

The result was a ribbon microphone that uses a Linear Systems LSK389 monolithic dual n-channel JFET to amplify the signal. Cloud wanted a dual JFET that would provide common-mode rejection in his amplifier, and settled on the LSK389. “It's just the JFET that we like the best. The power requirements for the circuit are low. We tested all of them, and this is the one that seemed to give us the results we were looking for.”

Wiley Ross, who runs the recording studio at the University of Arizona, is a big fan of the Cloud's gear. “I use the Cloudlifter almost always when recording from a passive microphone,” he wrote in a recent email to me.

“I don't use it because I need the gain, I have plenty with my Millennia Media HV-3R, but because the sound is better; richer, more depth,” Ross wrote. “With the Cloudlifter Z it's like having your mic locker expand to do all the different sonic possibilities afforded.”

Ross added that Cloud's JRS-34 is a “main-stay” ribbon microphone. “I love it on all wind and brass instruments. Listening tests with other fine ribbons with faculty, students, and staff confirm its superiority.”

The Cloudlifter

The Cloudlifter

Using a JFET-based circuit wasn't an obvious solution at first, Cloud said. “We were looking at several options with the microphones, and I was the one who suggested that we try the JFET technology. We were also looking at chip-based technology, different ways we could activate the mic. One of the problems with a chip-based design was we were going to have to use a couple of nine-volt batteries. We just couldn't figure out a way to get it to work on the phantom.”

Using the Linear Systems' LSK389, which combines ultra-low noise with low IDSS , enabled Cloud and Sank to build a phantom-powered active ribbon microphone. Cloud said the prototype of the active circuitry was set up in a test box to enable testing various microphones. “And it only took a few moments to realize that this box is cool. And that's how the Cloudlifter accidently came to be.”

The Cloudlifter is the active part of the company's ribbon microphone broken out into a format that can be used with many microphones people already have, and it soon became the company's best-selling product. “It's a game-changer, or more accurately a 'gain-changer,' because it allows people to use microphones that they normally would not even consider for softer sources, acoustic instruments, because the problem in the past has always been defeating the noise floor of the preamplifier.”

The Cloudlifter is a standalone product customers could use to boost the gain of other microphones. It plugs into the line coming from a microphone and uses any standard phantom powered microphone input device to provide up to 25dB of “ultra-clean, transparent gain.”

The Cloudlifter is recommended for: recording direct into a digital audio workstation interface; working with noisy or low gain preamps; using mixers/preamps that impair the sound at higher gain settings; working with passive ribbon microphones or low-output dynamic microphones; capturing softer sound sources such as acoustic instruments and voice; and when using long cable runs.

When Cloud and his colleagues were first designing the circuit, they used Toshiba JFETs. “We found out about Linear Systems' products so we tried and we compared it,” says Cloud. “We liked it better than Toshiba. It was still an improvement, and I liked what it did for the Cloudlifter — it seemed to drop the noise floor even further, which is always a good thing when you're dealing with really low-sensitivity microphones.”

Cloud said that after sales increased for the Cloudlifter, it became too labor intensive to match the single LSK170 JFETs. “It was becoming almost a full-time job to hand-match these little JFETs, so we started looking at the duals, and we went to the surface-mount technology, and I've got to say the Linear Systems duals are awesome. We can't match them that closely by hand, and the failure rate of the entire card as a whole is less than one in a thousand now. And we couldn't achieve that making them by hand.”

The common-mode rejection enables the Cloudlifter to “pass clean audio in the worst of circumstances, next to a radio tower, in EMI hell,” according to Cloud.

Related post:

21 comments on “Keeping the Family Business Going

  1. eafpres
    June 27, 2013

    Hi Tim–really nice article.  The music industry, both professional and amateur, seems chock full of really cool gear.  It is amazing to me the range of products available.  It seems to me that there are lots of folks who just really love music, and make really neat products even if they don't sell tons of them.  Take amps, for example.  There continue to be a broad range of boutique tube amp makers, so if you are a guitar person, the range of options is amazing, and relatively affordable.  

    These mics and mic-amps you describe appear to be in that category of being done by people who love the music and want to give people the best possible choices when recording.  In this time of all-out mass production, it is really nice to see this kind of business. 

  2. mccunets
    June 27, 2013

    Many thanks.  Yes, high-end audio lends itself to small, innovative companies.  I ran across a company in Romania a couple weeks ago that does specialized amplifiers.

  3. jkvasan
    June 27, 2013

    Hi Tim,

    Nice post.

    It was clever to have this Cloudlifter as a standalone module, if my understanding is right. You have two mangoes with one stone, as people in this part of world say. One, you got the product matched to present range and second, it also is compatible to legacy products. What more you could ask for as an enthusisast?

    I would be curious to know the Common Mode performance/Noise reduction improvement in terms of percentage between the original single FET design and the dual one. Just , for my interest, you know.

  4. Scott Elder
    June 27, 2013

    ” lends itself to small, innovative companies.”

    I think this is just an extension of the analog industry in general. Many ways to solve a problem and very difficult to prove that solution A is better than solution B because technical specifications are clouded behind human senses and emotions.  For example, distortion in just the right way is viewed by many as a good thing, and bad by others.  

    In digital, there is no hiding.  Tell me your Power, MIPS and FLOPS and a winner can be proclaimed.


  5. Netcrawl
    June 27, 2013

    @Tim nice article, RCA, the second-oldest recording company in the US history, I believe its owned by GE, its introduce us to the world of music- and the microphone.  Its the company that brought us the power of radio and microphones, yes, I remember this one, she's old, well I'm not sure about her age pretty old about 40-50 years old (my best guess). 





  6. Vishal Prajapati
    June 28, 2013

    It is very true that specialized amplifiers are still made and they are expensive too. Some of the examples I found here




  7. mccunets
    June 29, 2013

    Good question, I'll check with LIS's president, John Hall.  Hall first started working on duals with Jean Hoerni at Union Carbide in the early Sixties.  The big question then was whether tapering in the epitaxial layer would cause the duals to be mismatched.  This problem (as I understand it) was solved mostly by interweaving the two transistors on the die. Also, the epi layer inconsistencies often averaged out. Matching two singles takes a lot of time, so having duals that match well helps in production efficiency.  It's probably also not possible to get the same temperature drift with two singles as with a well-designed dual.

  8. mccunets
    June 29, 2013

    Great stuff.  The Burning Amp conference, www dot burning amp dot org, has a lot of cool audio gear, prototypes and small production run.  

  9. D Feucht
    June 29, 2013

    Interesting, nostalgic article, Tim – and kudos to Mr. Cloud! (I'm glad the company name has nothing to do with cloud computing.) These are the microphones that became the symbol of radio broadcasting. When I was in high-school at Benson Polytechnic School in Portland, Oregon (1964-68), we had a 250 W AM (1450 kHz) radio station, KBPS, that was run by students. It covered all of Portland. I remember one of those ribbon microphones in the studio.

    For my own (slightly illegal!) AM station at home, I bought a cheap (crystal, not ribbon) look-alike microphone and still have it. In the transmitter, I used 6L6GC tubes for the final stage and could light 100 W light-bulbs with the output power but did not have an AM antenna system that could deliver it. And that might have been to my advantage.

    After I got my first-class FCC radiotelephone license my senior year (age 17), I gave up home radio stations and turned to building oscilloscopes. I joined Tektronix at age 17, and my broadcasting “career” wound down quickly after that, though Bill Vandermay, chief engineer of KATU, channel 2 TV, in Portland, and an aavid “ham” who built the first satellite weather receiver of any TV staion in Portland, lived 3 blocks away from my (parents') house and my radio station in SE Portland, though I don't remember telling him about it. He also set the national high-jump record when he was an engineering student at the U. of Washington.

  10. David Maciel Silva
    June 30, 2013

    These amplifiers are specialized valved?

    The best and most expensive usually, but not very common to find, could share the data for this company?

  11. Vishal Prajapati
    July 1, 2013

    Thanks for the link. Cool stuff on the website. Really interesting.

  12. mccunets
    July 1, 2013

    @Maciel: Sure, would love to hear more about it

  13. David Maciel Silva
    July 1, 2013

    I found this site that sells plus valve amplifiers.

    Apparently the price is not so high ..

  14. Rodger Cloud
    July 1, 2013

    “I would be curious to know the Common Mode performance/Noise reduction improvement in terms of percentage between the original single FET design and the dual one. Just , for my interest, you know.”

    In answer to your question, the original single FET design is the same, except that it used 2 sets of hand-matched JFETs. The common mode noise rejection is theoretically and functionally the same, at least on an audible level, it's just that it was very time consuming and there would always be a few that did not pass our common mode noise rejection test. With the LSK389 duels from LIS, the sound is the same, but the pass rate for our -very strict- noise reduction test is pretty much 100%. The “match” is much closer than what can be accomplished consistantly by hand, and the fail rate on the assembly line is near zero. 

  15. jkvasan
    July 2, 2013


    Hand matching components could be very tricky. As you said, factory built devices have a better chance to match. The differences also may not be that wide. Pass rates can be higher in the line.

  16. RedDerek
    July 3, 2013

    As I posted in a comment on Brad's 'Very Low Gate Threshold FET', obtaining a dual MOSFET would afford the best potential matching since the FETs are most likely to be located next to each other on the wafer – exception is when moving from one row to start the next. Also, to best match the thermal issues, having MOSFETs co-packaged would allow them to share the same heat.

    I wonder how the low Vth parts that Brad mentioned would be of benefit in the circuit?

  17. jimtech14
    February 20, 2014

    Certain countries seem to have engineers more prolific in designing valve amps, the obvious standouts being China and Italy – this Unison Simply Italy has EL34 output tubes to generate just 12 watts, but my friend has one in his system and the quality is just amazing.

  18. kendallcp
    February 20, 2014

    Matching FETs for medium-scale production quantities of amplifiers is not that difficult.  I was building preamps with JFET input pairs in the early 80s, and computerized measurement technology was already feasible then (HP-85, anyone?).  You need one bag of suitable JFETs, one enthusiastic lab tech, one test fixture that sets the Ids and measures the Vgs, and a set of numbered baggies.  You measure the Vgs of all the transistors and log the values against the serial numbers of the baggies.  Then you just run a simple sort/match process in the computer, and get the tech to pull out the pairs (or quads, or…) that match.

    It's great that we can still get JFETs for this kind of circuit, because sometimes they are just the best.  I had a soft spot for my JFET-input switchable moving coil / moving magnet phono preamp.  The reason I used single matched JFETs (usually J108 or J106) was that they were only a small fraction of the price of the only available low noise duals at the time.  If that price skew has gone these days then yippee!

  19. kendallcp
    February 20, 2014

    Sounds like you're talking about the ALD MOS devices.  It's quite unlikely that these are anything like quiet enough for these microphone applications.  Of course, someone with the time to read the datasheets may demur…  Low frequency noise can be a pernicious issue with MOS front-ends in audio, and not many opamp manufacturers have that under control without chopping techniques.  The TI OPA2376 seems particularly good for low 1/f noise corner for a MOS-input amplifier.

  20. chirshadblog
    February 20, 2014

    @jimtech: Is it a Chinese product ? Normally the Chinese product quality is really good even though they copy and make low cost products. I feel that in Chine you get the best designers. 

  21. jimtech14
    February 21, 2014

    @chirshadblog Unison Research is an Italian company, they also make Aria loudspeakers 🙂 I agree about Chinese reliability !

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.