Hi Karas,

L1 and L2 forms the output transformer. K means coupling and 1 means 100% coupling between the inductors. This is the representation used by LTspice.

Please excuse me that I am too lazy to draw the tranformer symbol properly.

Johnny

Thx Johnny
What is this method called? It has no OPT in this amplifier?
＞＞The L1 and L2 forms the output transformer with the command "K L1 L2 1".
L3 is the plate choke for the 5842 tube.

Hello Niceday,
其實我冇300b，得px25, Px4，我只是用mh4 or 6j5做放大
px4我就未砌過，所以想做直結試試

The #5109 schematic use also 1K 300B cathode resistor utilizing the same two individual voltage supply, also direct couple by means of plate choke load in driver stage. I think Johnny was referring #5144 schematic using 3K PX25 cathode resistor.

Hi Kara's, 你可能需要知道你砌部300B SE要有幾多倍电壓增益先. Johnny #5145 線路電壓放大有10倍，你#5144 大約得7倍，呢兩部只可以做後級，你接部前級入去一樣要多一層交連元件例如交連电容或交連牛

而下图是3級直接交連有50倍以上可以做合併機，等於整個放大除咗輸出牛外全部放大級是直結



最後修改時間: 2025-07-08 12:49:07

Hi Karas,

Thank you for your question. The L1 and L2 forms the output transformer with the command "K L1 L2 1".
L3 is the plate choke for the 5842 tube.

Here is the schematic with more voltage readings.

Please note the circuit requires two power supply of 150Vdc and 430Vdc in series. Yes, this is better than the one #5109. The cathode resistor of the 300B is 1K ohm instead of 3K. That means the heat dissipation is much less. Also, any changes with the 5842 driver would not upset the bias of the 300B. Then, the 5842 tube can be tuned for minimum distortion as required.


Johnny



最後修改時間: 2025-07-08 05:05:45

#5137
HI Johnny
L2 is opt？
L1,L3 must be air gap choke?
how much is the cathode voltage of 300b?
看來直連好多voltage divider, 要學學用lt spice了....

我仲消化緊#5109張圖，亦同呢款又有點相似
獨立一組b+ for driver tube is better?

我嘅經驗係只用grid choke會出現bass boost情況. 低音量大增,令到我的機出太多低音. tonality吾太平？ (當時使用6SL7 mu-stage, cap couple to 300B. grid choke使用在300B grid)

當用Rg和grid choke並在一起,bass boost情況消失. 聲音同冇加grid choke差不多. 文中話Rg可以damp overshoot, 可能係指low frequency overshoot. 至於grid choke的好處是支膽被over-driven嘅時候, bias保持stable. 在我的機上感覺吾到這好處,可能我機設計上吾會over-drive支300B.

以上是多年前的diy經驗. 自己做實驗找結果正是diy的樂趣.

最後修改時間: 2025-07-07 13:44:17

我是用AI,幫手獲取圖片的文字。這樣就不需要打字。發現有誤差。


原文是這個。

Rg 用來damp overshoot,

如果choke 的質量好。可以不用Rg, 前提係沒有overshoot

https://www.jacmusic.com/lundahl/html/grid.html



最後修改時間: 2025-07-07 13:00:14

#5139

".........Note1) The input impedance at the RCA connector must be defined. This was the old DIN standard, at 47K by the LL170 choke, and this is still acceptable. If one wants 47K, this can be used as well. Without resistor, the LL170 coil has 4K impedance, which is too low. With a 47K resistor, the input impedance is 47K, as is normal. The real value is not critical. It is only important to have a resistor, and 47K is a standard value. 100K or any lower value......"

點解佢話 "...without resistor, the LL170 coil has 4K impedance, which is too low...."? (應該係LL1670)

Grid choke 應該有高AC resistance, 點解得4K impedance咁少嘅?

佢又話 "...With a 47K resistor, the input impedance is 47K, as is normal. ...."
47K resistor並4k impedance嘅choke, 得到input impedance 47K?

Grid choke 接在輸出膽。 gride choke 是可以解決信號過大overload 之後的 Grid blocking 的問題。



最後修改時間: 2025-07-07 07:30:42

5130-----用隻choke並落支Rg. 我覺得後者接法聽落同冇加choke差不多＞＞

查到網上的資料：
Standard Auto Bias Circuit
This is the standard Auto Bias Circuit. The grid choke, like all chokes, has low DC resistance, and high AC resistance. The AC is blocked from the grid circuit, so the grid is at DC ground via the choke. The bias is a very stable, predictable bias. If the input signal is excessive, the grid current will flow to ground at the top of the grid, so any music current from the grid will not affect the bias voltage. The problem of bias shift due to input overload is strongly reduced, as the grid is always at zero Volt DC.

Note1) The input impedance at the RCA connector must be defined. This was the old DIN standard, at 47K by the LL170 choke, and this is still acceptable. If one wants 47K, this can be used as well. Without resistor, the LL170 coil has 4K impedance, which is too low. With a 47K resistor, the input impedance is 47K, as is normal. The real value is not critical. It is only important to have a resistor, and 47K is a standard value. 100K or any lower value.

Note2) 47K was only choosen to match crystal record player to Phono input. Today, with MC and MM cartridges, 47K is not the best load any more. However 47K is rather sensitive, and 10K will reduce the input sensitivity by half, which is very significant, whereas 100K will only reduce it by 1dB or so.

Normal application for ultra stable bias

#5135


Grid choke 和 Anode choke 是不能共用的。

Anode choke 要過"直流"的， 有air gap

Hi Karas,

Using plate choke for the driving tube and NO grid choke required for the power tube in direct coupled SE amp would be just great. I worked on the schematic some time ago.

It requires one 150Vdc and 430Vdc power supplies.. The circuit is really simple and fail safe.

Johnny





最後修改時間: 2025-07-07 03:08:03

2 從接在柵極的膽去看。

之前的日本diy機都用在grid choke，會發出輕微音樂聲
這隻choke可否用在plate choke做直交？





最後修改時間: 2025-07-07 00:00:58

後級交連牛我己買齊(Tango NC20和ISO NC20F),不會用Grid choke/Plate choke/Cathode choke了,前級打算買YS-Audio的牛入牛出前級,因見到有用交連電容所以打算加Grid choke/Plate choke/Cathode choke(不知道它的線路),因未有地方放未買住不好意思打擾店家故上來問.

綠色代表。AC load line

紅色代表。DC load line

load line 越是水平斜， 就失真越低， 例如choke load 和 恆流源。基本上就是水平。

在這些ac和DC。兩條線。DC load line 失真更低。因為更接近水平的斜率。


由於Grid choke 交流的阻抗非常之高。所以DC load line 和 AC load line 的斜率基本上是一致。所以失真更低。對比柵極電阻。推動膽去推Grid choke 比推柵極電阻更加輕鬆。


例如 12AX7 的交流loading 是301k並470k=約180K, 但如果用grid choke 替代470k, loading 就變成約301K, 這樣的話12AX7 就會推得更加輕鬆。




最後修改時間: 2025-07-06 23:32:46

DC load line 主要由 R4 組成 301K

而 AC load line 是 R4 同 R5 並聯組成。

Grid choke

初步， 電學上可以從兩方面去分析。如果有遺漏。就請指正。

1 從 AC load line， 對前一級的影響。
2 從 膽自身去看。

小弟玩過的MagneQuest grid choke.....

一般用法是代替grid leak resistor, 但也有人用隻choke並落支Rg. 我覺得後者接法聽落同冇加choke差不多.

用手頭的Fluke 75已度不到任何微量的殘留交流電壓。另一方面，如果是用317或338等等的穩壓IC來取代以上的LC濾波的話，我相信一樣可以好靜，不過我亦相信201A陰極的表現無可避免地或多或少被劣化了。
加料
今年5月底，在市面上見到一些電感數值非常大的扼流圈，全部來自拆機，全部都是美國造，全部都是500H，數值大得驚人。這麼大的電感值，憑腦袋的有限無線電常識來判斷，這些一定是聲頻扼流圈，而非射頻扼流圈，所以必定適合音響器材用。另一方面，這些扼流圈所標示可行的直流均細得可憐，有些標0.05mA，另一些則標0.06mA。基於其大H和細mA，便可以斷定這些choke一般是用作柵負載，取代柵漏電阻。
這些柵負載扼流圈，多年來我還是第一次見到，當然機不可失。自己有限的鋁機常識告訴我，用choke一定靚過用電阻，如是者用choke作柵負載一定靚過落柵漏電阻。因此，就像92年買Mullard ECC 32般，買完之後，心思思，又再去掃過。這些500H choke，一共去掃了三次，共十六隻，包括四隻Stromberg Carlson、四隻Radio Receptor、四隻Freed、兩隻Kellogg和兩隻Federal。以上大部份都是四隻一款，原因是那四件頭並聯推挽KT66的推動級需要落四隻（上、下半波各一隻，因此左右聲道便一共四隻）。同一道理，功率管亦要落四隻500H。
執筆前的一個星期六（執筆此刻為7月6日），趁家裡無人，便替500H扼流圈開聲。用在前級的影響力比用在後級為大，這是音響常識，因此便決定將500H扼流圈用落今次的牛前級。將201A那枝680K柵漏電阻錫掉，然後飛線上盤面接500H扼流圈。今次揀了對Freed, 500H-0.06mA，放在201A和LS-10之間，飛好入和出，再用鱷魚線夾落殼作屏蔽。現時這500H的擺法純粹是作為試器材之用，估計將來很大機會是將500H兜底，藏在底盤內。
開聲之後，500H的表現力確是不同凡響，有實力。約九年初次提到GEC A2900之時，覺得A2900充滿音樂生命力，這隻500H同樣地是充滿音樂生命力，確實厲害。我下一步要做的，就是聽其餘四款500H，再下一步就是要決定把那一款500H兜底固定位置，來服侍201A。201A的柵極有了500H的照顧，今次這對牛前級當然是如虎添翼的了。