Building a brand new valve amplifier using the RH84 design and off-the-shelf components.
In Part 1 I outlined the design that will be used and choice of components. I will now go on to cover the build of the prototype and a pair of suitably high-efficiency loudspeakers to use with the amp.
The sound produced by valve amplifiers is often referred to as “warm”. I think this is partly because of the visual aspect — the glow that the valves produce — but a valve amplifier definitely sounds different to a solid state one. All things being equal, a valve amp will produce a softer sound as it produces a slight distortion that generates musical harmonics.
I thought I had better do some reading up on valves to better understand what I was dealing with. A valve, or tube as they are referred to in the US, is a device to control the flow of electric current. It consists of two or more electrodes in a vacuum maintained by a glass envelope.
Our amplifier employs 2 valves per channel: one double triode and one output pentode. The simplest type of valve is a diode, which contains a heated electron-emitting cathode and an anode. Current can only flow in one direction through the device — from the cathode to the anode. The triode valve has an additional “control grid” which, as the name implies, allows the flow of electrons to be controlled and hence used to amplify signals. The pentode valve works in a similar manner.
Valves have a limited lifetime, usually due to the filament or heater burning out, so they are made as replaceable units; the electrode leads connect to pins on the bottom of the valve which plug into a valve base.
The next stage was to build a prototype of one channel. We could use a bench power supply for testing, so at this stage, I was just interested in the amplifier circuit and not the power supply.
My first task was to “translate” the schematic and work out how things were connected. I laser cut a mount from MDF for the valve sockets, transformer and a length of tag board. Then, using the schematic for the amp and schematic legends for the 2 valves and their data sheets, I started trying to work out how all the parts were connected and how best to arrange them.
I initially laid the components across the tag board. This felt like a good way for me to keep things organised and get clear in my head how everything was connected.
To make my life easier I labelled all the resistors – I know it does not look very pretty but as I am relatively new to electronics it makes it easier than trying to remember all the colour codes. Once I had done that I needed to rationalise the layout and reduce the amount of wiring by connecting the components directly where possible. The use of an MDF base also meant I could draw, and re-draw the connections before I committed myself.
Once I was happy with the arrangement it was time to start soldering.
This is the first time I have attempted “Point to point“ wiring, so I spent some time looking at pictures of the insides of other people’s valve amps and got a good idea of how things should be connected. As this build was for testing purposes and would be dismantled once we were happy it was all working as we wanted, the joints needed to be soldered well enough to make a good connection, but not so well that they could not be taken apart later. To this end, I abandoned my usual practice of hooking wires onto connectors before soldering.
I tried to keep the layout as neat as possible, which would make it easier to identify any faults and ultimately abandon the tag board altogether. Although the prototype shows signs of having been reworked a few times, I should be able to make it tidier on the final version.
As the amp was going to be used to keep us entertained in the workshop it needed some matching speakers, that would work well with the relatively low output of a small single-ended valve amplifier. As detailed in the first post in this series, I had the Fostex FE83En drivers and some cabinets in kit form, which I just needed to paint and assemble.
I assembled the cabinets by glueing and clamping the MDF panels, then leaving the adhesive to set. I left the back panels off so as to make it easier to wire the drivers up to the input sockets, although in hindsight I would have achieved a neater finish if I had glued and painted the whole cabinet, even if it had made the wiring a bit fiddly.
The “Tuff Cab” coating was applied with a roller, starting with a fairly thin coat and then several more to build up a good protective layer, letting each one dry overnight before applying the next.
The drivers were then screwed in place on the front of the speakers and the covers fitted.
I then drilled the back panels inserts to accommodate the Neutrik Speaker Connector plugsand fitted them to the back panels of the speaker cabinets.
They then need connecting to the drivers before the back panels were fitted, followed by the cabinet corners.
We now need to test the single channel amplifier before assembling two for the final stereo version. As this involves high voltages, I will be taking extra care and making sure there is someone else there at all times. I also found this web page outlines some sensible precautions.
We then need to sort out the power supply and fit the whole thing in a suitably attractive enclosure, hence there is still a fair bit of work to go, so look out for future posts!