Lighting strings of C9 lamps with a NST / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Lighting strings of C9 lamps with a NST

Move Thread

LAN_403

klugesmith

Sun Dec 24 2017, 05:42PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Finally got around to a plan from last January. Back then, at a post-holiday clearance sale, I bought some strings of 25 screw-base lights.

Current measurements on a whole parallel string confirmed that the C9 lamps are 7 watts each at 120 V. Combined resistance: about 8 ohms cold, 82 ohms hot. Snipping some wires converted the string to series, for a 625x increase in resistance.

The IV curve was converted by relabeling the axes:

I lit the series string with a NST, and then with two 30 mA NST's in parallel on a Variac. Stopped at about 80% of nominal input voltage, when the NST knob voltages reached 1000 V on one side and 935 V on the other side. .

Next steps: a voltage divider to measure higher AC voltages. Then measurement of 60 Hz current in the 10-100 mA range without having to float the meter. Perhaps the current under test has multiple turns passing through an inductive sensor. Or put a bridge rectifier & LED side of an optoisolator in series. Are the current transformer toroids in GFCI's made of special stuff, to be usefully magnetized at 0.005 ampere-turns? How about high fidelity microphone transformers?

For the same or less money, one can get strings of C7 lamps, typically 5 W, with smaller screw bases. I've seen C7's in 4 W (for night lights) and 3 W, but never at seasonal-merchandise clearance prices

Dr. Slack

Mon Dec 25 2017, 06:06AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

Or put a bridge rectifier & LED side of an optoisolator in series.

Hmm, most cheap obtainable optos have an absolute max of 'several 1000v', with a working voltage of a fraction of that. Which makes them marginal for an NST on full bore.

You could make an opto with a LED, a photodiode, and a length of black biro barrel. Rubbish CTR, but designable isolation voltage.

Or perhaps make a series current powered current to frequency circuit to flash a LED or transmit an RF pulse, which can be received at a reasonable standoff distance.

klugesmith

Fri Jan 26 2018, 05:39AM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Here to report some progress on making an analog optoisolator good for 20 kV or more.
Along the lines of what Dr Slack said a month ago.

Was looking for LEDs that could handle 100 mA continuously, to avoid having to divide down the current under test.
Opportunity knocked in the form of LED flashlights on sale for $2.99, each having three AAA batteries and a single LED.
Batteries alone are probably worth half of that price. White light isn't ideal for silicon photodiodes, but where does one get a red or IR LED at that current level without having to wait for delivery?

I damaged my specimen by taking it apart wrong. Apparently the guts had been press-fitted into the machined Al tube. Guessing wrong, I first tried pressing them out toward the tail end. That mangled the star-shaped LED module and the black plastic spacer behind it.

The LED still worked, intermittently, for a little while.

A few days later it was convenient for me to buy a few more of the same model flashlight.

There may be an elegant solution to managing the luminous flux. Built-in paraboloidal reflector seems to direct most of it in a collimated beam.

An identical reflector, from the broken unit, ought to do a good job of condensing that back to a 5-mm-diameter spot. Maybe with less loss than simply placing the detector close to the bare emitter (not too close for 20 kV).

I think the full voltage will be present between the big ends of the two reflectors, because of their metal coating. Hoping to find an easy and elegant way to make field-shaping electrodes, to avoid corona or full breakdown where the metal ends. Any ideas?

Dr. Slack

Sat Jan 27 2018, 08:30AM

Registered Member #72 Joined: Thu Feb 09 2006, 08:29AM
Location: UK St. Albans
Posts: 1659

I'm not sure there's much heavy lifting that needs to be done at only 20kV.

One overkill that comes to mind is to make the spacer a potential-grading stack of single-sided Cu-clad, or alternating alli foil and plastic sheet, with a light hole though the middle.

You could peen the edge of the reflector back on itself to increase the radius of curvature of part facing the other reflector, ie the part that sees the strongest field. Or build up the edge radius with hotmelt and make it conductive with paint or tape.

At 20kV, surely all you need is to stick them one each side of a shard of window glass, or a few thicknesses of PET bottle, and move them to a new location if you notice any erosion due to corona.

radiotech

Sun Jan 28 2018, 05:10AM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Series street lamps were rated in amps and watts, and fed with a constant current
supply, typically 2300 volts,

Coincidentally prior to good bulbs, that was how arc lamps were wired.

A blowout cell is a holder of a paper spacer. When the lamp burns out, it shorts.

A new paper, and a new bulb are replaced together.

klugesmith

Sun Jan 28 2018, 02:37PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1714

Thanks, Radiotech. The shorting cell is new to me. Wonder if they used special paper, to develop a nice low resistance & stop burning after being punched through by overvoltage? I think today's miniature series-connected lights use the same concept. But the shunts (built into lamp) need to switch to low resistance when exposed to only 120 V.

An intermediate voltage, 600 V, was common for electric streetcars around 100 years ago. My late friend Jim Tangney told me about a different solution for series-connected cabin lights. When one burned out, an ordinary argon-filled lamp would go into arc mode, with damaging results. The solution was to use special (even then) vacuum bulbs. Then whole string would go dark when one failed.

radiotech

Wed Jan 31 2018, 09:42PM

Registered Member #2463 Joined: Wed Nov 11 2009, 03:49AM
Location:
Posts: 1546

Here is the 'paper' used in the General Electric lighting system. Edison inspired the culture of
General Electric.

Pity he had a limited education, and was only able to solve issues in the 1000 or so patents
he held.

1517434958 2463 FT181053 Novalux Cutout Film Paper

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