Friday, December 27, 2019

Tune A Room For Perfect Audio 27Dec19

Here I'll briefly describe how to "tune a room" for better audio... using only digital signal processing of source output signals with equalization filters. Here I don't consider altering the speakers or their placement, using wall treatments like bass traps, or even the audio source (except I assume it's digital and can be adjusted through equalization). This process is basically equivalent to a pro sound engineer's technique as he/she sets up any PA system in any venue... but one can simply apply it to home audio and theater systems and greatly improve "the sound." The simple tools I'll use are the donation-ware program REW and the $75 acoustic measurement mic miniDSP's Umik-1. The room/speaker setup I will adjust is the Nest study consisting simply of JRiver MC25 playing FLAC files from my MacBook via USB through a NAD digital amp to a pair of NHT SB1 bass-limited bookshelf speakers. This setup has been presented before.
Up front, I know deep bass is missing here and that, from simple listening, the whole low end is bloomy and distorted. It is the latter that I especially want to correct. First I connect the measurement mic to the Mac and REW and place it at my listening position - at the fly tying desk.
After assuring that REW is setup to capture measurement input on the Umik-1 and that my NAD amp is selected for output of test signals through the speakers, I shoot a simple 20-20K Hz freq sweep signal and capture it. I do that a couple times and then smooth the average of the responses. Next I load the desired "target" house curve - an ideal response is mostly flat, and somewhat declining throughout the 20-20K Hz range. I "line up" the house curve with my measured response. Yuck... see the weak sub-bass and see the blooming occurring in a number of spots in the 100-1.5K Hz range.
To adjust for these "problems" with my room and speakers, I use REW's EQ tool which "automatically" creates EQ filters. These will need to be applied to digital source material before it reaches the DAC in the NAD... one can alternatively apply the same filters to frequency bands of an analog signal on its way to amplification - but the digital domain is often easier and most of the audio sources we play these days is digital. Here's REW's EQ tool showing the filters and the resulting new, flatter, non-bloomy anticipated frequency response. The filters are actually a complex set of overlapping adjustments (as shown in the 2nd picture below) which mathematically combine to flatten the source material when played through the SB1s in the Nest study.
I then exported these filters as a text file. I am going to apply these filters through JRiver's parametric equalizer in the DSP section. Unfortunately there is no direct importer in JRiver so I have to type in all the filter adjustments and save the set as associated with Nest Study Listening. Of course I use entirely different DSP settings and adjustments when listening to headphones using JRiver on my Mac and there I employ a fully calibrated adjustment scheme via Sonarworks (this scheme has been presented previously). Unfortunately YouTube or Amazon Music output won't be equalized when played using this JRiver parametric equalizer because it is not applied system wide on my Mac. I could use a Mac sound plug-in like Boom 2 or Hear to apply equalization system-wide, and general solutions probably should do it that way but I really just use JRiver MC for playing music in the Nest Study through the SB1s.
The proof in the pudding is that the sound is so much sweeter and flatter and it's now much easier to listen for longer periods. Sure there are choices and tweaks one might want to make along the way here but in ~30 min I 1) measured, 2) analyzed and 3) adjusted to tune my room for the best audio the system can produce.

Thursday, December 26, 2019

Machinations With Marinick, Measurement Mics And Magnificent Magnetics 26Dec19

Michael Marinick was in Breck to visit his ski instructor daughter Gabrielle for Christmas and we took some time together to set some records straight and formulate machinations for the future of America... and drink some beer. Of course I introduced Mike to ORB whether it was out of the can on our deck or on the patio of the spot itself. Very good times with an old friend.
The dynamic range of my originally purchased USB acoustic measurement mic, the Dayton Audio OmniMic, was insufficient for my needs and not suited to task. I had to exceed my THX reference levels a lot to make valid measurements and that required me to use earplugs and duck and cover during tests. I cross-graded to the miniDSP Umik-1 which is kind of the standard and works perfectly. I accessed my serialized factory calibration curve and imported that to REW for corrected measurements. Much better.
The Russians got the ViaWave SRT-7 ribbon super-tweeters to me via FedEx. Wow, they weight 3.5# each. I must say FedEx failed on several occasions to know when they'd deliver and missed by several days ultimately. Not good for them and I've heard of their weak 66% on time delivery record this holiday season from news coverage. Anyway, I wasted no time in fashioning a sophisticated package (not) for them and went right to listening. They are sealed in their own enclosure so nothing more was needed to really make them sing. They are awesome but I am glad I have an attenuator as else they'd take over the high end - somewhat as expected/planned. The next steps are to package it all up into a cleaner add-on for the Blodgett Theater sound system. I have all the parts now but need to cut, assemble and integrate my final DIY SuperTweeter add on.

Wednesday, December 25, 2019

Merry Christmas 25Dec19

Merry Christmas all. Kala and I enjoyed a visit from Matt & Alexa in Silverthorne this year for the holiday. We celebrated, broke bread, played games and watched some bad TV. Matt worked before and must work after this Wednesday Christmas so the visit was brief but fun.

Monday, December 23, 2019

Runs On All Peaks With Sheldon 23Dec19

Dave Sheldon & I hit Breck today. It was very nice weather and the snow, while not new, was pretty decent. We were able snag rides on all five peaks and did plenty of bombing and bumps. It was surprisingly crowded for a pre-holiday time. I even met up with Mike Sheldon, one of Matt's classmates, a guy I haven't seen in 8 years... cool. Too, the grocery was a mob scene as I pulled duty to get the holiday food. Merry Eve all.

Saturday, December 21, 2019

Wow... Paste Studio In Manhattan 20Dec19

I'm trolling my channels, as I do, of strong radio streams, and I come to Paste Studios in Manhattan and they killed it today. Three spectacular concerts from the NYC uber-skilled: Common Jack, Samantha Fish and BRKN LOVE. Nods to great songwriting and sweet southern blues. Not a perfect style for me but these performers were great - Samantha Fish completely shreds a guitar and writes so down low.


Thursday, December 19, 2019

Blodgett Theater Super-Tweeter Update 19Dec19

I have acquired the ViaWave SRT-7 ribbon super-tweeters. They arrive from Russia in a week or so. I requested and have been supplied the specific frequency response curves of the serialized drivers. Below I have taken the Blodgett Theater's current frequency response and added the SRT-7's response to it. The resulting simulated new Blodgett Theater response curve is much flatter and sparkly in the 8K-20K range. Observe it's better matching to my House Curve than my current measured response. I will add a variable attenuator (L-Pad) to the super-tweeters so as to be able to adjust their influence and ensure the high end is not too bright (too far above the House Curve line).

Wednesday, December 18, 2019

20th Anniversary Of Wheatus @ Paste Studios 18Dec19

Paste just dropped a full 30min session of Wheatus/BBB and his band at Paste Studios in NYC. Pretty cool; I often troll Paste for good stuff. Dunno who the new male percussionist is but he plays plastic wrap and bell with Klass. Click through to the full session on YouTube.
I also discovered a recent release from the BBC whereupon HAIM sing accompanied just by piano... it's awesome.

Monday, December 16, 2019

Wilderness Medicine Certification & Boulder Stuff 15Dec19

Kala and I headed to Boulder for the weekend for class and fun with Wilkes, but we did miss an epic dump in Summit County... I got a number of pics of two feet of powder. We both attended the NOLS Wilderness Medicine class for backcountry 1st responders held at Chautauqua. The two days were packed with process and diagnostic info and rescue scenarios... it was really excellent, and we are now certified. There's no way to retain all the knowledge without some practice and while we don't want to encounter things to which we were exposed, we are going to have to practice and review somehow. So... be aware I can treat you if you need help, but you'll still need to carry me out.
Oh yea... unbeknownst till after, in our class was Eric Larsen, a highly decorated polar explorer from Boulder. On one exercise he was my scenario patient. Later, reading the latest issues of Outside, I recognized him from a double page ad for Ford and one for Citizen. I then looked him up and wow he's done some stuff. Sheesh.
While in Boulder we partied with our great friends Tom & Leslie Wilke. They took us to a keen new restaurant called Rosetta Hall. In the hall there are 8 mini-restaurants fashioning good food in styles from all around the world. There's also one nice central bar - it's work a stop. I had Korean chicken & cucumber salad and Kala tasted the French duck cassoulet.
Upon returning to COS, we did encounter snow as expected. Kala did the honors  of snow-blowing in the morning despite the frigid temps.


Wednesday, December 11, 2019

New Audio Streamer And Room Acoustic Measurement 11Dec19

I built a new HD audio streamer from a $30 Lenovo nettop to deliver high def streams from my music store to my lower level hifi system at Blodgett. I'd previously only been delivering transcoded 320K mp3 streams through Roku media capability (with a horrible UI and navigation), or through Plex with a weak UI. I wanted full 192KHz/24bit streams from my JRiver Media Center and a great UI. So, I put JR MC on a Q150 (1.6Ghz Atom D410/1Gig/nVidia ION/WinXP Home) nettop and futzed with it for a day (graphics drivers, networking issues, etc.) to get it running well. Many have failed or given up attempting good networking from Win7 to WinXP Home but I persevered. Kala even commented that IT really persisted too strongly. The nettop is connected to my TV via VGA, to my network via Ethernet, and to my DAC via optical. JRiver has a keen "theater mode view" which looks great on the TV and has simple navigation of my entire 2TB library. I've also integrated the new APC (audio PC) with my Harmony universal remote via a USB bluetooth adapter - allows slick control of the simple theater UI. 
I also updated my acoustic measurement technology through use of a great donation-pay app called REW, Room Equalization Wizard. REW allowed me to measure my Blodgett Theater audio sound pressure levels (SPL) and distortion (I did so at listening position A and B with a USB measurement mic connected to the Mac. It even allowed me to create a correction equalization curve (which I entered into JRiver MC's parametric equalizer) to improve/flatten response. I focused on reducing the "peaks" as raising the "troughs" would significantly increase power required during playback. I matched the theater frequency response to a "House Curve" as opposed to a flat response as humans do prefer a slightly higher bass and slightly attenuated treble as my House Curve describes. My room and speaker system are decent but there's more work to do, at least at the sub to mains transition and at the 10K-20K high end. I'll boost the sub a bit and then EQ it further if needed. I'll probably need to add a super-tweeter for the high end "repair." I've known of this high end rolloff just from my subjective listening tests but now I have more evidence I should do something about it.
I like this sealed ribbon super-tweeter from Viawave in Russia, the SRT-7. It's flat past 20KHz and is very high sensitivity @ ~100 in the frequency range in which I'd use it. The alternative would be a Fostex 90A, a similarly flat, high sensitivity super-tweeter but using horn technology. I could add a pair of either and adjust the gain using a DIY L-Pad and crossover to compliment the tweeters from my main Paradigm Reference Studio 100s. A good next audio project.


I have been studying and listening to bolster my ancient EE training. I initially thought I do speaker design and sound production but then computers, software, networking & products became a my thing as a pro.