Post by big_electron on May 25, 2011 10:39:43 GMT -5
I'm into analog electronics. Right now I want to make a guitar amplifier. Never studied tubes, so I'm stuck with silicon. I'd thought about using a germanium transistor for the amp. I need to make it so that the first transistor has a low operating voltage, so that it saturates, therefore distortion, overdrive, fuzz, whatever you call it.
Another thing I've wanted to do is to make an FM receiver using a PLL, instead of ratio detector. I just don't know what values of resistors and capacitors to use here. I once found something online where you can make an FM discriminator using logic gates.
Meanwhile, I plan to hit some garage sales, buy old radios, take them apart, and salvage away.
ADD: I own an oscilloscope, and 2 DVM's capable of testing resistors, diodes, capacitors, and transistors. I am trying to decide between a signal generator or LCR meter.
All those common-X configurations were designed before transistors came along.
There are of course some big differences: the high operating voltage for one, and for another there's no PNP-NPN equivalence.
The voltage thing means you'll need some transformers: the power supply will need two secondaries, one delivering 6.3V for the heaters, and the other around 200-300V (and centre-tapped, eg +300/0/-300 because of the way valve rectifiers work). You'll also need an output transformer to step down to a voltage that won't fry your speakers. And some hi-fi designs use additional transformers between amplifier stages.
If you're building an amplifier with a push-pull output stage, you'd want a pair of transistors: one PNP the other NPN. In the valve world you can't do this, so you need an additional valve to work as an inverter.
A working amplifier can be made with 5 valves: a rectifier, a triode as a pre-amp stage (which is what you'd overdrive for your guitar distortion), two pentodes for the power amp stage, and an additional triode (or two, I forget) for the inverter. Actually, I have a low-power turntable amp with just 3 valves, using a single pentode for the output stage.
Valves work by having a heated cathode which emits electrons. Normally these electrons would return to the cathode, but by applying a -ve voltage and placing a +ve anode nearby, a they will cross the gap, and a current will flow. The current can only flow in the one direction on account of the anode being un-heated, and the resulting device is call a diode because it has two electrodes (the heater doesn't count).
The current can be modulated by inserting a grid between the cathode and anode and applying a voltage to it, which impedes the flow of electrons. The result is a triode, which is broadly similar to a transistor. Its performance can be non-linear, however, so additional grids are used to help out, producing tetrodes, pentodes, etc.
There is a lot of information about this on the internets, including many circuit diagrams!
The serial number does give some info on the type of device. In the UK the Mullard-Philips system is pretty much standard. It uses two or three letters followed by a number, normally two digits. The first letter is a code indicating the heater voltage: E is for 6.2V, for instance. The next letter(s) indicate the type of valve: C for a triode, L for a pentode, Z for a rectifier. (No, I don't know why they picked those letters!). The numbers differentiate between devices' electrical characteristics. So the ECC81, ECC82 and ECC83 are all dual triodes, the EL84 and EL95 are pentodes, and the EZ80 and EZ81 are rectifiers.
The American system will similarly encode info on what the device is, but all I know about them is that they begin with a number, which indicates the heater voltage. So the US equivalents of the above valves will all begin with 6, except for the dual triodes whose heaters are wired for either 6 or 12 volt operation so their US numbers begin with 12.
Are your unidentifiable tubes unidentifiable because their numbers are illegible? If you can read them, post them: I like a challenge.
Post by big_electron on Jun 21, 2011 23:41:39 GMT -5
One thing that makes me want to run back to transistors is that tubes require a negative bias, that is, the input terminal, the gate, is placed at a voltage lower than the cathode. With the use of resistors, you can achieve said bias.