[OT] Hyperikon Retrofit LED Tubes Without Ballast

[Martin McCormick-3]

We just bought 4 Hyperikon retrofit LED tubes to replace some
25-year-old tubes that gave up the ghost.  Both pairs of tubes
worked just fine so we got 4 more for another fixture that has 2
RF-type ballasts of different design.  Both work properly with 2
pairs of old-school T8 fluorescent tubes but not with the new LED
tubes.
        The new LED tubes are 48-inch drop-in replacements for the
t8's and I seem to recall that the marketing information for the
LED tubes indicated they could work without ballasts if one
rewired the sockets.  The LED tubes are stamped with information
that they can work over a range of 100 volts to above 200 V so
they could be fed directly from the AC line without ballasts.

        In the ballast configuration, two gas discharge tubes are
in series on each ballast.  It sounds like the ballasts must double
the voltage so that each gas tube sees half that voltage or
somewhere about or slightly above the AC power line voltage.

        The LED tubes don't need the cathode heat voltage which
is somewhere around 1.5 to 3 volts and do need to see an
end-to-end voltage within the recommended range  for proper operation.

        Each old gas tube draws 40 watts and the new LED
replacements draw 18 watts each.  They also only shine in one
direction so that you orient the tube so that the clear side
points where you need the light.

        I was thinking about hooking one of the new tubes up to a
variac and transformer to see if my understanding is correct
before  ripping out the 2 electronic ballasts in the fixture that
does not work with the new LED's.  

        In theory, each tube has it's own current regulation
which is why the marketing literature mentioned ballast bypass
operation.

        The tubes come with stickers one can place on fixtures
indicating that old-style fluorescent tubes must never be used
here.

        Has anybody tried this already or is my understanding
flawed?

        When I discovered that the electronic ballasts in one
fixture didn't work correctly with the LED's, I picked out 4
newer glass tubes and was able to prove that both ballasts work.

        I would rather cart all the glass tubes off to our local
recycler, though, and use the LED tubes which save energy and are
not as fragile.  At 48 inches or 1.2 meters, they are just
begging to be accidentally hit/broken against some object when
one is climbing a ladder or carrying them around.  I've been
working with these for 30 years and finally bumped one against
the inside of a wooden box built around one fixture.  The blow
wasn't much but the tube shattered and made quite a low-grade
hazardous waste incident in our house.

        Any constructive ideas as to documentation are
appreciated.

        Martin McCormick
WB5AGZ
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[James Cameron-2]

I can't confirm the data you have on your product, but I've changed two
fittings over to LED tubes and have another nine not yet changed.  The
change for my tubes was to wire 240V to each end of the tube,
isolating the ballast and starter.

I halted my upgrade because;

* emission was not omnidirectional; with slanted white ceilings and
  the fittings suspended about 500mm below them, it felt oppressive
  not to have the ceiling lit,

* total lumens emitted to the floor area was not as great,

* the colour rendering index was poor.

Then I overspecified on solar panels and battery to compensate.  ;-)

--
James Cameron
http://quozl.netrek.org/
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Harold Hallikainen-3]

This discussion reminds me of just how interesting fluorescent lamps are.
Wikipedia has a nice article at
https://en.wikipedia.org/wiki/Fluorescent_lamp .

Many years ago, I did some work on fluorescent lamps for television
lighting. I always thought the ballast was interesting. An oscillator with
an H-bridge drove the lamp. The H-bridge drove the lamp through an
inductor. The H-bridge with series inductor was connected to one filament
pin at each end of the lamp. A capacitor connected the other filament pin
at each end of the lamp. Before the lamp fired, the LC was resonant
causing a high current through the filaments, lighting them. The resonance
also created a high voltage across the capacitor and across the tube. The
filament heating and high voltage across the tube would ionize the gas,
shorting out the capacitor. The inductor then became a current limiter.

I always thought that was VERY clever!

Harold



--
FCC Rules Updated Daily at http://www.hallikainen.com
Not sent from an iPhone.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Martin McCormick-3]

James Cameron <[hidden email]> writes:
> I can't confirm the data you have on your product, but I've changed two
> fittings over to LED tubes and have another nine not yet changed.  The
> change for my tubes was to wire 240V to each end of the tube,
> isolating the ballast and starter.

Thank you.  That sounds exactly like what I plan to do except
that it will be 120 volts at both ends due to this being North
America and 120 volts is within the voltage range specified on
the tubes, themselves.  First, out of cowardice on my part, I
will connect one LED tube to an isolation transformer and variac
set to around 120 volts to see if the tube both lights properly
and draws the amperage necessary to dissipate 18 watts.  If both
these things happen without anything going poof! or catching
fire, I'll be assured that I can junk the ballast which also contains
the starter circuit.

        All that does is briefly give a shot of a low voltage
of around 1.5 to 3 volts to the cathodes at each end so they
start producing electrons and the gas ionizes.

        I am thinking the LED tubes probably leave 1 of each end
pair of pins unconnected and the other caries 1 side of the
high-voltage pair.
The 120 or 240 volts potential is found at either pin on one end and
either pin on the other.

        If my work bench mockup turns up anything potentially
dangerous to us or things, I will post a message describing what
not to try.

        Martin McCormick   WB5AGZ
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Anthony Toft]

The Hyperikon tube I used came with new "tombstones" for the fixture. The deal was you took the ballast out and connected the new tombstones directly to the 120v, replacing the old existing fixture at one end. One end of each tube had a dummy fitting for mounting, with connectors at the other end. It was labelled so you knew which end to put in the connected tombstone.

As I remember it, the hardest part was getting the fixture abart to get the wiring sorted and the ballast removed.


On Wednesday, November 25, 2020 13:13 EST, "Martin McCormick" <[hidden email]> wrote:
 James Cameron <[hidden email]> writes:
> I can't confirm the data you have on your product, but I've changed two
> fittings over to LED tubes and have another nine not yet changed. The
> change for my tubes was to wire 240V to each end of the tube,
> isolating the ballast and starter.

Thank you. That sounds exactly like what I plan to do except
that it will be 120 volts at both ends due to this being North
America and 120 volts is within the voltage range specified on
the tubes, themselves. First, out of cowardice on my part, I
will connect one LED tube to an isolation transformer and variac
set to around 120 volts to see if the tube both lights properly
and draws the amperage necessary to dissipate 18 watts. If both
these things happen without anything going poof! or catching
fire, I'll be assured that I can junk the ballast which also contains
the starter circuit.

All that does is briefly give a shot of a low voltage
of around 1.5 to 3 volts to the cathodes at each end so they
start producing electrons and the gas ionizes.

I am thinking the LED tubes probably leave 1 of each end
pair of pins unconnected and the other caries 1 side of the
high-voltage pair.
The 120 or 240 volts potential is found at either pin on one end and
either pin on the other.

If my work bench mockup turns up anything potentially
dangerous to us or things, I will post a message describing what
not to try.

Martin McCormick WB5AGZ
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
 
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Martin McCormick-3]

"Harold Hallikainen" <[hidden email]> writes:

> This discussion reminds me of just how interesting fluorescent lamps are.
> Wikipedia has a nice article at
> https://en.wikipedia.org/wiki/Fluorescent_lamp .
>
> Many years ago, I did some work on fluorescent lamps for television
> lighting. I always thought the ballast was interesting. An oscillator with
> an H-bridge drove the lamp. The H-bridge drove the lamp through an
> inductor. The H-bridge with series inductor was connected to one filament
> pin at each end of the lamp. A capacitor connected the other filament pin
> at each end of the lamp. Before the lamp fired, the LC was resonant
> causing a high current through the filaments, lighting them. The resonance
> also created a high voltage across the capacitor and across the tube. The
> filament heating and high voltage across the tube would ionize the gas,
> shorting out the capacitor. The inductor then became a current limiter.

        Wow! that is clever.  Were the fluorescent lights being
driven at an RF frequency to avoid heterodyning with the
not-quite 30-HZ frame or not-quite 60-HZ field rate?

        I understand that monochromatic or black-and-white video
of the late 1940's and early fifties was meant to be exactly 60
HZ for fields and 30 HZ for the whole frame but color TV was
required by the FCC to take absolutely no more spectrum than
black-and-white so one way to make it all fit was to
ever-so-slightly slow down the frame rate so that 525 lines plus
the color burst at the beginning of each line didn't splatter in
to the audio carrier of the channel below the video carrier or
the guard band between the top of Channel X and the start of
Channel X+1.

        Conventional Fluorescent lighting has a lot of flicker at
120 HZ which human vision can't see but I bet it louses up video
pickups, both the old videcon tubes and probably CCD pickups used
today in solid-state cameras.

        People who are blind use so-called light probes to "see"
if lights are on or off such as anything from room lights to
indicator lights on panels.  These are RC oscillators with the R
being a photo cell such as a solid-state device or one of those
cadmium dysulphyde photo cells which are also called light
dependent resistors.  Light makes the resistance drop from almost
infinity in a dark room to around 1 K-ohm or less.  Light in the
mid spectrum range causes the oscillator to rise to several
kilohertz if seriesed with a .1 UF capacitor.

        If you hold one of these under an incandescent lamp, you
hear a little power-line frequency modulation of the whistle.
Under a fluorescent lamp, it's a regular growl at 120 HZ or 100
HZ in 50-HZ land.  You can even hear some 60-HZ modulation if you
move the photo cell near the ends of the fluorescent tube where
the cathodes are since each cathode alternates between being a
cathode and an anode every half cycle.

        Yes, Fluorescent lamps are fascinating and gas discharge
tubes such as the little NE2 and NE51 lamps of yesteryear are
really interesting gadgets since they do not conduct any
electricity until one reaches the breakdown voltage of the gas
and begins knocking those electrons out of their orbits which
make Ions and current flow.  

        Well, I guess I have gon off topic of my off-topic post
so I better sign off now.

Martin McCormick WB5AGZ
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Harold Hallikainen-3]

On the TV fluorescent lamps, yes the frequency was much higher than the
frame rate (though I don't remember what it was).

NTSC television is interesting! It's described at
http://mai.hallikainen.org/org/FCC/FccRules/2021/73/682/ . Prior to color,
the field rate was indeed 60 Hz, the power line frequency, to avoid
rolling "hum bars." With color, it appears everything is based on a
precise 5 MHz with the chroma subcarrier being 5 MHz * 68/88. The
horizontal scan rate is 2/455 * the chroma subcarrier frequency. The field
rate is then 2/525 * the horizontal scan frequency. I think that the
frequencies were chosen such that dots created by the chroma signal would
be white on one scan and black on the next, letting the eye cancel them
out. Similarly, the European PAL system switched the chroma phase 180
degrees on adjacent lines (since the image is interlaced, perhaps this
works out to just being a phase switch on each field). Chroma phase error
resulted in a hue shift, but in opposite directions on adjacent lines. So,
the eye averaged them out to avoid hue shift due to chroma phase drift.
Clever!

On light flicker, movie projectors used to use an incandescent "exciter"
lamp to light the sound track. Light would pass through the film to a
photocell. At first, the film density was varied to carry the audio, but
later the black to white area (width of a white or black stripe on the
film) was varied to carry the sound. Of course, if the lamp was powered by
60 Hz, you'd get 120 Hz hum in the sound. So, the obvious solution would
be to run the lamp on DC. But, many projectors instead ran the exciter
lamp on high frequency AC. There was a power oscillator to drive the
exciter lamp.

On the NE-2, one of my earliest project was an NE-2 based relaxation
oscillator running off a B battery. Choosing component values, I could get
a light flasher or get audio out of it.

Harold
http://w6iwi.org





--
FCC Rules Updated Daily at http://www.hallikainen.com
Not sent from an iPhone.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Jim]

The color burst frequency is 3.579545 Mhz.

Regards,

Jim

> -------- Original Message --------
> Subject: Re: [OT] Hyperikon Retrofit LED Tubes Without Ballast and
> fluorescent lamps and TV!
> From: "Harold Hallikainen" <[hidden email]>
> Date: Wed, November 25, 2020 5:41 pm
> To: "Microcontroller discussion list - Public." <[hidden email]>
>
>
> On the TV fluorescent lamps, yes the frequency was much higher than the
> frame rate (though I don't remember what it was).
>
> NTSC television is interesting! It's described at
> http://mai.hallikainen.org/org/FCC/FccRules/2021/73/682/ . Prior to color,
> the field rate was indeed 60 Hz, the power line frequency, to avoid
> rolling "hum bars." With color, it appears everything is based on a
> precise 5 MHz with the chroma subcarrier being 5 MHz * 68/88. The
> horizontal scan rate is 2/455 * the chroma subcarrier frequency. The field
> rate is then 2/525 * the horizontal scan frequency. I think that the
> frequencies were chosen such that dots created by the chroma signal would
> be white on one scan and black on the next, letting the eye cancel them
> out. Similarly, the European PAL system switched the chroma phase 180
> degrees on adjacent lines (since the image is interlaced, perhaps this
> works out to just being a phase switch on each field). Chroma phase error
> resulted in a hue shift, but in opposite directions on adjacent lines. So,
> the eye averaged them out to avoid hue shift due to chroma phase drift.
> Clever!
>
> On light flicker, movie projectors used to use an incandescent "exciter"
> lamp to light the sound track. Light would pass through the film to a
> photocell. At first, the film density was varied to carry the audio, but
> later the black to white area (width of a white or black stripe on the
> film) was varied to carry the sound. Of course, if the lamp was powered by
> 60 Hz, you'd get 120 Hz hum in the sound. So, the obvious solution would
> be to run the lamp on DC. But, many projectors instead ran the exciter
> lamp on high frequency AC. There was a power oscillator to drive the
> exciter lamp.
>
> On the NE-2, one of my earliest project was an NE-2 based relaxation
> oscillator running off a B battery. Choosing component values, I could get
> a light flasher or get audio out of it.
>
> Harold
> http://w6iwi.org
>
>
>
>
>
> --
> FCC Rules Updated Daily at http://www.hallikainen.com
> Not sent from an iPhone.
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Alan Pearce]

>  the European PAL system switched the chroma phase 180 degrees
> on adjacent lines (since the image is interlaced, perhaps this
> works out to just being a phase switch on each field).

Only the phase of the colour burst was changed on alternate lines so
that the phase lock of the colour subcarrier generator in the TV set
averaged out any phase shift in the transmission link. This gave a
more stable colour display than the NTSC system which could get
accumulated phase shift between the colour signal and the colour burst
(resulting in the moniker "Never Twice the Same Colour" for NTSC).

On Wed, 25 Nov 2020 at 23:42, Harold Hallikainen
<[hidden email]> wrote:

>
> On the TV fluorescent lamps, yes the frequency was much higher than the
> frame rate (though I don't remember what it was).
>
> NTSC television is interesting! It's described at
> http://mai.hallikainen.org/org/FCC/FccRules/2021/73/682/ . Prior to color,
> the field rate was indeed 60 Hz, the power line frequency, to avoid
> rolling "hum bars." With color, it appears everything is based on a
> precise 5 MHz with the chroma subcarrier being 5 MHz * 68/88. The
> horizontal scan rate is 2/455 * the chroma subcarrier frequency. The field
> rate is then 2/525 * the horizontal scan frequency. I think that the
> frequencies were chosen such that dots created by the chroma signal would
> be white on one scan and black on the next, letting the eye cancel them
> out. Similarly, the European PAL system switched the chroma phase 180
> degrees on adjacent lines (since the image is interlaced, perhaps this
> works out to just being a phase switch on each field). Chroma phase error
> resulted in a hue shift, but in opposite directions on adjacent lines. So,
> the eye averaged them out to avoid hue shift due to chroma phase drift.
> Clever!
>
> On light flicker, movie projectors used to use an incandescent "exciter"
> lamp to light the sound track. Light would pass through the film to a
> photocell. At first, the film density was varied to carry the audio, but
> later the black to white area (width of a white or black stripe on the
> film) was varied to carry the sound. Of course, if the lamp was powered by
> 60 Hz, you'd get 120 Hz hum in the sound. So, the obvious solution would
> be to run the lamp on DC. But, many projectors instead ran the exciter
> lamp on high frequency AC. There was a power oscillator to drive the
> exciter lamp.
>
> On the NE-2, one of my earliest project was an NE-2 based relaxation
> oscillator running off a B battery. Choosing component values, I could get
> a light flasher or get audio out of it.
>
> Harold
> http://w6iwi.org
>
>
>
>
>
> --
> FCC Rules Updated Daily at http://www.hallikainen.com
> Not sent from an iPhone.
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Martin McCormick-3]

"Harold Hallikainen" <[hidden email]> writes:
> On the TV fluorescent lamps, yes the frequency was much higher than the
> frame rate (though I don't remember what it was).

It could have been 40 or 50 KHZ and probably would have been just
fine.

> NTSC television is interesting! It's described at
> http://mai.hallikainen.org/org/FCC/FccRules/2021/73/682/ . Prior to color,
> the field rate was indeed 60 Hz, the power line frequency, to avoid
> rolling "hum bars." With color, it appears everything is based on a
> precise 5 MHz with the chroma subcarrier being 5 MHz * 68/88. The
> horizontal scan rate is 2/455 * the chroma subcarrier frequency. The field
> rate is then 2/525 * the horizontal scan frequency. I think that the
> frequencies were chosen such that dots created by the chroma signal would
> be white on one scan and black on the next, letting the eye cancel them
> out. Similarly, the European PAL system switched the chroma phase 180
> degrees on adjacent lines (since the image is interlaced, perhaps this
> works out to just being a phase switch on each field). Chroma phase error
> resulted in a hue shift, but in opposite directions on adjacent lines. So,
> the eye averaged them out to avoid hue shift due to chroma phase drift.
> Clever!
>

>From 1981 until January of 1989, I worked as a technician in the
Oklahoma State University Audio Visual Center where we fixed a
lot of 16-MM projectors and some video equipment.  

        One of our most interesting technical jobs was one we did
for a client who had a daughter who married a citizen of
Australia in the mid eighties and the family had a camcorder
that, of course, was built for the Australian version of PAL.

        The American members of that family, here, were eagerly
awaiting a new TV that that was available that would play videos
from about every standard available plus they had bought a
Panasonic VCR that would play everything, record most formats
except SECAM, but they still had no monitor.

        Their daughter and her husband bought a small monochrome
TV at a shopping center in Canberra and brought it all the way
hear, only to find that the VCR's remodulator wouldn't deliver a
signal on a frequency that the Australian TV would receive.

        What we did in the interim was to take a Commodore64
video monitor and adjust the vertical and horizontal hold to make
it sync up with PAL.  The lower vertical frequency of fifty
fields and 25 frames per second made for some overscan due to the
stronger magnetic field of the deflection coils at that frequency
but the picture was otherwise crystal clear except for no color,
Oh, I guess that's colour.


        I remember us commenting on the fact that it was June
26TH stamped on the screen of the recording from Australia which
was video of a local park as a little girl played outside.  In
the Northern hemisphere, this is normally a time of lush
vegitation but this was Australia and just at the start of Winter
so all the leaves were dead on the ground.

        I also remember examining the power plug of the
Australian set and noticing it was just like our power plugs
except that the two prongs are at 45-degree angles from each
other.  It was an interesting experience.

        We got the little TV to work by plugging it's mobile
power cord in to a 12-volt DC supply I had built that had a
mobile socket.

        The TV worked but I think the American side of the family
got their multy-standard TV and the little portable was
orphaned.


> On light flicker, movie projectors used to use an incandescent "exciter"
> lamp to light the sound track. Light would pass through the film to a
> photocell. At first, the film density was varied to carry the audio, but
> later the black to white area (width of a white or black stripe on the
> film) was varied to carry the sound. Of course, if the lamp was powered by
> 60 Hz, you'd get 120 Hz hum in the sound. So, the obvious solution would
> be to run the lamp on DC. But, many projectors instead ran the exciter
> lamp on high frequency AC. There was a power oscillator to drive the
> exciter lamp.

        Ah yes.  I Believe that one of the Bell&Howel 50's-era
projectors like the 399AV had a 50C5 oscillator which performed
that function.  I personally thought that projector had some of
the best film sound.  There was even a little lever that tweaked
the focus of the sound optics.

> On the NE-2, one of my earliest project was an NE-2 based relaxation
> oscillator running off a B battery. Choosing component values, I could get
> a light flasher or get audio out of it.

        At around 20 HZ, you could have a light flasher you could
hear.  Since the wave form was abrupt, you could listen to your
light flasher tick at even lower frequencies.

        If my 69-year-old memory serves me, some of the
electronic organs of by-gone days had a master oscillator octave
around High C and then got their base and mid-range notes by a
series of NE2 lamps wired as relaxation oscillators which
phase-locked on to the note above so only the master octave had
to be tuned and all the rest acted as binary dividers before
integrated circuits came along.

        I once tried to improve the squelch circuit on a tunable
VHF radio in 1968 by putting a NE2 lamp in the squelch circuit.
It worked beautifully if the under side of the chassis was
exposed to light but light helps ionize the neon and changes the
breakdown voltage so the ne2's trigger point kept changing after
I closed up the case.

        Martin
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Dave Tweed]

Harold Hallikainen wrote:
> NTSC television is interesting! It's described at
> http://mai.hallikainen.org/org/FCC/FccRules/2021/73/682/ . Prior to color,
> the field rate was indeed 60 Hz, the power line frequency, to avoid
> rolling "hum bars." With color, it appears everything is based on a
> precise 5 MHz with the chroma subcarrier being 5 MHz * 68/88.

That would be 3.863636 MHz, not even close to the correct value.

Jim wrote:
> The color burst frequency is 3.579545 Mhz.

The exact math is 30*525*455/2/1.001 = 3.579545e6 Hz

30 frames per second (60 fields)
525 lines per frame
455/2 cycles per line

The factor of 1.001 was chosen to make the sound subcarrier (fixed at 4.5 MHz
for backward compatibility) an integer multiple (286x) of the horizontal line
rate (15734.266 Hz), which reduced the effects of interference between sound
and both the Y and chroma components of the video.

-- Dave Tweed
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Harold Hallikainen-3]

>> NTSC television is interesting! It's described at
>> http://mai.hallikainen.org/org/FCC/FccRules/2021/73/682/ . Prior to
>> color,
>> the field rate was indeed 60 Hz, the power line frequency, to avoid
>> rolling "hum bars." With color, it appears everything is based on a
>> precise 5 MHz with the chroma subcarrier being 5 MHz * 68/88.
>
> That would be 3.863636 MHz, not even close to the correct value.
>

Oops! Typo. Copied from cited FCC rule, it's 63/88 * 5 MHz.


Also, thanks for clarification on PAL. I had only read about it and had no
experience with it.

Harold
https://w6iwi.org




--
FCC Rules Updated Daily at http://www.hallikainen.com
Not sent from an iPhone.
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Martin McCormick-3]

I did connect one of the tubes to 120 volts with the 2 pins
shorted together on each end.  It works fine with the current
draw at 120 volts being 15 milliamps for 18 watts as advertised.
it appears that 1 of the 4 pins at one end is unconnected.
Putting 120 volts  on the tube such that Neutral (white) connects
to the two pins on one end and Hot (black) to the other 2 pins
appears to be all that is necessary.

        If one lowers the voltage below 100 volts on the tube,
the current starts to rise so no dimmers.  The LED's do start to
dim but this apparently confuses the regulator circuit and the
current rises which doesn't sound like a good situation but I
feel confident in wiring up a fixture with all 4 sockets
parallel.

Martin
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[Alan Pearce]

> Also, thanks for clarification on PAL. I had only read about it and had no
> experience with it.

You are welcome. Thinking about it after sending I have a feeling that
the colour burst is +/-45 degrees on the required colour subcarrier
oscillator phase.

But you were correct, each alternate line does have the colour phasing
reversed. See https://en.wikipedia.org/wiki/PAL for a description of
the development and theory of PAL.


On Thu, 26 Nov 2020 at 16:48, Harold Hallikainen
<[hidden email]> wrote:

>
>
> >> NTSC television is interesting! It's described at
> >> http://mai.hallikainen.org/org/FCC/FccRules/2021/73/682/ . Prior to
> >> color,
> >> the field rate was indeed 60 Hz, the power line frequency, to avoid
> >> rolling "hum bars." With color, it appears everything is based on a
> >> precise 5 MHz with the chroma subcarrier being 5 MHz * 68/88.
> >
> > That would be 3.863636 MHz, not even close to the correct value.
> >
>
> Oops! Typo. Copied from cited FCC rule, it's 63/88 * 5 MHz.
>
>
> Also, thanks for clarification on PAL. I had only read about it and had no
> experience with it.
>
> Harold
> https://w6iwi.org
>
>
>
>
> --
> FCC Rules Updated Daily at http://www.hallikainen.com
> Not sent from an iPhone.
> --
> http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist

--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

[James Cameron-2]

On Thu, Nov 26, 2020 at 12:27:10PM -0600, Martin McCormick wrote:
> If one lowers the voltage below 100 volts on the tube,
> the current starts to rise so no dimmers.

You might try a leading- or trailing-edge dimmer rather than testing
with variac.

Variac will be giving you a sine wave with low amplitude.

Leading- or trailing-edge dimmers give you a cut in an otherwise full
amplitude sine wave.

Some retail packaged LED lamps render this more properly.

--
James Cameron
http://quozl.netrek.org/
--
http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist