RE: Ion Generator

["Dale Hall" <Dale.Hall-at-trw-dot-com> (by way of Terry Fritz <twftesla-at-uswest-dot-net>)]

Hi Brad,

(what exactly do you have, 
a flyback xfmr from a monitor/TV 
or a florescent ballast or ?)

If you have no idea what it is, and can't get to the inside
to assess, how do you plan to power it up ?
Send a private email pict of it if you are able.

If the topology is such that it is protected from short circuit 
such as NST's simply measure the output short circuit current.
This represents the max current available.

Another useful test if you have a relatively non loading HV probe 
  (harder to achieve if output is AC, even harder if high frequency like
Khz and up,
   however then use a rectifier and test at DC)
is to add resistive load until the output voltage is half that of noload.
This represents the internal resistance of the xfmr/power supply
and may be used in calculation.

Here is a copy of my response to Benjamin & List 
re: Ion Generator testing 7/31/00 which can be applied 
to flybacks or unknown xfmrs:

9kV at 30mA from 12dc would be great, 
IF it can do that continuously.

Let me and the list know the your results.

More likely it is a converter that is designed for much less 
continuous output Voltage, typical for solid state replacement
for the "NST" but capable of running only at ~1kV at 30mA continuous 
after a trigger pulse of 9kV which ionizes the neon tube.

Two investigative approaches:
1.) Find a spec sheet or inquire as to exacts specs from the mfr
2.) Set up bench test (suggested anyway to verify mfr statements/specs)

If you have the converter, run some bench tests. 
Required is a large likely custom HV load resistor.
Rload=9kvrms/.03= 300,000 ohms with
power = .03*9000= 270W so you need substantial heat dissipation capability.
like 10ea 30k 30W, or 30 ea 10k 10W etc.

If you can attach this load and the measure 9kVrms you are in business !!

(and we want to know where to get them, i.e. distributors, cost, etc.)

Note that this measurement may not be easy. 

The converter output probably switches at high frequency (~20kHz) 
to keep the size and weight (inductor) down.

If one side of the converters HV output is ground referenced (unlike 60Hz xfmr)
the load current may be more easily instrumented - 
measure the voltage across a part of the load. 3k = 100:1 so you'd expect 90V 
however response time becomes an issue.

High frequency reactance upsets the divider, 
and the need for compensation network
adjustment such as oscilloscope probes use.
Use of a fast HV rectifier, conversion to DC (averages) removes the
frequency problem.
Expect a much lower like output, like ~1kVrms with period(s) of HV pulsing.
Be pleasantly surprised if its 9kVrms !

There are individuals on this list with SS Ballast with hands-on 
knowledge and experience which I do not have (speaking from what I've read).
---------------------
-----Original Message-----
From: Brad Davidson (Kiloman) [mailto:iamkiloman-at-hotmail-dot-com]
Sent: Tuesday, August 01, 2000 5:50 PM
To: Hall, Dale
Subject: Ion Generator

Hey, thanks for the response! DO you have any idea how I coudl determine the
output characteristics of the transformer I found? It has no markings on it
besides lead lables, and all of the electronics are sealed inside a ceramic
capsule that is filled with a hard gel. Also.. if I were to procede with
this project as an ekperiment (I now have a 15 KvA NST to play with) do you
have an recommendations on the dimensions of coil parts? Thanks,
Brad D