# Re: Line filter -- where to ground it?

```Original poster: "Barton B. Anderson" <bartb@xxxxxxxxxxxxxxxx>

Hi Forest,

Ground the filter to mains grounds.

```
The 36" spark length is based on the famous John Freau equation of sqrt(input watts) * 1.7 and is obtainable. Your coil may go a beyond this if your losses are less than John's. You will likely end up in the 25" to 30" range without any changes to the coil. Spark length with this equation is based on input watts (measured at the input to the NST). If you have a greater amount of losses than John Freau had, you will have less power to the sparks and a smaller spark length (it's that simple). But, it's not always "that simple" to find losses and reduce them. That is always a challenge.
```
```
One very simple example of losses is the NST itself. If John's NST (used during those measurements) was more efficient than your NST, you will lose out no matter how well the coil is built. Coilers always upgrade this gizmo and that gizmo trying to stretch those sparks, but sometimes still can't attain John's efficiencies. When that happens, the hv transformer is an obvious reason. But, so what if John could get 36" and you got 30". No big deal. But, if your getting only 10", well, then there's a problem.
```
```
If you get to a point where you want longer sparks and don't care about the watts, you can always remove shunts in the NST (a lot of dirty work, but it's been done many times). If enough shunts are removed to double the current, you will end up with a 15/60 (900W) with a spark length of 51" obtainable. If there's one thing I've learned in this life-sucking hobby, the bond between power and spark length is very strong, not always linear, but still very strong.
```
Take care,
Bart

Tesla list wrote:

```
```Original poster: "Forest Darling" <fdarling@xxxxxxxxx>

I am going to wire a line filter behind a neon sign transformer in
backwards, but where should I ground the filter to? The NST case, or
the mains ground?

The NST case is going to be grounded to the RF ground (I'm using a
Terry filter on the high voltage side.)

Also, out of curiosity, how long of a spark would you expect a 450
watt 15kV 30mA NST to realistically be able to put out?
DeepFriedNeon's calculator said 36" but that seems very long.

http://www.deepfriedneon.com/tesla_f_calcspark.html

Thanks,

--Forest

```
```

```