A sharpened rod on top of the topload, pointing at nothing in particular. It looks like an afterthought and it is one of the few parts you can actually control the discharge with.
What it is and why
Electric field concentrates at a point. A smooth toroid holds its charge with the field spread evenly over its whole surface, and the voltage has to get very high before anything breaks down. Put a spike on it and the field at the tip is several times higher than anywhere else, so the air there breaks down first, and at a lower voltage than the toroid alone would have needed.
Two things follow. The discharge starts sooner, which is often what you want. And it starts where you put the spike, instead of wherever the toroid happens to be dirtiest or has the sharpest fingerprint.
It is a consumable. The tip carries every discharge and changes shape as it does, so its behaviour drifts over an evening. Builders who chase length change them the way you change a drill bit, and where the tip is being damaged by heat, Mads Barnkob's topload guide puts tungsten at the top end.
What you decide
How far past the toroid it stands
Criterion: whether the tip has cleared the field that hugs the topload's own surface.
Far enough to be out of it, and that is further than it looks. Barnkob sets the figure and gives the reason: the point "has to stand out far enough from the very close and heavy magnetic field that is close to the topload surface. At least 10 centimeter would be a good start, even longer and upwards pointing can be used to get arcs further away from striking down in the primary coil, earth rail or just get longer arcs to ground from the bigger distance."
Three things are bought there and only one of them is length of arc. The first is breaking out at all, because a topload large enough to do its other jobs will not break down on its own, which is the whole reason the point exists. The second is where the arc goes: pointing up and standing clear keeps it off the primary and off the earth rail. The third is the extra distance to ground, which is arc length, and it is the one that gets quoted alone.
How the tip is finished
Criterion: how many channels you want, and you want one.
A smooth tip gives one large channel. A rough one gives several short ones. All the energy in one channel gives the maximum length, while the same energy split between three gives three short arcs, so a single channel is a design goal rather than an accident. Barnkob asks for "a good conductor with a even surface to avoid a large corona spray all around the wire, rod or what is used for it".
Smooth is not blunt, and the two are different axes. The geometry wants to be sharp, because concentrating the field is the entire point of the part. The surface wants to be even, because roughness anywhere along the rod sprays corona instead of feeding one channel. A sharp tip on a rough shaft is the common mistake, and it is an easy one to make with a file.
The material is a choice rather than whatever was in the drawer. Against copper, Barnkob puts aluminium at about twice the resistance, brass at about four times and stainless at about forty, which he does not recommend. Where the trouble is heat rather than resistance the tip goes to tungsten, "if heating or even melting damage to the tip is a problem".
What will get you
The numbers
- Length: at least 10 cm clear of the topload surface, longer and upward if arcs are striking down into the primary or the earth rail.
- Tip material: against copper, aluminium about 2x the resistance, brass about 4x, stainless about 40x and not recommended.
- Capacitance to the topload: far larger than the topload's own capacitance to the world. In the taped case above it was not, and that is what made the divider.
- Material, at the top end: tungsten, where heat is damaging the tip.
Where these come from. The taped case was carried here with a measured capacitance in picofarads, and that figure could not be traced to any source we can fetch, so it has come out. What the case establishes is the order of magnitude, comparable rather than far larger; the measured value is not established on this page. The live what a topload costs still prints it, and it is the same untraced measurement there. The tungsten tip belongs to Daniel Eindhoven's RAIKIRI, remade in tungsten after the copper ones kept melting. That row is on the records wall flagged stale: his page is still up, and no longer carries any of it.
What goes wrong
- Arcing inside the structure and an unstable coil. The capacitive divider above.
- Three short arcs instead of one long one. A rough tip, or several competing points.
- The arc goes to the nearest earthed object rather than up. Point it upwards, which is the maker's own remedy, and check first that the earthed object is not simply the nearest thing in the room.
- Corona spray along the rod instead of one arc off the tip. The surface rather than the geometry: the shaft has to be even too, not only the tip.
- Performance drifts over an evening with nothing changed. The tip eroded. It is a consumable.
- Nothing comes off it at all. That is its own checklist, and it starts a long way from the tuning.
Where next
- What a topload costs, whether you want a toroid under it at all.
- An arc grows as fast as charge arrives, what happens once the discharge leaves.
- No sparks on breakout.
The breakout on the department diagrams is a needle rather than a ball, and it is short, for the reason in the middle of this page.