What kind of battery simulator does BMS choose?
Latest update time：2019-09-06
Reads：
In September, the Basketball World Cup held in China begins, and we can finally feast our eyes on it. This is also an important test for the reform of the national team in the past few years. Come on, the Chinese team.
Relay adhesion detection is a bit "brain-burning". I'll put it off for a while. Let's talk about something else today and change the topic.
Let’s talk about the selection of battery simulator.
As mentioned in a previous article, there are requirements for the voltage accuracy of BMS collected battery cells; whether it is self-test or customer acceptance, we need a relatively accurate
reference source
to test the accuracy of our hardware and verify whether it meets the requirements. If you use a battery as a reference source, firstly, it is troublesome to adjust the voltage. Secondly, the voltage of the battery will gradually change. Thirdly, it is unsafe and you need to be extra careful during testing.
Therefore, the battery simulator appeared. The most basic function is to set the target voltage value arbitrarily, with stable accuracy, a certain load capacity, and meeting the measurement requirements (as shown below, a relatively common foreign battery simulator).
I still remember that when I first started working, the first board I made with my master was a battery simulator tooling. The function I made at that time was relatively simple, just a DA output.
But now there are a large number of mature products on the market that can meet various testing needs. The test system framework is as follows (picture from NGI official website):
In fact, we use battery simulators more to simulate
functions such as
cell voltage and temperature
, simulated equalization and fault simulation (the picture below is from the ETAS official website), and other things like simulating the charging and discharging process of real batteries. In fact, I am It doesn't feel very reliable.
Let’s analyze several key requirements for choosing a battery simulator.
1. Analog cell voltage
This is the most important requirement. First, you need to decide how many voltage analog channels are needed. Generally, products have more than 12 channels. You can just mention this based on the product requirements.
The channels are electrically isolated from each other. If they need to be connected in series, they need to be connected through a wiring harness outside.
The principle of voltage simulation is as shown below (picture from
https://www.docin.com/p-835651570.html
). The output target voltage is set through the DA chip, and then the input voltage at the feedback end is collected through the AD chip to form a closed-loop system. , adjust the output voltage in real time and provide output accuracy; therefore, the general simulator voltage output has four lines: two output lines and two sampling lines.
Regarding voltage accuracy, in fact, this is also related to AFE. For the same battery simulator, the results collected by AFE from two manufacturers may have a deviation of more than 1mV. Excluding these interference factors, I think the accuracy must be
below
1mV
to meet general needs (especially during production line production).
Generally, manufacturers will give their accuracy parameters as shown below (sourced from NGI official website), and choose the appropriate product mainly based on the two parameters of setting resolution and setting accuracy.
2. Simulate temperature
A common and effective solution for temperature simulation is a resistor array, as shown in the figure below; we determine the required resistor value range based on the actual NTC resistance table.
The simulator sets four resistors with resistance values 1, 2, 4, and 8 multiples of this range on each resistance range. Then based on the piecing together of these four resistors,
any resistor in this range can be obtained. value
; in this case, by piecing together different gears, we can get any resistance value we want.
The resistance range in the figure below can meet most of our actual needs.
What needs to be noted in the application is that each temperature sampling channel in the BMS must be configured with a separate temperature analog resistor. Different channels cannot be shared, otherwise it will cause internal voltage division and sampling deviation.
3. Analog equalization
The battery simulator must also have a certain output capability, including current sourcing and sinking (the picture below is from the Internet), which is used to simulate the battery balancing function.
If a simulator is used in parallel for multiple AFEs at the same time, the output capacity of the simulator should be larger, basically about a few amps; if the output current selected is too small, various problems will occur during testing. Strange question.
Summarize:
Someone questioned: If I use a battery simulator to test the BMS and get the conclusion that the accuracy of the BMS meets the requirements, does this conclusion mean that the accuracy of the BMS also meets the requirements when testing real batteries? This is a good question. Facts speak louder than words. It is left to everyone to verify for themselves.
京公网安备 11010802033920号