300b lovers

Hi Lynn,

I found online only one filament transformer for 5V - Hammond 546-166MS. This transformer has only one 5v 3A tap. So I need 2 such transformers. I use Hammond in my DIY phono stage. But IMHO Hammond transformers are built cheap compared to Lundahl, Hashimoto, AN or even James Audio. 
Is these Hammond good enough for this task?

I ask again this technical question again. Which rectifier is better: full wave or bridge?

My driver stage power transformer has three taps: 250 0 250 and supports both - full wave and bridge. Full wave affords use x2 times bigger DC current. Why do many people prefer a bridge rectifier over full wave?

Hi @lynn_olson ​​​​@atmasphere ,

Thank you a lot for detailed answers!

I have another question:

In balanced push-pull self bias output tubes is not a big issue because the cathode resistor and capacitor are out of the main signal path. But what to do in my case with a 300B SET. 

Now I use self bias with 50uF AN Kaisei NP bypassed by 4750uF Nichicon and two Ohmite Gold 10W resistors in parallel.

So the question is what is the best solution for SET? Fixed bias? Is fixed-bias reliable and safe? What is important to know for building a negative power supply for fixed bias?

I continued tuning my amplifier for the Hashimoto A-107 interstage transformer (IT).

I changed the load resistor from 57K to 39K Ohm. The difference in square wave overshoot is very small, but frequency response peak is just +0.5dB instead of +1.3dB.

https://photos.app.goo.gl/tdDuNM15cziP4o5g9

But the most important - the rough upper mid coloration was gone with 39K.

I also increased 6sn7 idle current from 7.5mA and 9mA. When it was RC coupling before the idle current was 6.2mA.

In general, sound became more refined and organic with much better low level details. Now in addition to advantages that IT gives versus RC that I talked about before (like faster transients, better more detailed rhythm reproduction and bass control). I got more low level details, much more realistic drums kick/punch reproduction, piano is more real, as well as timbales, better instrument separation (less congestion on complex music), and more real musicians presentation in my room. 

Next step moving from an old amplifier box + power supply external maquette to two box solution (stereo amplifier + power supply).

https://youtu.be/Hro6eYBvb68

Hi @atmasphere ,

The bottom end is predictable. ~8K 6sn7 (260v, 9mA) internal impedance and 70-80H transformer primary inductance gives -3db around 20Hz. Actually I measured -2.37dB at 20Hz and -3.62 at 15Hz.

According to plate voltage-current characteristics, 6sn7 has very small distortion (0.5-0.7% 150 Volt peak to peak) with both 39K and 57K Ohm load. To drive 6f6 it needs 40 Volt peak to peak. I thought about trying a 27K load, but with 27K we get ~2% distortion.
But I use Linlai 6sn7 and it doesn’t meet the classical 6sn7 spec plate voltage-current characteristics. I have a bunch of classical vt231 40x productions. It will be interesting to make measurements with these tubes.

Hi @lynn_olson ,

I have a couple of technical questions:

1. Is it a good idea to feed indirect heated tubes AC filament with a separate dedicated power transformer? In my integrated SET 300B amp input tubes receive relatively low level signal after the input potentiometer.   

2. I use interstage transformers and separate PS for driver and input tubes. As a result, I have two separate GNDs: 1st - for output circuit and PS  2nd - for input/driver  circuit and PS. Should I connect both these GND together to the amplifier chassis at one point? Or can I leave one of these GND floating?

Hi @lynn_olson ​​​​@atmasphere ,

Thank you for useful information.

Does aluminum provides any shielding? I will use an aluminum chassis.

There is a beautiful picture of Blackbirds. But I don't see inter-stage transformers. Are they under the hood?

Hi @lynn_olson @atmasphere

I’m going to use a separate power supply (PS) box for my amp project. It is too big and too heavy for one box. I don’t want a monobloc solution, because it is an integrated amplifier.

I have 2 GND. One is a power cord GND (1) and another one a circuit GND (2). GND1 will be connected to PS and main chassis, while GND2 will be connected to GND1 thought resistor.

1. What value of resistor is recommended?

2. In which box GND1 will be connected to GND2 through the resistor? In the PS or the main chassis?

Hi @lynn_olson ,

The Hashimoto PT-100F power transformer that I'm going to use for drivers/input B+ and  drivers/input filament. But I'm not sure if it has an electrostatic shielding between HV and LV sections.

I can see some strips between high and low voltage windings. But as far as I can see they are made from a kind of cardboard.

I have had a DIY 300B integrated stereo amplifier since 2006 built by one guy.

But in all this long period of time I did dozents upgrade myself.

The amplifier has 6sn7 input, 6f6 (in triode mode) driver and 300B. Cathode bias and C coupling between stages.

I upgraded: output transformers, power supply choke, interstage capacitors, power supply capacitors, resistors, driver and input tube idle current...

My experience shows that big value power supply capacitors (after the choke), big value cathode capacitors, correct driver idle current gave me much better improvement versus change of output transformer from James Audio to Hashimoto H30 3.5 (3 times more expensive!).

Coupling capacitor quality and sound signature are very important too and cathode resistors as well. Also the quality of power supply capacitors is very important. To get big values and quality without spending too much I use a mix of electrolytic, polypropylene film and vintage German oil capacitors.

I don’t understand why producers put such small capacitors after a choke in 300B amplifiers (typically 100-200uF per channel). My experience shows x10 value gives huge improvement in bass control, transparency, focus, soundstage.

I don't have as much experience as @atmasphere @donsachs have. I also don't understand how audio designers make decisions. Probably a much cheaper change of schematics can lead to great sound improvement without spending too much on parts and size of amplifier.

I heard that Sakuma used 300b driver tube for 300b. He always used transformer for coupling.

To get 27 watts from the 300b push-pull it should be class AB.

Hi @lynn_olson ,

Thank you for very interesting information.

Do you have any data high order distortion in 6F6 and 6V6 in a thiode mode? A friend of mine use 6BX7 double triode for driving. Is there any information about 6BX7?

I heard 2 SET amplifiers built by one person using James Audion iron. Both have 6sn7 input. One had a 6sn7 driver and a kt88 in triode mode output. The second amp used 6F6 in triode mode driver and 300B output. Both amplifiers used RC coupling between input, driver and outputs. The first amplifier sounded too slow. The second one sounded more fast and dynamic. It lead me to the conclusion that 6sn7 is not strong enough to drive even kt88! And to drive 300b even harder. I don't talk about the cathode follower connection of 6sn7; it is a totally different output impedance of the stage plus direct coupling.

The output transformer for 845 SET should be at least 10K Ohm. And it can't be as wide a bandwidth as a 3K output transformer for 300B. Another issue is that driving 845 tubes is much more difficult.

A well designed 845 SET will always be inferior to a well designed 300B SET in sound quality.  

Hi @charles1dad ,

I don't have practical experience. But theoretically 300B is winn-winn vs 845.

1. Output transformer has wider bandwidth.

2. Easy to drive. Driver tube and driver transformer.

3. Easy to build a high quality PS for 400v vs 1000v.

The only one but significant advantage of 845 is more power. But it looks like a parallel 300B SET or 300B Push-pull can be a better solution if you need more power.

I have a question to @lynn_olson 
I have 300B SET 6sn7 input, 6f6g in triod driver with RC coupling. One friend of mine who reads DIY forums and tries all this thing by himself, recommended me to enlarge cathode capacitors up to 100,000uF in input and driver tubes.
So I did 10,000uf in input first and because I like the results I did 100,000uf on input and 50,000uf on driver. So I get even more improvement - deeper and more controlled bass, bigger soundstage, more low level details, transparency, bigger more dynamic sound.
Then I increased driver capacity to 87,000 uF and it gave a big improvement in dynamics and sound quality. Despite intuitively it shouldn't do too much!
In all cases I use a bunch of 10000uf-15000uf (CDE, Nichicon KA) medium quality capacitors bypassed by one 50uf AN Kaisei NP.
Can you explain why these huge value cathode capacitors work?

Hi @donsachs ​​​​@lynn_olson ,

I have 300B SET with 6f6 driver in triode mode. I did 2nd harmonic distortion calculation using 3 point method and 6f6 plate characteristic graph. So with RC coupling that I have right now I get more than 2% distortions with voltage swing 140v pick to pick and when I go to interstage transformer coupling I can get around 0.7% distortions with 160v pick to pick. The difference is huge!

I have a couple of question about using transformer. Do you use load a resistor after the interstage transformer? What kind resistors do you use? I'm going to use Hashimoto A-305 transformers. As I understand Hashimoto recommends 100K load resistor for this transformer.

Regards,

Alex.

I upgraded my DACT VC by using constant serial resistors and "half" of DACT itself like a shunt resistor. I used Shinkoh 0.5W resistors before and now I use AN Silver 2W resistors.

Dear Lynn,
Thank you for convincing me to use an interstage transformer for driving 300B in my SET. I installed Hashimoto A305 into my amp. I put 82K Ohm Kiwame load resistors.. First I measured the 6f6 driver stage with an oscilloscope. There is no overshoot on a square wave. (Thanks to Ralph for noting how to check the correct load that prevents transformer ringing). The frequency response -3db on small signals is 6Hz to 95KHz. The driver stage gives 150v peak to peak without visible distortion (with 300B working point Vg -65v). The output stage now has almost symmetrical clipping (and it was very asymmetrical with RC driver coupling). As a result, maximal power is significantly bigger than before.
Most importantly, the sound quality jumped to a different level. Much more transparent, higher resolution, bigger soundstage, more air and separation between instruments and together with its sound is less harsh, less grained. Even my DAC Chord Qutest sounds much more "analogue".

Dear Don,

I'm also going to make an independent power supply for the driver and input stages. Now I use one 5u4g rectifier for both channel tubes 2x 300B 2x 6f6 2x 6sn7. The total current is 180mA. Is to much for one 5u4g. Now I have very big B+ capacitors, 2200uF for each 300B and 900uF for each 6f6 and 6sn7. But the independent power supply for the driver gave me better PS separation between output and driver plus gave me the option to increase driver tube current and decrease load on the rectifier tube.
I think I will continue to use the same 5u4g rectifier for 2x 300b and SS rectifier for driver and input tubes. Can you give me any advice, what is important to do for driver + input power supply?
I am also going to use the Lundahl plate choke LL1667 for 6sn7 input tubes instead of resistor load.

Regards,
Alex.

I use 1/2 of 6sn7 in the input stage. The driver stage is the 6f6 tube drives the 300b though interstage transformer. The coupling between 6sn7 and 6f6 is RC with Vcup Cutf capacitors.

I have a question about the first 6sn7 stage in my 300B SET. What is the best way to make load and coupling?
1. The drawback of the interstage transformers is a narrow bandwidth in combination with 6sn7 that has high internal impedance. But low distortions.
2. RC - coupling has wide bandwidth but higher distortions.
3. LC - coupling looks good, but what is the optimal inductance for 6sn7? The issue can be not deep enough low frequency bandwidth and high LC resonance Q.
4. What are the drawbacks of SRPP with C or direct coupling? What is for and against it?

Hi Lynn, 
Reading your previous post encouraged me to order a power transformer and a pair of 15H chokes for input and driver tube. It will give me not just PS separation but also reduces the load current to the 5u4g rectifier that I use for all tubes in this stereo amplifier (180mA total). It also gives me a freedom to increase drivers idle current and probably I will try 6v6 instead of 6f6. 6v6 needs more current.
Now I use very big capacitors for B+ - 2000uf for 300B and 1000uf for drivers. What is the advantage of separate power supply for drivers over big capacitors? How do behave big capacitors during the clipping of 300B?

Hi Lynn and Don,
What do you think? Is it a good idea to use driver-input power transformer taps to feed 300B filament? This power transformer should be less pushed then the 300B power transformer.
I don't use only electrolytics capacitors for B+. My DIY friend taught me to use a mix of electrolytes, polypropylene and vintage industrial oil capacitors (Siemens MKV, Tesla, KBG-MN and new Obbligato). All this capacitor bank is bypassed by 0.1-0.22uF PIO and 0.001-0.01 uf Soviet silver mica. This bunch of capacitors doesn't smare sound. The only drawback I can hear is the relatively long warm up time (at least 2-3 hours). Probably a 200+uF set of good quality vintage industrial oil capacitors will sound better. But it will be more expensive.

Hi Lynn and Don,

It is very interesting to hear more details about IT versus LC coupling between 6sn7 and 6v6. What is a difference in sound?

What is the minimal inductance is acceptable for 6sn7 interstage transformer ? 80H, 70H, 60H?

The problem is transformers with high inductance have narrow high frequency bandwidth. But if interstage transformer doesn't have high enough inductance with 6sn7 the low frequency will be cut off. 

Hi @atmasphere , 

McIntosh MC30 has the similar cathode follower driver 12ax7 to drive 6L6 directly. The output tube works in fixed bias in this schematics. Isn't it? Can you explain how the output tube bias is self adjusted?  

300B - the total input capacity is 82pF, voltage swing peak to peak 150 volt.

45 - the total input capacity is 35.5pF, voltage swing peak to peak 100 volt.

There is a difference in number but it is not a huge. Why is so much harder to drive 300B? 

Hi @charles1dad ,

I am not a tube amplifier design guru like Lynn, Ralph and Don.

But here is some information about your amplifier.

https://www.audioasylum.com/cgi/vt.mpl?f=set&m=75278

The 6EM7 is a very strong tube for the driver with a very low output impedance 750 Ohm. The interstage transformer is 2:1. So on the one hand, it requires double voltage swing from the driver tube but on the other hand the bandwidth of IT in high frequencies will be twice wide and driver current twice as big. As I understand, 6EM7 can give 300 volt peak to peak output, that is good enough. I don’t know how 6EM7 is compared to 6L6 in terms of linearity and sound quality and signature.

@charles1dad 

45 and 6v6 tubes that Lynn and Don use to drive 300b also designed and mostly used as output tubes and also give around 2 watts output power. 

Hi @donsachs ,

I have a question. I bought and received a power transformer for the driver tube's power supply. This transformer has a central tap for the rectifier. I also bought 4 schottky diodes. What type of rectifier will sound better: a central tapped full wave rectifier or a bridge rectifier?

Hi @atmasphere ,

I will use this rectifier for the driver and the input stages. So i don't need a very high current and power capability. 

Here is Eleckt schematics:

http://4tubes.com/2-SCHEMATICS/BY-BRAND/ELEKIT/TU-8600/TU-8600R-Amplifier-300B-Schematic.jpg

Global negative feedback, parallel 12au7 driver, 12ax7 input, semiconductor bias control and power supply stabilizers.

chrome-extension://efaidnbmnnnibpcajpcglclefindmkaj/http://4tubes.com/2-SCHEMATICS/BY-BRAND/ELEKIT/TU-8600/TU-8600R-Amplifier-300B-leaflet.pdf

The question is about of calculating C-Miller for a pentode tube in triode mode. If we have no data of C input at triode mode, or also no data of Cg1-g2, then how do we calculate C-Miller?
How is big the total input capacity in 6V6 in triode mode compared to real triods like 45 or 6EM7, 6BX7,...?

Following Lynn and Don recommendations, I added a separate filament transformer for 300B. It made the sound clearer. Then I added a separate transformer and rectifier for driver and input tubes. I use a Hexfred bridge (Ralph's recommendation) and C-L-C-R-C filters. With 30H chokes for each channel. Input tubes B+ is connected to the capacitor after the choke with R-C. I’m going to connect it to the driver B+ capacitor later when I move from RC to IT coupling between input and driver tubes. The 300B output tubes are fed by an old transformer with the 5u4g Linlai rectifier with CLC and each channel has separate 15H choke and B+ capacitor. This upgrade doesn’t break in yet. Two weeks and around 30 hour is not enough (despite all capacitors being previously used). But what I can hear is that the bass control, speed and rhythm accuracy are significantly, radically better than they were before. I can’t hear improvement in the midrange yet. I also can hear the tone of instruments loose a little bit "tube magic". I understand I need more break in to make a more accurate conclusion.

Hi Lynn,

In my SET amplifier is IT coupling between driver and 300b and RC between an input and a driver stages.

I think I understand why the tone issue appeared. I played with driver tube B+ voltage and current. The cathode resistors 680 Ohms weren’t changed. The driver is 6f6 in triod mode. In begging, the idle current was 20mA. Then I increase it to 23mA and then to 26mA. So, in the middle setting 23mA, the sound character was faster and rough with flatter musical instruments tone. With 20 and 26mA, the amplifier sounds more like SET, the tone is more reach, and speed is more realistic and natural.

Hi @donsachs,

Lynn told in this thread that you prefer schottky diode in your amplifiers.

Why do you prefer schottky over other diodes?

Hi @lynn_olson ,

As DIY I can adjust my amplifier to my own taste preference. But compared to experienced Hi-End equipment designers I have less time, money, experience and knowledge. My BCs in Electronic Engineering helps me a little bit. I do digital ASIC design for my living that is far away from analogue electronics.

Hi @donsachs ,

Will you present your new amplifier in the Montreal Audio Show in March?

Hi @lynn_olson ,

According to your comparison between RC coupling and IT coupling, IT coupling between input and driver stage should increase slew rate (dynamics, trancients, speed) similar as it does between driver and output stage. Isn't it?
On the other hand, the input stage doesn't have a big voltage swing, and as result of it a linearity issue. It shouldn't drive the driver stage in class A2. 6sn7 with IT has less wide bandwidth compared to 6v6, 6f6, 45 that can be used in driver stage (in my amplifier 6sn7 18Hz-35KHz vs 6f6 6Hz-95KHz -3db ).
Is It wars to try interstage transformer after the 6sn7 input stage in my 300B SET?

Hi @donsachs ,

I have two questions about your new amplifier:

1. When you decided to go to an interstage transformer after the input stage, why did you stay with 6sn7 and didn't move to something like 5687? With 5687 you can get even wider bandwidth especially on bass. What is the advantage of 6sn7 over 5687?

2. Did you try direct coupling between input and driver tubes?

Here is an example of direct coupling:

http://www.single-ended.com/Lagarto/shishido/811A.png

Is 6f6 grid load easy enough for SRPP?

ECC82 can be changed for 6sn7.

What do you think about this schematic?

Hi @atmasphere ,

I meant that tube with less internal impedance with a transformer load with the same inductance has wider bandwidth in bass. I agree unfortunately 5687s OTOH. 

Hi @lynn_olson ,

Thank you for the detailed lecture about tubes!

Here is Shishido version with 300B. No DC on secondary of IT. 300B is in a regular cathode bias.

http://www.vestnikara.spb.ru/vestn/n4/shishido1.gif

What do you think about SRPP is this schematics?

Hi @lynn_olson ​​​​@donsachs 

What is changed in sound with the driver replacement from 6v6 to kt88?

On the one hand kt88 is much more powerful and low output impedance tube. On the other hand 6v6 is more linear. Did you try kt66 that is more linear and musical vs kt88.

I have a technical question for tube gurus @lynn_olson @donsachs @atmasphere

I put the intrastage transformer Hashimoto A-107 between the 6sn7 input and 6f6 driver tubes in my SET amplifier. The load resistor connected to the 6f6 grid was 520KOhm. I measured a square wave output and it had a big overshoot. I changed the load resistor to 120KOhm, and the overshoot decreased by amplitude and attenuation time. But still, there is a notable overshoot. I measured a frequency response and there is a hump +1.7dB at 35KHz. There is -3db at 19Hz and  47KHz.

Should I decrease the load resistor more to remove overshoot completely?

If yes, in which value range should be this resistor? For example, if I take a resistor less than 50K it can increase distortions.

http://www.tube-amps.net/HP_A107.htm

Hi @donsachs ,

Yes Dave Geren at Cinemag do great job!

I use Hashimoto A-305 IT between 6f6 (driver) and 300B. The square wave was almost perfect at 10KHz, 20KHz, 30KHz. I use an 80K Ohm grid resistor on 300B. But I’m not sure this resistor is needed! The -3dB frequency response is from 6Hz to 95KHz. This IT is extraordinary on measurement and sound.

For 6sn7 I use Hashimoto A-107 that is not as perfect but has more primary inductance that is essential for higher input impedance 6sn7.

Hashimoto A-305 is designed specially for SET, but A-107 is universal IT that can be phase splitter or SET 1:1 or SET 1:2.

I just installed A-107 and it has zero break-in time. So I can’t comment how does it sound compared RC (V-Cap CuTF, AN Silver Tantalum) that were before. It need more break-in time.

Hi @atmasphere ,

How does excessive transformer ringing can influence on sound?

Does it make it too sharp or bright?

All classic vintage PP amps from 50-60 have very weak PS in terms of chokes, power transformers and especially capacitors values. If you use the same power supply for SET with no feedback with a such PS will be a disaster. But many of these classic PP still sound fine! Why? Because feedback or because PP has PS noise cancelation?  

Hi @donsachs ,  

I changed a load resistor 56K Ohm (Alan Bradley 2W), in my SET amplifier. And still have a 2-3 times bigger overshoot vs Blackbird has. However the sound is more balanced compared to the 120K (Kiwame 2W) load resistor I had before. I'm still waiting for more break-in time.

So far IT compared to RC coupling between 6sn7 and 6f6 - IT sounds more fast, more precise rhythm and pace, faster transient, more clean in upper frequencies. But I don't hear a big difference in terms of low level resolution. The tonal balance is a little colder too. Bass is more fast and controlled but less deep.

Which resistor will you recommend for power supply? Now I use Ohmite gold 10w and Milles 12.

Hi @donsachs and @lynn_olson ,

Here is square wave 1KHz on my amp output:

https://photos.app.goo.gl/jcdjn88Q71WaEvoR9

Output transformer and driver IT have much wider bandwidth compared to input to driver IT. During the measurement the 6sn7 current was 7.2mA. I just increased it to 8.75mA. Load resistor is 56KOhm.

In my case, I compared the sound of my amplifier with different IT resistor load. 120K (Kiwame 2W) vs 56K (Alan Bradley 2W).
The sound with 120K resistor was too bright and rough, with 56K tonal balance was more natural and sound was more refined.
But with RC coupling the sound was even less bright vs IT with 56K resistor. I think the resistors material AB vs Kiwame can make a difference too.
I am afraid if I completely remove the load resistor the sound will be even brighter. So I want to try 39K AB 2W resistors (that I have in my stock) to make the sound a little bit warmer.