SMD part "VG73" (SOT-23-6) - Solar lamp circuit

[thymjan]

Try to repair a defective solar lamp. I have already replaced a slide switch with a loose connection.
Still doesn't work. Now I identified a defective SMD part (compared with a functioning lamp).
Does anyone know what this VG73 could be?


Only four of the six pins are used.
B+ and B- is battery
S+ and S- is solar cell

Any help appreciated.

[Konkedout]

Why not try to draw a schematic diagram as well as you can?  From the looks of that pcb, drawing a schematic ought not be so difficult.

  If you can draw and post that, I think you would be more likely to get a helpful response.

[thymjan]

Thank you for your reply.

I tried to do a part schematic:

[prints-f-darkness]

Hi,

   It might be a battery protection IC, your schematic looks similar to the app notes for the AP9211.

https://www.diodes.com/assets/Datasheets/AP9211.pdf

Which protection IC, I have no idea.

Its most likely a Chinese brand IC, like the DW01A https://www.lcsc.com/product-detail/Battery-Management_HL-DW01A_C18164398.html
 

[thymjan]

Hi, thank you for your reply.
This makes sense. There are three NiMH AA (in series) batteries inside.
I'll have a deeper look at the website.

[inse]

Usually there is D for Diode, Q for transistor and U for IC.
This circuit confuses me

[thymjan]

This was my guess. The parts look like transistors, but are marked with U. Should I treat them as ICs you think?
I didn't find data sheets for "42JD" either.

[inse]

I appreciate your effort but once you have to order parts like the ominous VG73 IC, the repair is futile from economical point of view.
Or is it a high-class decoration?

[fzabkar]

This was my guess. The parts look like transistors, but are marked with U. Should I treat them as ICs you think?
I didn't find data sheets for "42JD" either.

There is a Q1 at the right hand edge of the PCB ...

[thymjan]

I already spent some effort to repair the lamp. In some point of view it’s useless, but the other way to throw away a hole lamp only because of a defective little 3mm part – wouldn't a repair be nice?
I didn't know it would be such a struggle to get a spare part.

[thymjan]

There is a Q1 at the right hand edge of the PCB ...

So you think I should draw my schematic sketch once again?
Till now I didn‘t understand how this circuit functions.
Are MOSFETS signed with Q or U?

[inse]

Unsolder one and test it.

[prints-f-darkness]

*strokes my beard in thought*

- you want to save the lamps.
- you need a working circuit for that.

why not design a new pcb?
or buy a driver board online? https://www.aliexpress.us/item/3256805471942913.html

I've actually become quite invested in your post, it would be really neat if the forum came together and helped you design a new pcb for your lamps. It would help reduce garden light waste all around the globe, quite the environmental project if you ask me.

I can say, the BQ2971 looks like a winner https://www.ti.com/lit/ds/symlink/bq2971.pdf
app notes / EVM, https://www.ti.com/lit/ug/sluuaz3/sluuaz3.pdf

I assume U1 on your PCB to be a led driver of sorts, and you may find some uber simple garden light design online that you can adapt https://www.instructables.com/Simple-Solar-Circuits/

That's just my advice anyway, ill leave ya to your devices and or components.

(also I would like to see a high res image of both front and back of the PCB and a pic or two of this awesome lamp we are trying to save)

[fzabkar]

There is a Q1 at the right hand edge of the PCB ...

So you think I should draw my schematic sketch once again?
Till now I didn‘t understand how this circuit functions.
Are MOSFETS signed with Q or U?

Could the U parts be digital transistors (with resistors in the base)?

[thymjan]

Here is a link to the instruction manual:
https://www.solarversand.de/media/pdf/cf/d3/03/102530-GB-2023.pdf

The other side of the pcb:


I allready asked the manufacturer for help to identifying the part and awaiting his answer.

U1 is a NAND Gatter with Schmitt-Trigger (74HC132D CSM54806 TXD1803D).

There are also to slide switches wintermode/off/summermode and warm/cold white and a motion detector.

[inse]

Why are there no cables at the positions S+ and S- which you identified as solar cell connection?

[thymjan]

I had to take them temporarily off to make a decent photo.

[inse]

If identification and/or sourcing of the VG73 part should prove impossible, either a workaround needs to be implemented or the lamp discarded.
I would be surprized if there was an elaborate charge circuit implemented.
Usually those solar lights only limit the voltage on the battery terminals, an undervoltage cut off would also be beneficial.
A 4.7V zener should serve as overcharge limiter, undervoltage cut-off would require a comparator or a reset generator as proposed in this thread:
https://www.eevblog.com/forum/projects/tl431-rgb-led-for-power-on-low-battery-indicator/25/
Or you get a protected LiIon cell that includes all the above and replace the NiMHs

[fzabkar]

U6 is driven by U1 via a 100 ohm resistor. That suggests that U6 is an N-channel MOSFET with logic-level gate drive.

Perhaps U = MOSFET, Q = BJT?

[inse]

U=MOSFET and IC? This appears strange…
thymjan, did you give up on the lamp or shall we continue?

[thymjan]

[inse]

I guess that U3 is a kind of undervoltage detector to prevent the battery from being deep discharged.
Little anything else than a N-MOSFET makes sense for U5 as interruptor for the battery negative terminal.
Is there a connection from battery plus to the K+ terminal? Otherwise I could not figure out how the battery would power the circuit.
And where is the switch connected that you mentioned (SUM/WIN/K1?)
I wonder how U3 should have failed but if apparently so and no success identifying AND sourcing it, we need to find a work around.
As proposed before, a protected LiIon cell could be placed between K+ and S-, bypassing the onboard protection circuit.
Voltage would be similar to three NiMH cells.

[thymjan]

Is there a connection from battery plus to the K+ terminal? Otherwise I could not figure out how the battery would power the circuit.
And where is the switch connected that you mentioned (SUM/WIN/K1?)

Sorry, this part I missed. Here are the schematics of the slide switches:

I wonder how U3 should have failed but if apparently so and no success identifying AND sourcing it, we need to find a work around.

I measured a sort of short between the two-pin-side of U5. I removed R9 and C7. Nothing happened. Only when I removed U3 the short circuit was gone.

As proposed before, a protected LiIon cell could be placed between K+ and S-, bypassing the onboard protection circuit.
Voltage would be similar to three NiMH cells.

I don't know if a LiIon cell will perform well at -10° Celsius. Is this a better use case for NiMH cells?

I try to recover the schematics with scheme-it (DigiKey). I have a bit of a struggle to draw the unknown parts there. Do you have a better recommendation for beginners?

[inse]

I don’t know the program but I do schematic design on a professional basis.
For side projects, colleagues now switch to KiCAD, which I haven‘t used yet, I used Eagle as well.
All batteries suffer at low temperatures but as the solar light is only consuming a relatively low current and the solar cell won‘t do fast charge as well, the increased impedance is probably of less concern.
The idea behind my proposal is that protected cells include over- and undervoltage protection, so you could skip the defective circuit.
Otherwise you would have to repair it or design a new one…

[thymjan]

For a better understanding:
U3 is measuring  the battery voltage. U5 is a normally opened switch, which only closes if U3 is measuring a certain minimal voltage. If battery voltage is ok, S- and B- are connected.
What could be the function of R10?