近乎完美的1000MHz信號源

hknightlife

終於calibrate好信號源的OCXO頻率基準, 誤差少於1E-8  (0.01ppm)
跟普通crystal既20ppm實在強好多.

以下係用9位既頻率計實測.  其實那個信號源有可能準到1E-9級, 不過我既無咁準既frequency counter, 所以無得再進一步調校.
如果有台1秒出12位既frequency counter就好了.

下載 (84.01 KB)

2010-7-29 23:18

morris

......

hknightlife

The frequency counter has Rubidium reference, while the signal generator has OCXO reference without any discipline.

Hard to use GPS disciplined reference at the apartment.

GPS disciplined?
morris 發表於 2010-7-30 00:38

hknightlife

The stability of the OCXO reference i used is rock solid. It can barely achieve 1E-9 class (use with absolute confidence in 1E-8 range over the normal temperture range).  The only drawback of OCXO, when compared with Rb, is the inferior aging rate.  Actually, many high quality OCXO has slightly better phase noise than Rb.

morris

......

electronic_s

玩咩要到咁準的CLOCK 呀??

hknightlife

High quality OCXO既aging rate係少於 1E-9 part per day. 1 year aging rate 少於 0.1ppm
假設你唔係長開, 你個中古OCXO基準drift左0.1ppm或以上都唔出奇.

我唔太喜歡GPS disciplined frequency reference, 因為屋企window收得唔好, 好多時都收唔足數既衛星, 有時下大雨就仲死.  攪RF野, 一般用到1E-10既Rubidium reference就夠有餘了, 因為我通常調到1E-9級準確度. Unless 你有台12位既counter, 就要攪個GPS disciplined來驗證.

我都有部中古頻率計，好似都係9位，用OCXO做基準。另外也有遠古訊號源，都應該係OCXO基準。但係兩部野readi ...
morris 發表於 2010-7-30 10:16

rutellor

雖然有0.01ppm
但有無相位噪聲的分佈圖看看?

hknightlife

1E-8級既OCXO唔係話D乜野好準既野. 好多通信設備既standard reference都係用佢. 攪microwave communication, 就起碼要OCXO. 若果要求aging rate低既話, 就通常用Rubidium reference.

1E-10級既Rubidium才算發燒級, 再上係GPS disciplined frequency reference, Caesium, and finally Hydrogen reference (好似去到1E-13)

玩咩要到咁準的CLOCK 呀??
electronic_s 發表於 2010-7-30 10:26

morris

......