Honda Insight Forum banner

1 - 9 of 9 Posts

·
Registered
Joined
·
5 Posts
Discussion Starter #1
Hello everyone. I have an Honda Civic Hybrid 2008. I want to know how much current is used from the HV batteries when the electric motor is assisting the car at an specific acceleration rate, and compare it with many variables such as speed, time, acceleration and batterie‘s SOC. For that purpose, I’ve been thinking about using the HV battery junction board’s current sensor (reference SU14779) to know how much current is coming out from those batteries when the car is being assisted, but the problem is: I don’t understand how that sensor works and how the information is transmitted to the MCM. I have some of this ideas:

1. Using a logic analyzer in the OBDII and figure out its CANbus code (but I don’t know how to do it)
2. Understanding how does the sensor behaves, so I can connect it to an arduino and figure out the current (but I couldn’t find any datasheet.) What kind of sensor is it? and, What kind of signal do I get in the terminals? Voltage? Current? Pulses?

I hope you guys can give me some orientation about that. If you also know about someone else who made it already and you can share it with me I’d really appreciate it.

Thank you!

84915
 

·
Premium Member
2001 5S "Turbo"
Joined
·
10,716 Posts
This is an Insight owners/community forum. Not many Civic owners here. Where are you located? Canada?
 

·
Registered
Joined
·
5 Posts
Discussion Starter #3
Hello Willie, thank you for your response. I’m sorry about that, I’m new in the forum and that’s why I posted this in the “Other Hybrids Discussion“ section. My apologies if my post should not be here.

Yes, I’m in Canada. Best regards!
 

·
Registered
Joined
·
5 Posts
Discussion Starter #5
Hi Sr. Thank you for your time.
Hybrid and electric vehicles has been my interest for the last two years and now that I recently got this HCH I’m studying the theory behind it. Understanding how energy works in the car allows me to thinking in posible upgrades for the system in the future, such as bigger batteries or implementing a manual assistant (this is a great idea I saw in a website called 99mpg.com, and I’d like to implement it in the car someday. they call it a ‘mIMA’).
For that reason, I want to start by knowing the total amount of current coming in and out in the HV batteries. With that information I can make different curves like voltage-current and current-acceleration, and compare their changes when fixed values like mass, road angle and SOC changes.
I know maybe it is crazy or not worth, or maybe somebody else already did it, but I just started to diving into a lot of theory and at least I want to understand how does the car behaves through experimentation by taking measurements. Maybe, in the future I could make a mathematical modeling and/or simulations (because I studied some physics and computer simulation at university), and compare it with the measurements.
Also, I feel encouraged to publish all my results and advances here.
I hope I answered your question.
Have a great day sr. Best regards from Canada.
 

·
Administrator
Joined
·
11,800 Posts
That all sounds good.

MIMA is widely discussed on here, as is a newer variant IMAC&C that I make.

I've done a fair bit of work on the HCH2 but the fact it is a CAN network based car makes hacking messages much more difficult than with the GEN1 Insight or the GEN1 Civic HCH1 hybrid as they used analog or simpler serial message systems.

You will first want to acquire a full set of the Honda HCH2 (HELM) workshop manuals.
Next read all the threads on here about HCH2 research and ideas.

You can use a variety of methods to current hack and//or voltage hack the IMA cars, many are discussed on here in depth, search for them all .

You could tap into that sensor, it's probably a 20ma current loop or some sort or analog 0-5v output.
Test one on the bench or take readings from it in operation.

You could get the current sensor data from the OBDII CAN bus by sending it the current PID etc and working out what the data returned means. Lots of CAN snooping and analysing required, you will also need an OBDII device capable of sending the correct PID so you can watch the responses. A clone Honda tool fly100 is what I use.


Once you have done some of the background reading then we can talk in more depth.
 

·
Registered
Joined
·
5 Posts
Discussion Starter #7
Thank you very much! I’ll follow your advice! Once I have any update I’ll post it here! Have a great day.
 

·
Registered
Joined
·
5 Posts
Discussion Starter #8
Hello everybody. In the meantime I am reading all the information in this forum, I started by making a simple experiment to understand how the HCH2’s current sensor works, and I will share what I got with you.

I have an old 120W DC source and found 4 50 Ohm 10W ceramic resistors in my inventory. Assuming that the sensor output is a linear 0-5V with low hysteresis, I proposed the next circuit.

84944

By connecting those 4 resistors in parallel we got an 12.8 ohm 40W (max) equivalent resistance (Real value measured with fluke multimeter).
I proceeded to connect the fluke clamp meter to measure the circuit voltage, the red multimeter to measure the circuit current and the flukes multimeter to measure the sensor’s output, as shown here.
84945
The current sensor is powered with an isolated variable dc-dc converter to watch and change the voltage supply. As Honda works with 12V (or 14.4V when its dc-dc converter is working) I will setup sensor Vin to 12V.
Now, the sensor has 2 voltage outputs called Vout1 and Vout2. First, I’m going to connect the multimeter as shown below.
84947
Voltages didn’t changes in the sensor output as the current changed. (0A—>1.225), (1A—>1,227V). This suggest me that maybe the voltage outputs are independent, or the sensor works by current. So I tried this connection.
84948
In this case, voltage didn’t changes either. (0A—>11.45V), (1A—>11.47V).
When Vout2 is measured the same way, voltage didn’t changes either. (0A—>10.26V), (1A—>10.26V).
But when changing Vin from the variable DC-DC converter, there’s changes in both outputs.

I think many theories can be considered:
1. Changing from 0A to 1A is too few for a sensor that measures higher changes, so that’s why the voltage differences are too small.
2. The sensor works by current, not voltage.
3. Sensor should be powered with a different voltage value, or it’s broken.
4. Sensor’s ground may be connected with DC source ground (I’m not sure about that. They’re actually completely isolated).
5. Sensor is supplied with current it is measuring, not an isolated source.

What do you think about this? Any suggestions? Tomorrow I will connect a fixed resistor between Vout1 and Vout2 and then figure out any voltage changes in that resistor, so I can see if that works by current instead of voltage.

Best regards! And stay safe at home.
 

·
Administrator
Joined
·
11,800 Posts
Have a look at the Insight battery current sensor explanation on www.99mpg.com
Read up on 20ma current loops.
Are you sure the HCH2 sensor is powered by 12V?
What does the HCH2 manual say about the current sensor connections?
 
1 - 9 of 9 Posts
Top