Blog posts of '2014' 'April'


I recently met Steve Burke - Hare who is a Mortuary Manager and Medical Enbalmer with a penchant for valves.   Steve has engaged the talent of Monika, a Polish tattoo artist working from a studio in Macclesfield to illustrate his devotion to the thermionic cult.    

Here you can see Monika at work and Mrs. Mullard Magic is a little in awe of Monika as she likes the idea of ladies inflicting pain and discomfort on a man: - 

And now, you can see the fruits of her labours and I think you will agree, the results, in startling clarity, detail and colour are both beautifully executed and very impressive indeed: -

I remarked to Mrs. Mullard Magic rather smugly that I might get a valve tattoo done on some part of my body,  but then decided perhaps not as with my being a stickler for detail, I wouldn't like to make a KT66 look like an 813.  

Quick as a flash she retorted, " More like making an ECC83 look like a 956 acorn!."  Thus humiliated, that was an end to that particular idea, dejectedly, I am destined to remain unadorned.


Today, I thought we would talk about AC ripple and Pi section power supplies for as we all know, one of the most important considerations in an amplifier is the minimisation of AC ripple that can be superimposed on the DC voltage supplied by a rectifier.

The value of ripple is dependant for a unit voltage on the reservoir capacitance and the smoothing network - let's take a typical Pi network comprising of C1, a reservoir capacitor, and C2, a smoothing capactor with either an inductance L or a resistance R: - 

The ripple voltage is inversely proportional to the values of C1 and C2 - so if you want low ripple then get BIG capacitors.  Some people economise in the design of equipment by substituting a resistance in place of a more expensive choke with little appreciable rise in perceived ripple current.  Of course in this world 'tha gets nowt for nowt,' and the resistance unfortunately reduces the DC output voltage dependant upon current delivery loading and resistance value.   This disadvantage can be minimised by feeding the PA stage output valves from C1 rather than C2 with the ripple being nullified by a push-pull output stage, particularly if this stage is balanced.

Taking the above example: -

For C1 = 8 uF; C2 = 16 uF; L = 10 H then Irip = 210 mV for V0 = 251 V @ C2

For C1 = 50 uF; C2 = 50 uF; R = 500 Ω then Irip = 210 mV for V0 = 221 V @ C2 or 265V @ C1

So there you have it but my take on it is don't be cheap, use a choke and preferably one bought from Mullard Magic!!!!!!