Hi, BM2CAJ

Sorry to reply you with late. You seems find your answer. Please be noted, most of the characters in the measurement screen is controllable, you just double click them, scroll middle mouse wheel or click right mouse key to work on them.




In addition, you can just pull the measured S-parameter file(S1P, S2P in Touchstone format) to the screen and it will be processed. By changing the unit, you can get the transferred parameter immediately between S/Y/Z....




With its powerful software feature, I collect the S-parameter to file with HP8753E and do the post processing over it sometimes.




ps. You can use the RFSim99(freeware) to generate the S-parameter for filter, attnuator..., then pull them to VNWA for manipulation.




73,
BX3AAQ

[ 本帖最後由 bx3aaq 於 2012-11-27 18:21 編輯 ]

BX3AAQ, OM,

謝謝回覆!

請問 Y軸的刻度如何讓它顯示出來? 謝謝!





73 de BM2CAJ

已經找到了, 謝謝!

Hi BM2CAJ,The VNWA program can have only two Y-axis label. Maybe you can request to DG8SAQ to have 4 Y-axis label  - left inner/outer line and right inner/outer line.
73,
de bx3aaq

Hi BX3AAQ

謝謝您的回答!

11月中測量 KVE-60B, KC901H, NWT500, 與 VNWA 3E 的輸出信號,
想看看如果做為信號產生器的效果, 發現 VNWA 3E 輸出超低, 在 -13 至 -20dBm 左右,
其他三個儀器都比較強, 在 1 到 10dBm 左右,
而且 VNWA 3E 諧波似乎超多, 超不規則, 我很納悶!
如果 VNWA 3E 沒有中頻 (看說明書, 似乎沒有) 那為什麼會準?
不知你有沒有測過?

VNWA 3E 輸出 10MHz 約 -13dBm  頻率與示波器差 100Hz


有 30MHz -30dBm 的三次諧波, -60dBm 8.363636MHz 的怪諧波



VNWA 3E 輸出 50MHz 約 -20dBm  波形不漂亮, 示波器 Auto 抓不太到...


有 30幾MHz -60dBm 的怪諧波



VNWA 3E 輸出 100MHz 以上, 示波器已經抓不到了,
用 KC901H 看頻譜, 感覺很亂, 很多非整數倍諧波...  (照片還沒整理)

不知為什麼會這樣?

謝謝!

73 de BM2CAJ

[ 本帖最後由 BM2CAJ 於 2012-12-19 00:51 編輯 ]

繼續

VNWA 3E 輸出 100MHz 時, 有幾乎等幅 10MHz 的低頻雜波


VNWA 3E 輸出 120MHz 時, 有幾乎等幅 -30dBm 30MHz 的低頻雜波


VNWA 3E 輸出 140MHz 時, 有幾乎等幅 -30dBm 50MHz 的低頻雜波



VNWA 3E 輸出 200MHz 時, 有幾乎等幅 110MHz 的低頻雜波


感覺奇怪? ? ?

現在再看, 都差 90MHz 耶! 不知是如何移頻的結果?

73 BM2CAJ

[ 本帖最後由 BM2CAJ 於 2012-12-19 01:05 編輯 ]

Dear BM2CAJ,
Some pictures included and explained


1. VNWA3 Specsheet

I highlight the information about signal generator. Its level is -17dBm(maximum) and adjustable to lower value. The other key information - No harmonic filtering, and I will explain below.


2. VNWA3 architecture

It use two DDS which refer to same crystal oscillator, ie the two DDS's output frequency is phase align together and that is the key point for keep the phase information in VNA. Two of the DDS can be named as RF DDS(Reference) and LO DDS. The RF DDS signal will deliver to DUT(Device Under Test), some signal will be reflected(Reference) and some signal will pass through the DUT(Thru). Now, there are three signal - Reference, Reflect and Thru. All of them will mixing with the LO DDS signal(the mixer is SA612, most used by the ham hobby) to audio band, then the audio band signals will transfer to TLV2462 OPAMP for amplifying and filtering(Sallen-Key BPF). In the last, an audio codec(PCM2900B) do the A-D conversion, and deliver them to the computer through USB interface(TUSB2036). The PCM2900 have only two analog input(L/R channel), so it need to mux the Reflect/Thru signal with a selective switch. If you buy the latest model, a second codec is installed, VNWA3 can get the Reflect/Thru signal same time, ie. the S11/S21 measurement is done same time.


3. DDS output spectrum



The DDS output stage is a DAC to transfer the digital data to analog waveform. The pulsed data will get a sinc frequency response over DAC output.
If we want to get a clean and flat spectrum over DDS output, the reconstruction filter should be installed for correcting the sinc distortion and removing the aliases.
Does these aliases useless ? not necessary. VNWA3 borrow them to extend its operation frequency range.


4. VNWA3 output spectrum

That is the most trick part of VNWA3, it use the idea of "Aliases mixing" to extend the VNWA3's operation range to 1.3GHz. How can you expect the DDS ICs to work over GHz band several years ago! In above picture, the reference clock is 300MHz(F_ref). In an un-filtered DDS output, you can see the fundamental signal in red color(F_fund=80MHz) and the aliases signal over N*F_ref +/- F_fund in blue color(300 +/- 80=220/380MHz, 600 +/- 80=520/680MHz...). That is why you see many strange signal over VNWA3's TX OUT as a signal generator. Does it impact to VNWA3's measurement accuracy ? No. After mixer downconversion, the TLV2462 BPF and additional digital filter over computer will select the target alias signal to process.


Now, we know these aliases signal is normal in VNWA3, but I can't calculate exactly how these frequency components you got. Maybe you can paste your instrument setting as follow, and I have more clue to count for them.



ps.
About the DDS(Direct Digital Synthesizer), it is much used in the modern transceiver. If you check the service manual of FT-897D, the AD9850 and AD9835 is used as the local oscillator and side-band selection. In the ICOM IC-7000, you can also find the AD9951 and AD9833 work as 1st and 3rd local oscillator.

The DDS's output frequency will limit to 1/2 reference clock in theory(1/3 in practical). For example, a reference clock 400MHz to DDS, its maximum output frequency is 200MHz. The primary clock is generated with crystal oscillator most of the time. The crystal oscillator seldom to operate higher than 100MHz. So the DDS will have xosc multiplier(AD9859 used in VNWA3 with multiplier ratio from 4x to 20x). A 37MHz xosc will generate 37*20=740MHz reference clock.

73,
de bx3aaq

[ 本帖最後由 bx3aaq 於 2012-12-21 02:42 編輯 ]

DR BX3AAQ,

感謝精彩詳細的回覆,

我想, 如果輸入接的 SA612 乘法器線性度佳, 混調變不多的話, 不用中頻應該可以, 這種狀況反而輸出位準不太高才比較好.
我的 setup 條件沒改, 是 auto



今天用 VNWA 3E 輸出 200MHz , 用 KC901H 再量頻譜

1MHz ~ 3GHz 頻譜, 看到 110MHz 與 200MHz 的 3, 5, 7, 9, 11, 等奇數倍頻
與原理推算一致



但 110MHz 與 200MHz 各自還有 106MHz 與 196MHz 的頻峰
懶得再去追根究底了...





TU 73 de BM2CAJ

[ 本帖最後由 BM2CAJ 於 2012-12-22 23:43 編輯 ]

S21 動態區間 (Dynamic Range) KC901H  VNWA 3E  NWT500 大評比

之前測雙工器, 三個儀器測出的曲線不大一樣,
這次用 " 固定衰減器 " 組合成 -20dB ~ -106dB 測三個儀器來個 PK 評比一下:

衰減器是 MACOM DC - 4GHz 1-2W  -40dB 二個, -20dB, 與 -6dB (剛好沒有 -10dB)
組合成 -20dB, -40dB, -46dB, -60dB, -66dB, -80dB, -100dB, 與 -106dB 測三個儀器的 S21曲線

測量頻率:
NWT500   50KHz - 505MHz
VNWA 3E  1KHz - 505MHz
KC901H   100KHz - 505MHz

NWT500





NWT500   50KHz - 505MHz




看來 NWT500 在 -60dB 以下就不大準了!
不過以它的價格而言, 我認為已算是性價比相當好的儀器了!


73

[ 本帖最後由 BM2CAJ 於 2013-1-25 00:29 編輯 ]

VNWA 3E


已做好校正 (Full Calibration, Master Calibration Activated)



VNWA 3E  1KHz - 505MHz



另我訝異的是 VNWA 3E 在 -60dB 以下也是不大準的!
而且在 300MHz 有一個不連續切換
大致上, 在 UHF 約有 -3dB 左右偏低
到了 -80dB 以下原則上是不行的! ! !
上圖是圖片軟體剪接, 所以 -60dB -66dB 顏色較淡, 另看下圖:


-60dB -66dB 不怎麼好



-80db -100dB 很差



很訝異...


73

[ 本帖最後由 BM2CAJ 於 2013-1-25 00:16 編輯 ]

KC901H   100KHz - 505MHz


0dB


-20dB


-40dB


-60dB



待續

-80dB


-100dB


-106dB


PK



三個儀器測出的曲線與數據看來,

KC901H 大獲全勝!

無論是 0dB 至 -106dB 動態區間的線性度,
或是 100KHz 至 505MHz 頻率區間的波動, 都是最好的!

以上是測試結果

73 de BM2CAJ

[ 本帖最後由 BM2CAJ 於 2013-1-25 01:11 編輯 ]

Dear BM2CAJ,
OK, I will take time to check the VNWA3 in my side. It using the alias mixing for working to 1.3GHz with some performance degrade in higher band. In your testing frequency range, I am also wonder why it not work so good.
If I can't find good explanation to the result after doing the experiment, I will pass my doubt to DG8SAQ and see his feedback.
ps. For the attenuation test, I will do the same experiment with HP8753E in my side.

73,
de bx3aaq

Dear BX3AAQ

Thank you very much for the comment.
But don't worry. I will not regret to buy VNWA 3E. It is still a very good instrument for me.
It has lower power output level -27dBm. I guess this will be much better for pre-amp measurement,
which will be the next step for me. I played these instruments a lot. For the next step,
I hope to find time to make and test pre-amps.

Thank you very much!

73 de BM2CAJ

左岸 BG8AAS 給了一些評論, 有興趣的 OM 可以看看:

http://www.hellocq.net/forum/read.php?tid=309797

8樓 ...

73