Thursday, March 29, 2018

Tracking Tests on the ELP Laser Turntable

An ELP uses laser beams to pick up the signals (no digitization!) without physically touching the groove surface. One may think the distortion level would be much lower than using a stylus since there's no friction involved, and there would be no tracking issue at all. However, in reality there are other factors involved and the test results could be somewhat unexpected.

Using the eight tracking tests on the Hi-Fi News test record, the results are shown below. Some key observations:

1) Harmonic distortions, especially the 2nd and 3rd-order, are still quite visible on the spectrum analyzer, mostly similar to the case of using a regular stereo cartridge.

2) The strongest test signal in Test#4 leads to very high distortion, as can be seen in the deformed sine wave form. This level of distortion is in fact worse than some of the better tracking cartridges. But the good thing is, there's no skipping like in some cartridges' case.

3) From Test#3 to Test#4, the distortion level increases drastically, even though the difference in signal strength is only 2dB higher. Seems like somewhere between +16 to +18 dB, the threshold of the ELP's tracking capability has been crossed.

4) Tests#5, 6 and 7 corresponds to outer, middle and inner part of the record surface, respectively. The difference among them is smaller than most cartridges, showing the lack of side force interference is an advantage for the ELP.

5) Some unusual spikes in the spectrum are likely from dust particles on the record surface. The ELP cannot push away dirt in the groove like a stylus.


用 ELP 播放 Hi-Fi News 測試唱片的8個循軌測試軌,目前的觀察:

1) 播放時諧波失真也是普遍存在,所以這些失真的來源並非因唱針與溝槽摩擦而產生。觀察 TESTS#1到4,比起抗滑/側向力調整得好的唱頭/唱臂,失真率並沒有較低。

2)強度高的訊號如 TEST#4,ELP 也是會有嚴重失真,可見內部的雷射換能機構還是有遲緩的弱點。但優點是完全不會有跳針現象,而且左右波型都還相當平衡。

3)從 TEST#3 到 #4,失真的差異很大,猜測是剛好超跨過了 ELP 循軌能力的 threshold。

4)從 TESTS#5,6,與7看來,內外圈的差異相當小,較一般支點臂 consistent。

5)頻譜裡有些奇怪的雜訊,應該是唱片表面灰塵造成雷射的反應,判讀時應該將之忽略。



TEST#1

#2

#3

#4

#5

#6

#7

#8


Tuesday, March 27, 2018

How to Deal with Skewed Zenith?

Stereophile 的 Michael Fremer 在這文提到這個鮮少被人注意的參數:

http://www.analogplanet.com/content/stylus-zenith-angle-heretofore-neglected-set-parameter-considered#Pq2fSHWlKpbXrJBT.97

基本上,它關係到的是針尖與唱片接觸的那個「面」,與針桿之間是否為平行。一般用針桿來調循軌角時,都假設這不是問題。但實際上,並非總是如此。

目前我的看法是,如果沒有太歪,而且知道角度差多少,應該可透過循軌角的改變來做修正。從 Adjust+ 測量許多唱頭的數據來看,除非與標準值偏離得離譜,否則循軌角與zenith主要的影響,是在左右聲道相位延遲的差別。如果測出來左右聲道的延遲差異很大 (我見過差超過100度的名牌高價頭),那最好用顯微鏡看一下 zenith 是不是歪很多。

以下是我自己照的幾個唱頭的 zenith 。

這是圓錐形的針尖,所以不必考慮 SRA 或 zenith 的問題。





這是一個 Nagaoka 的唱頭。看起來 zenith 是很正。





這是我的 Lyra Delos,因為廠方用了太多粘膠來固定針尖,所以不容易看清。



這是一個老的 Lyra Skala,似乎是有點歪。



Wednesday, March 21, 2018

Intrinsic Advantages of Mono Cartridges: Low THD%

Two MC cartridges of the same model, one a mono and the other a stereo version, are measured here using the same 1000 Hz sine-wave test track. The mono version (1st chart below) shows a much lower THD% .

One can argue that, since the test track is in mono with only lateral modulation, any vertical movement of the stylus, which could be from imperfect record cutting, etc., is counted as distortion. A mono cartridge will pass on most if not all of the vertical movement, hence the measured distortion will be smaller than a stereo cartridge.

Mono 唱頭的好處之一,是THD% 測起來會比立體頭低很多。因為失真率是以 mono 的 1000Hz 訊號測出的,這代表立體頭在播放 mono 音軌時會付出較高失真的代價。所以,用 mono頭聽 mono 唱片,應該是有失真較低的優勢。
所有mono 唱片,不管是以前用老mono刻片頭或是近代立體刻片頭刻出來的,溝槽難免會有一些垂直向的立體運動成分。但既然原始訊號裡沒有立體資訊,這些都算是刻片的失真。用 mono 唱頭播放,感應不出這些不該有的失真,理論上會比較接近原來母帶,也是一種好處。
以下兩個圖,是同一型唱頭的 mono 與立體版本,裝在同一唱臂/唱盤。測量出的失真率。可以看到相當明顯的差異。



Monday, March 12, 2018

Why use X-ray to inspect a cartridge in its unopened package?

Why do we need to use X-ray to inspect a cartridge in its unopened package? Recently I've encountered several high-priced cartridges with very steep cantilevers, far greater than the 20-degree spec'ed in most user manuals. When communicating with the manufacturers, they either insist these are normal cases, or the anomalies have to be the results of user errors, asking for a service fee to make adjustments to the angle. If we can identify such issues before the box are even opened, then for sure we can say it's not from improper use of the product.

Many thanks to Alan C. Pan's willingness to test with the equipment in his clinic, and 黃崇斌's post processing! The results seem to indicate this may be a viable approach.

為何要用 X光來檢測未拆封唱頭的針桿角度?主要是因為,我最近看到好幾個還很新的高價唱頭,側面的針桿角度非常陡,遠超出多數廠方規格裡所標示的20度左右。這意味著絕大多數唱臂無法將VTA/SRA修正到正常工作的範圍。唱頭上緣水平時,SRA會到95度以上,測量IMD失真率也會很高。如果在還沒拆封之前就可以發現這問題,就能確認這完全不是使用者造成的,立刻可以退回要求廠方處理。

感謝 Alan C. Pan 願意嘗試用他診所的器材測試,還有黃崇斌先生的影像處理,下面的圖片顯示,這方法似乎是可行的。

Image may contain: drawing

Friday, March 9, 2018

Verifying AnalogMagik's Load Tuning Function

I was somewhat doubtful this would actually work before running this set of tests. In my prior experiments, I've found using a MC step-up transformer with various loading values in the phono stage would definitely impact the measured frequency responses. But I had been of the belief that, for the case of MC cartridges without using a step-up transformer, even though the impedance settings would impact the perception of overall tonal balance, the effects should be only in the harmonic compositions or even in the ultrasonic frequency range, presumably expecting no difference in normal frequency response measurements. However, the test results below quite clearly indicate the 500-ohm value indeed leads to the best "flatness" measurements in AnalogMajik. And in fact, this is the value I had found most agreeable with my ears before running the measurements.

I have run the tests using the Avid Pulsare II phono stage, with easily adjustable resistance settings of 10, 30, 100, 300, 500, 1k, 5k, 10k and 47k ohms. Only the most relevant data is shown below. All other values clearly result in inferior flatness data to the 500-ohm setting.

在還沒作實驗前,我也很懷疑這功能是否真的有用。以前的實驗數據顯示,MC唱頭搭配升壓器,不同的唱放阻抗或容抗值是會影響觀測到的頻率響應。但如果直接進入唱放,並未曾觀察到負載改變影響到頻響測量的例子。這裡是用了 Avid Pulsare II 唱放,10,30,100,300,500,1k, 5k, 10k 與 47k的阻抗負載,以及AnalogMagik 來觀察頻響曲線的平直程度。只列出3個最有意義的數據點,可以看到 500 歐姆的結果最佳,其它阻抗值包括未列出的都比較差。而這個阻抗設定,正是我之前用耳朵判定所得的最佳值。


500 ohms:

1000 ohms:

300 ohms:



Wednesday, March 7, 2018

Confirming anti-skating setting with AnalogMagik

I used to use the HiFi News Test Record with Visual Analyser to observe the THD% and set the anti-skating force. Without touching anything, I use AnalogMagik's anti-skating function to check if there's any major difference. Turns out the two approaches mostly agree with each other, showing almost identical THD% numbers in both channels. 


這是用AnalogMagik抗滑功能驗證我之前用 HiFi News 測試片及 Visual Analyser 調出的設定,算是非常符合。所以如果不想用先前跑8個測試的方法,用 AnalogMagik 來調整,也算是比較便捷的方案。

No automatic alt text available.

Sunday, March 4, 2018

How sensitive can the IMD numbers reflect the SRA changes?

How sensitive can the IMD numbers reflect the SRA changes? Just about 1/4 of an mm's adjustment on the Graham Phantom II, the resulting differences in measurements are quite obvious as shown below. The sound also becomes much more refined with darker background and lower perceived distortion.

IMD反應SRA的改變有多敏感?我長久未去變動設定的 Lyra Kleos 裝在 Graham Phantom II,很早以前就用顯微鏡調到SRA約92度。圖1是顯示調整後唱臂的高度微調刻度在7.75。而未動之前在9左右,所以大概是降低1/4個 mm,下方的粗刻度則是幾乎看不出差別。圖2是還沒動之前,用 AnalogMajik 測得的 IMD數據。圖3是微小調整唱臂高度後,測得的較佳IMD。這時,音質的細膩度與背景安靜度的確是得到相當的提升。


No automatic alt text available.


No automatic alt text available.
No automatic alt text available.