Building a brand new valve amplifier using the RH84 design and off-the-shelf components.
In previous posts in this series I have covered planning the layout of the amplifier, and building and testing a single channel version. Now that has been successfully completed I could proceed to build a 2-channel stereo version and make the power supply.
I had decided that all the components apart from the 3 valves and the 2 audio transformers were going to be housed in the base of the enclosure. I could put the circuit together on a piece of laser cut MDF for testing and that could then be used as a template for drilling the necessary holes in the enclosure.
I left my single channel prototype still soldered together so it could act as a reference for the first channel. I could then dismantle it to re-use the components and have the channel I had just put together as a reference. I also had the spreadsheet that I had made listing how everything connected that I wrote about in the previous blog post.
I was determined to make a neater job of this than I did of the prototype and although it is far from perfect I do not think it looks bad for a second go at point to point soldering.
I had read a couple of forum posts from people who had built a number of these or similar amps that recommended keeping wires as short as possible and using a single ground so that was the route I took. Although I think I may have taken the short wire thing a bit too far and made my circuit a bit crowded. I will give myself a bit more space in the final version.
Once the 2 channels were completed it was time to test them using the bench power supplies as we had done previously for the single channel. I connected the audio transformers, speakers and stereo audio input.
I followed the same procedure for testing as I had for the single-channel version. I first connected the 6.3V supply to see if the valves light up which, thankfully, they did. Once that was established we could give it a proper test with audio input and a couple of “disposable” speakers, just in case.
After we initially got a very low audio signal and high power consumption, I tracked the fault down to a couple of resistors that were the wrong value due – I think, due to the auto-complete in my spreadsheet. Something to be aware of in the future. (I have updated the attached spreadsheet with the correct values).
Once I had put that right the amplifier sounded great – even through the low-quality speakers with no housings. It was now time to sort out the power supply.
I have amended the spreadsheets detailing the components and their connections to include the power supply. The components are as follows:
- 1 x Transformer, 6.3V AC
- 1 x Transformer, 2 x 115Vac, Toroidal, 200VA,
- 4 x Diode, IN4007
- 1 x Resistor, 270R 3W
- 2 x Capacitor 220nF 400V
- 2 x Electrolytic capacitor 100uF 350V
- 1 x 2-row tagboard
It was decided to go with a solid-state power supply. Rectifier valves are actually not that great at doing the job they are intended to do – changing AC to DC. If the amplifier is loaded – by a sudden crescendo in the music for example – the DC voltage will drop before recovering, which is referred to as “sag”. This is because the rectifier simply cannot keep up.
In contrast, a diode rectifier will barely sag. Guitarists have come to like this sag effect and so valve rectification is common in many valve guitar amps. Although purists may argue that we should go all valve, this effect is not something we are looking for in our HiFi amp, hence the choice of solid-state.
A schematic for the EL34 amp that included a solid-state power supply on the DIY Audio Projects website, from the book "Build your own Audio Valve Amplifiers" by Rainer zur Linde, provided an example that we thought could be slightly adapted for our purposes. Referring to the schematic I sketched out my circuit as it would be constructed on the 2-row tagboard RS Stock No: 043-3703 that I had decided to use. It was then fairly simple to solder it together and fix the circuit and the two transformers to a piece of MDF for testing.
The first version I built blew the mains fuse every time it was powered up and, after some research into full-wave diode bridge rectifier circuits, I concluded that the schematic was at fault. I amended it, rotating the arrangement of diodes by 90 degrees clockwise and ended up with a working power supply.
There was some difficulty sourcing a suitable choke and so it was decided to substitute a 270Ω 3W wire wound resistor RS Stock No:and see how the power supply performed. If it proved too noisy we could replace it with a choke later.
This is the schematic for the final version of the HT part of our power supply with the components labelled as per the spreadsheet:
The two capacitors between the transformer secondary and the bridge rectifier were omitted but could be later added if deemed appropriate.
The design we found also had a circuit for the valve filament supply, which was rectified and with voltage regulation. However, we decided instead to, initially at least, just use 6.3v AC direct from the transformer secondary.
So as to avoid any potential damage to the amplifier circuit, the power supply first was connected to mains voltage and then the outputs tested with a meter. Satisfied it was producing the correct voltages it was then connected to the amplifier circuit and, when a sound source was connected, we could confirm it was doing its job.
Now that we have a fully working amplifier complete with power supply, the next stage is to tidy up the circuit some more, add a volume control, power switch and the necessary input and output sockets, and then fit all of this into our enclosure. I will be covering this in the next post.
WARNING: The circuit designs are shown here for informational purposes only, this is not a tutorial and the designs are not warranted to be safe. Dangerous voltages are involved and you should not work on such designs unless competent and safe to do so!