2026年2月24日 星期二

Rebuiding MFA Luminescence

 A few years ago I got the chance to have a set of MFA Luminescence Preamplifier. I call it a set because it is a two-chassis preamp; one power supply and one amplifier. In the power supply chassis there are two tubes, one EF184 (6EJ7) and one 6L6GC. In the amplifier chassis there are ten tubes: four 5691(6SL7), two 6EM7, two 6SN7 and two 6DL7.

The circuitry is hand-wired on soldering frame. If you search the internet, you can see the internals of the Luminescence. It is a very "hand-made" preamp as far as I know; since my Luminescence is different from the internet pictures of the amp. The schematics online also have many versions, and they are different from mine. I don't think it was "modified" in the middle since some differences are big. However, their circuitries are pretty much the same.

The power supply chassis has four transfomers, three for heaters and one for high voltage. All the elctrolytic caps in my heater supply are gone. The markings are 10000uF and 15000uF, all dried out and need replacing. There is a big 6.3V zener for 5691 heaters, that is gone, too. The high volatge regulator is using EF184 as error amplifier and 6L6 as current pass device, it gives 310V output. Apperently the HV works fine, but there is another transistor regulatorin series after the tube regulator, This transistor regulator is using two resistors and two capacitors as volatge divider and just one transistor as pass device. I removed that. In the tube regulator there is a zener as reference voltage. The rating is 82V but it measures 42-43V. It is sitting under the cathod of EF184. I wonder the current for the zener is too low to turn-on the zener so it is not working as a reference voltage. Since the output has a VR to adjust the voltage, the output voltage is alright.

Checking the plate resistor of EF184, I found the current going through EF184 and zener is about 0.25mA. It is not enough the turn-on the zener. Although bigger plate resistor can help on the gain of the error amp (EF184), the zener beneath it is not working properly. I reduce the plate resistor to 300k and the zener voltage goes back to 82V. Of course, I adjusted the output voltage resistor to make it 310V as before.

I removed the big 6,3V zener for heater. It is a TO-3 device but mine doesn't work. The heater voltage was very low, around 5.1V. after adjusting the resistance in the heater circuitry, I raised the voltage to 5.9-6V.

The heater voltage is wired negatively, I should say it is -6V, The negative voltage is used as bias voltage supply for 5691(6SL7). I replaced the resistors and capacitors in the bias dividers, just for the peace of mind.

By the way, I don't have four 5691s for this monster. I am using 6SL7 and the heater current is half of 5691.

It works fine now. I am using it to enjoy the music. No hum, no buzz. I am very happy with the sound.

我的Model 2 DAC的I-V轉換器

 在我設計DAC的初期 選定了AD1955之後 就是要想I-V轉換怎麼做了 由於DA晶片的輸出是電流輸出 不是一般放大電路習慣的電壓輸出 中間得經過一個I-V轉換 也就是電流轉電壓的線路 這在DA晶片的規格書最後 都會有建議的電路設計 架構不外乎是用OP amp加上大量的負回授 以及補償電容 就可以做出這個功能

我試了這個參考電路後發現 它的音色取決於用甚麼OP amp 跟後面的真空管沒太大的關係了 這有點出乎我的意料 我本以為用了最"平直"的OP amp做I-V轉換後可以用比較"不平直"的真空管來調音色 沒想到是OP amp對音色的影響比較大 而真空管反而差別不大了

為了驗證這個結果 我用了很多種OP amp來試聲音 的確 OP amp有差 但都有一種晶體聲 比較生硬 老的OP amp會圓潤一點 但是再怎麼圓潤都不像真的音樂那樣滑順 後面的真空管負責電壓放大 雖然可以再把音色調整一下 但沒有辦法扭轉OP amp的聲底

這使得我必須放棄OP amp做I-V轉換器 得用更基礎的辦法解決 變壓器就是一個天生的電流轉電壓元件 一般變壓器是以初級線圈產生電流 經過磁場 讓次級線圈產生電壓 初級和次級的線圈圈數比就是電壓比 現在已經有了初級電流 我要設計變壓器反而變難了 因為我得從最大輸入電流和最大輸出電壓去考慮 還有輸入阻抗(電感量)去考慮低頻截止點 這些都不是一般變壓器的規格

還好經過黃國文師傅的幫忙 我們做出了DAC用的I-V轉換變壓器 用的鐵芯是烏克蘭買來的環形Amorphous鐵芯 電感量非常高 最後的做出來的低頻可以到20Hz -0.5dB

聲音更是一掃OP amp 的晶體聲 音色自然純淨 後面的真空管可以發揮應有的特質 換上不同的管子可以聽出不同的音色 這才是我想要的真空管器材 可以享受換管的樂趣