Off the shelf inrush limiter / input surge limiter

[Roswell Bob]

If it is built into the BMS it is even easier than that. Just dedicate a FET or two to a pre-charge circuit and put it in series with a resistor that limits the current. Turn the pre-charge on first and once the voltage gets high enough turn on the main bank of FETs. The capacitors do not have to be fully charged. Even charging them to half the battery voltage will probably reduce the surge to a low enough level to eliminate all possible problems. (For safety, you would want to limit the time the pre-charge is allowed to be on and if it is exceeded, shut everything down with an error. This would prevent the resistor from getting too hot. Alternatively, use a high wattage resistor and heat sink so even if there is a dead short on the output, the resistor does not overheat)

Sweet design!

 

[corn18]

SL22 0R516 Ametherm | Circuit Protection | DigiKey

Order today, ships today. SL22 0R516 – Inrush Current Limiter 500 mOhms ±25% 16 A 0.866" (22.00mm) from Ametherm. Pricing and Availability on millions of electronic components from Digi-Key Electronics.

www.digikey.com

J115F31AH24VDCS61.5U CIT Relay and Switch | Relays | DigiKey

Order today, ships today. J115F31AH24VDCS61.5U – General Purpose Relay SPST-NO (1 Form A) 24VDC Coil Chassis Mount from CIT Relay and Switch. Pricing and Availability on millions of electronic components from Digi-Key Electronics.

www.digikey.com

Invalid Request

B41821F7478M000 EPCOS - TDK Electronics | Capacitors | DigiKey

Order today, ships today. B41821F7478M000 – 4700 µF 35 V Aluminum Electrolytic Capacitors Radial, Can 70mOhm @ 120Hz 2000 Hrs @ 85°C from EPCOS - TDK Electronics. Pricing and Availability on millions of electronic components from Digi-Key Electronics.

www.digikey.com

Use 3 relays, otherwise 1 each. figure another buc or two for wire and crimps and so forth. Maybe we can afford your $7 box

Do you still need a contactor?

 

[Roswell Bob]

Do you still need a contactor?

No. The relays can be wired in parallel for a total current of 150A. I am using a 125A fuse.

The bus is charged with the 1 ohm NTC thermister. The 400ohm resistor and 4700uF capacitor provide a short delay to the relays.

I've used the NTC and relay approach on 20kW inverters. I probably shipped over a million and haven't heard of any problems.

 

[corn18]

No. The relays can be wired in parallel for a total current of 150A. I am using a 125A fuse.

The bus is charged with the 1 ohm NTC thermister. The 400ohm resistor and 4700uF capacitor provide a short delay to the relays.

I've used the NTC and relay approach on 20kW inverters. I probably shipped over a million and haven't heard of any problems.

Can it go to 400A? Unfortunately, I am running a 3kVA inverter on 12V.

 

[toms]

BTW: I really don't like using contactors/relays like the REC system requires......but that is a whole different thread.

I’m on the other side of that equation - i’ve seen too many FET based BMS systems fail to be interested in using one.

But yes, the precharge is ONLY to protect the contactor - if you don’t have a contactor, you don’t need one.

 

[toms]

I've used the NTC and relay approach on 20kW inverters. I probably shipped over a million and haven't heard of any problems.

Only problem will be when the system tries to disconnect the relay for a LVD and the relay contacts are arc welded shut.

 

[Roswell Bob]

I’m on the other side of that equation - i’ve seen too many FET based BMS systems fail to be interested in using one.

But yes, the precharge is ONLY to protect the contactor - if you don’t have a contactor, you don’t need one.

No problem with the fast fuse clearing on power up without an inrush limiter? And yes, I could see the contactor having problems opening up under load. Maybe a RC snubber across the contacts would help.

 

[Horsefly]

But yes, the precharge is ONLY to protect the contactor - if you don’t have a contactor, you don’t need one.

Yeah, I don't quite agree. If you have a fast-blow Class T fuse, the precharge also prevents it from blowing due to the inrush current on the capacitors.

 

[Roswell Bob]

Only problem will be when the system tries to disconnect the relay for a LVD and the relay contacts are arc welded shut.

I've given this some thought. The BMS is supposed to open under a fault condition. (I am using an Overkill Solar with integrated fets.) I would expect the BMS to open long before the precharge contactor. So there shouldn't be a problem with contactor opening up while passing current. I hope the BMS can handle turning off the fault. What is LVD? Undervoltage?

 

[toms]

I've given this some thought. The BMS is supposed to open under a fault condition. (I am using an Overkill Solar with integrated fets.) I would expect the BMS to open long before the precharge contactor. So there shouldn't be a problem with contactor opening up while passing current. I hope the BMS can handle turning off the fault. What is LVD? Undervoltage?

If using a FET based BMS, unless as horesfly mentioned you are running a fast blow fuse - you don’t need a precharge.

I can’t think if any reason to run a contactor in conjunction with a FET based BMS.

LVD is low voltage disconnect. My auxiliary system (bore pump / desal plant / RC aircon) often runs into the night until the battery reaches 50V, then it opens the inverter contactor until the sun comes up.

Then the contactor is switched back on and requires automatic precharge (done via a REC precharge module).

In a full off-grid setup, it is much more efficient to have two complete systems, one for non-essentials, and one main.

If you are not planning to run your system to LVD, it is not important to have automatic reconnect.

You just need to understand what you are trying to achieve.

High Inrush current is very likely to damage contactor terminals.

Depending on your wiring, inverter type and main battery fuse style, high inrush current may blow your main fuse.

If neither of the above apply to your system - high inrush current is not an issue.

 

[Roswell Bob]

If using a FET based BMS, unless as horesfly mentioned you are running a fast blow fuse - you don’t need a precharge.

I can’t think if any reason to run a contactor in conjunction with a FET based BMS.

LVD is low voltage disconnect. My auxiliary system (bore pump / desal plant / RC aircon) often runs into the night until the battery reaches 50V, then it opens the inverter contactor until the sun comes up.

Then the contactor is switched back on and requires automatic precharge (done via a REC precharge module).

In a full off-grid setup, it is much more efficient to have two complete systems, one for non-essentials, and one main.

If you are not planning to run your system to LVD, it is not important to have automatic reconnect.

You just need to understand what you are trying to achieve.

High Inrush current is very likely to damage contactor terminals.

Depending on your wiring, inverter type and main battery fuse style, high inrush current may blow your main fuse.

If neither of the above apply to your system - high inrush current is not an issue.

I will be running a Class T fuse or a semiconductor fuse- both are likely to have a melting I^2t less than the charging I^2t of the bus caps. That is what this discussion has been centered around since it started.

The contactor is what shorts out the precharge resistor.

LVD (Undervoltage) is just one of the faults that the BMS will respond to. I am as concerned with bumping into an overcurrent/overload event.

The contactor shouldn't see any inrush. Inrush event will be over before contactor is closed.

I am capable of modeling inrush based on the system parameters. I did it for a living a few years back. Signature Solar is supposed to be supplying input bus capacitance values for a more accurate model. I am using a 0.1F value for now.

 

[toms]

In your situation i would use a slow blow main fuse capable of handling the expected inrush current.

No need to complicate the system with a precharge circuit and contactor.

 

[Roswell Bob]

In your situation i would use a slow blow main fuse capable of handling the expected inrush current.

No need to complicate the system with a precharge circuit and contactor.

Without a precharge circuit the BMS fets will be turning on into a lot of capacitance. Do you think BMS will survive that?

 

[Roswell Bob]

In your situation i would use a slow blow main fuse capable of handling the expected inrush current.

No need to complicate the system with a precharge circuit and contactor.

I'm going to use a Class T or semiconductor fuse. A slow blow fuse wouldn't offer much protection in the event of a fault. A dual element may be a good choice. The simulations I have done show that a 125A Class T will clear on inrush, while a 200A will survive. I'm leaning towards a larger Class T fuse to protect against short circuit with a 125A breaker to protect the wiring. I've designed a bunch of precharge circuits for inverters- so for me it is very easy. It's not complicated and the benefit is hugh.