Is this email a scam?

General Discussion
21

Some are very difficult to spot - o vs ο would be a good one (latin lower case o vs lower case Omicron), or c vs ϲ - different yet almost identical.

As has been said - have a high index of suspicion and always type URLs by hand.

The problem is that LED is much more efficient in terms of power in vs light out, the perceived quality of the light is probably due to the spectral content - incandescent is continuous, but because “white” LEDs are actually blue LEDs with phosphors added for the red and green components there are discontinuities - though there are now “wide spectrum” LEDs which overcome this (mainly for specialised applications at present).

LED is *way* better than CFL though.

They are much better than they used to be - cheap ones can still be a bit naff though.

I’m not generally in favour of these YT videos which encourage people with no knowledge of mains electricity to work on equipment with non-isolated power supplies (or poorly isolated ones which rely on little more than a wing and a prayer and very little clearance to keep mains voltage away from the low voltage side).

That said I did fix the light in the loo in more-or-less exactly this manner as the soldering on one of the LEDs had failed and I could not get an identical replacement for the lamp (it was the only one that we found which would fit exactly into the space available). Shorting out the LED restored the lamp to almost 100% functionality (it’s now a 5.5W lamp, not a 6W lamp).

1 Like
22

There are several factors with LEDs that make them less ideal (they’re still pretty good, if selected carefully). FWIW I’ve been using LEDs for quite a while, originally ordering separate units and drivers from US companies, building my own bike lights, then as bulbs started to become available for mains voltage being an early adopter.

Colour temperature is one, as already said. They tend to emit a spectrum that can seem to be quite harsh on the eyes, even when ‘warm’. Temperature may also be inconsistent especially between manufacturers, with some 3000k units being distinctly orange. Obviously the solution is to find a reliable make, though getting the right recommendations can be tricky too, if you don’t want trial and error.

Point source rather than a broad source can make for harsh lighting and unexpected patterns from shades. We use a lot of spots and finding LED spots with a pleasing beam pattern can be difficult, since makers used to hide beam angle in the specs - it’s better these days, but can still be challenging. Also the actual spot pattern may not be great, with a sharp fall off at the edges making the light harsh again. Some spots are better than others. Then there’s the odd pattern you get with the vintage style bulbs that have a few thick strands of emitter, shining through shades made of open-weave materials.

Finally there’s the near constant over-rating of output, where a 50w halogen needs a 6-7watt LED and a 100w incandescent needs about 11w to match perceived output. I’ve seen figures around half that often quoted, and while a lot depends on the binning of the LEDs themselves, there is little indication of the real output of a pre-assembled unit.

We have LEDs pretty much through out the house, but I did like the light from incandescent bulbs more.

2 Likes
23

The filament style bulbs tend to be *much* closer to the behaviour of a traditional incandescent although the construction lends itself mostly to candle style bulbs and larger globes. I don’t like the ones (as depicted in Graham’s video) which mimic a standard ampoule bulb with a ring of LED’s perpendicular to the axis of the lamp - the light output is in a completely different pattern and direction to the bulbs that they are supposed to replace. The “corn cob” style ones are better.

GU10’s are trickier, there is a type with three high power LED chips each with individual lenses which are excellent replacements for a halogen GU10 but they have been getting harder to find - probably because they have absolutely terrible reliability

They tend come in two power ratings 7W and 10W (though I’ve seen everything from 3W to 15W) none really match their claimed rating and the 10W ones I suspect are the same as the 7W ones, just run at higher current which explains the high failure rate - but they are a bit brighter and a very good match for a 50W halogen.

This is the style I try to find

image

The style which has a flat plate and multiple LEDs like this

image

are nowhere near as good at being spot lights.

eBay was my “go to” for these lamps but it looks like AliExpress is now the only place I’m going to have any luck.

One way to improve the life of some LED lamps is to de-rate them by decreasing the current output from the power supply - but we’re into “don’t try this unless you know how to keep safe around mains electricity” territory.

24

PS, on the subject of bulbs in general I presume everyone knows this story.

25

I use spots as general lights. Halogens offered a broad beam with gentle cut-off, and to replace them requires around 100 degree beam angle, give or take, so the second type you show are better for me. In a house with low ceiling like ours, the 36 and 60 degree spots just give a small ‘hot-spot’ on the surface below while the rest of the room is dark.

We briefly had some of the first type you show, but as you say, they were desperately unreliable and the output was pretty low too.

Somewhere I still have my original LED bike lights, with the LEDs glued to a large heatsink with thermal cement. They were nominally 120 lumen output, which was absolutely stonking at the time, but were very much brighter still run off 4 X 1.2V NiMH cells with no current control. The heatsink needed quite a bit of airflow to keep it cool, and you couldn’t use this as a torch for more than a couple of minutes because it got too hot to hold. This is relevant because the focussing lenses were available in various beam angles, and I had a 4 degree spot on one of the 2 LEDs to let me see ahead on the road.

1 Like
26

That’s only 1 year continuous use. Even a mediocre cap like that will do 50,000 hours. For out STBs, we specified a minimum of 100,000 hours continuous use, and they were very cheap … they had to be. Also, we specified a minimum of 10,000 power cycles.

1 Like
27

Forgive my scepticism, if you have a link to a capacitor datasheet which offers > 10k hours, I’d love to see it.

For instance, here’s an excerpt from a typical datasheet

image

The PJ series is a low impedance capacitor designed for switch mode applications - specified lifetime is 5000 hours, lower for some of the specific values.

To be fair this doesn’t mean caps are going to fail after 5000 hours - it means that if you operate the capacitor within it’s absolute maximum limits for temperature, voltage, ripple current etc Nichicon *guarantees* that after 5000 hours it will still meet its other specifications. After 5000 hours bets are off - there’s a good chance it will continue to meet its specs for quite some time but there’s no guarantee.

Indeed it’s probably going to be a lot longer than 5000 hours if operating conditions are well within absolute maximums.

Looking at my file server the “oldest” disk has a total head flying hours of 41292 - nearly 5 years of pretty much continuous uptime during which the caps in the PSU, on the motherboard and in the hard drives are still doing just fine. I have an Audiolab 8000A that has been basically powered up continuously for 30 years (to be fair I probably do need to think about re-capping it).

But cheap Chinese switch mode PSU’s won’t be using Nichicon PJ series - they will be using no-brand capacitors which are probably only rated at 1000 hours @ 85°C and are operated close to their voltage and temperature limits. Unsurprisingly they fail rather often.

28

The data sheet you have attached specifies ‘load life’ , which is not lifetime. This is just a guarantee that the capacitance value will stay within the specification if the cap is loaded at it’s maximum rated temperature and voltage for a fixed period of time. Actual expected lifetime is many times the load life.
Also, component specification doesn’t measure any sort of guarantee that every component will fullfil it’s role for a specified time. The specification includes a maximum expected annual failure rate along with other factors. Manufacturers are happy to work with customers to provide this information and many manufacturers will guarantee 100,000 hours on that basis.

29

I said that.

I also said that.

I also pointed out that cheap Chinese lamp PSUs don’t use Nichicon’s, and do tend to run their (very cheap and nasty) caps outside their temperature and close to their voltage ratings.

30

That’s fair enough in private, if you are buying enough components and if you de-rate the maximum ratings.

But, small aluminium electrolytics in cheap lamp power supplies don’t tend to be running in “benign” conditions.

31

When you’re buying several hundred million dollars of components a year or more, component suppliers will do everything they can to get your business, including selling you uprated components at the cost of the one you specify if it will meet your reliability spec. However, you have to make it clear, through action, that if their components fail unexpectedly in the field then you sue. We’ve had to do this several times, including a large (tens of millions of dollars) settlement with a very well known hard drive manufacturer. That was a really interesting one that I was very much involved in.

Crikey, alu caps are the pits for reliability compared to MLC caps, but they only need to be a tenth of a cent cheaper to make it worthwhile for the manufacturer. Sad really but true.

32

Yes, in that sense I apologise for my scepticism - I think we might have been talking about different types of caps as well (MLC vs alu electrolytic).

We’re also really talking about quite different ends of the spectrum of capacitor reliability - high quality components from reputed manufacturers running well within maximum ratings vs low quality ones stressed close to their ratings.

This is the sort of thing you find

image
image1600×1000 120 KB

It’s a “GSK” brand, 4.7μF 400V, 105°C is claimed.

I can’t find any data for that brand, 400V is pretty close to the working voltage (up to 357V for in-spec UK mains) and I’ll bet operating temp is 70°+ - the lamps run too hot to touch (though, obviously not as hot as halogen) and . You can see this cap has failed from the domed top.

Indeed - and if you were thinking about MLC lifetimes we were definitely  talking about different things :slight_smile:

MLC isn’t really appropriate for the high side reservoir cap - well, I found a KEMET 4700nF 500V MLC but the Mouser price for those is £27.65 (+VAT) a pop :open_mouth:

Another problem with this end of the market is that construction of cheap PSUs often leaves room for improvement, even if the component choice is not terrible (I did find some lamps with caps that claimed  to be Rubycon’s - not 100% convinced as it’s a frequently counterfeited brand).

This is not from a lamp directly but a USB charger which came with an illuminated magnifier - the magnifier itself was OK but the PSU (which has “KSJ” brand caps, also one I can’t find anything about - there’s a Koshin KSJ series - but I’m not sure that’s what I have; possibly “KSJ” was chosen by whoever made the capacitor for deliberate confusion).

image
image1600×1000 128 KB

Positioning the low side output between the mains input pins was never going to be a great idea safety-wise but I suppose there isn’t exactly a lot of room in a BS1363 footprint - there’s a token attempt at isolation with some slots routed into the PCB, but it has been almost completely defeated by the metal tab that crosses the gap to wind up barely 1mm from the metal body of the USB socket - I’ll remove the terminal and file off the tab which will improve things a bit - the creepage on the back of the PCB is 2mm or so, somewhat on the low side for standards compliance but probably OK in practice.

It has a nice UKCA mark on it, probably fake. I’ll bet they do a line with fake CE marks for the EU market :rofl:

33

Yesterday I had a problem logging into Amazon. Each time I tried, and I tried several times, Amazon said an SMS had been sent to my phone with a code that I should use to log in. But I was receiving no such SMS messages. Unknowingly I had accidentally blocked SMS messages on my smartphone.

Have since unblocked it and all of their previously sent SMS messages magically appeared. Problem over.

But this morning I received an email, below, which effectively is saying that Amazon has blocked my accounts, meaning I can no longer log in and must ring them to resolve the problem.

I have seen a few YouTube videos explaining how scammers work, and one of them is to encourage you to ring them, using a fake Amazon number, to help resolve problems, like a blocked account.

I think that because I tried to log in unsuccessfully several times it got logged as genuine suspicious behaviour, hence the issue of a valid - but rather draconian - email blocking my accounts.

But I don’t know if this is a scam, or if Amazon does block access to your account due to suspicious behaviour.

Happily, when I opened my 3 Amazon accounts this morning, they all opened with my name already showing on the top menu. I have full access.

But is the email a scam or not a scam?

is this an amazon scam
is this an amazon scam700×1258 119 KB

34

It looks genuine, apart from the lack of reference to your name, but I would never contact a company on a phone number (or URL) provided in a message.

You’ll be able to find the number you need on Amazon’s website.

35

I’m not worried and am not going to ring Amazon as my accounts are not blocked. I’ve reported it to the Federal Trade Commission USA (advised by Amazon) because the email address they used to send it to me is ancient and unused for Donkey’s years.

1 Like
36

If only.

Though Google does come up with a number the only way on the Amazon site seems to be to start a web chat then request phone back.

1 Like
37

Difficult to tell just from a screenshot.

If you’re interested and have nothing better to do with your time then you could look at the email headers which shows the geeky info about how the email got from wherever it was sent and into your inbox. Security wasn’t a priority when email was first invented, but various things have been implemented since to try and improve it. In particular you can use the email header to look up something called SPF, DKIM & DMARC - which allow you to determine if the sender’s email address is genuine or has been spoofed.

38

Incidentally I found a Tweet from a genuine Amazon account that posted one of those numbers so perhaps it is legit… :person_shrugging:

https://twitter.com/AmazonKDP/status/1737902632293786087?t=9PWUx3IskimnR8kC1vm_fQ&s=19

39

I rang the Amazon number for outside the USA & Canada and it was answered by an American robo-voice service which sounded genuine. I think scammers would prefer to answer themselves so they could persuade you with sweet words to get what they want. Don’t think robo-voice could do that, so I think my Amazon email was genuine.

I looked at a few forums where people said they rang these numbers, and were given assurances that their problems would be dealt with, but none mentioned being scammed.

40

How on earth could you tell?

Whenever I’ve needed to speak to Amazon I’ve used the online chat and requested a call back, giving them a code word that they must use when they call me.

2 Likes