Power transformer HT/B+ voltages

[vivid]

My question is further down in bold.

Hello everyone - I'm new to these forums and hope they'll prove to be useful as it seems to have helped a lot of you.

I'm also (relatively) new to electronics. While in high school I took some classes in electronics, so I have a basic understanding of it. And I'm also studying an engineering program with a pretty good amount of physics in it. So I have a theoretical understanding of what's going on in electrical circuits. Up until now I haven't really seen any benefit of knowing electronics.

What's different now?

I want to build myself a tube amp for guitar.

I'm trying to understand the design of the schematics I've read and how to alter them in a proper way for my own projects. I understand there's a certain level of complexity in these circuits - and I'm not here to learn how to develop similar circuits from scratch. I just want a working knowledge of how to alter them for my own needs.

Ultimately I want to build a guitar amplifier with a preamp based on channel 2 from a Mesa Boogie three channel Dual Rectifier, vintage model/gain/voicing setting with a solid state rectifier and "bold" power settings, paired with a ~5 watt, or so, preferably 6L6 power amp (based on the fact that I've had my best experience with 6L6 amps). The key to this is the word ultimately, as in "not next week, or month". I do not wish to do this to save money, but as a learning project.

Question: I've been studying power transformers. The primary 230 VAC, secondary lets say 190-0-190. I imagine three leads. Two with voltage peaks at 190 VAC, and one center with 0V. When rectifying the power transformer there are loads of schematics that measure ~300VDC (unsmoothed). How does that work? How can I get a higher DC voltage from a lower, but rectified, AC voltage?

Thanks

[Alastair E]

Full-Wave Rectification of 190V AC will yield, 268.66V DC after smoothing, and assuming no losses in rectification/chokes/etc....

AC volts is measured as RMS value, to obtain the DC smoothed voltage, you Multiply the RMS AC voltage by 1.414....

[vivid]

Is there a common practice in how many times you're supposed to smooth the DC voltage? I'm thinking RC smoothing (resistor-capacitor) , and that one RC stage is using one resistor and one capacitor. How stable does the voltage need to be?

I've seen four stages in most guitar amp power sections. Any comments on that?

[boywonder]

Originally posted by vivid
Is there a common practice in how many times you're supposed to smooth the DC voltage? I'm thinking RC smoothing (resistor-capacitor) , and that one RC stage is using one resistor and one capacitor. How stable does the voltage need to be?

I've seen four stages in most guitar amp power sections. Any comments on that?

One RC section will certainly smooth your rectified DC, and an LC section works even better if you want to spring for a choke. More RC or LC sections yield more smoothing, but drop volts and increase the output impedance of the supply. Some folks go for a low impedance output with a little more ripple, and some go for extra filtering sections to get the DC as flat as a pancake. If you add RC or LC sections for a given output B+, you'll need to start with more transformer volts.

BTW: 1.414=squareroot of 2

Download and play with Duncan amps power supply designer II (PSUDII) here:

http://www.duncanamps.com/psud2/index.html

It's free, easy to use, quite accurate and there are lots of people here to help you with PSUD. A great learning experience. It will show you what you will end up with as far as ripple, B+, etc. Use a current load instead of a resistive load.

[ray_moth]

Yes, good advice, use PSUD2. To get the best results, know and enter in the program the DC resistances of your power transformer primary and HT secondary windings. A lot depends on these values.

Put in a delay of a few seconds before reporting, because your filtering capacitors need time to charge up.

The 1.414 factor from AC RMS voltage to DC voltage assumes that the smoothing filter begins with a capacitor, by the way (e.g. C-L-C or C-R-C). The rectifier charges up the firct cap to the peak voltage (which is 1.414 x RMS voltage). The first cap slowly fall as it discharges into the load, until the next peak comes along and it gets charged up to the peak voltage once again. This happens 50 (or 60) times a second. So, the ripple has a 'sawtooth' waveform, which subsequent filtering stages help to smooth out.

[vivid]

Thanks for the answers. I'll make sure to play around with PSUD.

I've got another question on the B+ voltages. There are several B+ voltages in the power supply (B+1, B+2, B+3, etc). I understand that the voltages are different after more stages of filtering, but what deciedes which B+ voltage is used in a certain gain stage?

[Robert McLean]

but what deciedes which B+ voltage is used in a certain gain stage?

Any ripple in early stages will be amplified by subsequent stages. Therefore the supply for early stages must be the most filtered, and therefore will be of the lowest voltage, assuming the usual sequential arrangement of filters.
In other words, lowest voltage for the first input stage, highest voltage for the final output stage.

[vivid]

What kind of characteristic does the ripple sound have when amplified?



Also, how do I know how high I want my high B+ voltage to be?

[bigwill]

Originally posted by vivid
What kind of characteristic does the ripple sound have when amplified?

Buzzzzzzzzzzzzz or Hummmmmmmm depending on the amount of filtering (buzz indicates higher frequencies present)

[boywonder]

Originally posted by vivid
What kind of characteristic does the ripple sound have when amplified?



Also, how do I know how high I want my high B+ voltage to be?

Ripple will be low level hum at your mains frequency X2 so either 100Hz or 120Hz depending on where you live.

Your required B+ depends on your tube choice, operating mode (triode, pentode, UL) etc etc. Do you have a schematic that you can post?

It may be worthwhile for you to find an appropriate schematic of a 6L6 based guitar amp circuit, build that, and then modify as you gain experience. There are loads of 6L6 based designs around, I'm sure. Since you only need one output channel (mono for guitar), costs should be very reasonable since you only need one output transformer, etc. The schematic will drive your requirements for B+ for each stage.

Build it as a breadboard on a chunk of plywood, that makes it easy to modify and if you decide you want a different topology, just unsolder and start over. The tubes, sockets, transformers are completely recyclable for different 6L6 based designs, although you'll probably need different R and C values for different topologies.