[OR-Carl]

Hey wjbitner, thanks for the input! I read the documentation again on the balancing for the BMS, and you are correct that it should be possible to run it whenever the BMS is on - I was thinking it would only do its thing while charging. Anyway, I decided to run some numbers:

The BMS uses 24ohm resistors, so at 3.7v it will shunt about... 0.56watts

It splits the pack into groups of 6, picks the highest one (you can set a lower limit so it ignores the ones that are already low) and starts discharging. (this means discharge time will need to be multiplied by 6)

So, how much power are we talking about? Here the math gets a little fuzzy... Assuming the modules are 3kwh nominal - and that discharging from 4v down to 3v is 90% of the capacity, 1 volt would be (3000*.9) = 2700wh. Assuming a linear discharge curve, 0.035v would be 95.5wh - which is lot. At 0.56w, that would be about 172 hours... per cell. So do that 6 times and you are looking at 1032 hours, or about 43 days. I tried to be generous with my assumptions, so it could be a decent bit less.

Soooo, I had a thought. Part of the reason I bought these modules is that they are a good match to my off-grid stuff - so I am going to wire a little 300watt inverter through a power meter and then connect it to each high module one at a time. Running a 300watt load should bring the voltage down in about 20 minutes if my numbers are close. If I make sure to draw the same number of watt-hours from each module, the BMS should be able to handle the rest in a couple days.

 

[wjbitner]

Hey wjbitner, thanks for the input! I read the documentation again on the balancing for the BMS, and you are correct that it should be possible to run it whenever the BMS is on - I was thinking it would only do its thing while charging. Anyway, I decided to run some numbers:

The BMS uses 24ohm resistors, so at 3.7v it will shunt about... 0.56watts

It splits the pack into groups of 6, picks the highest one (you can set a lower limit so it ignores the ones that are already low) and starts discharging. (this means discharge time will need to be multiplied by 6)

So, how much power are we talking about? Here the math gets a little fuzzy... Assuming the modules are 3kwh nominal - and that discharging from 4v down to 3v is 90% of the capacity, 1 volt would be (3000*.9) = 2700wh. Assuming a linear discharge curve, 0.035v would be 95.5wh - which is lot. At 0.56w, that would be about 172 hours... per cell. So do that 6 times and you are looking at 1032 hours, or about 43 days. I tried to be generous with my assumptions, so it could be a decent bit less.

Soooo, I had a thought. Part of the reason I bought these modules is that they are a good match to my off-grid stuff - so I am going to wire a little 300watt inverter through a power meter and then connect it to each high module one at a time. Running a 300watt load should bring the voltage down in about 20 minutes if my numbers are close. If I make sure to draw the same number of watt-hours from each module, the BMS should be able to handle the rest in a couple days.

Cool. I was thinking since you really need the other batteries to move down .05 volts, this shouldn't take too long ( I was thinking that you'd do it while you were working and could monitor every 1/2 hour or so). I was guessing a 2-3 work session. A complete discharge would take quite some time.. Glad you can have fun and play with it.

 

[OR-Carl]

I was thinking since you really need the other batteries to move down .05 volts, this shouldn't take too long

Haha, yeah, it surprised me just how long it would take for the BMS to drop a mere 0.030 volts off the rest of the pack - but it speaks to how balanced these packs should tend to stay. I did some tests today, and my numbers actually came up surprisingly close!

Here is my power dumping rig:

I connect the ring terminals to each module, and then I have been running a box fan on high- which dumps about 200watts. The first couple modules I did not drain far enough, as the voltage sags while under load, and then recovers when I disconnect everything. The first module (cells 36-42) I drained out 87wh, the second (22-28) 82wh, and finally on (15-21) I pulled 96wh. That was pretty much dead on with my estimate.

pack 1 ----------
c1 - 3.764v >>DIS c13- 3.763v c25- 3.739v - c37- 3.739v -
c2 - 3.764v c14- 3.763v >>DIS c26- 3.739v - c38- 3.741v
c3 - 3.764v c15- 3.731v - c27- 3.739v - c39- 3.742v
c4 - 3.764v c16- 3.731v -- c28- 3.739v - c40- 3.741v
c5 - 3.764v c17- 3.731v - c29- 3.732v - c41- 3.740v
c6 - 3.764v c18- 3.732v - c30- 3.730v -- c42- 3.739v -
c7 - 3.764v c19- 3.732v - c31- 3.731v -- c43- ------
c8 - 3.764v c20- 3.732v - c32- 3.732v - c44- ------
c9 - 3.764v c21- 3.732v - c33- 3.729v -- c45- ------
c10- 3.764v c22- 3.740v c34- 3.729v -- c46- ------
c11- 3.764v >>DIS c23- 3.739v - c35- 3.729v -- c47- ------
c12- 3.764v c24- 3.737v - c36- 3.743v c48- ------

Here you see the BMS is manfully trying to discharge (>>DIS) the high cells in the rest of the string. Looking at the values from yesterday, It looks like it managed to shave .001v off a few of the cells .

I will need to go back and drain a bit more off, but I am confident that I will be able to get the cells down to a pretty close match. The really good news here is that my capacity seems to be pretty darn close to the nominal rating. It should mean that I have a solid 30kwh or so of useable capacity. I will try and crunch some more numbers when I get a chance.

I am going to try and get my charger hooked up next, and maybe just trick it into charging the string I am balancing up to the voltage of the other string. I have some questions about charging parameters, but I might start another thread for that.

 

[OR-Carl]

I did some work today on getting my HV wiring in place. I made some coverplates and cut some holes to align with the new position of the contactors. Once I got everything fitted, I tore it down to paint the covers, and will silicone them in place for the final install.

Part of me feels like there ought to be a service disconnect on the HV + at this spot, but I am not really sure what to use. As long as the contactors do not both weld shut, unplugging the 12v battery should completely isolate the voltage within the box.


I also have 2 conduits going between the cab and the box, one carries power to the contactors, power to the BMS, and CAN wiring plus the serial cable to interface with a computer. The other conduit will be for the power from the charger. How touchy is CAN about being next to power conductors? Should I move the power for the contactors (2-5watts of 12v) into the other conduit just to be on the safe side?

I had some problems with the connector that schelle63 warned me about - the plastic had become super brittle, so trying to plug in the spade terminals just snapped off the internal retainer. The outside retaining clip also snapped off. Anyway, I will keep a close eye on that, and will replace the whole thing at the first sign of trouble. I will say though, I contacted Thunderstruck Motors about the clip - and they sent me out a free replacement right away. They have been a great company to work with - and their products have been really well documented and easy to use.

I have also now run a conduit from the charger area up to the dash, so I can start pulling that wiring, and getting my switch panel online. Oh yeah, and I wanted to ask about CAN wiring - there are going to be 4 nodes on this network, BMS, Charger, Charge Controller, and Dash Display. The two blue butt-splices in the picture just combines all of the like colors in parallel, and then I am going to run another pair up to the dash. I am pretty sure my BMS has a terminal resistor, and the charge controller does as well. Does that wiring scheme sound about right?

Another day of wiring and I might be able to get those main contactors opening and closing, and the charger up and running.

 

[OR-Carl]

I got my main power leads all hooked back up today:

I was curious if anyone had any opinion on charging and contactor polarity. I have my contactors oriented to match the polarity of the pack, but while charging, the current will flow in the opposite direction, right? Since the charge controller and contactors share a power lead, a loss of power to one will also shut off the other - so I dont think it should be a problem. I moved the fuse for the charger into the battery box, which made it possible to move my HV junction box forward into the engine bay.

I also finalized the wiring for the charge controller and charger, and pulled the wiring through to the dash. The charger says something about making sure the earth ground from the J1772 is connected to the vehicle chassis, so I need to figure out if that means I should splice into the ground wire between the charger and plug, or if it is enough to make a good connection between the charger case and the chassis.

I got the 12v signal brought in to the dash panel, and there are a bunch of wires now that I can start getting squared away for that. I want my main power switch to turn on a relay that feeds the keyed ignition, so I need to figure out how all that is wired. I had the thought today that I might add a rotary selector and some sort of simple temperature display so I could check a variety of temperatures just by selecting different thermocouples. I might fiddle with that while I am waiting on drive train parts...

 

[OR-Carl]

I did a little more work today on wiring up my dash. I looked up a diagram of the ignition switch wiring, and spliced relays into the 12v inputs. I am not sure why there are 2 of them, or why they are so beefy. It seems like just using relays would have made more sense instead of routing big power leads through a switch, but what do I know... At any rate, now the key will not work until the main switch is flipped. That will allow the system to be brought online for charging with the main switch, but to turn on the motor controller and drive away one will need the key.

Ive got a couple more holes to drill in the dash panel, but then I can start getting all the switches and monitoring stuff installed. Most of the wiring is in place, so I am hoping it will come together pretty quickly

I cleaned up the frame brackets for the motor mounts today, but it seems I have discarded the rubber mounts. I am going to start working on the motor/transmission layout so I can build a shelf for the controller and HV junction box.

 

[OR-Carl]

I hooked up enough wires to hear the satisfying clunk of my main contactors closing today!

Goal #1 Mount batteries
Goal #2 wire main contactors to my power switch
Goal #3 Check voltages with BMS

I also finished making the cutouts for my dash panel:

Once the paint dries I will start populating it and get the wiring finalized. It will feel good to wrap up the work in the cab, and move on to the engine bay.

Goal #4 Finish wiring up dash panel.

 

[OR-Carl]

Goal 4 is not entirely complete, but almost. There are a few loose ends to wrap up, but the main circuit is working. The warning buzzer is indeed working, and it is very irritating. This should make the car pretty intuitive for others to drive - nobody would want to drive very far with it beeping and flashing red lights! The default settings in the charger did not like my battery voltage, so it it beeped untill I could reset them. The can network seems to be working. At first the SOC display would not show any info, but I had forgotten to enable it in the BMS.

I made a resistor array for the rotary switch, so I can select one of four charging modes. I will probably set up one for 220v, one for J1772, and then a couple of 120v modes for charging off my off-grid shop. The charger lets me set up a time-out limit, so I could run a quick charge while the sun is shining and not draw down my bank too much if I forget about it.

The amp meter is apparently not extremely accurate, but hopefully will be close enough to give me an idea of what I am pulling when driving. I have not gotten the thermocouples hooked up for the coolant heater or the temperature display, but I will work on that at some point here.

Goal #5 Mock up the motor position, and build a mounting rack for the controller.

 

[OR-Carl]

I am sitting in the cab of my truck with a little AC heater going, and the cheery buzz of the charger going behind the seats. I have my first string plugged in, and I am playing with the settings to try and get the charger dialed in. The voltage sag/rise is going to be an annoyance for trying to get the strings to line up perfectly. I am probably going to have to charge very slowly, and try to figure out where to set my end point so that when it settles, it will be lined up with the other string. At least I got pretty close on equalizing the cells in the string - they are all within 7milivolts, and most are within 2 or 3. I am currently only pushing 1.5amps in, which is maybe just as well, as the cells are pretty cold. I will make a note of what the final voltage of the cells is, and then see how much they drop back tomorrow. Baby steps.

 

[OR-Carl]

Another hour and 15 of low current charging today, and the cells are now very close. Once they sit overnight, I think the charged string will settle down to within my target range.

While I was waiting on the charging, I decided to mock up an electric motor, and wrestle the transmission back into the vehicle.

I was wondering how critical it is going to be for the transmission to sit in the exact same spot/orientation as in the original vehicle. If I understand correctly, the driveshaft can slip forward and back on the output splines to account for travel in the rear axle - will that play mean that I have a little leeway for where it winds up? Should I put the driveshaft back on to help get the alignment correct? Right now it is just sitting on its little rubber stub at the back, and the front end is free to swing around any which way.

Anyway, I made a little mock-up of an electric motor:

I am going to have plenty of space for all the other parts. I might try and fab up some mounting points for the controller shelf next.