This is one of the eternal arguments among vinylphiles. In belt drive world, we have the light platter/high torque motor crowd vs the heavy platter/low torque motor acolytes. Then we have the idler/DD aficionados who jeer (privately) at both. Modern methods of speed control have leveled the playing field somewhat. My dirty little secret is that I finally couldn’t stand the ritual of having to push start my otherwise perfectly good Nottingham TT and so was driven to experiment with idler and direct drive. (On one occasion I nearly knocked the Nottingham off its wall shelf with my overzealous push start.)
A Few Turntable Measurements using the RPM Android App
I found this Android phone app for TT rotation. Phone is Pixel 4a. Thought I'd try this app out. I'm skeptical of these phone apps. Accuracy is always an issue.
I have four tables. I took 5 readings for the first table in order to see what the repeatability is. The "absolute" RPM, RPM peak to peak, and 2 sigma range readings were very, very repeatable. Consequtive RPM readings differed by a max of 0.01 RPM. Two sigma varied by 0.01% ( 2 sigma means that 86% of the readings were within the stated value). I personally would use 3 sigma, but that's a personal quibble.
I've measured all four of my tables. I am very certain that the results are very repeatable. I measured with no LP, LP rotating, LP on and Stylus engaged, and phone offset from center. RPM was the same for all cases, The 2 sigma showed a 0.01% rise (really small). The reading at the edge of the LP was different. And scary to do!
Here's the results:
1. DD-40 #1, RPM = 33.32, 2 sigma = 0.07% (63 dB)
2. DD-40 #2, RPM = 33.27, 2 sigma = 0.09% (61 dB)
3. Acoustic Signature WOW XXL, RPM = 33.17, 2 sigma = 0.10% (60 dB). This varied 0.02% from reading to reading (after running the table for 10 minutes, this noise diminishes), but the 2 sigma stayed the same.
4. Denon DP-57L, RPM = 33.25, 2 sigma = 0.02% (74 dB).
I then went back to DD-40 #1. Using the RPM app, I set the mean speed to be 33.25. The strobe on the table was slowly moving! I checked against the strobe on the Cardas test LP and yes, the RPM speed accuracy was wrong. I reset TT speed using the strobe. The RPM app measured 33.23 again. I must conclude that although the RPM app is very repeatable, the absolute accuracy is not. The wow result (2 sigma variation) remains the same.
I measured the 45 RPM on DD-40 #1. RPM = 44.91, 2 sigma = 0.05%, so the 45 RPM is fairly accurate and the 2 sigma is lower.
This app makes no distinction between wow and flutter. It's all reported in the wow reading (wow and flutter are the same thing by nature, the only difference is the frequency range).
I'm surprised by the poor performance of the WOW XXL table. This a modern, belt driven table, with a massive platter. It is 5 years old. There's no way for the user to adjust the RPM. The variation in the speed is similar or slightly higher than the 40+ years old Micro Seiki DD-40 tables, which don't have crystal oscillator driven speed control. The WOW XXL takes about 10 minutes before the very high frequency variations settle. Now, I don't know much about the internal workings of the app. Helpful would be better accuracy (or the AC frequency in my house is not 60 Hz). Bandwidth is not reported.
The DP-57L performance is outstanding!. This TT was made in the 80s. And the DD-40 tables are not bad, but are as good as or better than the WOW XXL.
In summary, in my opinion, the RPM Android App is very useful. The absolute accuracy is a bit off, but the repeatability is very good The wow measurement is also quite good.