Posible cause flor NST voltage reduction / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Posible cause flor NST voltage reduction

Move Thread

LAN_403

rulmismo

Sun Aug 17 2014, 12:14PM

Registered Member #4187 Joined: Fri Nov 04 2011, 08:08PM
Location: Spain
Posts: 43

Hi, i have had a NST 8kv 20mA for some time and used it for a SGTC, a Marx and some jacob ladders.
I always use a Terry filter but without the MOVs, just resistors, caps and SG.
I use a 1000:1 divider to measure the output, with about 40M impedance (a number of 1500V 2.7M resistors in series)
The output voltage that I measured when I started using it was near 8kV,but presently I only measured 5.7kV.
I had to change the divider due to some problems but it seems to work and impedance load is low for the NST.

Could the NST by damaged in any way so the output is lower? Or do you think about another explanation?

Thanks for the feedback
Raul

Ash Small

Sun Aug 17 2014, 12:52PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

The way NST's are laballed is misleading. Output is 8kV OR 20mA. Open circuit voltage is 8kV, to get the neon tube to light, but then voltage drops as current increases to maintain the discharge.

It may be that you are drawing more current, so the voltage has dropped.

johnf

Sun Aug 17 2014, 07:59PM

Registered Member #230 Joined: Tue Feb 21 2006, 08:01PM
Location: Gracefield lower Hutt
Posts: 284

Raul
It may have developed shorted turns
does it get hot with nothing connected to it after half an hour

Signification

Wed Feb 04 2015, 07:28AM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

CONNECTING YOUR DMM TO THE HIGH VOLTAGE OUTPUT OF A NEON SIGN TRANSFORMER!
About 15 years ago when I was most into TC's I remember having several tough problems with neon sign transformers (NST's). One big problem was that they were potted in tar which carbonizes easily creating secondary shorts. I think these NST's are pushed to their limit (label ratings). They are center-tapped to the case which goes to the third plug (earthed) prong which should always be used--otherwise greater voltage stress is placed on the unloaded NST. When powering a neon sign, the NST output can actually drop to several hundred volts. I found that the labeled current is actually the short-circuit (ZERO output voltage) current. It would be very useful to have a way to easily measure this current on the high voltage winding.

Here's how: use a DMM! If you do it right it works great. Since you can get DMM's now for about $5, it's not much risk at all anymore (not like risking your $250 meter back in the 90's). OK, set your meter on the 200mA reading (if it makes you less nervous, and is your first go at this, use the 20A/10A setting) It's much safer if the meter uses a shunt here and no fuse. If a spark jumps the meter is toast. So do this carefully and make sure all connections are very tight and secure. With power off, connect the meter leads to the high voltage leads of the NST. Make sure they are tight connections--double connections (two wires per side) is not a bad idea. BE SURE THE FUSE IS NOT BLOWN! The high voltage will arc across an open fuse, destroying the meter instantly!! I learned this one the hard way--on my good meter! NOW, power the NST, the meter should read the label's value. Unplug the NST and remove the meter.

I found that knocking off the insulators and removing the tarred transformer from it's case allowed de-tarring--I used gasoline, soaking and occasionally scrubbing with a toothbrush until ALL the tar was finally gone revealing the beautiful shiny transformer underneath. It takes several days immersed in gasoline, but well worth it when done patiently! Now you can remove several magnetic shunts from each end to get a higher current capability--measure it with your meter! If you really want to get maximum current, you can remove a few more shunts and operate the NST under oil. I did this to an NST rated at 15kV @ 60mA. I managed to get 75mA from it. Instead of messy oil, I potted it in a plastic container filled with paraffin. This was nearly 15 years ago and still no trouble--I use it often--still get a nice loud hot long arc. This single NST can power the primary of an impressive Tesla coil! I haven't tried this with a microwave oven transformer yet--that just now occurred to me.

Antonio

Wed Feb 04 2015, 12:31PM

Registered Member #834 Joined: Tue Jun 12 2007, 10:57PM
Location: Brazil
Posts: 644

This measures the short-circuit current. You can measure the voltage by inserting a string of resistors in series with the meter, designed so they drain just a small fraction of the short-circuit current. A good way to destroy a meter: Prepare the setup, make the measurement, turn everything off, including the meter, and then try to make another measurement, forgetting to turn the meter on on the current scale FIRST.

Signification

Sun Feb 08 2015, 03:03AM

Registered Member #54278 Joined: Sat Jan 17 2015, 04:42AM
Location: Amite, La.
Posts: 367

I have a homemade high voltage probe that I use to read capacitor voltage with (I have had this thing for ~20 years) but have never tried on an NST (or any other high voltage *AC*) output. It is a divide-by-10 probe made for a multimeter with a 10 mega-ohm input impedance. It is a series array of nine 1meg 1/2 watt resistors potted with paraffin in a large plastic veterinarian's syringe holder. I used 1/2 watt resistors simply for the increased 'high voltage' rating. As far as wattage is concerned if I had used 1/4 watt resistors here then, power-wise, they would handle 20,000 Volta! I had one of those "100 projects you can build" books that told how to make one of these x10 voltage probes. It said that from the resister 'voltage divider' equation, for a x10 voltage attenuation you make the "x10 high voltage probe" with a series of resistors that are ten times the value of the meter's input resistance value. Here is that formula...

Vo = Vi*R2/(R1+R2);

where, R1 and R2 are two series resistors with the top resistor (R1) of the string connected to Vi (the input voltage that is to be divided), the bottom resistor, R2, is connects to ground, and the output voltage, Vo, is the (divided) voltage at the node of the resistor pair. Well, if you solve this equation for the ratio of the input voltage (Vi) to the output voltage (Vo), you get:

Vi/Vo = (R1/R2) + 1

This says that the resulting divided voltage ratio (Vi/Vo) is NOT simply equal to ratio of the divider resistors (R1/R2). To illustrate, the equation can be rearranged as:

R1/R2 = (Vi/Vo) - 1: So, if you want to divide by TEN you must use a resistor ratio of NINE (ten minus 1)...etc. My probe contains nine 10meg resistors and connects to a standard multimeter probe wire. To use it, I simply remove the meter's red test probe wire and replace it with the 90Meg probe wire. Now just multiply the displayed voltage by ten to get the actual value.

**OK** "hold please... I'm too curious! Before sending this post, I am going to get up right now and go measure the voltage output of an NST using this probe, I will come back with the results current time:(19:45) ................................................

OK: Time is 20:21...
I hooked the NST (I used a variac) to one of my dirt cheap ($5 on ebay) meters through the divider to the NST and slowly increased voltage--it did NOT seem to work! Small numbers were jumping eradicately--I got brave and turned up the variac (probably eventually to 3 or 4kV)--NO CHANGE--but also no damage. I got a feeling that there must be something about this cheap meter, I attempted to read it's input resistance by measuring 10VDC with it 10meg resistor in series with the 10V source. The meter read 0.9V, if it's internal resistance is 10meg shouldn't it be closer to 5V? I re-tried the reading with one of my better meters that I knew had 10meg input impedance--it read the expected 5V. So, the input resistance of the $5 meter must be lower. I found that a 1meg resistor gave the expected 5V reading--I guess the input impedance of this 'cheapie' is 1meg. BUT, that doesn't explain the strange behavior with the NST high voltage--same story with low
voltage (140VAC). BUT the 1meg "cheapie" works fine on high voltage DC inputs with the expected "divide-by-100" (a 5000Vdc input gave a 50Vdc reading).

Then I got really brave and used one of my regular 10meg meters with the divider on the high voltage NST (with fingers crossed)...IT WORKED PERFECTLY! ...and, nothing burned out! So it works with high voltage AC as well.

THE QUESTION...
Why won't the 1meg meter work with the divider on AC but work fine on DC? ...'haven't yet had time to think about it.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob