electrickery.

[Phil Pascoe]

We had a period of low voltage yesterday for a few hours. Some things kept working, some went off. When the power came back on one of the kitchen LEDs didn't, and curiously one of two different LEDs wired together in the workshop didn't. Strangely, though, the one that came on later didn't and the one that didn't later did (still does). Anyone any suggestions? If the low voltage has knackered anything why didn't they just cut the power altogether instead of seriously reducing the voltage?

 

[John Brown]

Should've stuck with incandescents!

Seriously, I can't think of a reason for this, unless the lower input voltage made the current excessively high in some part of the circuitry. How low did the voltage go?

 

[Phil Pascoe]

I don't think they actually said - it was announced as a power cut. The whole area had trouble with some things working and some things not.

 

[John Brown]

Well since LED drivers tend to rely on switching techniques, I'd guess that either too much current damaged something, or the circuit stop working properly and something stayed on when it should have been switching rapidly. Either way you'd expect some sort of safeguard, but things are built to a price.

 

[Doug71]

Similar experience in my workshop last week, can't answer your question but feel the need to have a little moan.

When I turned on my workshop lights only one came on, it's and LED all the rest are fluorescent tubes and they stayed off. I tried the 240v sockets , some things ran slow and some didn't work at all, nothing worked on the three phase. My workshop doesn't have any windows so it was too dark to do anything so I used the opportunity to nip to the local Screwfix etc as I needed a few bits and pieces. On my way I passed some pylons where there were 6 Northern Powergrid vehicles parked, I thought "Ah, there's the problem." When I got closer I saw there were 2 blokes sat in deck chairs and about 10 blokes having an impromptu cricket match using a bit of wood and a tennis ball. Now maybe they were on their break or maybe they were awaiting instructions but they could have tried to look a little more industrious while the likes of me were losing money because of the power problems

 

[Spectric]

it's and LED all the rest are fluorescent tubes and they stayed off. I tried the 240v sockets , some things ran slow and some didn't work at all, nothing worked on the three phase.

When the voltage drops, reffered to as a brownout so does the available power and because LED's are low energy devices they will work, in some the PSU being a switching device will have a wide input votage range from 96 to 240 and so can still function. With fluorescent tubes that use a lot of energy to trigger they will not funtion and in electronic systems you have what is called a brown out circuit that responds to a dip in supply voltage to reset the device to ensure no damage is caused and it is in a known state.

 

[BucksDad]

Perhaps like water, they are now conserving electricity for the tourists in Cornwall

 

[Phil Pascoe]

They have advised us to be economic with the water to ensure there's enough for tourists - in 1976 they advised tourists to be careful with the water to ensure there was enough for the locals.

 

[John Brown]

When the voltage drops, reffered to as a brownout so does the available power and because LED's are low energy devices they will work, in some the PSU being a switching device will have a wide input votage range from 96 to 240 and so can still function. With fluorescent tubes that use a lot of energy to trigger they will not funtion and in electronic systems you have what is called a brown out circuit that responds to a dip in supply voltage to reset the device to ensure no damage is caused and it is in a known state.

Like you, I would have expected LED lights to function over a range of 90 to 240 or thereabouts. It would make sense in a global market, but could be cheaper to manufacture two versions, with a one or a few component options.
The bottom line is that Phil's LED units have apparently failed after being subjected to a low voltage, which seems counter intuitive, but without seeing a schematic it's hard to know what might have happened.

 

[Phil Pascoe]

It's also odd that two out of eight identical units have failed. Same lamps, same age. House wiring all checked since they've been fitted.

 

[guineafowl21]

I’d guess a transient/voltage spike from the switching event.

If the LEDs are capacitive droppers, maybe a shorted LED or rectifier took out the little fusible resistor.

if switching type, should have coped better but a shorted component still possible.

Send the duff units to BigClive.com - he might do a teardown and analysis video. Just his sort of thing.

 

[Spectric]

We don't know the make or cost of these lights but this can be something that happens if you buy a cheaper product which

could be cheaper to manufacture two versions, with a one or a few component options.

It might not be the low voltage that has caused the issue but if there was a spike, like if you switch something off and back on to fast which again comes back to fewer components like not having anything like a MOV on the input.

 

[Spectric]

It's also odd that two out of eight identical units have failed. Same lamps, same age. House wiring all checked since they've been fitted.

Maybe those two took the spike and saved the others, are they first in the supply ?

 

[MorrisWoodman12]

Like you, I would have expected LED lights to function over a range of 90 to 240 or thereabouts. It would make sense in a global market, but could be cheaper to manufacture two versions, with a one or a few component options.
The bottom line is that Phil's LED units have apparently failed after being subjected to a low voltage, which seems counter intuitive, but without seeing a schematic it's hard to know what might have happened.

Just had a look at the packaging on a LED ball lamp. It has tick boxes for 220V, 110V, 24V and 12V. 220V being ticked. This implies that the voltage range is not similar to lots of today's switched mode PSUs. I remember when early SMPSUs were 190V to 240V and not full range 96V to 240V; maybe LED bulbs are similar.
Remember that the LED PSU will try to supply the LEDs with constant power so the power drawn from the supply will be constant more or less. As the voltage drops the current will increase so something may have blown due to too much current.

 

[Stuart Moffat]

I wonder what the probability is that all the domestic feed in systems are all 100%effective in turning off when the network fails. Is there a possibility that very low voltage could arise from a defective domestic PV system locally? I know it shouldn’t happen in theory.

 

[John Brown]

Just had a look at the packaging on a LED ball lamp. It has tick boxes for 220V, 110V, 24V and 12V. 220V being ticked. This implies that the voltage range is not similar to lots of today's switched mode PSUs. I remember when early SMPSUs were 190V to 240V and not full range 96V to 240V; maybe LED bulbs are similar.
Remember that the LED PSU will try to supply the LEDs with constant power so the power drawn from the supply will be constant more or less. As the voltage drops the current will increase so something may have blown due to too much current.

Do the check boxes also imply a common PCB with component options?

 

[Phil Pascoe]

Maybe those two took the spike and saved the others, are they first in the supply ?

No, they're in different places. One is wired with another different LED that's OK.

 

[Spectric]

As the voltage drops the current will increase so something may have blown due to too much current.

Again pointing to the possibility of not having full protection circuitry, if you respond to the under voltage then the overcurrent should not happen but also keep in mind that the markets are flooded with counterfeit components so more possibilities arise and it is a major issue for the OEM's.

 

[Scruples]

In an area near you there might have been a neutral inversion event where the neutral becomes disconnected at the substation. Depending on the load variations at the time, the voltages will rise and fall in different areas leading to high voltages in some places and low voltages in others. It results in some equipment failing, damage to others, lile fluorescent chokes, and motors having there windings loosened and some equipment remaining, seemingly unharmed.
Any damage is usually caused by voltage spikes which take out components of lower quality, or tolerance.

Rural areas use to be more susceptible to neutral inversions and I think there might well be better electronic protection against it happening for long, these days. It would be worth asking Western Power what the fault was, if you're that interested.

 

[pgrbff]

I disconnect fridges and freezers during a low voltage period. I believe the lower voltage means higher current which can damage the mtor/pumps.