Inrush Current Limiting Resistor - Rectified 528Vrms 1.5mF

[Magnethicc]

Hi everyone,
I need to choose a combination of sereis/parallel resistors to charge 1.5mF capacitor through a rectified 528Vrms 50/60Hz input voltage.
I was thinking about HPC5C by KOA which can handle very high peak energy but our assembly house is a bit wary about using this resistor due to assembly reasons.

Does anyone have any recommendation for different resistor to handle the task?
Thank you.

[moffy]

Hi everyone,
I need to choose a combination of sereis/parallel resistors to charge 1.5mF capacitor through a rectified 528Vrms 50/60Hz input voltage.
I was thinking about HPC5C by KOA which can handle very high peak energy but our assembly house is a bit wary about using this resistor due to assembly reasons.

Does anyone have any recommendation for different resistor to handle the task?
Thank you.

What particularly are the assembly reasons, and what load current and speed of charge is required?

[Magnethicc]

Load current - none, I short out the charging resistor via a semiconductor switch.
Operating Temperature from -10C to 60C - so thermistor is not convenient for me since its resistance changes alot just from the operating temperature range.
Speed of charge - lets say peak charging current from the rectified mains is ~60A - this usually corresponds to somewhere between 100msec to 200msec.

assembly reasons:
1. Ceramic Resistor - Very very sensitive and can easily break during transport or falling to the ground.
2. 2mm tolerance for the length of body of the resistor makes it hard to fit in place and make the forming - could potentially be harmed during the forming process.
3. manufacturer says to avoid using part feeder and care must be taken while handling this resistor.

[moffy]

So a very rough calculation, 10R and between 1-1.5kJ surge rating for the resistor, please correct the calcs if it's wrong.

[coppercone2]

what the hell, I have a bunch of these ceramic comp resistors. I think the assembly house is just making bull shit up.

Very  very sensitive? What? It's a solid piece of ceramic. Its strong. And they don't have jigs to bend stuff? they sound bootleg and uninformed. Why don't you buy one and play with it. The dude is planning to drain your wallet. Its a 5W resistor too. Are they planning on dropping them? I could understand if it was a pirate that had two hook hands. I got 1w, 2w, 3w versions here, and a 5w somewhere. its basically like any other resistor in terms of handling. And for precise bending you just need a jig (even a major cheap skate place would just make it out of wood), but they bend fine by hand. I abused the crap out of them when I was bending them every which way (with my bare hands) doing experiments on different impedance analyzer fixtures. 0 fails .

this a rugged part intended for high stress applications. It would not work if it was weak.

p.s. it sounds like some bullshit involving concrete floors and stuff falling on top of stuff that is left on the ground. maybe they should get some esd flooring. this place sounds shady.

[Magnethicc]

So a very rough calculation, 10R and between 1-1.5kJ surge rating for the resistor, please correct the calcs if it's wrong.

How did you get the surge rating?
resistance is better around 14~20R.

what the hell, I have a bunch of these ceramic comp resistors. I think the assembly house is just making bull shit up.

Very  very sensitive? What? It's a solid piece of ceramic. Its strong. And they don't have jigs to bend stuff? they sound bootleg and uninformed. Why don't you buy one and play with it. The dude is planning to drain your wallet. Its a 5W resistor too. Are they planning on dropping them? I could understand if it was a pirate that had two hook hands. I got 1w, 2w, 3w versions here, and a 5w somewhere. its basically like any other resistor in terms of handling. And for precise bending you just need a jig (even a major cheap skate place would just make it out of wood), but they bend fine by hand. I abused the crap out of them when I was bending them every which way doing experiments on different impedance analyzer fixtures. 0 fails .

That is assuring! I have some samples I will try to play around with them to see if there is any water in the claims!

[coppercone2]

its zero I got one out right now. Ceramic around that size (1/4 inch thick, alumina for example) starts to break if you drop it on concrete and its more then like 8 inches long and it falls bad

ceramic stock likes to roll though, so its trouble some. the resistor will not

[moffy]

So a very rough calculation, 10R and between 1-1.5kJ surge rating for the resistor, please correct the calcs if it's wrong.

How did you get the surge rating?
resistance is better around 14~20R.

The total energy in the cap when charged is 0.5*C*V*V = about 2100 Joules, so just a guestimate that about half of that power would be dissipated in the resistor during charging, just a rough guess, could be way off.

P.S. Oops, I had one zero too many for the capactance, so maybe 100 Joules surge rating.

[coppercone2]

and I had zero problems using them for dead bug stuff (RLC filter), where the leads were like looped, soldered and it hangs in air between bnc connectors in a box. and making near triangle shapes out of the leads so it fits in a slim impedance analyzer fixture (big part, small connector pitch)

[coppercone2]

and if you are really concerned, you bend stress relief bends into it. Special pliers for this. Just make sure they are sized right for the dimension of the lead

https://forum.digikey.com/t/bending-and-forming-leads/8184

[Wolfram]

So a very rough calculation, 10R and between 1-1.5kJ surge rating for the resistor, please correct the calcs if it's wrong.

How did you get the surge rating?
resistance is better around 14~20R.

The total energy in the cap when charged is 0.5*C*V*V = about 2100 Joules, so just a guestimate that about half of that power would be dissipated in the resistor during charging, just a rough guess, could be way off.

P.S. Oops, I had one zero too many for the capactance, so maybe 100 Joules surge rating.

When charging a capacitor through a resistor from a voltage source, the energy dissipated in the resistor is equal to the energy stored in the capacitor, so the energy rating of precharge resistors needs to be equal to the capacitor energy at the worst-case precharge voltage.

Also seconding that bulk ceramic resistors like this are excellent for this application, and the fragility is not a practical issue at all in my experience.

[MariuszD]

If HPC5C resistors don't suit your needs, check out other ceramic composition resistors.
https://eu.mouser.com/c/passive-components/resistors/ceramic-composition-resistors
Ceramic composition resistors are best for absorbing large amounts of energy. It's not true that ceramics are easily damaged. Most resistors contain a ceramic substrate, yet damage is rarely seen.

[Magnethicc]

Ok so I threw some samples on the ground, tried breaking it with my hands and slammed a small weight on it tried to break it in half... safe to say nothing happened to the resistor. I also tried scretching the insulation also did not do much so yeah.. this resistors are pretty tough and we have decided to go furth with this resistor.

Thank you for your answers everybody!

[PCB.Wiz]

I need to choose a combination of sereis/parallel resistors to charge 1.5mF capacitor through a rectified 528Vrms 50/60Hz input voltage.

You did not say if that was being shorted after ramp, but another common inrush limiter are the NTC ones

https://www.digikey.com/en/products/filter/inrush-current-limiters-icl/151

[moffy]

When charging a capacitor through a resistor from a voltage source, the energy dissipated in the resistor is equal to the energy stored in the capacitor, so the energy rating of precharge resistors needs to be equal to the capacitor energy at the worst-case precharge voltage.

Also seconding that bulk ceramic resistors like this are excellent for this application, and the fragility is not a practical issue at all in my experience.

Thanks for the info, I did a quick bit of math of a capacitor being charged by a resistor with a constant voltage source, integrated power across the resistor over time and ended up with Er_tot = 0.5*C*V*V, good to know.