Michigan Electric Company: Between The Obscure and The Unknown

 

While CPS1 is designed for all common cylinder formats, the primary impetus to develop it was the absence of a system to audition and transfer wax cylinder recordings optimally with minimal wear.  The earliest of these are made of brown wax, and they are of particular interest because of their rarity, fragility, and often murky provenance.

When affordable phonographs started to find their way into homes (mid/late 1890’s), the demand for cylinders took off. But record molding wasn’t commercially viable.  Records were produced in batches.  So, the technical threshold to go into the business was low, and many new companies sprang up.

Little is known about these early “labels,” most of which went under in a matter of years.  Typically, all we have is corporate/legal information, some advertisements, and perhaps a few short entries in The Phonoscope, the fledgling industry’s trade paper.  Actual discoveries of recordings have been few and far between.  But, over a century later, they still occur…

 

20 years ago, Michigan Electric Company was virtually unknown among early record collectors.  But in about 2010, a few Michigan Electric cylinders surfaced.  The records stood out: the cylinders have a blind emboss on the rim and their boxes are branded with a large rubber stamp.  A few years later, a Michigan Electric catalog from 1897 was found that included a list of releases.  Then, last year, a Michigan collector made an “attic find” of about 40 Michigan Electric releases still in their original dilapidated cabinet.

Working with Ken Flaherty, a CPS1 owner based outside of Detroit, we pieced together the story of Michigan Electric.  It appears in the current issue of The Antique Phonograph, the quarterly publication of the Antique Phonograph Society.

Following are transfers of a few Michigan Electric cylinders, premiering (again) 122 years after they were released!

A Hot Time in the Old Town Tonight, J.W. Myers
One of the most durable songs from the 1890s, is sung by one of the era’s most prolific baritones.  Myers recorded for many companies, including at least two he helped manage.  The bulk of his recordings, however, were made for Columbia.  Significantly, perhaps, this popular title wasn’t one of them.  That distinction was reserved for Len Spencer, Columbia’s leading baritone in 1898, when this cylinder was likely recorded.

That’s How the Ragtime Dance is Done, George F. Williams
Williams was a local singer and his Michigan Electric cylinders are his only found performances.  The song is an early, obscure composition by Harry von Tilzer, one of the most successful and prolific composers and songwriters of the Tin Pan Alley era.  Coincidentally, von Tilzer was born in Detroit in 1872 (né Aaron Gumbinsky).

The refrain is:

First you do the rag, then you Bombershay
Do the side-step, dip, then you go the other way,
Shoot along the line with a Pas-Ma-La,
Back, back, back don’t you go too far!

The Bombershay (also called the Fanny Bump) and the Pas-Ma-La are types of ragtime dances.

The primitive nature of Michigan Electric’s studio is revealed by the uneven speed of the recording machine, most noticeable at :35.

Chimes of Trinity
This sentimental ballad about New York City’s teeming populace and one of its beloved landmarks is beautifully sung by Edward M. Favor, a popular stage and recording star from the 1890s.  Favor never performed the song for another label.  Perhaps this version of it was made while a touring show took him to Detroit.

>2000 transfers and counting…

CPS1 serial number 002 just celebrated its third birthday at the University of California, Santa Barbara (UCSB), where it lives at their Cylinder Audio Archive – part of the Special Collections at UCSB’s Davidson Library.

About six months after CPS1 arrived, Stewart Engart – a third-year Ph.D. Candidate in Musical Composition – joined the Archive as Cylinder Audio Engineer.  Since then, he has transferred over 2,000 cylinders on CPS1 using our custom cartridges.  He is likely the most experienced end user of our technologies.

Stewart’s a great guy and, as you will see, he’s doing interesting things, too.  His work with CPS1 was recently featured on UCSB’s library website.  The posting includes two videos.  What stands out is his affection for these early recordings, for the cylinder format and for our technology.

https://www.library.ucsb.edu/news/student-spotlight-digitizing-cylinder-audio-archive

Notice Stewart’s comments about the use of cylinder audio – by himself as well as other modern musicians – as content in contemporary music compositions.  It’s a similar approach to artists who create installations from discarded materials.

What would Edward Issler make of all this?  What would he make of Stewart?

 

 

The Zenith of Cylinder Recording

After all the records and hours spent listening to cylinders, I still sometimes find myself in awe of the sound buried in their microscopic hills and dales.  The cylinders that most likely drop my jaw are early direct recordings on brown wax and Edison 4-minute wax Amberols that escaped the destruction of period reproducers.

The following is a case in point.  It’s Victor Herbert’s famous Badinage, recorded in 1910.  While many consider it little more than fluffy “salon music,” it comes alive here, thanks to the master’s touch and the quality of Edison’s recording technology.

 

  • Note the authentic and forward sound of each instrument relative to their counterparts on 1910 78s.
  • Note the dynamic range of the recording from delicate passage to thundering.
  • Note the subtle interplay of strings and celeste and how vTrace resolves and separates them.

 

 

vTrace Bakeoff – Part II

The rubber meets the road

Cartridges, like cars, have suspensions that dramatically affect performance.  Like car tires, cartridge styli need to maintain tight surface contact to deliver the best performance – even when the surface is uneven.

In the case of cars, the surface (road) varies with potholes, speed bumps and road dips. It’s the same with cylinders: there are media eccentricities, surface bumps, axial distortion, and more.  Car suspensions are tuned to absorb road surface variations.  But, conventional LP cartridges are not engineered for cylinder issues.

vTrace series cartridges have suspensions tuned for their specific cylinder format.  The following sound clips demonstrate.

We used a typical out-of-round celluloid record (Edison 28151), transferred dry (no filters or EQ).  The only adjustment is overall gain to facilitate comparison.

  • First we played the cylinder with a Shure M44 cartridge at 4.00 grams – the minimum force required to track it. The thumps you hear are the cartridge hopping over the out-of-round surface.  Its suspension can’t absorb the eccentricity so it gets transduced into a low-frequency sound.  In fact, the stylus actually leaves the groove for a fractional second every time it hits this spot.

 

  • Now listen to the same segment with vTrace, tracking at 2.15 grams. The audio is a bit “watery” as the eccentricity affects pitch.  But the improved tracking of our custom suspension is obvious.

 

  • Now compare the sonic quality of this clip with the Shure. Notice how the contralto’s voice and the instruments sound natural with vTrace and the Shure cartridge clip sounds grainy.

Bottom line: putting a different stylus on an LP cartridge to play cylinders is like putting high performance tires on a truck.  There are many differences between the two formats and vTrace is designed to address them.

vTrace Bakeoff – Part I

Hearing IS Believing

vTrace is the only cartridge specifically engineered for cylinder playback.  All cartridges used by others to play cylinders were designed to play LPs.

The question is: does it matter?  Does vTrace sound better?

This video/spectrograph answers these questions.  It’s a short clip from a 1910 4M wax cylinder (Ed 17077) played first with a conventional LP cartridge (Shure M44).  Then the same clip is played by vTrace.  Both are dry, unfiltered, unequalized transfers.  The only adjustment is overall gain to facilitate comparison.  The spectrograph uses orange to show sonic energy across the frequency spectrum.  The louder the sound, the brighter the orange at its frequency(ies).

 

To evaluate this clip, you need to use good speakers or headphones.  Careful listening will reveal important differences:

  • Compare the horn passages at :07 and :41. The Shure cartridge adds audible distortion: they sound grainy and the trumpet peaks are clipped. The vTrace clip has better instrument definition.  The horns sound like horns; the reeds sound like reeds.   You can see on the spectrograph that the audio is less distorted.
  • Compare the portion of the spectrograph to the left of both these passages. Notice that the image at :07 is more orange than the one at :41.  This is noise added by the imperfect extraction (L-R) of the cartridge.  vTrace doesn’t extract; it plays the groove directly.  (For techies, the overall noise improvement is about 3db.)
  • Look at the bottom of the spectrograph (under 150 Hz) of each clip. The two orange bars are the low frequency hum of the recording machine motor.  The space between the bars looks like groove noise and, clearly, vTrace is less noisy.  The difference is not groove noise.  It’s the noise generated by the cartridge –  again, the product of imperfect extraction and mismatches between cartridge and record.

No one is saying that cylinders are a full fidelity medium but sonically accurate transfer is always the goal regardless of the source.

CPS1 LISTED ON “PHONOGRAPH MAKERS” WEBSITE

Over the decades, I’ve been fascinated by the various facets of early sound recording that intrigue and attract us today.  Some collect phonographs while others create record archives.  Some collect needle tins while others – like me – focus on playback technology.  But Christer Hamp is in a category all by himself.

His interest in cylinder playback began decades ago while working in the huge record archive at the Swedish Broadcasting Corporation.  At the time, he was exposed to the many formats of discs and cylinders that have been developed for recording.  For cylinder transfers, the archive was using the player produced by sound engineer, Art Shifrin, in the 1980’s.  The experience inspired him to try his hand at building a player himself.

In 1998, Christer decided to write up his building experience and publish it on the Internet.  Then, he started researching other people’s efforts to build players and published them as well.  20 years later, his website is unique.  The Phonograph Makers’ Pages is the only resource with comprehensive information on the dozens of efforts to play cylinders using modern equipment.  Yet, despite Christer’s important contribution to advancing cylinder playback technology, he doesn’t own any cylinders or phonographs himself.  In fact, Christer is best described as a collector of inventors – specifically, those who have focused on unlocking the sonic secrets of this esoteric recording format.

Christer and I started corresponding when CPS1 went into production 2 years ago.  Here are his notes on my technology, including the new vTrace series:

http://christerhamp.se/phono/levin.html

 

MORE KUDOS FOR CPS1!

Brenda Nelson-Strauss at Indiana University just gave Charles A. Asbury – 4 Banjo Songs an excellent review that acknowledges CPS1’s contribution to its sonic quality.  The review appears in “Black Grooves,” a music review site on the Archives of African American Music & Culture (AAAMC) website.  AAAMC is the number one hit on Google searches for “African American music” and a major destination for scholars and students of black music.

http://blackgrooves.org/charles-a-asbury-4-banjo-songs-1891-1897/