wart overvoltage protection

I use a fair number of the 150-kHz ones--they're still nice and slow and stupid, but use smaller and cheaper inductors.

Yeah, and the predominant frequency changes depending on the trace lengths you're got attached. We had one horrible failure on a small board with three switchers on it--two worked fine, but the LMR23630 we were using to make -12 from +13 produced a VHF-fest like yours.

One one trace it was mostly near 180 MHz, but on another it was 125ish. We gave up.

Yecch.

Cheers

Phil Hobbs

I mirror his website at

. Cheers

Phil Hobbs

I think the circuit you mean is at

. One could maybe save the BJT by using an LM4041 instead, but it would depend on the threshold voltage of the second FET.

Cheers

Phil Hobbs

This is one of my favorite waveforms:

I think the scope limited the rise time.

We ship this particular box with a small 12 volt wart, but a customer might use some other supply in a system application. They do sometimes apply 24 and blow up the TVS.

The ideal product handles up to +-48 at zero source impedance and behaves gracefully for any combination of brownouts and fumble-fingers.

Any decent power supply should be able to handle half-short circuit. Put in a big heat sink and it might just work with 24V supply, shunt to 12V.

lørdag den 1. oktober 2022 kl. 21.23.38 UTC+2 skrev John Larkin:

TPS2400 add PFET for reverse voltage protection

TPS2400 has Vin max of 6.8V. TL431 can go up to 30V.

Resistor R3 of 100 Ohms may be a typo. This requires the TL431 to sink 246 mA at an input of 30V, which is typical of an unregulated 24V wall wart with little or no load. This is well over the limit of 100 mA of the TL431. Changing R3 to 1K gives a much more reasonable 27.8 mA max at 30V.

Also, the GS voltage of M1 limits the maximum input voltage to +-8V. I know JL has reached 70V or so without damage, but I tend to be sceptical of exceeding voltage ratings. However, the input of 24V that JL asked for far exceeds this limit.

IRLML6401 datasheet blob:

the gates just need a zener and a series resistor

I thought the usual go was to short circuit the current long enough to blow a conventional fuse. Of course, you have to handle the case where the PS cannot supply enough current to blow the fuse, but sits there and fizzles instead.

Or is even fuse replacement beyond the customer's competence? Would you need to confiscate everything conductive and fuse sized first?

Sylvia.

On a sunny day (Sat, 01 Oct 2022 12:23:27 -0700) it happened John Larkin snipped-for-privacy@highlandSNIPMEtechnology.com wrote in snipped-for-privacy@4ax.com:

From datasheet quote: LM2576/LM2576HV Series SIMPLE SWITCHER ® 3A Step-Down Voltage Regulator General Description Features The LM2576 series of regulators are monolithic integrated n 3.3V, 5V, 12V, 15V, and adjustable output versions circuits that provide all the active functions for a step-down n Adjustable version output voltage range, (buck) switching regulator, capable of driving 3A load with 1.23V to 37V (57V for HV version) ± 4% max over excellent line and load regulation. These devices are avail- line and load conditions able in fixed output voltages of 3.3V, 5V, 12V, 15V, and an n Guaranteed 3A output current adjustable output version. n Wide input voltage range, 40V up to 60V for Requiring a minimum number of external components, these HV version

So HV version, preceded by bridge rectifier and insulated wart input. Used this chip, did not see ultra short spikes but I have no femtosecond capable scope...

10 to 12 GHz sat reception was still OK though.

Obviously. However, there is a much more serious problem. Jim's circuit does not protect against reverse polarity due to the body diode of the mosfet. Neither does the SLVA163. JL had requested this feature.

The obvious solution is to add a diode in series with the input. This will add a slight voltage drop, which can be accommodated in a number of ways, or simply ignored.

Here is the complete TL431 OVP circuit including reverse polarity protection:

Sylvia Else wrote: =============

** A 16amp triac with a 12v zener provides full protection.

Will trip at 13V with normal polarity and about 1.5V reverse. Anyone stupid enough to use the wrong voltage/ polarity wall wart deserves to see it die.

..... Phil

I don't see a problem with the body diode in Jim's circuit. I do agree that R3 must be a typo though.

John

søndag den 2. oktober 2022 kl. 07.09.19 UTC+2 skrev Mike Monett VE3BTI:

look closer at the schematic, the reverse protection does work

the first FET is reverse so initially the body diode will work as a series diode and when the FET turns on so there is no voltage drop

12V LDO regulator with thermal shutdown.

Very nice catch. It works. Thanks.