Hello everyone and thanks for reading this post.

I'm by no means skilled in electronics and I really don't know much beyond very simple circuits. I am a plumber by trade and have a sewer camera (basically waterproof camera with lights mounted on the end of a fiberglass cable) that I'm attemping to rebuild on my own.

My question lies in the PCB for the LEDs of the camera. The original LEDs were red which work fine for a B&W cam but it will be replaced with color and I needed to switch these LEDs over to white. Well, I removed the LEDs and cleaned up the PCB, but I failed to make a note of how the original LEDs were installed and now I'm a little confused.

From what I could tell, there may be one ground in this whole sequence and three power sources. (3 white wires, and one green - green was attached together with camera ground while white wires each had a separate connection point).

Excuse the crude MS Paint lines, but can anyone make sense of this?

Thanks so much in advance for your help and ideas!

 

Let me be the first to agree that you screwed up.
So let's see if we can back track. Are there resistors on this board and what's the supply voltage?

 

I'm not 100% sure on supply voltage but I am almost certain that resistors are already inline on a separate PCB. I can't remove that PCB from the piece that houses the connector (that attaches to the fiberglass cable), so I can't take a look at the underside of that. I attached 6 LEDs in series (I think they were 17000mcd) to what I believed was the +/- of the other PCB and they work, but are VERY dim, which leads my to believe there is a resistor there but available current may be low.

Let me try to get a pic of the other PCB and test voltage. Will report back in 10-15 minutes.

 

Turns out there is 14.7V available on each one of those 3 power sources.

I'm assuming the voltage is so high to compensate for LEDs in series (15 in total, so 5 per power source?).

Here is a pic of the PCB.

The leads are circled (red for +/black for -).

 

We may have a problem here. Red Leds drop about 1.5V but this spec can be higher with some models. White Leds will drop 3.0V min but can be higher (3.6V). Because your Leds are series connected 3.0V x 5 = 15V drop and this is excluding the series resistor. If your white Leds are 3.5V the problem worsens.

http://www.oksolar.com/led/led_color_chart.htm

 

Is the voltage for the leds the same across all manufacturers? That really may be a problem.

I think I may have the PCB figured out though.

Does this make sense?

As far as voltage and the LEDs go.....what if I cancelled some of the LEDs out and just ran a piece of wire to bridge those points in the circuit? I think I could do with 4 LEDs per circuit rather than 5. That would give me 14V if I use high power LEDs or even 3 per to give me 10.5V.

What I'm trying to say is that I can do without using as many LEDs as long as I can make the same circuit and pcb work for white LEDs.

 

Yes, you may have to jump some Leds. You're going to want about 20mA per Led string. It's quite late on this side of the pond so I'll be hitting the sack. By the time boot up in the morning you will have bunches of replies and suggestions. See ya then.

 

The higher voltage is pretty much a consequence of physics. Typically white LEDs are based on blue to near UV diodes exciting some phosphors which give longer wavelengths, giving an apparently white result.

The shorter wavelength of blue light compared to red means a bigger photon energy, which in turn implies a bigger voltage for blue LEDs than red. Different manufacturers products will vary, but I would doubt whether you are going to find any white LED much under 3V.

 

In some ways it may be simpler to build one from scratch.

 

Please tell us what the values of the three current limiting resistors are. They should all be the same.

BTW, the resistors are SMD's (leadless). Are the LED's too?

 

I forgot to ask.. Is this battery operated?

Edit: Just looked back at one of your posts. You said the measured voltage is 14.7V. My guess is that there's a 15V regulator somewhere.

 

The LEDs are not SMD (thank god because I'd never be able to solder them in).

My guess on where the voltage reg is that it would be in the control unit of the camera before if even gets to the fiberglass rod.

The whole thing is 110V, no batteries.

Building from scratch isn't very practical for me considering my limited skills.

And I now understand why white LEDs will always be higher voltage. Thanks

 

Oh and the resistors for that circuit are on the back of the pcb in the 2nd pic I posted and I can't get that apart (I belief potting epoxy plus a solder joint is holding it all together).

 

Building from scratch isn't very practical for me considering my limited skills.

Don't underestimate yourself! Making a PCB that shape and size would be daunting, but there could be other options. Like ultrabright LEDs mounted on another disk of some kind, keeping only the existing power supply.

Have you thought about the beam spread of the LEDs? Some are wide angle and some are highly focused. I'm guessing a tight beam might be better for you, but it would be worth checking the diameter of the spot that your current LEDs make at say, one foot away.

I also recommend determining the current draw of your existing LED string. I don't think you want to design for much more than that amount of current, because of the unknowns.

Maybe somebody here knows if this is a good approach: Could it be more efficient to use several colors of LED? Stage lighting consists of various colors mixed together to make white. So I wonder what the most efficient strategy is. What gives the greatest illumination for a given current?

 

Oh and the resistors for that circuit are on the back of the pcb in the 2nd pic I posted and I can't get that apart (I belief potting epoxy plus a solder joint is holding it all together).

Since you've already removed the original LED's (bad boy ) it's going to be difficult for you to do as wayneh asked,

wayneh: I also recommend determining the current draw of your existing LED string.

so what are the values of those resistors?

 

Since you didn't post the values of the current limiting resistors I had to make some assumptions. The attached schematic and resulting curves show the problem at hand. With R2 selected to provide about 20mA @ 14.7V, VM2 and AM2 show red LED voltage and current. VM1 and AM1 curves show resulting curves when the LED string is replaced with only 3 white LEDs. As you can see without replacing R1 we can produce about 16mA. This may be sufficient for you but only you will know.

 

I replaced the string of 6 LEDs I had used to try the circuit with only 4 (I only have blue on hand, so that's what I tried with) and they got up to full brightness (I think).

So, my next question is, if 4 blue LEDs (17000mcd) work, will 4 white LEDs of the same brightness work just the same? Or is there a significant difference in forward voltage of blue and white LEDs?

 

White LEDs are blue LEDs.



Confused?

There is no "natural" white-producing LEDs. SO, to cheat, manufacturers use blue LEDs and place a dollop of phosphor onto the blue LEDs' die.

When the blue light hits the phosphor, it puts off white light.

So that is why, voltage-wise, blue and white match.