Maximum reverse voltage of AZ34063A step-down from 24V solar PV to 13V battery ?

[eneuro]

I did not find any specifications in the MC34063A documentation of maximum reverse voltage Vout->Vin ?
I modified the AZ34063A circuit so that the output is now 13V instead of the original 5V/500ma and connected to a 24Voc 150W photovoltaic panel but I plan to charge the gel battery to about 13Vmax, so it may happen that at night the input voltage will be very low, so on the plus side of the PV panel I have diodes protecting against current flow from the battery to the panel  I am thinking about adding an additional 1N4007 bypass diode in the dc-dc circuit from output to input, because there are BJT transistors in the MC34063A chip diagram in the documentation    What do you think about this solution, can this bypass diode be useful there or is it necessary in this case? 

Warning: in attached photo of bottom PCB bypass diode was connected from output ground to input plus which is a mistake of course, but it's not a problem, it is corrected right now...

[Andy Chee]

Reverse voltage rating V_EC is omitted on discrete transistor datasheets as well. 

In practice, you can simply use the V_CE figure, which for the MC34063A is 40V

That said, I would include the diode anyway to prevent the battery from discharging through anything else in the circuit (e.g. through the feedback resistor voltage divider).  BUT, not as a bypass diode as you've described it.  I would install it in series with the battery, as you would a polarity protection diode.

[eneuro]

In practice, you can simply use the V_CE figure, which for the MC34063A is 40V

I am talking about the protection of internal BJT transistors Q1 and Q2 in the MC34063A, when the voltage at the Vin input will drop practically to 0V, because after some time the 100uF input capacitor will discharge, while on a charged battery the Vout 13V will still be supported by the battery, so looking at the diagram of a typical buck converter from the MC34063AB documentation, it seems that a better place to install the bypass diode may be between the Q1 emitter, the SWE2 pin to the connected DRC8 and SWC1 pins, because it should not disturb the normal operation of the MC34063A circuit in buck converter mode.
In addition, the input capacitor will be charged via the resistor Rsc and the inductor L1 ...

[Andy Chee]

In practice, you can simply use the V_CE figure, which for the MC34063A is 40V

I am talking about the protection of internal BJT transistors Q1 and Q2 in the MC34063A,

So am I.

I am saying that in practice, the maximum reverse voltage of a BJT is the same as the maximum forward voltage of the BJT.

The maximum forward voltage of the MC34063A internal BJT is 40V.

[Manul]

I am saying that in practice, the maximum reverse voltage of a BJT is the same as the maximum forward voltage of the BJT.

The maximum forward voltage of the MC34063A internal BJT is 40V.

What about reverse bias of BE junction then? It's usually very low, maybe 5-7V and then it starts to zener. So it sets the limit of how positive emitter can go relative to base (npn).

[Andy Chee]

What about reverse bias of BE junction then? It's usually very low, maybe 5-7V and then it starts to zener. So it sets the limit of how positive emitter can go relative to base (npn).

For a discrete BJT transistor, you're correct.

But here we're talking about the internal NPN switch transistor inside the MC34063A, whose base is connected to a flip flop (probably at least two more transistor junctions).

[Manul]

What about reverse bias of BE junction then? It's usually very low, maybe 5-7V and then it starts to zener. So it sets the limit of how positive emitter can go relative to base (npn).

For a discrete BJT transistor, you're correct.

But here we're talking about the internal NPN switch transistor inside the MC34063A, whose base is connected to a flip flop (probably at least two more transistor junctions).

I understand that, but strictly speaking we don't know what exactly is there. There could be circuitry not shown in datasheet. I've never done these tests myself, nor I was particularly interested, but I remember reading that some BE junctions don't like zenering and may experience degradation. Although I occasionally zenered them by design, and they seem to hold.

Anyway, in this case I might try a high value series resistor like 100k and slowly increase voltage, observing with a multimeter if it starts to increasingly leak and clamp at some point. Then judge the margin of how far you are from leakage in normal intended operation, aka do the missing characterization yourself.

[magic]

These are NPNs so they will likely break down near 7V as a few people pointed out. Then BC junctions will forward bias and supply power to Vin, possibly causing the chip to start operating and draw some not entirely negligible current through those output transistors.

A diode from pin 2 to 6 should protect the transistors effectively. A diode between pin 2 and the load would protect them too, and also prevent pointless power consumption.

Hard to tell more because I have no idea what larger circuit this converter is part of...

[PCB.Wiz]

I did not find any specifications in the MC34063A documentation of maximum reverse voltage Vout->Vin ?

Because you are talking about a working system, you are best to measure the actual current levels, day/night, to decide what the actual issues are.
You may be better off with an ideal diode

[eneuro]

But here we're talking about the internal NPN switch transistor inside the MC34063A, whose base is connected to a flip flop (probably at least two more transistor junctions).

You are right, it looks like the base of one of the internal BJTs Q2 in the diagram from ST's MC34063AB is connected to the flip flop, but ...    Is it possible to supply voltage to the Collector-Emitter in reverse?  https://www.rohm.com/products/faq-search/faqId/232

[eneuro]

Hard to tell more because I have no idea what larger circuit this converter is part of...

As in the photos from the first post, a typical buck converter on a noname AZ34063A chip, reused from a 12V-24V car converter to 5V/500mA, mostly compatible with most of the documentation for MC34603A, and I am now designing my own converter but on the MC34603AB chip from ST

[eneuro]

A diode from pin 2 to 6 should protect the transistors effectively.

Yep, I soldered a FR107G diode between the Switch Emitter of the output BJT Q1 pin 2 to the Drive collector pin 8, in practice this also protects the BJT Q1 because in a typical buck converter topology the Switch Collector pin 1 is shorted to pin 8.
Tomorrow the moment of truth when the sun comes out 

[Andy Chee]

By the way, I assume you're only using the MC34063A because you obtained it for zero cost from salvaged recycled e-waste?

Otherwise I'd recommend looking for a more efficient buck controller to conserve every watt from your solar panel.

What is the size of your gel battery?

[eneuro]

By the way, I assume you're only using the MC34063A because you obtained it for zero cost from salvaged recycled e-waste?

No, I am using a noname AZ34603A 5V500mA based car converter PCB modified by myself to 13V500mA because Dave from EVblog encouraged me to do so    EEVblog #110 - Let's Design a DC to DC Switchmode Converter

[eneuro]

You may be better off with an ideal diode

Nope, diode FR107G 1kV 1A DO41 1.3Vmax is sufficient as a bypass between pin 2 SWE and pin 8 DRC of MC34603A in this project, modifying the purchased car converter to 13V instead of 5V500mA  Fits perfectly on the bottom of the PCB 

[Andy Chee]

No, I am using a noname AZ34603A 5V500mA based car converter PCB modified by myself to 13V500mA because Dave from EVblog encouraged me to do so 

Have you performed the efficiency calculations that Dave shows in the video?

How big is your gel cell?  If you have a small battery, then maybe you can afford to waste efficiency.

But even a LM2596 module from eBay is more efficient than all the 34063A variants.

[magic]

This still has the potential problem that if you use this module to charge a battery, it will be discharging it while not charging.

To prevent that, you would need a diode or some sort of switch between the charger and the battery.

[Andy Chee]

I already suggested a series diode in the opening reply of this thread.

[eneuro]

This still has the potential problem that if you use this module to charge a battery, it will be discharging it while not charging.

After storing the PV panel in the garage, the current consumption from the 12.2V battery was, as expected, only about 5mA, 1mA voltage divider, 1mA green output diode and the AZ34063A chip itself has a quiescent current of 2.5mA  according to MC34063A datasheet

I already suggested a series diode ...

Such a 1Vf @ 500mA diode would be a 0.5W heater and -1V voltage drop, which in practice would make it impossible to charge this 12V battery to 13V

[Andy Chee]

I already suggested a series diode ...

Such a 1Vf @ 500mA diode would be a 0.5W heater and -1V voltage drop, which in practice would make it impossible to charge this 12V battery to 13V

You can compensate by increasing the output voltage of your converter to 15V

[eneuro]

You can compensate by increasing the output voltage of your converter to 15V

I don't have to, I want 13V, because such a 12V car or motorcycle battery with a capacity of tens of Ah usually loses about 10 times more mA than -5mA when using this 24V->13V converter 

Protecting a battery from discharging has nothing to do with the topic 

I guess you were wrong about the reverse voltage VEC of the BJT Q1 transistor, but I think I solved the problem by soldering a diode between pin 2 -> pin 8 of the AZ34063A circuit, because after putting the PV panel away in the garage, nothing burned out, the current consumption from the car battery -5mA is 10x smaller than the typical -50mA in this type of automotive applications 


Is it possible to supply voltage to the Collector-Emitter in reverse?  https://www.rohm.com/products/faq-search/faqId/232

[Andy Chee]

Protecting a battery from discharging has nothing to do with the topic 

If battery discharge has nothing to do with the topic, then go ahead and remove the solar panel diode as well.

I guess you were wrong about the reverse voltage VEC of the BJT Q1 transistor

Well, I have heaps of 34063A chips that I’ve obtained from heaps of car chargers that people have disposed of in e-waste. I can perform a reverse voltage test on a bunch of them and let you know when they breakdown.

[eneuro]

Protecting a battery from discharging has nothing to do with the topic 

If battery discharge has nothing to do with the topic, then go ahead and remove the solar panel diode as well.

Yes, your answers resemble AI prompts, while the topic of this thread is reverse voltage protection of the MC34063A family ICs. 

[NiHaoMike]

Why wasn't there any mention that the solar panel is 150W while a 34063A might realistically do 15W at most? I would go with a 555 switching a MOSFET buck converter at a duty cycle that approximates the MPPT curve, plus a voltage sensing circuit that simply stops the 555 to limit the output voltage.

[KerimF]

I noticed that the input power source for MC34063A is a voltage source. Let us recall that a solar panel is a current source.
I can't go on since I am not sure about the exact circuit which is of interest here.