LED as power on indicator with AC Mains Supply

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by MrAl » Fri Feb 12, 2010 7:42 pm

Hi there,


Here is a circuit that i have used before for a night light. It works equally well for an indicator light.

Note that this circuit uses either four diodes or a bridge rectifier for rectification, unlike many other
designs that only use one diode. This feature makes it possible to get twice the brightness from the
LED that would be possible with only one diode for the same value capacitor.
To get half brightness you can reduce the cap to half it's value shown on the schematic. 0.1uf will
get you about 1/3 brightness, which may still be enough for an indicator lamp.
The voltage rating for the cap should be at least 400v, or 250vac, but 600v would be even better.

Note that all the resistors are only 1/4 watt too.
It's also a very high quality and very safe design...safe both for the user as well as safe for the white LED.

Image
LEDs vs Bulbs, LEDs are winning.

jimmy101
Posts: 82
Joined: Tue Mar 24, 2009 11:08 am
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by jimmy101 » Sat Feb 13, 2010 10:25 am

jwax said:
jimmy101- I disagree about using the RMS value. The LED will indeed see the peak voltage.
And yes, I neglected to consider the reverse voltage breakdown of the LED, sometimes rated around 5 volts.
That being said, I connected a LED with a 56K resistor, applied 240 VAC, 60 Hz, and the LED works fine.
Open discussion!
I believe the RMS is the correct value for the power dissipated in the circuit (both the LED and any limiting circuitry like a resistor). Power needs to be time averaged in this application. For example, the final stable operating temperature of the resistor (and LED) is based on the RMS current and voltages and not the peak voltages. It is overheating that causes something like a resistor to fail, not power or current or voltage. Heat is a function of those three (P,I,V) and the ability of the device to dissipate heat and the devices heat capacity (power to temperature rise characteristics) and how long heat is being generated.

You see this often in digital control circuits for LED displays. For example, a CPU scans an LED display turning LED segments on and off in order. If a segment normally is operated at 20mA for 100% duty but because of the scanning is only actually on 10% of the time then the power through the LED is boosted significantly, the LED may well be fed 100mA so that it has sufficient brightness as it is blinked on and off. As long as the average power, over a sufficiently short time frame, is within specs then the LED will be fine. The LED can dissipate 20mA*1.7V=34mW (or so) even if it has brief peaks at much higher power levels, the average just need to be kept at or below 34mW.

Lets see, for your example, 56K ohm at 240 V (the LED drop is insignificant and within the tolerance of the resistor's resistance). There's 1 watt of power dissipated in the resistor at 240DC. If you use the peak voltage, say 320V then there is ~1.5 watts. There isn't that much power being dissipated by the resistor. Given the 50% duty and using the 240 RMS value there is about 1/2 watt being dissipated.

Now in terms of voltage, the LED (and resistor) will indeed be seeing the peak voltages. If you exceed peak voltage for a solid state device for a couple of milliseconds then you will almost certainly cook it. But in this case the peak voltage is still limited by the resistor even when the LED is reverse biased and you should be OK. The current will be spiking but the duty cycle is pretty low and the average power is obtained using the RMS value. The LED still drops ~1.7V unless the input voltage were to spike much higher than 240 or 330V.

User avatar
Bob Scott
Posts: 1192
Joined: Wed Nov 20, 2002 1:01 am
Location: Vancouver, BC
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by Bob Scott » Sat Feb 13, 2010 1:19 pm

I have a commercial night-light plugged into the wall upstairs. It uses white LEDs. It does not get warm, so it does not contain a high wattage resistor. Has anyone taken one of these apart to look at the circuit?

I saw a post about a year ago in the forum at electro-tech-online.com. Vaious posters gave various suggestions for driving a LED from line voltage AC. Darn, I can't find it! A poster named "pommie" had the best idea. Use a small capacitor in series in order to drop the voltage down. The cap has high impedance at line frequency. However, a cap tends to pass damaging high frequency pulses that can damage the LED, so he also uses a resistor pad in series, with another cap at the output to suppress transient spikes.

This ain't a lot of help, is it? I'd disassemble and examine the night light, or LED xmas lights.
-=VA7KOR=- My solar system includes Pluto.

User avatar
jwax
Posts: 2151
Joined: Mon Feb 09, 2004 1:01 am
Location: NY
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by jwax » Sat Feb 13, 2010 2:40 pm

OK jimmie101, I see what you're referring to now.

peterlonz
Posts: 8
Joined: Thu Feb 11, 2010 10:38 pm
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by peterlonz » Sat Feb 13, 2010 5:17 pm

Thanks guys,
I see some of the errors of assumption I made, & also how I came to calculate incorrectly the power dissipated by the 1M resistor.
The author of the circuit found by Externet stated:
"The resistor value was chosen to limit the worst case inrush current to about 150 mA which will drop to less than 30 mA in a millisecond as the cap charges."
So I calc Power = Resistance X current squared.
Acknowledging now he meant the 1K Resistor, the power dissipated here is still quite high at 0.9W & this resistor is specified as 1W, is that sufficient & how does the AC duty cycle affect things.
To Mr Al: thanks for your circuit, I thought you might have something up your sleeve & I will probably build your device along with the other circuit (Ex Externet).
One last comment; how important do you see the issue of spike suppression (in power strip) & has anyone any idea why my power strip employed a thermal fusible link?

Peter O

jimmy101
Posts: 82
Joined: Tue Mar 24, 2009 11:08 am
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by jimmy101 » Sun Feb 14, 2010 10:04 am

Looking at the circuit the MrAl posted and the rest of this thread, reminds that a few years back a friend that works for "Ginormous Gov't Military Contractor Of High End Electronics" says all the ubiquitous 120 VAC bus strips with the NE-2 type power indicator lights were being replaced in their labs. NE-2's have very long lifes but after 10 or more years of basically 100% duty the NE-2s start to dim and eventually burn out. Since the indicator light is a safety feature, and it is not fail-safe, the strips were all being replaced as being a safety hazard. (Admit it, we've all used the switch on a power strip to power up a circuit from a safe distance when we weren't 100% sure of what was going to happen :razz: )

Sounds like a potential custom chip/LED product opportunity. Build complete electronics similar to MrAls posted circuit into the LED package. Heck, put in multiple LED emitters to give life times up into the several hundreds of thousands of hours (many decades at 100% duty). Seems like a huge market so the costs per device should be pretty low.

I wonder, if you wired a red and green and blue LEDs in parallel would only the red LED light up? When the red LED eventually burned out then the green LED would come on? When the green burned out the blue would come on? IIRC, the color of the LED controls the voltage drop and the voltage drops are in the order red, green, blue.

peterlonz
Posts: 8
Joined: Thu Feb 11, 2010 10:38 pm
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by peterlonz » Sun Feb 14, 2010 3:32 pm

Jimmy,
I thought along exactly the same lines. But then realised until the issue of spike management is understood there is little point in building an indicator only module.
I can't see the need for providing an LED with a longer life than you'd expect from a de-rated standard 20,000 hr unit. Also I think failsafe in this case means you want the power strip to function even if the LED fails. That way you have indication that the strip needs replacing whilst it remains working. In my case the damned thing failed due to a components presence that I still can't understand & I nearly lost a freezer full of food.
BTW I replaced my unit with an identical strip costing Aus$7 (about US$6). I wanted to see what would happen, & I notice already that the end with the circuit board fitted is noticeably warm to touch which is not a great sign.

User avatar
CeaSaR
Posts: 1702
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by CeaSaR » Sun Feb 14, 2010 4:19 pm

peterlonz wrote:...the damned thing failed due to a components presence that I still can't understand & I nearly lost a freezer full of food.
BTW I replaced my unit with an identical strip costing Aus$7 (about US$6). I wanted to see what would happen, & I notice already that the end with the circuit board fitted is noticeably warm to touch which is not a great sign.
This begs the question, how many amps is the freezer pulling when in operation? If you are operating near the max
of the strip, you may be stressing the strip unduly, especially during start-up.

Next question, is the strip warm when the freezer is not running? If it is, then I'd take another look at the strip's
circuitry to see why.

Makes me want to check all mine now.

CeaSaR
Hey, what do I know?

User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by MrAl » Sun Feb 14, 2010 10:00 pm

peterlonz wrote:Thanks guys,
I see some of the errors of assumption I made, & also how I came to calculate incorrectly the power dissipated by the 1M resistor.
The author of the circuit found by Externet stated:
"The resistor value was chosen to limit the worst case inrush current to about 150 mA which will drop to less than 30 mA in a millisecond as the cap charges."
So I calc Power = Resistance X current squared.
Acknowledging now he meant the 1K Resistor, the power dissipated here is still quite high at 0.9W & this resistor is specified as 1W, is that sufficient & how does the AC duty cycle affect things.
To Mr Al: thanks for your circuit, I thought you might have something up your sleeve & I will probably build your device along with the other circuit (Ex Externet).
One last comment; how important do you see the issue of spike suppression (in power strip) & has anyone any idea why my power strip employed a thermal fusible link?

Peter O

Hi,

Spike suppression is built into power strips to prevent spikes from damaging equipment that is plugged into it.
The thermal fuse is there to protect against surges that may cause the MOV to heat up too much. If the
surge lasts too long this would happen, so the thermal fuse cuts out to stop current flow.
LEDs vs Bulbs, LEDs are winning.

jimmy101
Posts: 82
Joined: Tue Mar 24, 2009 11:08 am
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by jimmy101 » Mon Feb 15, 2010 10:04 am

Peterlonz

I don't think 20,000 hours is nearly enough. A single year at 100% duty, which is nothing for a bus strip, is 8760 hours. 20K hours is less than three years. Figure the lifetime of a generic bus strip (ignoring any surge protection capability) is probably 10+ years that means the LED will almost certainly burn out. I know the NE-2 will generally burn out after that long. Even if you operated the LED at a low power level and got the lifetime up to 100K hours that's still just 11.4 years at 100% duty.

"Failsafe" generally means a device fails in the safest possible mode. For a device with an "On" indicator I would expect that to mean if the indictor says the device is off then it is indeed off. Off is generally safer than on for electrical devices. An on state when the indicator says off is the reason why the Ginormous company was replacing all the bus strips. Failsafe generally does not mean that the system continues operating, it means it isn't a safety hazard when it fails. True failsafe devices are kind of uncommon.

In the case of a freezer "safe" means the power is off even if that means the food in the freezer rots.

A failsafe brake system locks up the brakes. That's what happens with air brakes on big trucks and trains, a significant leak in the hydraulic system means the vehicle's brakes are locked up and it can't be moved.

A prime example of a device that "fails-unsafe" is a typical Geiger-Muller radiation counter. When the GM tube experiences a radiation level that saturates the tube the meter's reading drops to zero instead of pegging off scale. The operator thinks there is zero radiation when in fact the radiation level is way off the counter's scale.

peterlonz
Posts: 8
Joined: Thu Feb 11, 2010 10:38 pm
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by peterlonz » Mon Feb 15, 2010 3:46 pm

I don't wish to drag out indefinitely this power strip design issue but I think some points raised are worth a reply.
First re safety:
Most early power strip designs had no mcb, no indication & generally no switch.
Later models introduced either single switches or individual socket switching & in some cases neon indication, again sometimes a single indicator & sometimes an indicator for each socket.
Then the resettable mcb ..... etc etc.
Now, most home switchboards employ circuit breaker protection, & many also feature RCD protection, circuits have generally been uprated from 10 to 15 amps where circuit breakers are employed.
This means that the switchboard properly offers an excellent level of protection as practical experience demonstrates.
So any power strip that trips at say a "marginal" 10 amps is likely to be a nuisance sooner or later, if it's the resettable mcb no problem, but if it's a failed component resulting in a non-working power strip, well that's a more fundamental issue.
In my case I might mention the power strip feeds a small freezer, medium sized refrigerator, rotary clothes drier, & two small battery rechargers. Sounds a lot but before this latest failure I have had no unwanted tripping of either power strip or switchboard circuit breaker.
"Failsafe" I guess is well defined by jimmy, can't argue with impecable logic.
And yes of course your large enterprise would replace aging strips, when the indicators no longer functioned, is that not a no brainer in that environment?
The packaging of my strip says: "If during use the surge indicator lamp goes out, the protection circuit has probably absorbed a large power surge & its surge absorption capacity (525 joules) is spent. This means the unit is no longer protecting your equipment & must be replaced, however it can still be used as a regular power board".
This claim is clearly not correct but indicates to me that the designer saw clear merit in not failing to safety, IE the board should remain functioning.
My replacement strip varies from mildly warm to blood temp at the circuit board end, & most times I check, the demand is zero, IE all appliances are dormant.
I don't believe it failed due to a power surge.
As to why anyone would want surge protection beyond that offered by their mains supply, particularly that which at best, could only be a meagre offering from something worth $7 is beyond me.
I simply liked the switch & indicator feature & I know that decent surge protection costs $$$$.
I'd go further & say the average buyer would not have a clue about surge protection, except perhaps since the advent of expensive home cinema set ups, quite sophisticated & expensive power strip style surge protectors have become available.
So it looks as though this design mas MOV's (don't know anything about their use in practice but they used 3 X GNR 14D471K) which somehow were protected with a thermal fusible link despite having a 10A mcb.
Looks to me like to much protection in a very marginal design.
Peter O

User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by sofaspud » Tue Feb 16, 2010 12:50 pm

Instead of a power-ON indicator, you could think in terms of a power-OFF alarm. The world's largest retailer has one consisting
of a few white LEDs and rechargeable batteries. My local store has it priced at just under US$6. You could easily add an audible
alarm to the unit. I'm not thrilled with the AC LED option, but please make sure the circuit is fused upstream if you decide to
implement it! Are you sure there's not a problem with the freezer itself? It's hard to imagine it's pulling that much juice at 240VAC
(unless it's a walk-in type). :lol:
"...the power strip feeds a small freezer, medium sized refrigerator, rotary clothes drier, & two small battery rechargers."
Wow. That's sounds like an awful lot to me. No wonder things heat up.

peterlonz
Posts: 8
Joined: Thu Feb 11, 2010 10:38 pm
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by peterlonz » Tue Feb 16, 2010 9:59 pm

The combined coincidental load on this power strip may be quite high.
The rotary clothes drier would almost certainly be the biggest draw.
The freezer is the smallest of the chest type.

I simply have not bothered to check max demand here because:
If the mains feed circuit is overloaded the main switchboard Circuit breakers can trip, but in 5 years that has not happened.
The power strip itself is fitted with a 10A mcb which also has not tripped in over two months of use with this particular (suspect) power strip design.
The 4-way power strip previously used in this exact application offered zero protection features but remains fully functional & "unscathed" in another application around the house.
So I remain confident that overloading is not an issue here.
Today I measured the outer temperature of the plastic case: 32 Deg C; I am confident that I have felt temperatures well above this possibly 40 Deg.
I am not sure why there is distrust about AC powered LED's, but nor can I claim to fully understand exactly how they work or what risks are attached to such designs. Please be a bit more specific about the size & type of an upstream fuse. If the fuse fails its going to be a problem. I believe too much protection just leads to unpredictable performance.
Love the idea of a power off alarm, BUT having just dismantled a Canon digital camera & Aiwa miniature tape player, both with hidden & very difficult to access internal rechargeable batteries, .......... well I'm very nervous about that option.
Peter O

User avatar
sofaspud
Posts: 531
Joined: Wed Feb 13, 2002 1:01 am
Location: San Antonio, TX
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by sofaspud » Wed Feb 17, 2010 8:19 am

"So I remain confident that overloading is not an issue here."
But I'm hopeful you can admit that by the looks of it, you're certainly pushing it. What I do is run a heavy duty extension cord from
the next nearest branch. I'd do that for at least the clothes dryer, or more probably everything but the dryer. I can't tell you how
many orange cords I have in use or at the ready at my home.
"I am not sure why there is distrust about AC powered LED's, but nor can I claim to fully understand exactly how they work or
what risks are attached to such designs."

It's not distrust (carefully implemented), I just think there's got to be a better way. LEDs are low-voltage DC components, but it is
possible to use them in a high-voltage AC circuit. Keeping in mind that a shorted-component failure is a serious fire hazard. Don't
want to cook a freezer full of food all at once! You could place the fuse in series with the LED circuit, remembering that the LED
indicator may be off but the circuit is still live. Instead, I'd find an old alarm clock/radio and recycle its small transformer to power
the LED.
"If the fuse fails its going to be a problem."
As it should be.
"...hidden & very difficult to access internal rechargeable batteries..."
The unit I saw was 6"x3"x1" in a plastic case. I doubt the internals are hard to get at. If desired, you could gut the innards and put
them in your own custom case, with or without the aforementioned sonic alert.

User avatar
CeaSaR
Posts: 1702
Joined: Sat Nov 08, 2003 1:01 am
Location: Phoenixville, PA USA
Contact:

Re: LED as power on indicator with AC Mains Supply

Post by CeaSaR » Wed Feb 17, 2010 1:47 pm

Here's a managerial thought:

Why not 2 circuits in the power strip - 1.) first one goes across the incoming supply to indicate that power is actually
applied, probably red to indicate high voltage, and: 2) second one is after fuse/circuit breaker to indicate if outlets are
live or dead, probably green to show status of plug-in devices.

Each should be it's own stand alone unit, i.e. uses it's own power supply. A pair of tiny flat pack transformers or whatever
other scheme you use, 1 for each. I'm still partial to transformers, but MrAl's solution looks to be smaller and less expensive.

CeaSaR
Hey, what do I know?

Post Reply

Who is online

Users browsing this forum: No registered users and 13 guests