[OR-Carl]

I am getting pretty close to starting on my high voltage system for my truck, and I have been kicking around ideas in my head for a while now. I wanted to get them down on "paper," and also see what people thought of my plan so far. Any and all feedback is welcome.

First off, I am going to keep the Tesla-designed battery box, but split the string in half. Each half will have a high voltage fuse (maybe 200A rating?) in the positive side before the main positive contactor (C1). My first question is about C0, the contactor that breaks the negative side of the circuit. This was how the pack was initially wired, and I suspected it was about redundancy and extra safety? It came with the modules, so I am inclined to leave it, as it is no extra cost to me.


The battery box is roomy once I pulled out all of the original brains, and the TSM BMS comes in nice small little units, so they will all go inside the box. The software can monitor parallel strings, I will just need more of their satellites than I had originally thought. The BMS will communicate with the charger over CAN, so I am thinking that I will not really need to use any of the contactors (like C6) to mechanically stop the charging, right? If anyone is using this charger and BMS setup, I would love to hear some input on this topic.


The main power cable will go from the battery box to the HV junction box, where I plan to tap the main feed to the motor to power a smaller fuse block. I have not searched locally for the cable, is it something that a welding supply place would carry? Is 2/0 enough for a modest system? Each circuit will have its own fuse and contactor, which means that there will be a degree of redundancy (if for example one of the main contactors welds itself shut, there should not be high voltage present anywhere outside of the main junction box when the system is shut down).


I think I am going to handle pre-charge with a little control module that Zeva makes - seems like a good way to not have to fuss with it.


On my dashboard then will be a main control panel. To start the car, there will be a switch with those cool red guards, and maybe some sheet metal tabs to make it hard to accidentally shut off your whole car while fumbling to turn on the defroster. So switch 1 will open the positive and negative main contactors, and run power to the ignition switch. My thinking here is that if I want to charge my car at the store, I dont want to have to leave the key in the ignition. When the main switch is on, and the key turned to run, the precharge will activate automatically, the DC-DC converter will come on, and I will get 12v power to my dashboard. The 12v power will fire up the vacuum pump, and start the coolant pump - which will always run. The coolant temperature relay controller will also come on, and monitor the coolant - if it is below 50 degrees it will feed high voltage to a coolant heater until the temp gets up to where I want it. I probably wont have the coolant heater and pump run when I charge without the key, as I will probably never charge it unattended without driving it first. Also, I am not going to be charging it at a very high current, so unless it is very hot out, there will probably not be enough battery heating during charge to worry about.



The motor controller does have CAN functionality, but it seems only to connect to a power usage display. They are fairly expensive, so i might just add a simple clamp-on meter to see how much current I am drawing. I have seen modules that output a signal to the tach, which I sort of like the idea of. We will see.



Anyway, does this all seem like it sort of makes sense? Am I forgetting something obvious?

 

[OR-Carl]

I thought of one thing today that I wanted to ask about - Should there be some sort of manual disconnect between the battery box and the junction box? Tesla used a very beefy 2-pin connector at the box, but sadly it came damaged, and did not have the mating half anyway. I suspect there will be plenty of projects to do under the hood going forward, and it seems like it might be nice to be able to quickly disconnect the traction battery and know that there is no continuity to the high voltage battery while I work. Would a high ampacity marine switch like this one work?


https://www.amazon.com/dp/B000MMC914?tag=duckduckgo-ffab-20&linkCode=osi&th=1&psc=1


The rating is only for 48v, but is that relevant if you are only switching it while it is not under load?

 

[kennybobby]

i like the idea of a manual disconnect, and most OEM EVs seem to have one that is recommended to pull before working on any of the HV systems. Not sure that 48V device is what to use though.

it's not clear what battery pack you are using, but are you splitting it into 2 parallel strings?

i would want to put C2 in parallel with C1 inside the battery box with a fuse on the output to protect the 2/0 cable.

Then a startup sequence would be C0, then C2 for sufficient time to pre-charge the caps, then C1 followed by cutting off C2.

i would think a good current sensor in the pack would be required; you will want to use that signal to assist in controlling the power-down sequence, e.g. when the key goes to OFF all the load devices are also turned OFF, but the main contactors remain closed until the current goes to zero, then switch off C1.

There should be some bleed resistors on all the big capacitors; so C0 remains closed until that bleed time has passed, then switch off C0.

just my 2¢,

 

[OR-Carl]

I think I will plan on adding some sort of large 2/0 socket that is rated for a lot of amps to manually disconnect my pack. It occurred to me that the redundant contactor on the charger (C6) would cause problems with my unattended charging mode, as the previous drawing would only show it closing with the key turned. Perhaps even better would be to have a switch that opens C6 for charging, and then it can be shut off the rest of the time.

I added some details on the battery - it will be two strings, and each string will have a fuse, so the 2/0 cable should be protected by those. The smart pre-charge module has a resistor and a timer/sensor (not really sure which). So instead of controlling C2 directly, you give power to the SPM which opens a resistive bypass around C2, waits, then applies power to the coil of C2. Should take the guesswork out of precharging, and makes it impossible for someone unfamiliar with the car to do it in the wrong order.


Kennybobby, I would like to hear more details on your thoughts about bleeding down the caps. Are you suggesting that there are resistors that will bleed them down, or that I should add them?



I also see that trying to shut down the system while large currents are flowing is going to be potentially damaging to my contactors - but it would seem like breaking currents would at least be less likely to weld the contacts than shutting a contactor onto a large unfilled capacitor. The proper shutdown procedure then would be to shut off the heater if it was running, then shut off the ignition- which would turn off the DC-DC and close all the HV contactors for the heaters. I would be calling on those contactors to potentially break the current being drawn by their loads, but the water heater will probably only pull like 12amps, and the DC DC even less, like 4 maybe. As long as I dont try and shut the car off while cruising at top speed, do I really need to worry about there being no current draw?

 

[kennybobby]

i would recommend that you do a search here for Wolftronix's pack build thread on how he planned to connect parallel strings and prevent internal current between strings in the event of a weak or shorted cell. Basically a means to take a string off-line if there is a fault and continue in limp-home mode.

Bleed-down resistors: i would expect that the inverter input and the charger output capacitors would have these resistors built into the circuit; but better to check and verify than to assume. If none exist then they should be added.

As shown the charger probably does need to have it's own contactors--it doesn't need to be energized when driving.

The main contactors are usually fairly expensive and if switched with load current flowing will suffer arcing damage to the precious-metal contact surfaces. The junction resistance is usually very very low, but a carbon coating that frosts the contact surface will increase the resistance and cause heating, voltage drop and power loss. Countermeasures to prevent arcing are highly recommended.

Battery pack voltage and current sensing for monitoring and control.