Thread: Garrard to BSR
View Single Post
Old 15th Sep 2018, 1:14 pm   #10
julie_m
Dekatron
 
Join Date: May 2008
Location: Derby, UK.
Posts: 7,735
Default Re: Garrard to BSR

SPICE is an old, but still very useful, computer program which analyses electronic circuits mathematically. The circuit was described by a deck of punched cards, one for each component. The text file that does the same job nowadays is still sometimes fondly referred to as a "deck"; and although my first experience of SPICE was on a microcomputer, our lecturers had used the old mainframe versions with actual punched cards.

Today, there are several Open Source and Proprietary circuit simulation programs based on the original SPICE and they all accept the same circuit descriptions, whether on punched cards or on disk. Each line describes one component; it has its name (which includes what type of component it is and therefore how many pins it will have), a list of the circuit nodes to which each pin is connected (in a fixed order) and its value or parameters. Lines starting with * are comments and are ignored.

In this circuit, V1 represents the mains supply (a sine wave with zero DC offset, 340V AC peak amplitude, 50 cycles per second). R1 is a surge limiter resistor. C1 and C2 are the voltage dropping capacitors. R2 is the discharge resistor. V2 is a fixed voltage source of 0V DC, and R3 is the lumped resistance of the valve heaters (=830Ω for a UY85 and UL84). This is all just one series circuit. There is always a node 0, which is the circuit's earth. We sketch out a simple diagram, numbering nodes as we go along:

(0) ---V1--- (1) ---R1--- (2) ---C1--- (3) ---V2--- (4) ---R2--- (0)

The reason for including V2 is purely to use it as a current measurement point. (Some later SPICE versions incorporate the ability to measure current through any component, not just voltage sources; but this is a proprietary extension and not part of the original Open Source version.) We should see a peak current of 0.1414A through V2, corresponding to an RMS current of 100mA, if we have the right capacitance values between points 2 and 3.

If you download your own copy of a SPICE-based circuit simulator such as NGSPICE (Open Source) or LTSPICE (Windows only, Proprietary but £0) you should be able to feed it that input, and see whether the output looks reasonable.
__________________
If I have seen further than others, it is because I was standing on a pile of failed experiments.
julie_m is offline