flavio81
Turntable technician
Where can I buy it? The homepage "buy it" button doesn't work.
New W&F Analyzer + Polar Plots
- Thread starter JP
- Start date
-
- Tags
- measurement turntable w&f
flavio81
Turntable technician
When you use a test record you are also measuring the flutter (or wow?) introduced by the interaction of the cartridge compliance, tonearm mass, tonearm counterweight suspension/damping, record surface evenness, and thus a needle producing variable friction on the groove on the drive system.
It is a system test and IMO the tonearm and cartridge should always be documented. That's how the B.A.S. (Boston audio society) did it in the 70s.
Last edited:
I sent A mail..
The value I use from Shaknspin is the 2s , the others can be strange, the most useful feature is the excel export where I can do this. A plot say so much more than a single value. A single DIN value can hide a lot of problems. Shaknspin makes it easy to collect data in a hifi shop or hifi event. Test Records can tell more but limited to a home setting.
Very nice, thanks for providing this. I saw your post in the ASR forum but haven't read through it. I have a test record with a 440Hz signal (standard tone "A") but I assume it will not work with this analyzer. What currently available test record would you recommend?My project this week was a web browser based W&F analyzer application. It complies with AES6-2008 / DIN 45507 / IEC 60386 / CCIR 409-2 standards. It runs completely within your web browser and processes whatever files you give it on your computer.
You can play with it here: W&F Analyzer
If you don't have your own W&F tracks recorded and want to play with it, this link will launch it with one of my test files.
Geek stuff here.
Thomas_AA
Well-Known Member
Cleaned out the bearing of my Linn Axis with sewing machine oil and replaced the Linn Black oil with the same sewing machine oil. It gave significant effects both on the spinning time of the inner platter, increased the speed and reduced W&F. Specifically reduced the 1 Hz drift peak. Now I will have to adjust speed on the trim pot, but that's the easy part.
Just an example of how useful this app is!
Before:
After:
Just an example of how useful this app is!
Before:
After:
flavio81
Turntable technician
As a suggestion, have you considered higher viscosity oil? Perhaps it could have a filtering effect on all the flutter noise we are seeing on the graph.
Belt compliance acts as a spring, and bearing lubrication should act as a damper for it, as far as I understand.
Thomas_AA
Well-Known Member
The 1.1 and 2.2 Hz is the record, stays the same. The problem is the LF drift, and the frequencies below 2.2 Hz (0.25, 0.5 and 1 Hz, which is belt related). If you have more friction you will most likely increase the varying tension of the belt. Flutter > 2.2 Hz is very low, IMO.
Have not tried yet but think low viscosity is the key here.
JP
7480 74111110101115
It has been tested for carrier frequencies from 50Hz to 5kHz - 440Hz will work, but your max bandwidth will be 175Hz.
No test record is great, and modern ones are generally worse. The value is more about the plots. You can figure out the error in a test record by trimming the signal capture carefully so you have an "index" point and then rotating the record 180deg on the platter. This will show what's from the record in the polar. No good way to separate the record contribution in the waveform or spectrum plots, though.
JP
7480 74111110101115
They're also 2x and 4x rotation harmonics. That 0.51/0.57 is odd - what does the histogram look like? Perhaps precession from the lighter oil?
Thomas_AA
Well-Known Member
The 0.50 and 0.55 (rotational) sometimes blend to one, since they are close. But I see 0.5 Hz as the belt creep or belt/platter resonance. Belt will creep and any inconsistency in belt every revolution show a slightly lower f than the platter rotational. Or it is the resonance f of belt/platter. Next experiment will be changing belt to see if it has any impact on these f.
Last edited:
Thomas_AA
Well-Known Member
If you see the sort of convex Linn pulley, it requires the belt to ride at exactly the same position not to change speed. It does not really stay exactly at the same position when looking close - there are tiny movements that comes ≈1 platter revolution. So it is not the pulley - it most likely is the belt that needs higher precision.
JP
7480 74111110101115
Thomas_AA
Well-Known Member
Too much work…If you have a model we can add belt effects to my motor calculator: Motor Frequency Calculator
The best performing TT I have in terms of W&F is a bottom-of-the-line Denon DP-15F.