DD Turntables

The goal of the following experiments is to develop a methodology for the in-home measurement of the basic mechanical distortions, like ones originated by motor's rumble, in order to study the turntable - plinth relations. I know that the right way to do it is to use a special precision testing equipment, like a "THORENS Rumpelmesskoppler" pictured here: http://www.flickr.com/photos/b.....363924053/, in the proper dedicated lab. But still, one could try.

This study was originated by the need to choose the best plinth construction for my new Technics SP-10MKII turntable. Many particular decisions were results of the discussion held at Vinylengine forum. After I started, I decided to test a few more direct drive turntables from my collection. Why direct drive only? For the belt tables the motor is much smaller, so there is no problem to insulate it to the any desirable level (at least theoretically). DD turntables have the source of vibrations in the heart of the table, so the problem of the motor immediately became a problem of the table as a whole. And the better (more torque) turntable, the more severe the problem is.
What do I expect to find? Manufacturer's specification for the Technics' SP-10MKII motor unit rumble is 70dB, when for the same motor in the SH-10 B3 plinth it is 78 dB. So I would like to see those 8 dB on the -80 dB level.

Disclaimer: I am not in the sales business, it’s all about entertainment and to pursue the knowledge

Testing DD Turntables. Round Two.

Testing configurations

Part I. Motor without plinth. Arm in the pod standing alone. Configuration Comments
 - Setup A. Motor and tonearm on the table.
 
I've used a few arms (Technics EPA 100. SME III, Signet XK-35) with a few MM and MC cartridges (Shure V15-III, Pioneer PC-401, Grado Ref Platinum), but the most of the tests were done using three Stanton 681 EEE cartridges mounted on the three arms.
- Setup B. Marble base under tonearm, rubber spheres under SP-10. Cartridge is lowered on the motor unit and needle is sitting on the aluminum block of 1"x1"x3" size.
 - Setup C. 60 lb granite slab on the table. Granit slab

Granite 30" x 20" x 1" about 60lb.

Underneath of the slab there are four rubber feet 1" high

Part II. Original Technics obsidian plinth SP-10B3, EPA-100 tonearm SL-1000 Turntable was placed on the granite slab without any particular purpose. It is simply too heavy to remove it from the table each time I did a test.
Part III. My budget “universal” plinth SL-1000  
Part IV. Modified Technics SL-1200SP2    
Part V. JVC Q50    

Software and settings

Initial tests done in the first round were inconsistent and not informative. The next logical step was to improve the resolution of the spectrograms in order to identify the key areas of the future tests.
First, I've decided to change a PC used for testing, and after a few tries found (in my house) the one with the noise floor below -100dB.
Second, now is a time to pay more attention to the software, so I've tried different setting and compared results from Audacity with spectra obtained using Spectrum Lab V2.75 b09 program.
Some details on procedure are posted here.

As the result, for the following turntable tests I’ve decided to use Spectrum Lab V2.75 b09 program, with following settings:
Sound card (recording sample) parameters: Sample rate 11K, bit-depth 16
FFT processing: FFT input size 65536

Testing results

Background for all future tests: the noise floor of the sound card itself (no signal on the sound card line input).

 

 

Fig. 1. Noise floor

 

 

If you average fluctuations, you will receive pretty much flat (within 3 - 4 dB) graph in the region from 4 to 1000 Hz all under -110 dB level.

 

Calibration

 

Cartridge Stanton 681 EEE has output valtage 3.5 mV. Test record provided two options: 1 kHz recorded at 7 cm/sec and at 1 cm/sec (average recording level). To "normalize" pick signal to 0 dB I used 7 cm/sec signal adjusting volume control on preamp. On the fig you can see 1 kHz pick at -5 dB, that is a maximum amplification I could get from my sound card.

Results
Fig 2. Test record : 1 kHz recorded at 7 cm/sec
   
   
   
   
   

 

 

To be continued .............

 

About the Rumble

 

What to expect? Technics uses as they call it "20 pins 3 slot 3-phase DC Brushless motor MJX-12", but to make it easy for estiamtions, it's 15 coils on the stator which push circular magnet on the rotor. At 33/60 rotational speed we can expect at least 15 pushes per one turn, or 33/60x15 = 8.25 Hz base frequency and multiples of that number. Rumble spectrum provided in the turntable's manual looks like:

rumble

Picks around 8.25 x 3 = 24.75 Hz and 8.25x4 = 33 Hz on that graph look like those multiples