Switching 36V@7A doesn't work after connection with Buck Step Down

[Crios816]

Good morning everyone.
I bought a 36V@7A switching power supply (this is the seller's link: https://it.aliexpress.com/item/1005002641974879.html?spm=a2g0o.order_list.order_list_main.22.1875369674TJ9G&gatewayAdapt=glo2ita)

As soon as it arrived I tried switching without a load and read correctly 36V out.

After that I connected a Buck Step Down Converter (this is the seller link: https://it.aliexpress.com/item/1005005274933083.html?spm=a2g0o.order_list.order_list_main.79.1875369674TJ9G&gatewayAdapt=glo2ita) to withdraw 12V. (in the project I have to make, part of this voltage is used to power an arduino nano).

After connecting the Buck Step Down Converter the switching stopped working. now supplies 5V instead of 36V.

The power supply, by specification, has these protections:
- overvoltage protection
- overcurrent protection
- short circuit protection.

Do you have any suggestions to clarify why it stopped working?
Do you have any suggestions to get it working again?
Thank you.

[T3sl4co1l]

You hot-plugged it (connected regulator to SMPS while active), right?

Put an electrolytic or TVS (say P6KE43CA) across the input, at the DC-DC input terminals.  This reduces the peak voltage overshoot due to inrush.

But because it's only rated 40V, this probably won't be enough.  Choose a 60V regulator instead.

Tim

[Crios816]

HI.
yes, i hooked up the Buck Step Down after turning on the switching power supply.

Why did the over current/over voltage protection or switching power supply short circuit protection fail?

There is a possibility of reset the switching power supply ? or do I have to look for and replace the broken component?
Thanks so much

[T3sl4co1l]

Most likely OC had nothing to respond to.

This is somewhat unimportant as far as complexity level versus functionality, but what's happening is: the equivalent circuit is the SMPS output capacitors (a few 1000 uF?), starting with some initial voltage (36V), switched into wiring inductance (maybe 0.1-2uH? depends on length), into regulator input capacitance (a few uF?).

Such a circuit has an RLC transient response of overshoot and ringdown, before eventually stabilizing at the supply voltage.

This is exacerbated by the ceramic capacitor having less C as its voltage goes up, and having fairly low loss so that the ringing proceeds more or less in full.

The loss of an electrolytic absorbs the ringing energy instead, or a TVS clamps the overshoot.

This is easily enough simulated in a tool like LTspice, though if you just want stuff to stick together, it doesn't matter, get a new reg, appropriate ratings, protection.

Tim