Help with 24 led light powered by 12 Volt car battery

[campus189]

That's good information. Apparently the Vf is a little higher than 3.2V even but something isn't right. With this setup running, what's the supply voltage from the converter?

5.22 Volts

 

[KJ6EAD]

The numbers suggest that you only have one resistor connected (≈5Ω).

 

[campus189]

The numbers suggest that you only have one resistor connected (≈5Ω).

Ok, double checked connections. With 1.5 Ω bypassed & using the two 5.1Ω resistors in parallel
0.67 amps for draw & 3.19 Volts across LEDs

 

[KJ6EAD]

Okay, that's too much current so Vf is going to be 3.1V and the correct resistor value will need to be 4.3Ω for that supply. The resistor would be dissipating almost 1W of waste heat.

That would work but as you know now, since someone spilled the beans early, you can reduce the output voltage of the converter to a value much closer to what your LEDs want and significantly reduce the amount of power wasted in a resistor. This modification would require you to remove and replace 1, possibly 2 SMD resistors on the converter board.

How do you want to proceed?

 

[campus189]

Okay, that's too much current so Vf is going to be 3.1V and the correct resistor value will need to be 4.3Ω for that supply. The resistor would be dissipating almost 1W of waste heat.

That would work but as you know now, since someone spilled the beans early, you can reduce the output voltage of the converter to a value much closer to what your LEDs want and significantly reduce the amount of power wasted in a resistor. This modification would require you to remove and replace 1, possibly 2 SMD resistors on the converter board.

How do you want to proceed?

I have no problem replacing the SMD resistors, there so damn small though
And question is, will we get the correct ones ? lol

 

[KJ6EAD]

I have no problem replacing the SMD resistors, there so damn small though
And question is, will we get the correct ones ? lol

Those resistors program the output of the converter for 5.17V. You'll need to find out what size they are (R2 & R3). They are actually large for surface mount parts so they're relatively easy to work with. Those look like they could be 1206's or 805's (size designations).
Surface-mount technology - Wikipedia, the free encyclopedia

The current values (4.7kΩ & 1.5kΩ) are just barely readable in your photo. A better one would help. If you include a millimeter rule alongside the resistors, the size will be readable at the same time. If you have a macro or close up mode on your camera, you should use it. You should also disable the flash and light the board from at least two sides. This will improve the depth of field and prevent bright reflections in the photo.

An advantage to modifying the output voltage of the converter besides saving a lot of power wasted on heat is that it doesn't have to be any precise voltage, just a few tenths of a volt above the Vf of the LEDs. The difference can then be made up by an ordinary ¼W resistor. For example, say we change the converter output to 3.5V, that means the resistor only needs to drop 0.4V while passing 480mA for a dissipation of 192mW or if we make the output 3.4V, its 0.3V for 144mW.

Since the power dissipation of the resistor is so low, you can use any low value potentiometer to dial the current you want after modifying the converter. Then just remove the potentiometer and measure it's value to determine the fixed resistor needed for the light.

It looks like the modification will only require replacing R3 (1.5kΩ) with 2.7kΩ to program the output to 3.43V but don't go buying resistors yet. You don't know the size and I need to check some other considerations when I get the higher resolution photo.

 

[campus189]

Well, now I'm ticked off.

I took a close up picture of one of them, didn't get a good view, so I took a picture the other pcb.
Some of the numbers are different from the dc to dc converter pcb
So, now I have to wait till i get sunlight outside
I may have to take pictures of ALL the pcb's and record the numbers & pictures
I will let you know..
Thanks

 

[campus189]

Ok, here we go..

I have two different sets of numbers.
4 PCB's have the following SMD resistor numbers

R1- R220
R2- 472 Has an underline under the number 2
R3- 188
R4- 511
R5- 753
R6- 513
R7- 433
R8- 513


**broken link removed**

I figured this would be easier than looking at tons of pictures.
I can include pictures if needed.
Thanks

 

[campus189]

The other 4 have these numbers

R1- R220
R2- 4701
R3- 1501
R4- 102
R5- 753
R6- 513
R7- 433
R8- 513


**broken link removed**

Heres the closeup for the 1.5k resistor


Thanks

 

[KJ6EAD]

Chinese foolishness. Can you measure R2 and R3 in circuit on the 472/188 board. Measure both directions as there may be some interference from other components.

Those resistors in your edit on post #49 are 1.5Ω, not 1.5kΩ.

 

[campus189]

Chinese foolishness. Can you measure R2 and R3 in circuit on the 472/188 board. Measure both directions as there may be some interference from other components.

Those resistors in your edit are 1.5Ω.

R2- 4.25k Ohms - Other direction 2.05k Ohms
R3- 1.45k Ohms - Other direction 1.31k Ohms

Hope that helps

Thanks for the reply

 

[KJ6EAD]

R2- 4.25k Ohms - Other direction 2.05k Ohms
R3- 1.45k Ohms - Other direction 1.31k Ohms

That's interesting. I kind of expect they're 4.7k and 1.5k. Take another look at R3. Could it be 153? I might need a picture.

 

[campus189]

All I did was to compare numbers on SMD resistors.
I will go over all 8 and let you know the differences.
Sorry to put you through all this.
Never expected it to be this difficult
Got a few things to do, will post later
Sabrina

 

[KJ6EAD]

Sorry to put you through all this.
Never expected it to be this difficult

I volunteered and nothing about it has been particularly difficult, at least from my end :wink:, just some small discrepancies in measurement and the inherent time lag of this form of communication.

 

[campus189]

Strangely enough, I put the original 1.5k back in original spot as well as connecting 2-10 ohm resistors in parallel for an output on the LED's @ 3.1 Volts
I was hoping for 3.2 Volts.
I'm still looking at the differences between them, found them, will post them later..

 

[KJ6EAD]

I put the original 1.5Ω back in original spot as well as connecting 2-10 ohm resistors in parallel for an output on the LED's @ 3.1 Volts. I was hoping for 3.2 Volts.

That would reduce the resistance to 1.15Ω which will allow too much current to flow to the LEDs, but what do you expect to happen?

 

[campus189]

Ok, found only 2 differences..

Read post 48 & 49 & look at the pictures...

Post 48 has the picture of the device with copper winding showing.
Post 49 has the picture of the device with the copper covered.

I triple checked to make sure and all the dc to dc converters that have the copper winding showing, use the same smd resistors.
The same goes for the once that have them covered.
I connected all of the dc to dc converters one by one to the breadboard to record voltages for each one.
I triple checked each one for accuracy.
I can post voltages if needed, but they are all close to each other..
Almost forgot, all the numbers are the same on the pcbs.
revision numbers and dates are all the same.

Thanks

 

[KJ6EAD]

Can you check the resistor that you thought was 188 and see if it's actually 153? Those pictures don't show any of the resistors since they're on the other side of the boards.

I went to my local Harbor Freight store and got one of those Ningbo lights just like yours so I can test anything that seems weird. I had been thinking of doing a similar conversion when you started your project which is part of the reason I took an interest.

 

[campus189]

Can you check the resistor that you thought was 188 and see if it's actually 153? Those pictures don't show any of the resistors since they're on the other side of the boards.

I went to my local Harbor Freight store and got one of those Ningbo lights just like yours so I can test anything that seems weird. I had been thinking of doing a similar conversion when you started your project which is part of the reason I took an interest.

Well, all you need now is the dc to dc converter,lol
If ya need that, I can send you the link

Ok, Heres the picture..

**broken link removed**

Looks like 188

Heres the other one..

**broken link removed**

I actually went out and got a couple hundred LEDs from flea bay from a US manifacturer so I had exact specs on LEDs, just in case these were oddball LED's.
Going to test the on the breadboard to see if theres a difference,lol

This is the flea bay listing...

Specifications:
Size (mm) : 5mm
Lens Color : Water Clear
Reverse Current (uA) : <=30
Life Rating : 100,000 Hours
Viewing Angle : 20-25 Degrees <--------- purchased the wrong ones, why they don't produce much light
Absolute Maximum Ratings (Ta=25°C)
Max Power Dissipation : 80mw
Max Continuous Forward Current : 24mA
Max Peak Forward Current : 75mA
Reverse Voltage : 5~6V
Lead Soldering Temperature : 240°C (<5Sec)
Operating Temperature Range : -25°C ~ +85°C
Preservative Temperature Range : -30°C ~ +100°C

On the bag that came it stated 3-3.4v 30mA Max 27,000 MCD

I'm going to try to duplicate using the LEDs I purchased.

BTW, **broken link removed**

Forget about them being from China, lol, I'm not purchasing from flea bay, I'm looking for more viewing angle & brightness.

 

[KJ6EAD]

I have several DC-DC converters similar to yours. They use the same circuit topology and control chip.

Millicandela or candela ratings are not very useful to me. I have to convert from millicandela and viewing angle to lumens to compare LEDs. Usually, the makers and sellers of cheap LEDs overstate the ratings and the LEDs have lumen maintenance problems anyway. If I'm working on a project that matters I stick with Cree, Nichia or Kingbright. I have no idea what those straw hat things are about.

You need to find out what size R3 (1501 or 188) is and replace it with the same size 2.7k. (details in post #46)