Precharge or no? Sunny Island
- Thread starter Maitake
- Start date
daklein
New Member
- Joined
- May 8, 2020
- Messages
- 190
I did set up my system so that the 48v bus could be precharged (with 2 large forklift FLA batteries). The SI6048US inverters would not start up correctly if the precharge circuit was used. I forget now exactly what they did, but I had to not use the precharge circuit. So beyond not needing to precharge, they don't work with precharging the bus.
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 20,600
Another member welded his GigaVac contactor with LiFePO4 batteries and SI, had to use precharge.
It should doable using suitable precharge resistance. That could be selected to supply the low 4W standby draw of SI. Maybe a maximum rise time is needed for power-on reset circuit?
If not a discrete resistor, I would consider a length of wire. That would reduce stress on capacitors from inrush due to low IR of lithium, and would reduce burning of contacts.
Out and back through a 50' coil of 12 awg for 0.159 ohms and 300A peak, or much shorted 8 wire LAN cable, could do it.
I have a 400 AGM bank, so no precharge in my system.
It should doable using suitable precharge resistance. That could be selected to supply the low 4W standby draw of SI. Maybe a maximum rise time is needed for power-on reset circuit?
If not a discrete resistor, I would consider a length of wire. That would reduce stress on capacitors from inrush due to low IR of lithium, and would reduce burning of contacts.
Out and back through a 50' coil of 12 awg for 0.159 ohms and 300A peak, or much shorted 8 wire LAN cable, could do it.
I have a 400 AGM bank, so no precharge in my system.
ChrisFullPower
Inverter Snob
Im assuming the DC caps are after the DC breaker?
ChrisFullPower
Inverter Snob
Good to know! I have a REC pre charge module for my contactor but it’s pointless if the breaker is off. I guess the ark will be in the breaker and not welding my contactor shut but I’d still like to be nice to the caps and avoid any arks if possible.
ChrisFullPower
Inverter Snob
I’m actually not so sure anymore….. I’m just now starting to test my system and wanted to confirm polarity on my main bus bars before starting the inverters. Had the SI DC breakers OPEN, allowed my BMS to pre-charge and close the contactor, checked my voltage with my meter - read 52v proper polarity, told the BMS to open the contactor. clunk - it opens but my meter still has 40-some-odd volts and is slowing bleeding down.
Maybe the main DC bus caps aren’t on the lug side of the breaker, but it sure seems like there’s at least something there? Maybe some low level logic that detects faults before start up? ?
Maybe the main DC bus caps aren’t on the lug side of the breaker, but it sure seems like there’s at least something there? Maybe some low level logic that detects faults before start up? ?
RCinFLA
Solar Wizard
- Joined
- Jun 21, 2020
- Messages
- 3,563
Inverter 48v battery input capacitors range from about 20,000 uF for low-cost HF 5kVA inverter to about 30,000 uF for LF inverters.
Total initial series resistance for BMS, batteries, cables, breaker, and capacitors ESR ranges from about 15 milliohms to 30 milliohms depending on particular system setup.
16S 48v Lithium battery with system of 15 milliohms total series resistance and 25,000 uF inverter input caps will have an initial peak current of a little over 3500 amps that decays off exponentially to low current level in about 1 millisecond.
You really don't want to subject any DC circuit breaker contacts to that initial surge current. Some BMS's will trip overcurrent protection and you really don't want to expose the BMS MOSFET's internal wire bonds to that much current surge. If you have a fuse instead of breaker system it may blow out fuse.
2 ohms resistor across breaker cable terminal lugs for at least 0.1-0.2 second before you flip the breaker ON will pre-charge inverter caps so you only subject breaker contacts to a couple amps max.
Nothing fancy really needed. I may be a bit unusual, but my inverters have had continuous applied battery power for over 17 years straight. That was the last time I actually used a pre-charge on inverters.
Total initial series resistance for BMS, batteries, cables, breaker, and capacitors ESR ranges from about 15 milliohms to 30 milliohms depending on particular system setup.
16S 48v Lithium battery with system of 15 milliohms total series resistance and 25,000 uF inverter input caps will have an initial peak current of a little over 3500 amps that decays off exponentially to low current level in about 1 millisecond.
You really don't want to subject any DC circuit breaker contacts to that initial surge current. Some BMS's will trip overcurrent protection and you really don't want to expose the BMS MOSFET's internal wire bonds to that much current surge. If you have a fuse instead of breaker system it may blow out fuse.
2 ohms resistor across breaker cable terminal lugs for at least 0.1-0.2 second before you flip the breaker ON will pre-charge inverter caps so you only subject breaker contacts to a couple amps max.
Nothing fancy really needed. I may be a bit unusual, but my inverters have had continuous applied battery power for over 17 years straight. That was the last time I actually used a pre-charge on inverters.
ChrisFullPower
Inverter Snob
I have the REC precharge module. So far (~10 cycles) I haven‘t welded my contactor yet ?
Hedges
I See Electromagnetic Fields!
- Joined
- Mar 28, 2020
- Messages
- 20,600
4W standby / 48V = 0.083A
0.083A x 2 inverters = 0.167A
0.167A x 300 ohm = 50V drop
50V battery - 50V = 0.0V
Doesn't seem enough to wake up SI, not sure how much SI draws playing tug of war with the resistor. i.e. what voltage is achieved before contactor closes?
I would consider 30 ohm for 1/10th the drop.
50V / 30 ohms = 1.67A
I think 3 ohm or 0.3 ohm would also be fine (possibly a bit of contact burning at 167A while bouncing, but should be well within needed cycle life. I think GigaVac has a graph.
The low resistance values would be high peak wattage (800 or 8000), depend on not having a continuous load. The 30 ohm resistor would be 83W, less than its 100W rating so would just get very hot if precharge never completed. (assuming heatsinked, which you likely wouldn't do.) OK if brief.
0.083A x 2 inverters = 0.167A
0.167A x 300 ohm = 50V drop
50V battery - 50V = 0.0V
Doesn't seem enough to wake up SI, not sure how much SI draws playing tug of war with the resistor. i.e. what voltage is achieved before contactor closes?
I would consider 30 ohm for 1/10th the drop.
50V / 30 ohms = 1.67A
I think 3 ohm or 0.3 ohm would also be fine (possibly a bit of contact burning at 167A while bouncing, but should be well within needed cycle life. I think GigaVac has a graph.
The low resistance values would be high peak wattage (800 or 8000), depend on not having a continuous load. The 30 ohm resistor would be 83W, less than its 100W rating so would just get very hot if precharge never completed. (assuming heatsinked, which you likely wouldn't do.) OK if brief.
