Building/modifying SV-S1616D 300B amp

Member
Joined 2021
This is the current state of my sunvalley 300b. I am back to using the red HV supply wires. Ultimately I found that I could not quite get the output tube bias high enough with the purple wires. So now with a 5u4gb I am at ~420V and 75-80ma cathode current. With a 5u4g I get a bit less B+, ~400V and 68-70ma cathode current.

Notice that I have changed the cathode resistors for 300b’s to ~935 ohms.

Also to help with turn on inrush I am using a current inrush limiter CL-40.
 
Member
Joined 2021
If you are referring to R107, that resistor is dropping the high voltage down for the preamp(12at7) and driver stage(12au7’s). It is fairly critical that this voltage remain at about what the schematic shows, which is 432V. If you have lowered the power supply B+ voltage to near 425V then you will need to lower R107 to keep the preamp section voltage up as much as possible. I am using 820 ohms at 3watts. There is only about 25ma current draw in the preamp circuit so the resistor doesn’t need to be huge. The driver stage will not work as well if the voltage drops too much below 400V. The 12au7’s will not be able to swing enough voltage to drive the 300b to full power.
 
Member
Joined 2021
I will also try to encourage you to check all of the voltage test points in the system. It’s critical to know how one change is going to affect the other in the circuit. If you let me know what the voltages are at each test point I can give you more info about what’s happening in there.
 
Member
Joined 2021
Ok I just realized you did post your voltages. I had not seen that you filled in the chart on the left. Let me absorb that and then reply.
 
Member
Joined 2021
Look at B2 and B3 test points. B2 is 388v. Personally I would not go any lower than that. I would decrease the resistance to get that voltage above 400v. You will begin to lose drive and power if you go lower than 388v. It will sound wimpy.

My biggest problem with this amp in stock configuration is that it sounded wimpy and hard-with glare in the midrange and lacked details in the high frequencies. It wasn’t open and full and rich and relaxed which is what I’m used to when using 300b. It was too tight and lean and lacked power. I found most of the problems were solved by rebiasing the input tube. I would encourage you to try a 5751 and 12ax7 as input tube. I never tried them because I wanted to have a little less gain not more and I wanted a tube that wasn’t so expensive to buy. 12AV7 is almost the same as 12AT7, but to me sounds a bit better.

All you need to do to rebias the 12at7 is to lower the 2 plate resistors. It’s easy if you don’t like it just put the 220k resistors back.
 
Member
Joined 2021
After rebiasing the input tube the next thing I wanted to address was the output tube bias. I see that your 300b’s are running at about 67-68ma. In my amp I changed the cathode resistors to make them draw 75-80ma. When you do this it will cause the B+ to sag down to about 420v. So now plate voltage 420 minus cathode voltage 72 is right at 350v. If your target voltage is 360, then maybe try 1k cathode resistors. I am at 935 ohms.
 
Member
Joined 2021
As you can see I have a lot to say about this amp. I want to point out that you should check the filament voltage on the preamp tubes. It supposed to be 12.6v mine was at 14v. This will prematurely burn out your tubes. Also it will sound better if you get it to 12.0-12.6v. I set mine at 12.3v.
 
Member
Joined 2021
I just had a closer look at the input tube resistors in my amp. I didn’t remember that I did change the plate resistors and the cathode resistors. So there are 4 resistors that need to be adjusted. If you’re interested to know exactly what I did let me know and I will give all the details. Cheers Thomas
 
Member
Joined 2009
Hello Thomas,

I haven't been checking this thread until just now. Sorry about that.

Have you checked your input voltage from "the wall"? I'm using a Variac and am back to using the Mullard GZ34 tube instead of the RCA 5U4GB rectifier. The GZ34 needs 116V input to get 5V on the 300B filaments, and the 5U4GB needs 120.7 V to get the same 5V. My wall input varies from about 123V to over 124V, depending on the time of day or night. So, this is a "silent" area where most everyone suffers, and I discovered this while working on this amp. Previously, I eliminated a line conditioner and used a Furutech NTX outlet box instead. The Furutech unit wasn't cheap, but it sounded better. I already had a Staco Variac to bring up new amps, but I never used it. Finally, a friend suggested that I put the variac on the 300B, and low and behold, my 5V on the 300B filaments became possible without changing parts.

You need to work with a Variac set at the voltage to get your filament supply at around 5V for each tube. From there, you can adjust. If you live in the US, you probably have a modern input voltage above 122 V. My larger dropping resistor of 7.5K allowed the GZ34 to generate a higher voltage for the 9-pin tubes. Depending on what you want to accomplish, you might change your resistors, but maybe not if you used my 7.5K dropping resistor rather than the original dropping resistor Sunvalley used (5.6K).

Now, my GZ34 operates with higher voltages, so the nine-pin tubes sound good, but the filament supply can be dialed in exactly. I'm now looking at replacing the Staco with an adjustable variac that limits the variance to an acceptable number and goes beyond the Staco in capability.

So, to generate 5V on the 300B filaments, the GZ34 requires around 116V of wall input AC, and the 5U4GB requires about 120.7V of wall input AC. Creating that situation made a huge difference in sound quality and increased the tension of the GZ34 tube to a point where it now sounds better than the 5U4GB. Previously, the 5U4GB sounded better but was too "airy." The GZ34 now has just enough air not to sound boring. Everything now sounds up front, and singers also sound up front rather than recessed as they did with too low a voltage on the input tubes. But I changed the input tubes to make the sound quality "pop."

Just my two cents regarding making changes.

I'm now using GE 6201 tubes for 12AT7 and a pair of GE 6189W tubes for 12AU7 tubes in the 9-pin sockets. A friend sent me a stash of his 12AT7 tubes, and I already had the GE6189W here. The GE6189W was too forward before, and now it is just right. It shows that the right circuit can change our idea of what works. My former Brimar and Mullard tubes don't sound anywhere near as good, but the GE6189 paired with the 6201 is perfect. The Telfunkens and other rare NOS 12AT7 I had also didn't sound as good as now. So I can now use this amp to fine tune my speakers.

That brings the sound quality to a point where I now really love the sound quality with my Voxativ 99db Hagen speakers. I just added a pair of 12" open baffle bass per channel 4 EMS drivers) for the low end, and I am now bi-amping using the 300B on the Hagens and a good Class D amp on the bass augmentation. My idea is to take the low bass reproduction of the 300B off and place it on a high-powered Class D amp. However, the bass is no longer as fussy, so all Class D amps can sound good when used only to drive the bass augmentation speakers. I sense that the Class D amp needs a pot, but I'm reserving this decision until I go through the next 11 modules that I've ordered from Sublime. I haven't received them yet.

So will my final Class D amp be a Hypex Nilai kit, or will it require a passive pot, in which case I'll use a Khozmo amp with a built-in pot?

I just ordered 11 more modules from Sublime Acoustics to change my crossover point and limit the 30f0B amp. Right now, I have the Hagens playing full range (no crossover), and the open baffle bass is limited to playing no higher than 80Hz. With the modules, I can experiment and go down to 45Hz in 5 Hz increments. Once I have the speakers dialed in, I might go back to change this amp further, but honestly, I will probably have a custom 300B built and use the Sunvalley amp as my backup should I ever need to service the main amp.

I now have superb sound quality and transparency. My 300B tubes are Western Electric 300B, and the rectifier tube is a NOS Mullard GZ34.

Even though the voltage on my nine-pin tubes might not be perfect, I'm now fine-tuning by listening and want to get the speakers dialed in before making any other changes to this amp. Changing the input tubes made the difference I needed.

Once I'm done, I'll consider purchasing another even better 300B amp with a substantial power supply, two chassis, and very few electrolytes. It will cost a lot more, but it will be worth it.

I hope this helps,
Richard
 
Top