How to measure TT rumble?


#1

First of all, this is pretty awesome that there is an analog section!! Very, very cool. Thanks Paul!77_gif

So in keeping with all things AOCD (audiophile obsessive compulsive disorder - soon to be a DSM V diagnosis, I’m sure), I want to be able to measure the rumble in my TT against a standard. I’ve heard that the Feickert Adjust+ Pro disc has this capability. I have the non-pro version which is about half the price, more or less. I’ve done a little research online and haven’t found anything that looks easy or reliable. Does anyone have a more DIY way to reliably measure rumble?


#2

It would seem to me that a disc would be necessary and I have the HiFi News disc that has a blank space track. If you have access to an NPC or similar type product, then it would be relatively easy to record the setup without the needle in the groove and with it as the platter turns. Once you have a digital version then it’s probably a simple matter to use Audacity or any other such program to filter out everything but low frequencies and see the rumble component.

Just an idea.


#3

It’s funny that you mention that since I’ve done exactly that with the NPC. I get a curve but what does it mean? I guess I need to reference it to a 1kHz test tone at a certain dB level. My Feickert disc has a 1kHz tone at like -20 dB. Is there a standard for this?


#4

I am going from memory, but there is an unweighted and weighted average for the specification. The unweighted is 0Hz to something like 100Hz. The weighted is 30Hz to perhaps 300Hz. This would need to be referenced to a tone, typically at 1kHz, at a known level. I do not know how to do the math for the resulting rumble noise curve.

This would be fun to play with and I have the equipment to do the recording and graphing. But the result would be only amusing to me and otherwise pretty useless.

The Hoffman forum and other locations may have specific information.

Please follow up if you learn more.


#5

Alright, so I recorded using the Feickert Disc which has a 1 kHz tone which is actually at 0 dB. I adjusted the NPC gain such that it matched the measured input in Audacity at 0 dB, so I guess that’s calibrated correctly. I then recorded the tone at 1 kHz and in a silent groove on the disc. I also monitored the input gain with NPC on and the motor off and the NPC on and with the motor on but stylus up. Here’s what I got:

NPC on with motor off: -68 dB

NPC on with motor on with stylus up: -59 dB

NPC on with silent groove (using Nordost Sort Kones): -52 dB on the meter but on analyzing I get a peak of -60 at 3 Hz and -62 at 9 Hz. If I were to average the peak for 0-100 Hz (unweighted), I get -71.5 Hz by averaging measurements in 10 Hz increments. For 30-300 Hz (weighted), I’d guess it’s around -78 dB

NPC on with silent groove (using Fern and Roby sorbethane footers): -52 dB on the meter but on analyzing I get a peak of -55 at 4 Hz and -67 at 18 Hz. If I were to average the peak for 0-100 Hz (unweighted), I get -74.4 Hz by averaging measurements in 10 Hz increments. For 30-300 Hz (weighted), I’d guess it’s around -84 dB

NPC on with 1 kHz: 0 on the meter and 0 on peak at 1 kHz.

So what does all this say? I’m not sure since I don’t have a frame of reference! On Hoffman forum several guys say that a really, really good TT system can achieve in the -60 dB range using this technique. For this really to be done correctly it must be done with a Thorens Rumplemesskoppler which is a device which attaches to the spindle and has an arm with a flat spot for the stylus. Obviously I don’t have one of those, but it seems like my numbers are at least pretty good? I don’t know. I did find it interesting the difference between the measurements in the silent groove using the Nordost Sort Kones and the F&R sorbethane footers which allow the spikes in the TT to insert onto a plastic receptacle under which sorbethane is sandwiched between another layer of plastic. The sorbethane had a higher peak dB at very low frequency but significantly lower average dB lever from both 0-100 and 30-300 Hz, which also bears out in the graph. So is the sorbethane footer better? They both sound similar to me, honestly, but I haven’t done much critical listening to compare the two. The Sort Kones are designed to drain mechanical energy away from the component which the sorbethane is more of an isolating material to shield from outside vibrations. I doubt too many outside vibrations were going on since my room was quiet during these tests. Maybe the sorbethane footers are superior by this testing? More testing is needed!!


#6

Or @amgradmd

you could accept that your tt is operating within expected specifications, as would most modern tt (you haven’t mentioned what tt you are using).

If you cannot hear ‘rumble’ - which was only really an issue 20+ years ago - the measurements are only of passing interest. I think you might be going down a self imposed rabbit hole.

Good luck.


#7

@frank7036 Of course I’m going down a rabbit hole!! In all seriousness, my TT is a Fern & Roby Montrose. Fern & Roby are a local outfit in Richmond and I’ve become friendly with one of the designers of the TT, Christopher Hildebrand. He has asked me to help in any way I can in determining the weaknesses of the table from an audiophile’s perspective so I’ve spent some time and effort in doing just that. Trying to figure out how this table performs in terms of rumble is just my way of doing due diligence. Plus it’s fun!


#8

Seems to me from what I’ve read so far that your table has a reasonable rumble level. I am no expert on tables and don’t know what to expect. Is it belt driven? My Clear Audio table has a completely separated motor arrangement and a belt to spin the platter. The belt no doubt isolates the motor noise from its problems but to what degree I don’t know.


#9

@amgradmd BTW, if you wouldn’t mind, maybe head over to the Analog forum on Hoffman’s forum and invite some of those folks to chime in here. Maybe they’ll like the weather on this forum. 105_gif


#10

Yeah, I think it’s functioning well within spec. It is a belt driven table abut the motor is attached to the plinth, albeit isolated by dampening etc. I would imagine that any time there is continuous mechanical attachment of the motor to the plinth that there is the possibility of increased rumble/motor noise etc.

And Paul, I’ll be sure to drop a hint over at Hoffman. Very delicately, of course!


#11

Thanks, and delicately is good. They seem a rather volatile crowd.


#12

Unfortunately, most enthusiast forums attract and cultivate aggressive interaction. I do not understand the appeal.

OTOH, I am curious as to how rumble figures are measured and calculated. I understand it is a measurement of the turntable’s inherent noise, but what is it exactly?


#13

It generally is low frequency noise from the motor and/or bearing. Often times it’s inaudible, it’s so low. Like Frank said, it was much more prevalent 20+ years ago before TT technology significantly improved. I think one of the issues is that it degrades SQ, even if inaudible, because of “woofer pumping” thereby affecting the speaker’s ability to accurately reproduce higher frequencies. Hence the inclusion of high pass filters in many phono preamps above 15-20 Hz.


#14

I’m sorry; I was unclear. What I am wondering about is how the measurement/specification is defined.

I remember watching a subwoofer flapping without the high pass filter engaged.


#15

Me Too!