- #4269 [johnnykmtang], 24-07-18 16:41
- #4268 [niceday009], 24-07-18 11:33若是全用Triode做部机,基於胆高电压特性有三种先天缺陷
1.級與級之間需使用coupling零件如牛或电容
2. 驅動級voltage swing及功率級要大比較難掌握高綫性(相对SS)
我諗到最精簡放大綫路為:
6SN7 兩sections 並聯 Active cascade MOSFET CCS loaded 3.8K 輸出阻抗
(*或用ECC99 兩sections 並聯輸出阻抗1.2K,唔好睇少這幾K分野,約把300B接fixed bias輸入阻抗由auto bias 250K变成50K, 雖然我用220K負偏压电阻可以令300B栅唔drift, 但此並非令300B在fixed bias 情況下有220K高栅阻,Wavelength老細曾揚言Auto cathode bias比fixed bias 更具備動力感就係因為此原因,我試過都係咁,如果要保持此出色動感,300B若用fixed bias,前面推動級屏阻最好2K左右,我曾用6CA7 triode strapped 900 ohm Z out推fixed bias 300B就係令動感出衆 ,300B fixed bias可以唔使用cathode bypass cap 純度明顯增加)
→300B fixed bias 接30H nano crystalline 1:1.5 分相牛,輸出阻抗約 1.5K
6SN7 及300B stage 只用LCLC 作choke input穩压,隻choke input 要120mA 15H DCR<50 ohm(好大隻),6SN7及300B全程A1動作用choke input 电压應該已好夠穩定
→3 cascade SS OCL power stage, MOSFET 1st → Bipolar 2nd → 5 Pair parallel Bipolar power stage,20W Class A, 200W Class B all with SS voltage regulator (好似頂班Krell 稳压制作)
全机只使用1個coupling 元件分相牛,而隻分相係有其主要用途作setp up 及分相作用
Voltage Amplification =20 x 3.3 x 1.5 = 99 倍完全滿整個放大要求
Target:
20W Class A, 200W class B等於90%以上係聽緊Class A声, 而voltage amp用Triode 單端設計可保持單端動作之獨特善性音質(或叫善性音染)
最後修改時間: 2024-07-18 12:32:14 - #4267 [WilliamsonAudio], 24-07-18 08:26#4264,#4266
我一直認為,膽石不分家,好聲就得。玩音響的人,膽和石也占了一定比例,並非一面倒。
電壓放大,最好元件是“三極膽”,當中直熱三極更好。例如Lynn Olson 及好多文章也提到。
電流放大,最好元件是“原子粒”
原子粒也是有優點的。低內阻的特性。膽是好難超越。
例如有一些MOSFET ,內阻低到 0.007 ohm. -
- #4266 [niceday009], 24-07-17 22:36My latest thought is too use soild state for current amplification, with this approach even need mush less voltage swing from the triode voltage amp, much less voltage swing stands for more linear
- #4265 [johnnykmtang], 24-07-17 21:26Hi Niceday009,
300A push pull. Simple and low distortion approach. The only issue for this antique amp would be the transformers which don't have the quality of today's.
Johnny - #4264 [niceday009], 24-07-17 18:22Found a good article
http://www.nutshellhifi.com/library/FindingCG.html
我特別喜愛佢咁形容
Triode operation is characterized by a low plate resistance (Rp), low to moderate gain (mu), medium to high Miller capacitance, and low distortion with rapid falloff of upper harmonics. Comparisons of triodes to other classes of device reveals that they have lowest distortion of any amplifying device ever made.
比較任何人類發明過之其他元件,三極管係最綫性最低失真(尤其是令人厭惡的高order失真) - #4263 [WilliamsonAudio], 24-07-17 09:47我第一次找到 300A 的電路圖。
古董 WE300B 我就聽過。
WE300A, 就還未有機會聽。 - #4262 [johnnykmtang], 24-07-17 03:31It has been some days no new message coming up. Most likely it is because nobody interested in putting out the tube amps during the HOT Summer Days.
- #4261 [johnnykmtang], 24-07-09 21:23I bought 2 pairs of LinLai 845 tubes. C$180 per pair with free shipping to Canada.
They are performing very well over a year now.
Johnny - #4260 [WilliamsonAudio], 24-07-09 19:55
- #4259 [niceday009], 24-07-03 00:45如果將nano crystalline 之磁滯参數(coercivity 矯頑力),nano crystalline 磁滯係少過普通矽鋼片(如Z11)33倍,40:1.2, 已經冇必要用hysteresis graph 去測試nano core, 我買呢間係西德大廠,佢地license 咗Hitachi 做Finemet product ,信廠方數字就是
最後修改時間: 2024-07-03 01:00:32 - #4258 [niceday009], 24-07-03 00:39typo
Lundahl 3 夾 2 x 2 , 兩個bobbin - #4257 [niceday009], 24-07-03 00:10設計及winding geometry, 用Teflon 及Kapton 紙,訂制Epoxy bobbin, 呢種winding 比lundahl 或Sowter 更細分綫組,Lundahl 1677 4夾2再X2(lundahl 用單C兩個bobbin) 20 to 34kHz, 我用5夾4, 20 to 44kHz
最後修改時間: 2024-07-03 00:38:53 - #4256 [niceday009], 24-07-02 19:00刚繞咗隻Nano Crystalline core interstage, 高頻响應唔錯,下面是25kHz方波(係25kHz唔係10kHz),大約去到40kHz, 27H, 27H俾300B load 低音可以好好
最後修改時間: 2024-07-02 19:06:42 - #4255 [johnnykmtang], 24-06-29 16:31This is my 6SN7 - KT66 full balance/differential push pull - the PURIST design.
The 6SN7 driver stage use the Lundahl LL1667 push pull plate choke and capacitor output to the KT66. This maximize the frequency response at both low and high ends. Also, it provides better balance and low distortion drive signal.
Complete constant current in both driver and power stages. Therefore, the power supply can be very simple one.
最後修改時間: 2024-06-29 16:35:57 - #4254 [WilliamsonAudio], 24-06-28 23:40佐久間的設計理念。分成幾點方便閱讀。
Design Concept
1 use the same (直熱膽)tube for driver and output.
2 Their tones are clear and powerful.
3 the driver's tone contributes the greater part of the amplifi-er's voice.
4 Few driver tubes can sing along in beautiful harmony with a good power tube.
他設計的 4212 poweramp, 當然也是用 4212
做推動膽。直熱推直熱,是“”全直熱設計理念“”。
我對佐久間的理解是。
在他的文章中,提到一個例子。就是WE300B, 當他用300B 推300B,就感受到真正西電300B 的聲音。因為整個信號就祇有300B,並沒有加上其他膽的聲音。
膽本身就有分層次。例如。小九腳膽,大8腳膽,直熱膽。五極,三極等。當然最好就是直熱三極。現代的設計主流都是應用高跨度膽和直熱三極膽。當中直熱三極更是貴價機中的主流。
由於推動膽。影響聲音的比例更加重要。所以在直熱膽前面用比直熱膽低級的膽,只會聽到驅動膽的失真和音色。down grade 了直熱輸出膽的純真度。所以好多時直熱膽機,我們是在聽前面推動膽的聲音。例如12AX7, 五極管。
想像一下。4212 用 12AX7 去推KT88 再推4212. 你就知道兩者之間的設計理念區別有多大。膽成本相差也好遠。
最後修改時間: 2024-06-28 23:44:40 - #4253 [niceday009], 24-06-28 00:11I also believe a good sounding amp must be equipped with reasonably low distortion but tuning to low distortion must not involve loop type NFB.
Wilson made a point that driving stage linearity before the DHT output power tube is crucial
最後修改時間: 2024-06-28 00:13:04 - #4252 [johnnykmtang], 24-06-27 21:1315 years ago, I built a single ended 45 tube amplifier for a friend with Sun Audio's schematic. I used expensive high quality components. After it was done, the amp was tested at my friend's place. He was so delighted with the wonderful sound from it. His wife came in and she asked if we can play an old Chinese CD album by Teresa Tang. After we played a song, she said it doesn't sound like Teresa. I just blamed it on the poor recording of the CD.
After I found my way in minimizing the distortion in my amplifiers, I played couple of the old Chinese CD albums including the ones by Teresa. I realized that the sound quality is real good indeed and I could recognize it is the real voice of Teresa that I familiar with.
I think this is a good way testing of the HiFi system with some old CD albums and see if it can reproduce the lively sound that you like and familiar with. - #4251 [johnnykmtang], 24-06-27 06:47So, how to determine the best idle current of the SE output power tube?
It is better to use the plate V-I curve graph, draw the load line, try out the lower distortion region you can get for say half power. Most of the time we just listen to 1 or 2 watts max (depending on the efficiency of your speaker).
Here is a good online program you may try.
https://trioda.com/tools/triode.html
My idea is getting low distortion at low power. High power??? No no.... I don't need that.
Johnny - #4250 [johnnykmtang], 24-06-27 06:41Regarding using 211 to drive 211, I believe the concept is providing ultra low distortion signal driving the output power tube. With use of smaller tubes to drive, usually its output signal has higher distortion than the large power tubes.
Therefore, if we can find a small tube with higher gain and low distortion, this would be the best solution. That's why I experimented using 6DJ8 differential PP IT drives the 300B SE.
Johnny
最後修改時間: 2024-06-27 06:41:59 - 下一頁 (1 of 214)
- 返回 ...
I would suggest the very last solid-stage can be built with separate chassis so that you may experiment with different ones later.
Johnny