Chargery Amperage wild fluctuation

[cinergi]

I just received a BMS-16T with a 600A shunt and 300A DCC. I have everything working well EXCEPT that during discharge (of a new 4s 12v 100Ah LiFEPO4 battery using a cheap modified sine inverter), the amperage readout on the chargery fluctutates wildly despite all my other measurements (DC and AC measurements with a Fluke; load on the inverter is a heat gun). I'm talking 2A, 3A, 3A, 3A, 7A, 59A, 37A, 3A, 59A, etc... and the real load is a constant 14-15 amps.
Charging the battery with my desktop power supply at 10 amps reads perfectly - 10 amps on the chargery.
When I take a voltage reading across the shunt, it's stable - yet the chargery is wildly inaccurate - but only during discharge with this inverter.
One thing to note is that since this is the BMS-16t connected to only 4s, I have an external power supply (plugged into household/street power) powering the chargery.

I haven't reached out to chargery yet. I also have two chargery units (and 2 DCCs) and I swapped out ALL components. Same results. I even calibrated the shunt. Any thoughts?

 

[cinergi]

I do see in the manual (FAQ) #4 for this issue that a discharge current ripple might be a problem. And to add a low-pass filter on current sensor ... hrm... I'll need help with that (like, exactly what to buy and where/how to install it)

 

[Ampster]

One thing to note is that since this is the BMS-16t connected to only 4s, I have an external power supply (plugged into household/street power) powering the chargery.

I am not sure what the purpose is of the external power supply or how a BMS designed for 16S will be accurate if only 4 of the balance leads are connected.

 

[Bob B]

Others have reported that same phenomena. Maybe someone will pop in with a link to this discussion about this.

 

[cinergi]

I am not sure what the purpose is of the external power supply or how a BMS designed for 16S will be accurate if only 4 of the balance leads are connected.

The BMS-16t requires 15+ volts to operate. It's designed to work as a 2s through 16s configuration but requires external power if less than 8s configuration.

 

[cinergi]

Others have reported that same phenomena. Maybe someone will pop in with a link to this discussion about this.

I tried searching but not having luck ... but I bet it's the current ripple from the modified sine wave inverter... @Chargery 's solution is a low-pass filter in the FAQ .. I'll need some help with that.

 

[Cal]

Yep, it’s inverter ripple current. It might be easier to leave it alone. You’re seeing extreme ripple current. A sine wave inverter would be a lot better.

I’m using a 100mH, 100uF filter in my diy current meter to eliminate inverter ripple.

 

[Steve_S]

Unfortunately some Mod Sine and even some really poor Pure Sine (cheapo knock-offs) have crazy ripples and filtering is needed. I know that a few who have experienced this had upgraded to a better Inverter and had no issue. I can't speak to how best to filter it nor the cost of doing so and if there is any trade-offs by doing so. Truth be told, Mod-Sine is a terrible choice for running most equipment, it's fine for short term Car Inverter use if our ina pinch and have to get a cheap inverter. Why Mod-Sine is being sold is because it's cheap !

If using the Chargery Supplied Shunts that came with the unit, Calibration is not required, they are precalibrated.

Correct, A Chargery can be Battery Pack powered IF the battery pack is greater than 15 Volts, so for an 8S and up.
A BMS-16T, can do 4,8 or 16 cell Packs. The BMS requires 15VDC and up to 3.0A if driving two relay/contactors. Less if using a DCC obviously.

!!! Be CAUTIOUS with the BMS, any contact with an electrical source on the casing can short it, these are quite sensitive.

 

[Craig]

!!! Be CAUTIOUS with the BMS, any contact with an electrical source on the casing can short it, these are quite sensitive.

Yes this should always be taken into account.

 

[cinergi]

WRT shunt calibration - I actually *needed* to calibrate one of them (I have two sets of BMS-16t's) .. it was not only reversed (reading -8) but was wrong (should have been +10). But yes, they're supposed to arrive calibrated.

I was hoping to use a cheap modified wave inverter for load testing etc.. not for anyhthing serious.. which is why I ddin't want to drop much money on it... Looks like I need a new plan

All in all, the chargery with the DCC is working well. I'm working on a video now ...

 

[cinergi]

I'm using 5 55 watt 12v lights to create 22 amps of load and the chargery is quite happy with that ... so definitely was the ripple current from the inverter. (picture of the silliness here: https://diysolarforum.com/threads/new-chargery-dcc-dc-contactor.12045/#post-135869)

 

[Bzzzt]

I had the same issue with a cheap gopower inverter leading to a DIY current monitor. Similar (but different!) to Cal's solution I added an RC filter between the shunt and monitor inputs which will average out the signal. I used 4.7k resistors followed by 10uF capacitors shorted to ground (assuming you have a low side shunt!) on both inputs. This reacts quickly enough to changes in currrent but smooths the noise in my case. It also stopped high frequency noise from knocking out my whole monitor, which it would happen occasionally.

 

[Steve_S]

A general rule which is often written into Inverter Manuals:
Do NOT keep the battery wires far apart. Keep them taped together to reduce their inductance and induced voltages. This reduces ripple in the battery wires and improves performance and efficiency. (Direct extract from Samlex EVO Inverter Docs and there is a LOT more on this topic in their docs)

I can attest to the amount of "noise" that can ripple through if the wires are NOT together and as I have some pretty long battery cable runs (14') from battery terminal to Inverter) and all of it 4/0 fine strand welding cable (Royal Excelene).