Designing the Power Amp
In order to design a tube power amp, you’ll need a load line chart using the plate characteristics of the tubes you are using. This is found on the datasheet.
I took my plate characteristics chart and laminated it so that I could draw multiple load lines on a sheet or start over without destroying a bunch of paper. You can follow my steps by reading this article at The Valve Wizard. The calculations below are a summary based on my amp design of what I learned at The Valve Wizard and from books on the subject.
I have decided to use an 8k primary output transformer as my starting point for this project. The first point on my load line will be the DC voltage, which is going to be about 300V. I put a dot at the bottom of the load line at 300V.
Next I find the anode current. These are drawn at 1/4Za-a (Class B) and 1/2Za-a (Class A). These correspond to Class A at 4k ohms and Class B at 2k ohms.
- 300V/4000ohms = 75mA
- 300V/2000ohms = 150mA
I will draw a dot at each of these points on the left side of my graph. Then, each segment will be connected with a line using a ruler. This isn’t the final operating point of the power amp. The next step is to find a comfortable bias point. This point will be about 75% of the plate dissipation point at 300V. I counted 6 squares up from the 300V/0mA point and placed a dot. I also placed a mark 6 squares up from the 75mA dot on the left side of the graph. I can calculate everything I need to know about my power amp from this line.
My bias point is at 30mA per tube (60mA total), and the cathode voltage will be at 9V. I now know that my cathode resistor will be 9V/.060A = 150 ohms. The power dissipated by this resistor is .060A*9V = .54W so I’ll use a 2 watt resistor for this. The cathode bypass capacitor will be a 100uF 100V capacitor.
The rest of the power amp design is pulled from most schematics that I’ve seen. Grid 2 is connected to the linear taps of the output transformer using 470 ohm 1W resistors. Grid 1 of each tube is connected to the phase splitter through a 4.7k ohm 2W resistors (grid stopper), and the grid reference resistors are 220k ohms.
In the next blog article I will design the power supply.
Defining Amp Specs
As stated in part one, reading up on the subject of tube amp building is a great place to start when you want to build something. If you use web resources, bookmark them in a folder in your browser because there will come a day when you’ll want to reference it. Sometimes I have an idea in my head, and I forget where I saw it. Having your stuff in order will help you locate something should you need it.
After you’ve become familiar with the process of building an amp, the next step is to figure out what kind of amp that you want to build. The easiest guitar amp to build is one with a single-ended output. I’ve build four of these in the last few years of which three still exist; I turned the parallel 6DG6GT into a single 6DG6GT in 2013, and I updated it in 2014.
The Specs of My Build
I’ve started a new project that will take me quite some time to complete due to the complexity of the build (in both the design and the cost). Here is where I started on this build. Every part of the project will be based on these findings. These specs may change over time given the availability of parts or feasibility of design, but this gives me a good starting point.
- EL84 Push-Pull Class AB power amp
- 15 to 20 watts power output
- Fixed bias on power tubes
- 8ohm output
- 300V power supply
- Single channel input
- Single 12AX7 preamp
- 12AX7 phase splitter
- No effects (tremelo, reverb)
Based on this I have my starting point. Now I have to decide what I want to design first. In the next post, I’m going to start by using the EL84 datasheet to draw load lines, and find the operating point of my power amp section.
It seems like this has been one of my favorite topics on Tech-Tut. I really love to build tube guitar amps. It is proven over the many discussions that I take great enjoyment in this part of my hobby.
There have been many ups and downs, and I have temporarily given up many times, but I rarely give up for good. I know when to cut losses, but a hobby is something that is for me, and I can revisit it again at any time. As with the second revision of my 6DG6GT amp, it goes to prove that persistence finally paid off.
Here I am once again working on my next challenge. I want to design and build a low power, but higher power than a single-ended amp, push-pull guitar amp using the EL84 tube.
I want to document most of what I have done for this project. While I am building this project for a profitable venture, I want to leave the whole project as open source for those who might want to try to build their own.
As with any “tutorial” on my website, sometimes I tend to narrate more than give step-by-step instructions. That’s what leads me to my first installment in how I construct the EL84 guitar amp. I want to explain my research.
There are many links and books available to explain how to design and build tube amps, but my experience with them is that a second opinion or angle helps to solidify the learning process. I didn’t get where I am today from one source. I have read Instructables, webpage after website, and several crazy expensive books to help me wrap my head around the process. With that, I pass to you some of my most favorite links explaining this hobby.
- Valve Amplifiers by Morgan Jones
This book has a lot of great info, but is more directed to HiFi.
- Design and Construction of Tube Guitar Amplifiers by Robert C. Megantz
This title does a great job of showing many stages of a guitar amp. I would have liked more discussion about how the design process happened, but it has been a valuable tool.
- Guitar Tube Amp by gmoon
I’ve mentioned this Instructable many times because this guy walked me through most of the processes. It’s a long read, and it will take many passes. From some of the rudiments of design to how to cover a cabinet, gmoon does a great job of documenting the process.
- The Valve Wizard
This website does a little better job at explaining the design process.
- The Cooperative Tube Guitar Amp Project
For Single Ended amps, this site has a whole document of theory. There is also a forum with people who can help you along the way.
Do Your Research
Ask questions. Google stuff! Check datasheets. See what others are doing. It’s the age of information, and you can find almost anything on the web. There’s no reason why you shouldn’t. Step one in starting any project is doing research. Start with knowledge bases that are directly related to what you want to accomplish.
Tune in next time when I show you where I’m starting with my latest guitar amp build.
Today, 6-20-2010, I took the amp to church to test it out in the real world. My main concern was the heat from the LM317 TO-220′s. I used the chassis as the heat sink because without a large heat sink, the LM317′s would go into thermal shutdown which caused the heaters to drop out, which effectively makes the amp stop working.
My goal was to get at least 1 hour of continuous use. I played the amp from 7:45am to 8:45am, 10:30am to 11:30am, and 4:45pm to 6:35pm. The amp was very hot, but never cut out. This is good news.
This amp has gone through quite a few changes. The reading might be interesting to see why I made certain choices, but the schematics here are for reference ONLY. They work, but they need work. Check on my tube amp page for the latest schematics.
[Brackets note changes and experiences after real-world testing had begun. Some things had to be ironed out during the perfection/acceptance process.]
Designing a tube amp circuit isn’t really that difficult, but it can be frustrating. The hardest part that I experienced was designing the output stage. I never really read anything that helped me understand why I did what I did. I’ll explain this when we get there. Also note that everything is designed to be within 10% of the expected value. Some things could be changed after other values are figured, but with tolerances of components falling between 5% and 25%, what’s the point in trying to get any closer on paper? Testing the actual circuit is the only true measure of its performance.