Measuring turntable performance.
Note: This page has been edited. Check back later for updates to methodology of the actual measure taking. Meantime this page can serve as a repository of known methods and techniques used in assessing the performance of the rotating platter and its drive mechanism.
A brief survey.
Back in the 1950's, 1960's and on into the 1970's turntable manufacturers routinely measured their products for rumble as well as for wow and flutter. These measures were published in both advertisements and product brochures and were then verified and re-tested in product reviews conducted by various audio magazines of those times. The reasons for testing the three parameters noted are easy enough to grasp.
Rumble, for instance, is the sum of the mechanical noises generated by the record player that does manage to make its way into the record groove / stylus interface. These noises accumulate from the vibrating drive motor, the platter bearing and any other associated drive train components. For instance in the case of an idler-driven turntable there is additional noise from the capstan/idler wheel interface. In a belt driven turntable there is a very minimal noise generated by the belt as it makes a stretch/contract style of contact with the driven platter. In a direct drive turntable mechanical noises are restricted to the bearing itself..... but other noises associated with the electromagnetic components do exist. The direct drive turntable, after-all combines the driving motor, the platter bearing and platter into the same component.
And this mechanical noise becomes part of our listening experience as it plays through our speakers along with the music signal. Generally, the lesser this noise, the more apparent all musical detail within the record groove. According to the Bruel @ Kjaer document; "Audible Effects of Mechanical Resonances in Turntables", when rumble figures measured are in the -65db range it is difficult to measure this due to the fact that other components within the audio signal chain, such as the phono cartridge and the preamplifier also produce a background noise at this level. And human hearing is usually not sensitive to these low (very quiet) levels.
Wow and flutter. These two parameters are used to describe rotational speed stability of the record player's spinning platter. When there is frequency modulation in the range of .5hz to 10hz (1/2 cycles per second to 10 cycles per second) it is called wow. When the frequency modulation is 10hz to 100hz it is called flutter. The human ear is very sensitive to wow. According to some studies, it is most sensitive to wow at modulations of around 4hz. Typically, wow and flutter levels are reported together and as a -- ±% -- from zero speed deviation. As an example, one study found that a wow/flutter performance of ±.06% was detectible to the human ear on a complex 5khz tone when the modulating frequency was 3hz.
Methods for measure:
Rumble. For rumble it used to be that an un-modulated track within a test record would be used by playing this track using the tonearm and cartridge on the record player. Then sending the resulting signal out of the cartridge via its phono cable into hardware such as a response test unit and then to a signal analyzer. These are very specialized equipment designed for the test. Over the years this method evolved. By the mid 1970's, progress in turntable bearing design and in new designs that more effectively decoupled drive motors from the chassis had resulted in dramatically lower rumble level performance. It had become evident that the turntable of that day was capable of producing less rumble than many of the mastering lathes that in years previous had been used to create the records being played. And as a result it was increasingly difficult to quantify these very low figures. For example consider a test record with an unmodulated track that contained more rumble within its grooves than the record player by itself would create. And so some new methods and devices were devised by various different parties.
One such device that came out of this period is the Thorens Rumplemesskoppler (rumble mass coupler)
A Thorens Rumplemesskoppler.
It was used in the following manner:
The device couples its base to the rotating spindle pin at the platter bearing by means of a reflex clamp, which attaches the jig firmly. With this device in place the turntable platter is operated at 33 1/3rd rpm with the phono cartridge stylus planted at the receiver as pictured. The signal is taken by means of the phono wire to analysis hardware where the signal is quantified. This method isolates the reading to that which is physically transmitted through the platter bearing of the turntable. Thorens, a prolific manufacturer of record players over the years, used this device to quantify the performance of its players afterward. Thorens did sell some of these devices to interested parties. And now we see them once in a while on the internet.
Here is another example of the Rumplemesskoppler in use that I have copied from a Japanese website: Text is in Japanese and English.
Weighting the measurements per different specifications. (more to come on this important component of the equation)
However, it has become popular in the last two decades to resurect and refurbish some of the older motor units from the early days of the "hi-fi" era. Idler driven turntables. These "motor units" were known to report rumble figures in the -30 to -50 db range. Some test records with unmodulated grooves might still be useful to measure and report rumble performance of a refurbished idler driven record player. When rumble charts start showing rumble figures in the negative 60 db range, and lower, it will be necessary to find a method alternate to the unmodulated groove. Also, belt driven players that indicate rumble at higher levels can still be assessed using the unmodulated groove of a test record.
Wow and Flutter measurements:
Typically wow and flutter have been measured using a test record track with a single tone reference signal and while using the record player's arm and cartridge. The signal, via the phono cable, is sent into more hardware; analysis equipment designed specific to the purpose and typically will report the amount of pitch variation over a specific time period.
There will be more to come on the wow and flutter testing.....I will soon acquire the 7 inch "platterspeed" record from Dr. Feickert in Germany. This test record emits a steady 3150hz test tone with which to analyze wow/flutter.
Acoustical and Mechanical Feedback. There is a description of method of measure for these parameters within the paper by Bruel & Kjaer. Link The methods described involve rather expensive equipment. There will be described less expensive methods and indicators that, to some degree, will discover the presence of the effects of acoustic feedback and mechanical feedback on the arm and cartridges as it plays within the listening room. Stay tuned. More to come.
Tonearm and Cartridge Resonance.
This can be measured using one of a few different Test records that have modulated tracks designed to test this parameter. For this at The Analog Dept. the test record "HFN001" is used.
More to come......
Software in use: Audacity (freeware)
hardware: a record player to be measured.
test record: any record that has an unmodulated groove (aka silent groove) to be used to assess evidence of rumble.
test record: with a steady pitch; say 3150 hz,
recording device. In this case I used a 2nd hand hard drive recorder. A 12 year old Alesis Masterlink 9600. These can be found on ebay at various low prices. Mine was purchased for less than $400 usd including shipping.
A blank CD-RW to record to. It is important to have a re-write type of recording disc for this operation.