Blog posts of '2012' 'August'


Many valve types don't use pinch type construction but instead rely on a PRESSED-GLASS BASE or button in which the lead in wires are  moulded into the glass base of the valve which then serve as both valve pin and electrode cage support.  The EF50 was the valve type that pioneered the use of the pressed glass base for Mullard at the start of WW2 so that the technique was perfected for the time when B7G and B9A valves became the staple of valve manufacture during the early 1950s.

Typically, a PRESSED GLASS BASE was made using a multi stage rotary carousel  base making machine like the one shown below: -

Rather than try to explain each stage in writing, here we have an X Ray representation of each of the main forming stages for you: : -

1: precut pins in holder                    

      2: glass collar surround 

3: heat

4: mill press

5: mould

6: finish

Things are slightly different for a pinch stem base as seen on the majority of octal valves and below we show a photo of a base pinching machine: -

The sequence of operations is quite complex and is done in two forming stations, the first to get the pinch glass shape and the second to embed the lead wires into the pinch.  Here's the sequence for the 'initial shape' formation:-

And following, the sequence for the second 'embed 'formation: -

To give a finished base pinch with embedded wires and an evacuation stem ready for electrode cage mounting and envelope encapsulation:-






Today, I want to look at LEAD IN WIRES.   Look in any PINCH & FOOT valve and you can see the LEAD IN WIRES sandwiched on all sides by the pinch glass.  In order to maintain that seal, the wire has to have a similar coefficient of expansion as the glass surroundings - a tall order indeed!

In actuality, the LEAD IN WIRES are of a composite construction with the top section being of sufficient diameter to make the electrode cage support rigid and similarly with the bottom section of copper wire that forms the valve pin connection.  The middle section of wire however, is made of an alloy that  has an elastic property as well as the same coefficient of expansion as the surrounding glass.  

Typically the alloys are complex and still proprietary information but although we may not know the precise proportions of the eutectic adjunct, we do know the constituent metals and the names of two of these alloys.   Alloys were made of Iron (Fe), Nickel (Ni) and Cobalt (Co) and we have exotically named wires such as Kovar, Fernico, Dumet wire and most impressive of all, Red Platinum. 


Let's chat about SCREENING CAGES today.   This is a natural progression from yesterday's topic of METALLISING as recent valve designs tend to use SCREENING CAGES instead of an external metallic coatings to prevent outside magnetic and electrical fields from interfering with a valve's normal operation.  

Screening cages usually comprise of a wire mesh cylinder which completely encircles the electrode cage.   The cage functions as an electrostatic shield by preventing external field interference and additionally serves to collect stray electrons that may escape from within the electrode cage, this is also economically beneficial as there is then no need to INTERNALLY CARBONISE the envelope.

So, now you know why EF86 don't have any INTERNAL CARBONISATION and you may have also realised that many sellers on EBAY erroneously peddle their EF86 as having a 'pin mesh anode'  - and they are  wrong - what they should say is a 'pin mesh screening cage.' 

So here's a salutory lesson, don't buy from an EBAY wide-boy who knows nowt but from a webshop wide-boy like me -  Mr Mullard Magic who knows just that little bit more.!!!!!



Today's valve component is METALLISING.   Many valves have an external metallic coating applied to their envelope which is either copper or more commonly zinc, often finished in a coat of coloured laquer so we see silver - grey, red, gold and occasionally green colour coated envelopes.   Typically, this coating is connected to a valve base pin, often the cathode by wire hoop and connecting wire placed near the junction of the envelope - base join.

Although small signal valves tend to be fully covered, many output valves have the metallic coating confined to the lower half of the envelope ala the Mazda 6P25.  This is done to allow heat to be dissipated which could otherwise shorten valve life.  The purpose of this metallising is not just to look pretty but to prevent external electro-magnetic fields from affecting a valve's correct operation.



With the birth of the Superheterodyne radio and the means to keep valve count and commensurate valve cost low, it turned to the valve manufacturers – particularly Mullard to come up with advanced valve designs to accommodate this requirement.

In early Superhets, the heterodyning frequency was generated by a local oscillator valve and was then super-imposed on the signal by a 'mixer' valve. A handy dodge was to incorporate both of these functions in a single envelope and developments saw successively the double-grid valve, the screened tetrode and the HF pentode being used where injection of the heterodyning frequency was achieved by the use of a coupling coil in the screen or auxiliary grid circuit. Cue the introduction of the Octode where the coupling of the emissive stream occurred completely within a single envelope.

The Octode, is an 8-electrode valve, having six grids and an anode concentrically arranged about a single cathode which with Grids 1 and 2, can be considered as a triode, and may be employed for the generation of the heterodyning frequency. The positive potential on the screen (Grid 3) causes the emissive stream to be hyper- accelerated and electrons will pass into the space between Grids 3 and 4, but by application of a negative potential grid bias applied to Grid 4, these electrons will be repulsed, resulting in a 'space charge' pulsating at heterodyne frequency.

From this 'space charge,' or 'virtual cathode,' electrons are drawn by the action of the anode of the Octode and will oscillate at heterodyne frequency then be mixed with the modulated radio-frequency signal applied to the control grid, Grid 4. Hence, the virtual cathode with Grids 4, 5 and 6 and the anode form a pentode mixer with variable-mu characteristics.





Today, I would like to raise the topic of SCREENS.  As we discussed last time, internal carbonising only decreses the risk of secondary emission but cannot prevent this unwanted electron straying.     As we have noted, secondary emission can play havoc with a valve's preformance, particularly in multi-electrode cages due to unwanted  coupling effects.

For some multi-electrode cages, metallic screens are fitted to keep the electrons on their true path.  Below, we have a photo of just such a deflection screen shown on the electrode cage of a dismantled EF91 pentode where you can clearly see a metallic screen plate between the anode lead as well as the grid and cathode connections.




Today, I want to discuss INTERNAL CARBONISATION sometimes also referred to as screeding or colloquially as grey glass.   From preceding blog articles, I have already explained that electron emissions  are drawn from the cathode to other positive electrodes within the cage and sure enough, the majority of emissions do travel within the electrode cage, however, like recalcitrant small animals and wilful children, a small proportion don't do what they should and escape from the electrode cage striking the glass envelope walls.

These 'loose' secondary electrons can build up a potential charge on the glass which can cause audible distortion and sometimes a striking blue glow on the envelope especially where the electrode cage is closest to the envelope or if the envelope internal surfaces were not scrupulously clean at the time of manufacture.   Many customers have sent me valves for testing and diagnosis, often recent manufacture Chinese devices showing this glowing phenomemon which they have mistakenly suspected is a 'gassy' valve whereas in actual fact they are seeing Jablonski shift glow caused by secondary emission due to poor internal envelope cleanliness.    

I have here at Mullard Magic HQ,  a pair of Chinese globe shaped 300B triodes that exhibit this effect, interestingly, I have also recently tested for my pal Andy Smith of Head-Fi fame,  a modern recreation of a CV181 twin triode that showed ridiculously high anode current which also had INTERNAL CARBURISATION that I suspect performed a dual role in both hiding the apalling electrode cage support construction and stopping it's resultant secondary emission from causing the valve to perform as a nightlight!!!!!!!!  .

In order to minimise this secondary emission, an envelope inner surface is coated with colloidal graphite which shows very low tertiary emission characteristics.  So now you all know why your KT66, KT63, KT61, 6J5, 6SN7 et alia have grey glass


Let's continue this story by today looking at GRIDS.  These are found between cathode and anode except for diodes which have just two electrodes - the anode and cathode.  The purpose of a GRID is to control the value of anode current.  In triodes, we have just one GRID - the control grid.  In pentodes we have three grids with the inner one being the CONTROL GRID, the next the SCREEN GRID and the outer one the SUPPRESSOR GRID.   In each case, the GRID is a cylindrical or ellipsoid spiral of fine molybdenum or molybdenum alloy wire wound onto two rigid longitudinal nickel or copper  struts, the ends of which are secured top and bottom within mica sheets to prevent movement.  Let's look in a little more detail at these differing grid types: - 

The CONTROL GRID - two separate voltages are applied to a control grid, the first being a steady voltage, often of the same magnitude as that applied to the cathode but the potential is negative to that of the cathode. This voltage, often termed the grid bias potential, determines the mean anode current under no-signal conditions.   The second voltage applied is the input signal voltage which takes the form of IF, RF or AF voltage inputs.  The effect of these applied voltages combined is to cause variations in anode current flow.

The SCREEN GRID - the anode current variations caused by applied signal voltage will affect the external load of the valve, however, as well any external effects there can be corresponding internal effects which can reduce available voltage at the anode which the consequent reduction of the valve amplification factor.  By introducing  a second grid situated between the control grid and anode and maintaining this at a positive potential with respect to cathode, unwanted internal voltage reduction can be reduced.

The SUPPRESSOR GRID - electrons reaching the anode at high velocity can expel other electrons from the anode itself giving rise to secondary emission.  These secondary emisions are attracted to the SCREEN GRID and thus flow in the opposite direction to the anode current which again has the effect of reducing anode current.  This unwanted effect is minimised by inserting a SUPPRESSOR GRID which is connected to the cathode such that secondary emissions are returned to cathode hence minimising anode current reduction.

And now we have a very nice little photo which shows from top to bottom the control, screen & supressor grids for, on the left a DF33 and on the right, the much bigger EL37


At Mullard Magic we are paragons of equality choosing to never discriminate against a customer for colour, creed, religious belief, sex, sexual orientation or any other basis.  However, not all businesses take the same stance.   Just look at what my webmaster and I saw pasted to an elevator door when we went out for a curry in Levenshulme near Manchester: -


 Make the wimmin do the hard graft whilst the men travel upwards in comfort with ease ............ it's a funny old world!