Fish Tank LED Power Supply Gets Hot then shuts down

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

before we get sidetracked,
where does the 5v and 12v from the psu go to?

So there are three aluminum heat sinks, each with 5 rows of LEDs attached with two screws. The PCBs of the LEDs are aluminum PCBs and the PCB has thermal compound grease on the aluminum side of the PCB.

Each PCB is wired to another PCB that has high-wattage resistors on one side and the four MOSFETS on the other (which I thought was odd because there are five rows of LEDs but only four MOSFETS).

Lets call those PCBs the LED driver boards...

The PSU lines come into the unit and are soldered onto the first driver board then they are connected to the other two driver boards in a daisy chain fashion but from a schematic perspective, they would be wired in parallel.

Then from the last driver board in the housing, that 5-volt feed from the PSUs is soldered to those two wires that melted and attached to the main controller board via J3 as shown in the photos. J3 feeds the input of the 7805.

From that main controller board, there are two voltage regulators. The 7805 and a 3.3v regulator (the input of the 3.3v regulator is fed from the output of the 5v regulator), which powers the MC. There is an RS-232 chip that feeds a connector which looks like it is used to program the MC, and there is a chip that is just MOSFETS (ST TL084IN). Then two small square chips next to it (CSI 5114V) that look to be digital POTs.

The outputs from the controller board consist of the outputs that go back to each driver PCB for the lights, which are connected via ribbon cables and those would be driving the MOSFETS that power on the LEDs, then the fan outputs which each of the four fans are home run straight to the main board.

I think that MOSFET chip (TL084IN) actually drives the fans ... I don't think it has anything to do with controlling the LEDs ...

The damage to the main controller board is primarily around the 7805 and along the ground strip on the edge of the board that rides along the MOSFET chip and the digital POT chips that are next to it and that damage even managed to burn off a piece of solder mask on the board.

I checked all four fans, and three of them work. At 12 volts they are drawing 140 ma ... the fourth one does not power on at all and it is not drawing any current.

Nothing I've seen so far, explains why the 5 volt line is drawing 5.5 amps from the main PSUs ... unless there is something about those two MOSFETS on that last driver board (showing some kind of damage to them) that is the same board used to feed the 5 volt power to the controller board ... I can't rationalize how those would be in any way related electrically so Im not really sure what to think about that ... I might need to remove that board entirely and trace it out to see how things are connected but mixing that 5 volt line with those mosfets makes no sense since the mosfets drive power to the LEDs only (at least that is the assumption)

Also, the more I think about it, it makes sense to me that the small digital pots are manipulated by the MC which are then used to control the gates of the TL084IN - giving the unit the ability to control the speed of the four fans... the only thing about that is that I cannot see any kind of temperature sensor that it would use to determine how much power to give to the fans. The only temp sensor that I can see is one that sits a few mm above the microcontroller. But there might be a temp sensor on the driver boards that feed back via the ribbon cables to the MC ... but I'm not sure about that yet.

[stj]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

but are the led drivers running from 5v or 12v?
the controller must be 12v or it wouldnt need the 7805 regulator.

i suspect you program the unit over the serial port and it can dim the lamps at different times using the clock chip.

[eccerr0r]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

And do we need to back up to the first post that the LEDs work just fine with one PSU and not the other, so technically the LEDs should be fine and just need to figure out why it needs 5.5A... when the LEDs are on or off? Is the MCU/LCD working?

Are those LED units 3S (so the individual 5-LED boards are actually 3S5P?) I still can't really see a good reason to run the LEDs off the 5V line but who knows how they designed it to balance the use of the power coming from the PSU ...

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

Just a side comment aside from my detailed response to STJ, is that after my basic examination of this unit, I remain kind of puzzled to understand how an aerospace company managed to get a patent that was obviously close enough to the design of this device that was then leveraged to put this company out of business.

Nothing I've seen about this design so far appears to me to be anything other than typical circuitry that drives power to these LEDs.

Now I don't know anything about the LEDs specifically, nor do I understand how anything about this unit makes it uniquely applicable to fish tanks, UNLESS there is something about the code inside the microcontroller that provides the "magic" that makes this device somehow a device that is beneficial to marine life. Because outside of the microcontroller, all we have is typical voltage controlling via MOSFETS as would be the case in any LED driver environment.

Yet, the patent and the action against this company did happen so clearly I'm missing something but that something must exist in the microcontroller since the LEDs - Im assuming - are most likely something that can be purchased by anyone and were not manufactured by this company.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

but are the led drivers running from 5v or 12v?
the controller must be 12v or it wouldnt need the 7805 regulator.

i suspect you program the unit over the serial port and it can dim the lamps at different times using the clock chip.

I have changed my original response to you several times ... go back and re-read it as you might have read an earlier version of it.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

And do we need to back up to the first post that the LEDs work just fine with one PSU and not the other

That is incorrect. My first post indicated that the unit cannot power on at all unless BOTH power supplies are connected, and even then, the unit only powered on for a couple of seconds, then it just turned off.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

i suspect you program the unit over the serial port and it can dim the lamps at different times using the clock chip.

You might be right about dimming the lamps ... that might be controlled from the access panel (the user interface LCD screen) which goes to the MC and then that would mean that the MOSFET chip (TL084IN) is actually driving the driver board mosfes and the gates from the TL084IN are controlled by those digital pots that are controlled by the MC

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

but are the led drivers running from 5v or 12v?
the controller must be 12v or it wouldnt need the 7805 regulator.

i suspect you program the unit over the serial port and it can dim the lamps at different times using the clock chip.

This might help you in understanding how the 5v line from the PSU connects to the main controller board

That line from the LED Driver to the main controller board are the wires that melted

[redwire]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

Wild guess is the +5VDC rail powers only the red LED(s) and one has shorted to the heatsink and thus chassis ground.
I would say you need to test/study the LED boards a bit more- 5 LED's, 4-wires is weird. It might have 2-red because they are dimmer than blue or green. What colour mix are on a board could help solve the puzzle.

It must have artificial sunrise/sunset capability, thus the need for an RTC and ability to control colour.

[stj]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

that cant work, a 7805 needs atleast 6v input
and you also have 12v fans - so i suspect the controller is taking 12v

[redwire]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

I meant the 5V from the PC power supply could be powering only the red LED's.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

I meant the 5V from the PC power supply could be powering only the red LED's.

What RED LEDs?

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

OK, forget everything I said about the PSU connections ... I just lifted the main heat sinks out of the unit and the PSUs are split up ... ill be uploading a diagram soon.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

.

.

.

OK, THIS is accurate in how the boards are wired to the two PSUs.

I should mention again, that when I first connected the unit to power to verify what the owner said happens after power up, I only saw TWO of the three LED panels light up, and we already determined that ONE of the two PSUs was in failure being that it made all kinds of noise and its own heat sinks heated up to where they were too hot to touch in less than 10 seconds after power on.

HOWEVER, when the PSUs were cold, and turned on, the two LED panels would light up for a couple of seconds (if that), then everything just shut down.

Notice that on those two melted wires that feed power to the controller board ... the ground of that connection is being taken from the WHITE wire from the PSU and the hot side of that wire is being taken from the BLACK wire on the PSU ... this makes NO SENSE to me at all because I thought black was ground.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

Also, here is a video I made showing everything in my diagram.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

I made the comment in my diagram post that I was confused over the fact that the RED wire that powers the controller board was tapped into the black wire from the power supply ... as it turns out, they decided for whatever reason, to use the red wire as the ground wire and the black wire as the high side of the line feeding the 7805 ...

So I mapped the actual power supply wires to the internal wires and this diagram is accurate to the POWER SUPPLY colors.

Notice that the power line feeding the 7805 is taken from the two HOT LEADS from the power supply. When I checked those voltages on the PSU with the meter, the delta between them is 6 volts, so they are feeding the 7805 with 6 volts taken from the +12 and +5 off the PSU.

Now to think about how the +5 line is drawing 5.5 amps ... the way they wired this up makes this even more complicated to troubleshoot. But that +5 line from the PSU would be connected to the ground side of the main controller board.

[stj]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

take out 1 led panel and it's driver
lets reverse engineer it to see how it works - then use that knowledge to test all 3.

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

take out 1 led panel and it's driver
lets reverse engineer it to see how it works - then use that knowledge to test all 3.

So are you thinking that one of the panels is causing the 5.5 amp draw on the 5 volt line?

Cause remember, out of the three panels, one of them has a severely dis-colored pair of MOSFETS and it would not turn on while the other two did. And its also the same LED panel where the 6 volts is tapped for the power on the main controller board...

[stj]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

the panel may drive some leds from 12v and some from 5v to keep balance/
so we need to look at them - also we are assuming they are white and not mixed colours

[EasyGoing1]

Re: Fish Tank LED Power Supply Gets Hot then shuts down

the panel may drive some leds from 12v and some from 5v to keep balance/
so we need to look at them - also we are assuming they are white and not mixed colours

Ummmmm ... They kinda looked like this when they turned on ...