Osram Parathom par16 80328 led voltage

[bushtech]

This light stopped working although it has extremely low hours, probably about 10. Now it's a pricey little blighter so I thought I'd take a look inside.
The light partially opened:


Looking inside, the two wires attached to the pcb. What voltage should I see there? I get about 1000mV dc which is probably low.

 

[bushtech]

Got it open:

The big cap on top is 6.8uf and the small 6 pin chip is GAG7. Haven't found any info on it yet

 

[bushtech]

This light plugs straight into 220V so it's actually a smps problem. I get 330V on the "big" cap.Here's a better photo with some annotations:


Is that white thing on the right hand side a resistor? I read 0.8Ω resistance (one side not lifted)

 

[bushtech]

First smps I've seen where the output of the ferrite TX directly supplies power to the device, although there is a third output connected to the pcb. Is that how they monitor/control output?

 

[Martaine2005]

Is that white thing on the right hand side a resistor? I read 0.8Ω resistance (one side not lifted)

It certainly has a ref des of R1. I would think it’s a fusible resistor. 500mA 240v.
I think some of the solder joints look terrible. I would reflow quite a few joints and see if that solves your issues first.

Martin

 

[73's de Edd]

Sir bushtech . . . . .


Is that white thing on the right hand side a resistor? I read 0.8Ω resistance (one side not lifted)

CONFIRMED . . . . . that is being your AC input line fusing .. . . testing as GOOD . . . since you ARE getting a +300VDC raw DC voltage output at the units main filter capacitor.

In just visually evaluating that unit, it appears that the brains is that IC101 that drives the POWER MOSFET which uses 300+ DC to drive that ferrite based transformers primary to then output a reduced AC voltage(variable pulsewidth squarewave) on its isolated secondary winding. Therein . . . . is being the safety factor of an isolated winding for the LED supply portion.
The two black secondary wires go to the heatsinked LED trio . . . .assuredley being wired in series . . . .JUST after it has been rectified by the sole diode onboard and then being filtered by the high density poly or ceramic filter capacitors. That " micro " filtering is SIGNIFYING . . . that this lil ' puppy must be ringing out at a 100's of kilohertz operating frequency.

I guess that this unit must be the LED equivalent to the smaller sized incandescant lamps that have the reflector spot / flood light glass shell profile and usually pull 60-75 watts. Usually they are found, used in recessed lighting fixtures in the ceiling.
How about additionally giving the printed on specs of the wattage, Ma of current and Kelvin color temperature of the unit.

I got your saying that the main raw 300+ VDC is present and you should find it making its way to the heat sink tab of the POWER MOSFET or its clipped center pin.
One might also suspect the circuitry to use a dropping resistor to tame the +300 on down to a sub voltage to then be filtered and Zener diode shunted for regulation, to then feed Vcc to the IC 101.

I will need some "hep" on some photo definition aspects:

I cannot see the pin # 1 indexing of the IC101. Nor can I work with the given "GAG7 " numbering scheme designation, as it just wont factor out, nor can I extract its square root.
Its drive output seems to be from the corner pin, closest to the POWER MOSFET . . . . . . which receives it, into its far left gate pin
I can see your PRINCIPAL main E-capacitor filter, for the +300 VDC on one side of the circuit board.
I can not make out the smaller part to its side . . . is it being a metal film power resistor, used as the POWER MOSFETS drain resistor. Or is it being another physically smaller E-capacitor, that is being associated with filtering the aforementioned IC101's, Sub Vcc DC supply voltage ?

Considering that the small unit is being a filter, and associated with the SUB supply, measure across the W7 Zener diode and see how much voltage is being present.
OR . . . if that that filter is being in a hot environs and CONSTANTLY having the HELL hammered out of it with those brutal square waves,. you might want to sub in another E-cap in its place, to compare, to be sure of its present condition.

TECHNO REFERENCE . . . . . Marking Codes


You might lift one black lead feed from the transformer secondary. Then use a 9V battery and computed current limitig resistor to test out your POWER LEDS, individually and then as the series arrangement, to see if all are good . . . . . seems like one domes colorization, appears a bit different from the other two . . . . . . maybe an overheated unit ? . . . open ckt ?

73's de Edd . . . . .


Just wondering ? . . . . How do they get deer to cross the road, only at those YELLOW road signs ?

 

[bushtech]

Thanks for that Martin. Busy retouching the soldering

 

[bushtech]

Thanks for also stepping my knight in shining armour St Edd.

Looking at photo in post #3 the centre pin on side closest to fet is connected to the fet leg closest to it ie the RH leg.
Correct, no pin indexing on the chip, and I have studied it

Apologies, bad picture, hope this one is better. I think it's an inductor



I want to probe the IC for voltages to try and find Vcc. Is it correct that I can use the neg leg of big cap for that ?

This is what it looks like: https://www.ebay.co.uk/itm/Osram-Parathom-par16-80328-Warm-white-/323750158139
Here's a photo of the box:
10 years later info leaflet long gone



I suspect the leds are ok. probing them with my DMM on diode mode lights them up with a dull red light.

Still trying to figure out the two black wires going to the led board exiting straight from the transformer secondaries. Never seen anything like this on a smps.

 

[bushtech]

St Edd
By W7 do you mean the diode marked D3?

 

[73's de Edd]

Still trying to figure out the two black wires going to the led board exiting straight from the transformer secondaries. Never seen anything like this on a smps.

In just visually evaluating that unit, it appears that the brains is that IC101 that drives the POWER MOSFET which uses 300+ DC to drive that ferrite based transformers primary to then output a reduced AC voltage(variable pulsewidth squarewave) on its isolated secondary winding. Therein . . . . is being the safety factor of an isolated winding for the LED supply portion.
Ergo . . .your two black wires.


By W7 do you mean the diode marked D3?

Yeth . . . . . unless that is a W1 ? marking . . . which would be lower voltage than is to be expected for a power supply voltage , unless its being an external reference for IC101 to be using.

I want to probe the IC for voltages to try and find Vcc. Is it correct that I can use the neg leg of big cap for that ?
In a powr down mode ohm out from main filter cap negative to the 6 pins of IC101, to see if there is being some direct connection . . . . therewith . . . . . a common "hot" ground.

 

[bushtech]

Thanks st Edd, still trying to get my mind around that one........... no filtering, ripple removal et al

You might be interested in two new arrivals:


Still too scared to use the scope in anger

 

[bushtech]

Oh horse manure, shorted my dmm probe on something, all 6 pins of mystery ic GAG7 shorted.

fin

 

[Martaine2005]

Dog crap is better. But you are in good hands with Edd. He seems to find schematics not in existance. He can route a pcb with a out of focus photo. Got to love this guy. Edd , do you want a new grandson?