Use of 120 VAC neon mini lamp on 240 VAC circuit

"Serious Machining" schreef in bericht news: snipped-for-privacy@z14g2000cwz.googlegroups.com...

Dennis,

(Almost) all neon indicator lamps fire at around 90V, light at around 70V and extinguish at around 60V. I guess the "firevoltage" to be the normal operating voltage. So the 120V lamp should already have a current limiting resistor inside. To account for the extra 120V at 270uA you need a

444k/0.032W resistor. So one 470k/0.25W resistor will do fine. If the light is too dim you can try a 390k resistor as well.

petrus bitbyter

Pieter, Thank you for your prompt and thoughtful response, I really appreciate your time and informative advice. Dennis A.

The Radio Shack 272-704 neon lamp has a clear plastic case (with the thread built-in -- don't overtighten!). If you look at the side of the lamp, you'll see the resistor. It's about 47K, I believe.

Here's how it works. The Ne bulb has a 90V trigger voltage, and after firing, it stabilizes at about 70V. For 120V line voltage, the lamp will see a peak of 170V. While the lamp is on, only 70V is across the lamp, which means that a peak of 100V will be across the resistor. That will mean you'll have about 2mA peak going through the resistor (which is a good, safe value).

With 240VAC, there's a peak voltage of about 340V. 70V of that will be across the Ne bulb, so you have to choose a total R so the peak current will be around 2mA. Ohms Law comes in handy here:

270V / 2mA = 135,000 ohms

Since you've already got a 47K internal to the lamp, try an 82K to get close to the above. It's possible 390K additional resistance at 240VAC might not look as bright as the lamp with 120VAC.

Good luck Chris

"Serious Machining" schreef in bericht news: snipped-for-privacy@z14g2000cwz.googlegroups.com...

889K

assembly.

the

Dennis,

(Almost) all neon indicator lamps fire at around 90V, light at around

70V and extinguish at around 60V. I guess the "firevoltage" to be the normal operating voltage. So the 120V lamp should already have a current limiting resistor inside. To account for the extra 120V at 270uA you need a 444k/0.032W resistor. So one 470k/0.25W resistor will do fine. If the light is too dim you can try a 390k resistor as well.

petrus bitbyter

I picked up a pack of the Radio Shack 470K ohm, 1/2 watt resistors (271-1133) they all checked out at 464K +/- 1K ohms, a deviance of only

1.3% for a tolerance of 5%. I pulled the red lens body off the 272-704 neon lamp and saw the built-in resistor. The four colored bands appear to be: orange, orange, orange, gold - which should yield 33,000 Ohms @ 5% tolerance. I checked the brightness of the neon bulb in four increments: 1) As is, no addt'l resistor, @ 120 volts: Medium brightness, detectable in lit room. 2) As is, no addt'l resistor, @ 240 volts: Very bright, calls attention to itself. 3) One 470K resistor added @ 240 volts: Less bright than #1, s/b usable in dim light. 4) One 470K resistor added @ 120 volts: Lit fine - just very dim, barely visible. I think I'll wire it up as per #3 since I expect it will be "ON" perhaps as much as 5 hours per evening, 6 months of the year. Despite this amounting to around 900 hours per year, the bulb should last 28 years - probably more since it won't be as bright as the "rated" output

- although the frequency of its ON/OFF cycles might offset this. I'll probably wire up the second neon lamp as per #4 to serve as a "light switch locator" on a 120 volt circuit. It will always be "ON", should still last a very long time, and should be easily enough to detect at close range in the dark of night. Thanks also to Chris and Ralph for their follow-up postings - Chris for reminding me to consider the actual 'peak voltage' associated with the 'averaged' 120 and 240 volt AC mains. The prediction that test #3 above would not be as bright as test #1 was right-on. Ralph's comments gave me the courage to conduct test #2 - very useful for the semi-quantifiable comparison of the four tests.

- Dennis A.

The neon bulbs are not very current (voltage) sensitive. As long as the resistance is low enough for them to light and high enough to keep the current under the maximum you are ok. Anywhere inbetween it is just a manner of how much light you want and how long you want them to last. The pocket voltage testers go anywhere from about 90 volts to 600 volts with the same resistor. They are just brighter with the higher voltage . As they are only used for a short period of time the life of them is not usually a problem.

>

"Chris" schreef in bericht news: snipped-for-privacy@o13g2000cwo.googlegroups.com...

Chris,

I don't have the lamp, only the figures provided by Radio Shack.

|(272-0704) Specifications Faxback Doc. # 34728 | |Wattage Consumption:..............................................32.4 mW |Insulation Resistance:................500 Volts, DC, 100 M-ohm., Minimum |Dielectric Strength:..........................1500 Volts, AC, one minute |Life:...................................................Over 25,000 hours |Unit:.......................................................120 Volts, AC |Fire Voltage:................................................70 Volts, AC |Fire Current:.....................................270 micro Amps, minimum |Hole, Drill:....................................................7/16 inch |Lugs:..........................................................Solderable |Housing:........................................................Black ABS |Body:..................................................................PC | |Specifications are typical; individual units might vary. Specifications |are subject to change without notice.

They state clearly 120V/34.2mW and that's where my calculations were based upon. Nevertheless you may be right. Can't trust those specifications, can you? After all they say that they may "change without notice". From my own experience I know these neon indicators come in current ranges from 0.22mA to over 1mA. The low current types are advised to be used with a 220k series resistor at 120Vac and with 750k at 240Vac. Accordingly the high current types are advised to be used with 56k and 180k. So I stick at my advise to try the 470k first and lower it when the light is too dim. If you can really see a 47k or 56k resistor inside, you can go down to 150k or even 120k.

petrus bitbyter

Pieter,

Thanks for your latest helpful follow-up to my inquiry, that's really nice of you ! I will experiment with some of the other 10K & 100K Ohm series of resistors offered by Radio Shack. It would be interesting to actually be able to meter the light output intensity, but I suppose one can settle for making an estimate of the brightness by means of mathematical formulas. Even that would be fun. Take care and have a good day.

Dennis A.

Pieter,

Thanks for your latest helpful follow-up to my inquiry, that's really nice of you ! I will experiment with some of the other 10K & 100K Ohm series of resistors offered by Radio Shack. It would be interesting to actually be able to meter the light output intensity, but I suppose one can settle for making an estimate of the brightness by means of mathematical formulas. Even that would be fun. Take care and have a good day.

Dennis A.

Ed -

Thank you for your detailed explanation on how to determine which resistor to add to the circuit - yes, I was wondering what sort of consideration in the calculation was to be given to the 70 volts.

Now I have a new question about knowing which end to start with when deciphering the color coded banding, but I'll save that for a later date. I don't want to get booted out of this group after being a member for only 3 days ! Peace be to everyone.

- Dennis A.

"Serious Machining" schreef in bericht news: snipped-for-privacy@g44g2000cwa.googlegroups.com...

Dennis,

Apparently Radio Shack *did* change the specifications without notice. This indicator will consume over 200mW instead of 32mW. To achive the same brightness you found at 120V you may will have to add a 82k or even a 68k resistor while using 220V. But if you're content with the current brightness, keep it as it is now.

petrus bitbyter

Why?

Why not? Isn't the neon dropping ~70V while it's illuminated? I'd subtract that from 120, giving 50V across the 33K, for ~1.5 mA. To get 1.5 mA through 240 - 70 = 170V, you'd need 113.333K total resistance. 120K is the closest standard value, or, to put in series with the 33K leaves 80K, which 82K is the closest standard value.

Cheers! Rich

Now that you know what's in there (the resistor value), you can easily determine the total resistance needed: double the existing resistance. The resistor acts as a current limiter. You compute it based on imagining the neon as shorted. The ~70 volts across the neon is not needed in the computation, because of your research (disassembling the indicator). So 33000 ohms at 120 needs to be changed to at least 66000 ohms at 240. I'd go a bit higher and put a 47K or 68K in series to give you

80K - 101K total.

Ed

--
?????

that'll work. but with neons there's either current flowing (in which case the voltge drop in your neon is well below 70V or no current flowing in which case there is no voltage drop in the resistor.

so reaching or maintaining 70V isn't going to be a problem

don't forget to use mains rated resistors.

>

--

Bye.
   Jasen

Yes, you do need the 70 volts. You substract 70 from either 120 or 240 to get the voltage to use in the calculations for the current.

Resistors do have voltage ratings, but any I have seen are good for a couple of 100s volts.

No, to both you and Rich. You guys are correct in computing the current drawn - you must use the ~70 volts in the computation. But we're not trying to *supply* a particular current through the neon with the resistor choice, we're trying to *limit* it to some maximum. Neons work over a wide range of voltage (and therefore wide range of current) but the current must be limited to some maximum.

Yes to both you and Rich, if the Radio Shack neon assembly resistor was improperly computed by its designer by the inclusion of the voltage drop in the computation.

Ed

I should add that you can use the 70 volts in the computation if you want to - I don't mean to indicate doing so will ruin the circuit.

Ed