Hi,
I'm doing up failed fancy garden lights!
The chip that is cycling and dimming the colors is under a black blob. Is that a Chip on board?
I am unable to determine the details as the blob conceals any details that may be there.
Hunting through Temu/ ebay has not yielded any answers.
I'm only after 10 of these chips the rest of the components I have been able to source.
Does anyone have a clue as to where I may purchase these elusive controllers?

 

It's mostly likely chip on board.

Even if you find a source of the chips, and even assuming that that's the only problem, how do you plan to replace them? Do you have the ability to do the necessary wire bonding and/or solder ball reflowing?

 

One possible solution: Buy several assemblies and spares, then rip out the LEDs and connect your own LED circuitry.

 

My interpretation of the post #1 is that the TS is seeking to repair or fix the failed assemblies. The bad news is indeed that what is under that black blob is the "chip on board", and that replacements are not available, and also, that if those replacement chips were available, it is not likely that the TS could do the replacement, nor could I do that replacement.

 

Any microcontroller can replace it if you know how to program it.

 

Most likely the chip below the black blob is one of those ultra-cheap Chinese microcontroller, or a custom ASIC made by a Chinese manufacturer.
In plain English: Unobtanium.
As Bob mentioned, if you really want to fix the lights, you may redesign the board with a low cost microcontroller.
Have to warn you that the project will cost much more than purchasing new lights. But if you like the challenge and want to avoid E-waste, it can be a rewarding experience.

 

The "small black blob" controller is actually quite complex, at least the ones that I have seen, which use PWM to produce much more than just the three colors. Certainly the display can be produced with a programmed processor and three transistors. But not in that small black blob package!

 

I recently converted a 3 different colour LED Lava Lamp for a member where he wanted to ramp in and ramp out the 3 colour banks sequentially.
I used a 8 pin micro-chip.

 

I recently converted a 3 different colour LED Lava Lamp for a member where he wanted to ramp in and ramp out the 3 colour banks sequentially.
I used a 8 pin micro-chip.

Max verifies my point!! The "8 pin microchip" was used INSTEAD of the black blob assembly..

 

Max verifies my point!! The "8 pin microchip" was used INSTEAD of the black blob assembly..

This idea of yours sounds very interesting.
The black blob that is in the circuit I'm dealing with has:-

(a) 3 negative inputs that are switched separately by the operator.
1. selects a 3 colour display from a single LED
2. selects a single white LED on or off
3. selects the colour showing when switching by the operator and locks the colour on.

(b) 3 negative outputs. These brighten then dim the 3 color led doing each color in turn taking about 20 seconds to complete the 3 colors. The cycle then starts again.

(c) 1 negative output for a separate for white LED

(d) 1 positive in

(e) 1 positive out to the common anodes on both the colour LEDs. (The lights have 2 of these 3 colour leds)

(f) 1 positive for the white Led anode

How involved is the circuit using an 8 pin microchip? as I would probably have to make 18!!!

I've dispensed with original battery and solar panel and switching day/ night circuitry on the garden lights.
I now have 18 of these garden lights all hooked up in a parallel 3.8V dc circuit fed from my roof solar panel.

 

I used an 8pin micro that had the usual single PWM generator, but had the ability to change the designated output pin on the fly, so I was able to sequence the 3 colours in turn.
Ramp up - ramp down.
Three mosfets to drive the LED banks.

 

I have seen one of those COB blob devices create combinations of the three colors. a very interesting effect.

 

Essentially what I did. Using 8pin 16F18313 micro.

 

I now have 18 of these garden lights all hooked up in a parallel 3.8V dc circuit fed from my roof solar panel.

Why do not open working one to check chip markings?

 

Why do not open working one to check chip markings?

As the TS stated the chip is under a black blob, the chip markings can not be read.

 

If they could, most likely a custom programmed processor

 

Most folks, myself certainly, would be unable to replace that "chip under blob" if we were given the replacement chip and the information as to exactly how it should be connected.
That is to say that those "are to be serviced as an assembly". That means replace the PCB and blob.

 

The OP could do as I did, if the rest of the components are visible.
A little bit of reverse-engineering up to the 'blob' .
I am betting that it uses a similar method that I found in the reverse engineering as per post #8.
In my case, the markings on the original IC had been removed.
The difference in my case, they wanted a different lamp pattern.

 

The OP could do as I did, if the rest of the components are visible.
A little bit of reverse-engineering up to the 'blob' .
I am betting that it uses a similar method that I found in the reverse engineering as per post #8.
In my case, the markings on the original IC had been removed.
The difference in my case, they wanted a different lamp pattern.

I've been researching your micro controller idea with the help of AI.
Would I require all the following to program the controller?

PIC16F18313	Your target microcontroller
PICkit 3, 4, or Snap	Programmer/debugger for flashing code
MPLAB X IDE	Development environment from Microchip
MPLAB XC8 Compiler	Converts C code into machine code (.hex)
Breadboard or PCB	For physical connections

 

That is the route I went
Except I used assembly rather than C