top of page

Don't miss a post

Writer's pictureLuke Petersen

Frequency Response – No Sparkle? All Boom?

Over the past eight years, I’ve answered countless phone calls and emails with questions about ribbon mics. When to use them? How to use them? Storage position? Preamps? Input impedance? Max sound pressure level? Phantom power? Wind? Dust? Foil thickness? Corrugation method? Ribbon tension? Polar patterns? Frequency response? Transients? If you do a quick online search, I can guarantee you will find answers, but I can’t guarantee that those answers will always be correct or based on more than intuition or blind tradition.

 

Before his passing, I was blessed to have my mentor, Clarence Kane, to answer any questions that only he could as RCA’s last living ribbon microphone technician. Over the years, I was fortunate to work on nearly a thousand ribbon microphones under his close supervision. I could fill a book with all of his tricks-of-the-trade and stories—many pertaining to microphones but many more to age-earned wisdom. Following my study of ribbon microphones in an academic setting, Clarence and I together performed many tests and experiments to prove or disprove the prevailing claims about these mics. Thus, my research and testing of ribbon microphones has continued for over eight years, and I’m just getting started. This, combined with the knowledge that was carefully passed down to me, leaves me with a great responsibility to ribbon microphone users everywhere.

 

I am surprised by the widespread myths surrounding ribbon mics that often have little to no basis in scientific testing. The primary parameters of ribbon microphone design have already been meticulously tested and optimized to death by legendary acoustician and ribbon microphone inventor Harry Olson. Although reading his academic work may be challenging (even I grumble at the words “dynamical analogies”), it is misleading when musicians or repairmen make claims that contradict Olson’s tried-and-true research. So, in the tradition of the great Harry Olson, I will be doing my best to dispel these myths one by one and shed light on the timeless technology that is the ribbon microphone—all through scientific testing. Beginning with...

 

Frequency response—the mother of all microphone specifications. It is your microphone’s unique tonal thumbprint. It tells you everything you need to know, that is, when it is telling the truth! Let’s quickly go over the basics of frequency response graphs…

 

The horizontal axis represents the spectrum of sound, with lower bass notes/frequencies to the left, and higher treble notes/frequencies to the right. Ideally, this axis should range from the lowest note humans can hear (20 Hz) to the highest (20,000 Hz or 20 kHz, although as we age this slowly drops). However, it’s not uncommon for microphone manufacturers to limit the range of this spectrum for two primary reasons: the natural limitations of their testing equipment, or hiding microphone performance anomalies they aren’t proud of. If you limit the range of your graph, you could deceive the viewer by making the range of your microphone’s response look broader at first glance (Fig. 2 compared to Fig. 1).


Fig. 1 – Frequency Response with Appropriate Axis


Fig. 2 – Frequency Response with Limited Horizontal Axis


Fig. 3 – Frequency Response with Inflated Horizontal Axis

 

Put simply, the vertical axis represents the sensitivity of a microphone in decibels. Decibels are an incredibly powerful yet often confusing scientific unit that come in many flavors for a variety of uses (Relative dBV, dB SPL, dB Z, dBA, dBC, dBV [re: 1V], etc.). For today, all you need to know is that higher sensitivity will sit higher on the graph and lower sensitivity will sit lower. The thing to look out for here is axis range. Some microphone manufacturers will inflate this axis to make a microphone’s response appear more flat (Fig. 3). Reasonable frequency response graphs will have a total range of 40 dB or less, say -20 to +20 dB.

 

The ideal, most versatile microphone should in theory have a smooth, flat response from 20 Hz all the way up to 20 kHz. So how do ribbon mics fair by this standard? Many say they are too “dark” or lack “high end sparkle/air”. This is another way of saying that the frequency response doesn’t extend high enough (to the right). But this is not necessarily true! While many ribbon mics are designed to sound “darker”, many ribbon microphone models extend well above 15 kHz and all the way up to 20 kHz at healthy signal levels. In fact, ribbon microphones exhibit an unmatched smoothness in this upper range due to their lack of pesky high frequency inharmonic resonances (which are inherent to the circular fully-bounded transducers of dynamic and condenser microphones). In other words, the physics of the ribbon element encourage smooth, harmonic sound reproduction—much like the way we hear in real life!

 

And what about the low end? That is where ribbon microphones really shine! The element in a ribbon mic is tuned near or even just below our lowest audible frequency, so ribbons are able to reproduce sound all the way down to these low notes with ease. Traditional figure-8 (bi-directional) ribbon microphones also exhibit a “proximity effect”, or enhanced low end as you get closer to the sound source. This is due to the logarithmic change of sound pressure level with proximity to the source, resulting in a greater pressure gradient across the microphone at closer distances. However, this is not inherent to ribbon microphones, but is exhibited by any microphone with a figure-8 polar pattern. In fact, ribbon mics of different polar patterns don’t even exhibit this proximity effect in the same way. In other words, proximity effect is solely a function of a microphone’s polar pattern. But regardless of polar pattern, ribbon mics still excel beyond other microphones in the low end due to their naturally low tuning frequencies.

 

Fig. 4 – RCA 44-BX Published Frequency Response...see if you can spot the error!


 Fig. 5 – RCA 44-BX Measured Frequency Response

 

Fig. 4 shows the published frequency response of one of my favorites, the RCA 44-BX. Fig. 5 shows the real response of a 44-BX measured in my lab, including the selectable M, V1 and V2 bass roll-off modes. This 44 (freshly re-ribboned to RCA specs) is an excellent example of the impressive range of a traditional passive ribbon microphone. The high end is subdued but still plenty sensitive for sound sources louder than a whisper. And the low end is thunderous, especially at closer proximity.

 

There are countless factors that can affect a ribbon microphone’s frequency response, but today I just wanted to cover the basics. I will begin to explore more of these factors soon and include their effects on frequency response along the way. For the technically curious, my testing setup includes a calibrated full range monitor and software that calculates an accurate near-anechoic response down to 77 Hz by rejecting known room reflections. Measurements were taken at 94 dB SPL (1 Pa) at 12 inches into a 22 kΩ load with variable smoothing.


Until next time,

–Luke

 

Curious about your own microphone’s frequency response? We now offer microphone testing as a supplemental service. We can test your personal microphone and provide a unique specification sheet featuring measured frequency response, measured sensitivity, measured output impedance, and more! Visit pitmanmic.com or contact us for more information.


Want to know more about the science, design, history, and misconceptions about ribbon microphones? Sign up for updates and tell us what you want to hear about next in the comments below!

2 Comments


Tom Booth
Nov 11

Hi Luke, nice article! I have noticed the error in the response curve from the Instruction Booklet (IB-24267-5) that I think you extracted the figure from. It is interesting that the response curve from an earlier version of the Instruction Booklet (IB-24267-3, for the MI-4027-B & -D versions) has the same error! Somebody must have blindly copied the earlier figure.


Keep up the good work, and I'm looking forward to more of your posts.


Tom Booth

Like
Luke Petersen
Luke Petersen
Nov 11
Replying to

Thanks Tom, nice to hear from you again! And good catch, you're absolutely right. Also, if you look in earlier catalogs (1941), in that case they actually graphed in units of VU at 1 Pa SPL into matched load, despite this matched load condition going against Olson's design recommendations. As you might expect the curves look noticeably different. Perhaps we'll tackle that next! -Luke

Like
bottom of page