Electronics > Beginners

Understanding circuit for inrush current limiting

<< < (7/8) > >>

spec:

--- Quote from: AngraMelo on December 22, 2018, 02:46:52 pm ---So for example,
I have a 700W transformer, 220ac primary and 2 secondary winding, each 45Vac capable of 7.5A each.
If I want to limit the inrush current for the primary side, what I have available is this:

Fist my math: Max allowed peak current on primary = 2A
Vpeak = (Vrms + 5%)x1.414 = (220 +11)x1.414 = 326Vpeak
Rminimum = 326/2 = ~162ohms

I would have to bite the bullet and go for the 100R NTC, which has no markings, no brand, is 5mm (I have no idea how the size comes into play, probably heat dissipation but with no datasheet no way to be precise)

And there is our candidate:
https://proesi.com.br/termistor-ntc-100r-5mm.html

--- End quote ---

Your calculations are  probably fine, but you are being over-cautious when you say that the maximum peak current allowable is 2A. Transformers, and many other components, will take quite high peak currents, for a short period without a problem. So, if I were you, I would try your circuit with just a fuse and see how it goes.

By the way, computer PSUs and the like, do not have a transformer input. Instead they have a diode rectifier feeding a large reservoir capacitor to generate 309V DC peak from 220V RMS AC input. The 309V is then fed to a switch mode PSU circuit, which switches at 50kHz to 4MHz and produces the various DC supply lines required by the equipment. In computer PSUs the SMPS is nearly always a fly-back type.

The bottom line is that the reservoir capacitor has a practically zero impedance so, with that architecture, it is essential to limit the inrush current to protect both the rectifier diodes and reservoir capacitor.

About your unknown thermistor, you can can determine its characteristics by measurement. The tricky bit is determining the thermistor maximum working voltage, but even that is not too much of a problem, but you would not get a definitive answer for all thermistors of that type.

ArthurDent:
Here is a circuit to limit inrush current to a transformer using an NTC/relay and it explains why the relay should be included in the circuit. Good explanation of the circuit and hints on how to adapt it to other situations. Most of the previous circuit was a line filter for connecting to a transformerless or SMPS.

https://www.instructables.com/id/Controlling-the-Inrush-Current-Required-by-Large-T/ 

AngraMelo:
For a beginner, it is hard to know what is actually critical and what isnt.
Great to hear that I if I pick more or less the right NTC I could get away with it, Ill get some, make the tests and come back to you guys.

I heard all sorts of things like: inrush current will slowly kill your capacitors and transformer...
So I ended up considering this a situation that needed a "precise" solution, when in fact, it doesnt.
I guess if Im in the right ball park of current limiting it would be way better than just having the Ac line dumping current directly.
So Ill make do with what we have on hand.

PS: Spec, there is no way around it, Ill have to learn Eagle. Im having a lot of problems with undesired oscillations and Im pretty sure it´s due to my horrific board layout tecniques that involve jumping everything with tiny wires. I quickly discored that some voltage regulators and Opamps are not happy with that.
I already have the software and Im downloading some libraries.

spec:

--- Quote from: AngraMelo on December 23, 2018, 08:26:52 pm ---
PS: Spec, there is no way around it, Ill have to learn Eagle. Im having a lot of problems with undesired oscillations and Im pretty sure it´s due to my horrific board layout techniques that involve jumping everything with tiny wires. I quickly discovered that some voltage regulators and opamps are not happy with that. I already have the software and I'm downloading some libraries.

--- End quote ---
Excellent move learning EAGLE (version 9.2.2 now). It is very easy to use for doing circuits once you get to know its funny ways.

Yes, a good layout and decoupling are vital for many high-gain functions like opamps, power amplifiers and PSUs. A good layout etc is also imperative for high speed circuits, both analogue and digital. And when you get to microwave frequencies, even 5mm of extra track/wire will wreck the performance. It amazes me that opamp circuits, for example, ever work when you consider that a typical opamp has a gain of 120dBs, which is a voltage gain of one million. That means that 1uV at the input would result in 1V at the output. You inject 1uV into a circuit just by looking at it. :)

Luckily, there are some simple rules that are easy to understand and do, that will ensure that you get the best performance and minimum problems.

But, apart from simple circuits, you nearly always get problems with the first build.

spec:

--- Quote from: soldar on December 22, 2018, 12:28:56 pm ---Pretty much all computer PSUs add inrush limitation and line filtering and I would include it in any design of mine. Filtering may be legally required but, in any case, protects the device from external transients and noise and also prevents transients from the device from going out.

I do not remember when I last bought any components for this because I have a crate full of old PSUs and I just get my components from there.

--- End quote ---

Computer PSUs do not have a transformer input- they have a rectifier/ capacitor input which has quite different inrush characteristics to the OPs circuit.

Yes, it is amazing the components you can get from junk, and computer PSUs are a rich source. One of my favorite sources of components are the bases of failed EL ceiling lights: capacitors solid and electrolytic, transistors, inverter transformer.

Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod