Spent probably 12 hours last week reading up on fluorescent starting methods. I think I’m about done with that. I think my autistic obsession was just fueled by the fact that fluorescent starting methods are just grossly over simplified. And for me personally, I just don’t like having an incomplete understanding of something. So when you have a starting method such as “rapid start” that is clearly described but doesn’t match the current starting method of my electronic ballast that is described as “preheat rapid start”, that just sent me down a rabbit hole. Thankfully with the wayback machine/archive.org I was able to find some old websites that had decent descriptions of the current evolution of ballasts in the late 1990’s.
The first question to answer, is why is a ballast needed? Unlike incandescent bulbs when switched on, have a resistance that rapidly increases, fluorescent lamps are the other way around. Once a fluorescent lamps begins to flow current, the resistance to current rapidly goes down. Hence a ballast is needed to provide a steady current consumption otherwise the lamp would just quickly burn up.
Now as to starting methods.
First there was the old “preheat method” which used the starting circuit everyone is familiar with. When switched on, current was initially directed onto the filaments of the lamps to heat them up. After a second or so, the starting circuit opens, which causes the magnetic field now built up in the ballast to collapse and if it lines up properly with the current A/C phase, a couple hundred volts of potential are placed onto the ends of the lamp with the end goal of “striking” an arc and the lamp turning on.
The next evolution of starting methods after that was the “rapid start” method. This did not require a separate starting circuit. Instead, when a “rapid start” ballast was switched on, current was applied to both the filaments and the lamp ends. As the filaments heated, the required “striking” voltage of the lamp would decrease and would eventually get down to the same potential voltage being applied at the lamp ends where the lamp would then “strike” and turn on.
But both of these methods are very old tech and they had two purposes: to be very cheap (both preheat and rapid ballasts are nothing more than wires wrapped around iron) and combined with red hot filaments, to bring the striking voltage of a lamp to something below 300 volts. (because there was no cheap way to get voltage above 300 at the time)
After 1990, things began to change. With semi-conductor price drops, it was very cheap and easy with just a couple of transistors and inductors to make a circuit that would take the 120 volt line input, convert it to DC, then convert that to an oscillating sine wave in the kilohertz range. Under no load, these types of circuits have the behavior of quickly producing hundreds and/or even thousands of volts. A prime candidate for starting a fluorescent lamp as when the lamp is first switched on, the circuit is seen purely as open, voltage quickly rises and nearly instantly strikes an arc and the lamp is now on.
The question now becomes, what kind of start up method is this? The term “instant start” actually refers to another starting method that was popularized in the 1940’s. But “instant start” in this context referred to an entire fluorescent system requiring an “instant start” ballast and “instant start” lamps. The lamps in this case had no filaments, just a single pin on either end of the lamp. This system is so old I can’t really find any kind of schematic of a true 1940’s “instant start” ballast but I believe it was just a transformer hooked up straight to the lamps in order to bump up the voltage. This system was simpler, but had a reputation of lamps not lasting as long in frequently switched off and on situations. As such, and this is purely my opinion, I believe ballast manufactures have avoided labeling ballasts as “instant start” to avoid confusion and reputation with the older system.
Also, there are many standards that ballasts and lamps must follow such as:
ANSI C82.2-2002, American National Standard Methods of Measurement of Fluorescent Lamp Ballasts
ANSI C82.3-2002, American National Standard Specification for Fluorescent Lamp Reference Ballasts
ANSI C82.13-2002, Definitions for fluorescent lamps and ballasts
So one must abide by the rules, and as such it was easiest to have a ballast fall under “rapid start” than any other. The standards are all behind paywalls so I can’t really verify anything, but I assume “rapid start” is simply defined as a starting method whereby one applies voltage to the filaments and the lamp ends simultaneously. Which with an electronic ballast, that is not hard to do as one just needs to connect the filaments in series with a small capacitor which will bleed a negotiable amount of current through them and thus meet the definition. But this is just my opinion.
There seems to be some folklore about starting up lamps without proper filament heating time. Proper heating leads to more on off cycles (but not necessarily longer lamp on hours). It seems both instant starting and rapid starting can give around 15k on/off cycles. There are a few sources proving this,
but this one seems to offer the most details as it seems to clearly be testing regular "rapid start" lamps with an "instant start" ballast:
With all that said and doing some math, I'm not usually turning this office light on more than once a day. Typically one cycle a day, sometime two or three. So being generous, a max of 1000 cycles a year would be 15 years to get to 15k cycles. The lamp it self it be dead long before then so I probably shouldn't be worrying about whether my filaments are properly heated or not
Took me like 5 hours to write this up
