Author Topic: Mains Powered LED - LED Lamp Reverse Engineering  (Read 30607 times)

0 Members and 3 Guests are viewing this topic.

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Mains Powered LED - LED Lamp Reverse Engineering
« on: January 08, 2022, 02:43:07 pm »
Hi Everyone. Re-visiting my little side project today.

I am learning how LED Light Bulbs (Lamps) work and I have designed my LED Power Indicator Lamp Schematic to run a 20mA LED at 10mA half brightness. This circuit is working wonderfully, please see attached schematic.

I recently took apart an old LED lamp that blew (from my hallway ceiling light) and I have drawn up the schematic, see attached (maybe I can fix it but that is a later problem).

It is very similar in design to my schematic but there are a few differences and a couple points I do not fully understand and I am hoping some of you can help me with this.

  • The fusible resistor value. I am still a little confused by this concept and I have come across a lot of mis-information online. Could someone help me understand how to calculate the appropriate resistor value?
  • In the Elektrek lamp the dropper capacitor is in series with the Neutral and not the Live. What advantage is there to this or does it not matter?
  • In the Elektrek lamp the resistor R3, which I assume is a load resistor, is in parallel with the LEDs where as in my design the resistor is in series. What are the advantages / disadvantages to these methods?
  • In the Elektrek lamp what is R4 doing?
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #1 on: January 08, 2022, 03:13:10 pm »
Another schematic from another faulty, but the same model Electrek lamp. Interestingly it is a little different....

Firstly I notice it is a regular film capacitor and not an X2 rated one!

A higher value load resistor across the output of BD1, 510K instead of 100K

And only 9 LEDs?? Not 10.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6211
  • Country: de
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #2 on: January 08, 2022, 05:59:29 pm »
Both are marginal designs with issues.
For an indicator lamp, why don't you just use a small neon lamp plus a resistor?
Much easier.
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #3 on: January 10, 2022, 12:13:54 pm »
Both are marginal designs with issues.
For an indicator lamp, why don't you just use a small neon lamp plus a resistor?
Much easier.

You're missing the point, I am trying to learn about the existing designs, the flaws and how to improve them.
 

Offline Terry Bites

  • Super Contributor
  • ***
  • Posts: 2498
  • Country: gb
  • Recovering Electrical Engineer
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #5 on: January 10, 2022, 10:29:47 pm »
X2 input caps are a must.

Great info at

https://www.tdk-electronics.tdk.com/en/373562/tech-library/articles/applications-cases/applications-cases/everything-for-capacitive-power-supplies-from-a-single-source/1380514

Yes of course! I am well aware about X2 capacitors!!

Thanks will take a read.

Anyone able to shed light on some of my previous questions?
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 12314
  • Country: us
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #6 on: January 10, 2022, 10:42:18 pm »
The fusible resistor value. I am still a little confused by this concept and I have come across a lot of mis-information online. Could someone help me understand how to calculate the appropriate resistor value?

It is mainly for safety, so the exact value is not critical.

Quote
In the Elektrek lamp the dropper capacitor is in series with the Neutral and not the Live. What advantage is there to this or does it not matter?

It does not matter.

Quote
In the Elektrek lamp the resistor R3, which I assume is a load resistor, is in parallel with the LEDs where as in my design the resistor is in series. What are the advantages / disadvantages to these methods?

R3 is a discharge resistor for the big capacitor, for safety. It is a moderately high value so it does not disturb the operation of the circuit.

Quote
In the Elektrek lamp what is R4 doing?

It is a current surge protection resistor, to prevent the capacitor appearing as a dead short across the rectifier.
 
The following users thanked this post: paul_g_787

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #7 on: January 10, 2022, 11:50:45 pm »
Thanks for that, it has certainly made a few things clearer!!

Although I thought R2 was the discharge resistor (in parallel with the dropper capacitor).

Is R3 not there to create a load on the output of the rectifier? otherwise the 3.3uF capacitor would have to be much larger to smooth the ripple?
 

Online IanB

  • Super Contributor
  • ***
  • Posts: 12314
  • Country: us
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #8 on: January 10, 2022, 11:55:03 pm »
Although I thought R2 was the discharge resistor (in parallel with the dropper capacitor).

Each capacitor needs its own discharge resistor. R2 cannot discharge C2.

Quote
Is R3 not there to create a load on the output of the rectifier? otherwise the 3.3uF capacitor would have to be much larger to smooth the ripple?

I don't think so. But C2 charged up to 340 V would give you quite a belt if you touched it. It definitely needs discharging for safety.
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #9 on: January 11, 2022, 12:03:35 am »
Yes of course, hence the R2 in parallel with the dropper cap.

In my little design I used a 1M resistor in parallel with the class X capacitor which when I measured, discharges it in a few seconds after power is disconected.
 

Offline strawberry

  • Super Contributor
  • ***
  • !
  • Posts: 1199
  • Country: lv
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #10 on: January 11, 2022, 09:21:33 am »
X2 because regular film capacitor (even >600V) will fail on mains for some reason
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #11 on: January 11, 2022, 09:17:15 pm »
Yes X2 is what should be used, although about half of the led light bulbs i have taken apart had regular film capacitors in them but it was always open circuit leds. Never seen the dropper cap fail. Yet.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6211
  • Country: de
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #12 on: January 11, 2022, 09:48:55 pm »
The reason for X2 caps is that they're designed and guaranteed to fail open. Not all film caps will do that, even though they seldom fail.
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #13 on: January 11, 2022, 10:05:48 pm »
The reason for X2 caps is that they're designed and guaranteed to fail open. Not all film caps will do that, even though they seldom fail.

No that is class Y you described.

Class X is for across the line (L-N) and is designed to fail short circuit so that an RCD will trip and cut off the power. So class X is the type to be used in the led lamp for the dropper cap because it is between Live and Neutral.

Class Y is for line to earth (L-E or N-E) and is designed to fail open so that the earthed chassis does not become live.

My source 12 different Marshall and Fender amplifiers I have fixed and also this link:

https://www.allaboutcircuits.com/technical-articles/safety-capacitor-class-x-and-class-y-capacitors/
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19885
  • Country: gb
  • 0999
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #14 on: January 11, 2022, 10:12:38 pm »
The ratio of C1 to C2 is important. C2 needs to be big enough to ensure it doesn't charge to too higher voltage. If forms a capacitive voltage divider, with C1. I made a post at the end of last year about this.
https://www.eevblog.com/forum/beginners/what-is-the-purpose-of-d1-diode-in-the-cap-dropper-circuit-here/msg3881156/#msg3881156
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6211
  • Country: de
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #15 on: January 11, 2022, 10:43:12 pm »
Class X is for across the line (L-N) and is designed to fail short circuit so that an RCD will trip and cut off the power. So class X is the type to be used in the led lamp for the dropper cap because it is between Live and Neutral.

If that was the case, your LED lamp would be deadly, should the dropping cap fail!

I discourage using All About Circuits for gaining knowledge, it's sort of the "Electronics' Anti-Vaxxer" site.

Here's information from a manufacturer of X- and Y-caps. The salient point is the "Self-healing" capabilities:

https://www.wima.de/en/our-product-range/rfi-capacitors/
« Last Edit: January 11, 2022, 10:47:06 pm by Benta »
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19885
  • Country: gb
  • 0999
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #16 on: January 13, 2022, 10:41:25 am »
Class X is for across the line (L-N) and is designed to fail short circuit so that an RCD will trip and cut off the power. So class X is the type to be used in the led lamp for the dropper cap because it is between Live and Neutral.

If that was the case, your LED lamp would be deadly, should the dropping cap fail!

I discourage using All About Circuits for gaining knowledge, it's sort of the "Electronics' Anti-Vaxxer" site.
Lol interesting metaphor. There are legitimate reasons to be cautious about the COVID-19 vaccines, especially in children and young men. It does annoy me how anyone with genuine concerns gets labeled an anti-vaxxer.

Anyway, back on topic: Electronics-lab, instructables (it should be called destructables) and many DIY audio sites are also dodgy. It's very important to look for multiple sources of information on the Internet, especially when it's safety critical.

Quote
Here's information from a manufacturer of X- and Y-caps. The salient point is the "Self-healing" capabilities:

https://www.wima.de/en/our-product-range/rfi-capacitors/
Yes, that's a better resource.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6211
  • Country: de
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #17 on: January 13, 2022, 08:18:23 pm »
Lol interesting metaphor. There are legitimate reasons to be cautious about the COVID-19 vaccines, especially in children and young men. It does annoy me how anyone with genuine concerns gets labeled an anti-vaxxer.

Never mentioned COVID-19 specifically, my scope included Polio, measles, Diphtheria, yellow fever, the lot.
I don't want to see something like this again. 'Nuff said:

 

Offline themadhippy

  • Super Contributor
  • ***
  • Posts: 2945
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #18 on: January 13, 2022, 08:24:48 pm »
Quote
Class X is for across the line (L-N) and is designed to fail short circuit so that an RCD will trip and cut off the power
Hows the RCD going to trip with a short between live and neutral?
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6211
  • Country: de
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #19 on: January 13, 2022, 08:29:24 pm »
Hows the RCD going to trip with a short between live and neutral?

Terse and to the point. Bravo!  :-+
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19885
  • Country: gb
  • 0999
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #20 on: January 13, 2022, 09:07:02 pm »
Lol interesting metaphor. There are legitimate reasons to be cautious about the COVID-19 vaccines, especially in children and young men. It does annoy me how anyone with genuine concerns gets labeled an anti-vaxxer.

Never mentioned COVID-19 specifically, my scope included Polio, measles, Diphtheria, yellow fever, the lot.
I don't want to see something like this again. 'Nuff said:
Yes, vaccines are good to protect against many illnesses, including CV-19, but you're right, discussing it here will get us banned.
Hows the RCD going to trip with a short between live and neutral?

Terse and to the point. Bravo!  :-+

Also note that even if tripping the RCD were the case, it shouldn't be relied upon, as older installations don't have RCDs.
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #21 on: January 13, 2022, 09:24:24 pm »
That is the reason for having a fusible resistor.
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6211
  • Country: de
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #22 on: January 13, 2022, 09:32:12 pm »
That is the reason for having a fusible resistor.

No.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 19885
  • Country: gb
  • 0999
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #23 on: January 14, 2022, 08:36:07 am »
That is the reason for having a fusible resistor.
A fusible resistor should be used because it will be guaranteed to blow open circuit and not catch fire. It's there in case the current limiting capacitor fails short circuit, which shouldn't happen anyway.
 

Offline paul_g_787Topic starter

  • Frequent Contributor
  • **
  • Posts: 489
  • Country: gb
Re: Mains Powered LED - LED Lamp Reverse Engineering
« Reply #24 on: January 14, 2022, 02:12:53 pm »
*begin rant
Please keep replies on topic and start a different thread for discussion about covid etc..... thank you  ;D

Before anyone jumps to conclusions, yes I know these lamps can be replaced for cheap and there is no financial benefit to repairing them and I also know I can buy other types of lamp.
However I am not repairing them for these reasons I am just taking them apart to see how they work. Any repairs are just a bonus!

The reason I am doing this is that I am trying to fully understand how these capacitive drop LED lamps work for no other reason than knowledge. There is a lot of incorrect information on this topic and I am trying to filter that out and find the correct information.

*end of rant

So I am obviously misunderstanding something still, maybe someone can help clear this all up?

Let me start by listing the faults with the dead lamps that I have taken apart so far, this information might help to work out what I am not understanding?  :-//

Out of 8 total 5W faulty lamps of different makes and types, 4 of them have failed dropper capacitors. 3 of which are class X capacitors which have no visible faults from the outside.

2 of the class X caps are dead shorts (8 ohm DC and 3 ohm DC) with blown fusible resistors and blown up black spot LEDs. I assume the LEDs killed the cap which took out the fuse?
The 3rd class X is open circuit. (which after replacing with a salvaged one, this lamp is now working fine and lighting my desk lamp.  :-+) Is this a manufacturing defect in the cap?

The 4th lamp has an open circuit (not class X) film capacitor which has burned and bulged and oozed, there are no other faults in the lamp and it is working fine after changing this for a class X of the same value. Not sure what killed this one?

The other four lamps are all normal film capacitors (not class X) but the fault with all of these is one or two of LEDs have become open circuit and have a black spot. The capacitive dropper supply circuits are actually still working fine it is just the LEDs that have failed.

So far this is what I have found:
8/8 faulty LED lamps I have seen so far are capacitive drop power supplies (which I assume means it must be the cheapest/simplest solution).
4/8 lamps have open circuit LEDs as the reason for failure.
7/10 lamps use film capacitors that are not class X rated.
2/3 of the faulty class X capacitors are dead shorts, the other is open circuit completely (>200MΩ).
3/3 of the shorted capacitors caused a blown fusible resistor.

In the case of the three with blown fusible resistors, I would have assumed (like Zero999 said) that due to the capacitor becoming short, this would cause the fusible resistor to blow?

I would like to add that these are all lamps bought from Wickes or Tesco and not some dodgy online auction site that shall not be named.


The 2 main things I need clearing up are these:

I am still not getting my head around is the fusible resistor malarkey. A previous poster said that the value doesn't matter however I don't think this is strictly true. Would one not use Ohms law and Watts law to calculate the value of the fusible resistor? And would there not be a trade off between either the resistor blowing at a lower current or the resistor wasting current as heat?

Additionally I am failing to understand the Class X capacitor fully. I have 2 out of 3 here that failed short circuit, but some people say they cannot, but they clearly can as I have seen first hand. I am very confused by this. Can anyone clear this up as to which conditions cause these to fail in this manner?  :scared:

I appreciate all the advice that anyone here can all give on this subject.  :-+

« Last Edit: January 14, 2022, 02:15:21 pm by paul_g_787 »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf