4Springs' Brumby

Post up a thread for your EV. Progress pics, description and assorted alliteration
User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Mon, 21 Oct 2013, 02:11

Ok, after discussing with TC Charger, I have ordered a 2kW TCCH-H175.2-12. This will shut off at 175.5V (48 cells x 3.65V). Price is $485US + $150US shipping. My bank charged $24 for the transfer. Grand total at today's exchange rate is $711.60AU.
I decided on the 0-5V current control option. I haven't been persuaded that going digital is worth the effort.

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Sat, 23 Nov 2013, 13:36

Hooray, my new lithium battery has arrived!
48 cells, 130Ah = approx 20kWh.
This is conservatively 1.5 times the lead-acid battery capacity, at about 0.5 times the weight and reportedly about 10 times the life. At more than $10,000 (including freight) it is about 2.5 times the cost.

I had my BMS module circuit boards manufactured by a place in the UK. I had the specs on the cells to tell me the terminal size & spacing, but I didn't have an actual cell. Then when I sent the circuit board file to the UK to be converted into reality I found out that circuit board files apparently don't include any size information. So there were emails back and forth with the helpful people to confirm exactly what size everything was.
Now I had quadruple checked every measurement, but I know how easy a mistake can creep in. So I have not started assembling the boards, knowing that it would be much easier to do any modifications (e.g. drilling out a hole) without the components on the boards.

So the cells have finally arrived, and the first thing I did was grab a board and bolt it on. Perfect fit! Whew! Now I can start assembling those BMS modules. Thankfully the delay has meant that it is now cricket season, so I can keep myself sane during that monotonous task.
(In re-reading this last sentence I find that some clarification may be needed for some readers. The "monotonous task" is the circuit board assembly, not the cricket. The cricket keeps me sane through the excellent ABC radio commentary.)

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Sun, 08 Dec 2013, 02:37

Image
Here are my 48 130Ah cells.
I checked the voltage when they arrived, and they were almost all exactly the same. But I decided to balance them anyway, by tying them all together in parallel with electric fence wire.

I have previously only tested about 3 of the BMS module boards strung together. Now I have completed the whole 48 boards, I can start to better test the communications between them.
I've found lots of bugs in my software (which is heavily modified from Nevilleh's), but the programmer utility provides lots of helpful tools to track them down.

One reason why I put the modules on the cells while tied together is that I could check the calibration on the voltage values. They read between 3.25 & 3.28V. From what I have read lately about cell balancing, I don't think this will be a problem. I'll be top balancing, so there won't be much difference in SOC between (say) 3.65 & 3.68V.

I can't seem to find a handy place in the dash for the BMS master display. I also think it will be very distracting, so I might put it in the tray or under the bonnet instead. I'll have a fault light (and probably buzzer) in the cabin. I've also modified the software to display a (persistent) fault code on the screen. Arber talked about writing an Android app to monitor the data via bluetooth, which might be an option in the future for a passenger.

User avatar
Adverse Effects
Senior Member
Posts: 971
Joined: Sat, 01 Jan 2011, 03:30
Real Name: Adverse Effects
Location: Brisbane

4Springs' Brumby

Post by Adverse Effects » Sun, 08 Dec 2013, 04:20

wow nice bunch of toys there problem it meens lots of work lol

it should run better with only 1/2 the battery weight in it

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Mon, 27 Jan 2014, 02:33

Here is an "in progress" picture of the pretty blue lithium cells in the back:
Image
There are 12 in the front and 36 in the back. The gap on the right hand side is to let the cable come up through the floor.
In all the box modifications weren't too difficult. There used to be 3 lead-acid batteries in the front, in three boxes. I managed to use two of those boxes to hold 6 lithium cells each. 6 lithiums are slightly larger, so the boxes had to be adjusted a tiny bit.
In the back, I was able to remove the two bottom boxes (under the tray) completely, which is a relief. Those were horrible boxes to design and build. Once assembled, they also had the disadvantage of not allowing inspection of the connections. When I took them apart I found one quite loose connection on one of the batteries. Luckily it didn't seem like it had been hot at all.
The box in the tray didn't need much modification. I took out the water heating arrangement, which was also a relief (I've been waiting for it to leak for the last 2 years), but kept the battery warming pads hooked up. So I still have heating under all of the lithium cells. It will be very interesting keeping a track of the temperature, since the BMS monitors every cell. This morning at one point it said that the highest was 26 degrees (in the sun), with the lowest 12 degrees (in a closed box).

I took Brumby 2.1 on the road for the first time today. It is a much better car to drive now, just like everyone said it would be!
First I did a slow and careful kilometre or so, got out with my non-contact thermometer and my nose to check for hot connections. All was good, so I did a bit more vigorous driving after that. It really takes off much better now, and I managed to get up to 120kph on our short stretch of slightly up-hill road. Previous top speed was 115kph, which I could only get to on a much longer road.

BMS is strangely fascinating to watch. The charger is supposed to shut off at 175.5V, but the cells are not balanced enough yet. I have the BMS shutting the charger off if a cell gets to 3.8V, and that is what is happening at the moment (i.e. that cell gets to 3.8V before the total gets to 175.5V). I'm only shunting 200mA, so it might take a while for the cells to balance, but it should happen eventually. If I get impatient I could tell the BMS to shunt that cell continuously for a while.


User avatar
jonescg
Senior Member
Posts: 1996
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

4Springs' Brumby

Post by jonescg » Mon, 27 Jan 2014, 02:59

WHile they were all rigged up in parallel you could have put a 3.65 V power supply on and charged them up that way. About a week later you'd have a completely top-balanced pack Image Still, lots of shallow discharge and recharge cycles should bring it up to speed.

Looking great anyway Chris. Once you have them all balanced and you can take it for a long drive, I'd be curious to know what your range and Wh/km is at highway speeds.
Last edited by jonescg on Sun, 26 Jan 2014, 16:00, edited 1 time in total.
AEVA National Secretary, WA branch vice-chair

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Mon, 27 Jan 2014, 14:16

jonescg wrote: WHile they were all rigged up in parallel you could have put a 3.65 V power supply on and charged them up that way. About a week later you'd have a completely top-balanced pack

I did think of this at the time, but was put off by the (roughly done in the head) numbers. It's too late now, but lets be a bit more specific. Maths for fun! Image
I bought myself a laboratory power supply to help with making the BMS. It is rated at 2.5A. This is the only thing I have that would produce 3.65V. If I was going to charge a cell completely (130Ah), this would take 52 hours. Times 48 cells = 2496 hours = 104 days.
But the cells were not fully discharged. They were at 3.25V. This is where my knowledge runs out. How much capacity is there from 3.25V to 3.6V? I didn't have a clue then, but now I can say that once connected it took the 2kW charger quite a few hours to make up this gap. I had it on and off a lot, but I'm guessing about 8 hours. So let's make a guess of 8x2kW = 16kWh. How long would it take my 2.5A, 3.6V power supply to supply 16kWh? Well 2.5*3.6 = 9W. 16,000/9 = 1778 hours = 74 days.
So I guess my gut feeling was correct! I could not have made much of a difference in the top balancing in the amount of time that I had with the equipment that I had.


User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Mon, 27 Jan 2014, 14:22

Here are a couple of extra photos.
Image
Some messy wiring in there needs tidying up, and I have to get some new warning labels made (the old ones had "contains lead-acid batteries").

Image
View of the tray showing off the new(ish) Rhino Liner. It does a very good job of hiding all the little dents collected over the years. Should help stop new dents being made as well!

User avatar
Johny
Senior Member
Posts: 3647
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

4Springs' Brumby

Post by Johny » Mon, 27 Jan 2014, 15:53

Nice clean tray. Good job!

User avatar
carnut1100
Groupie
Posts: 320
Joined: Tue, 24 Feb 2009, 16:39
Real Name: Greg Milligan
Location: Hobart
Contact:

4Springs' Brumby

Post by carnut1100 » Tue, 28 Jan 2014, 00:37

Sweeeet.......

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Sun, 16 Feb 2014, 17:03

FINALLY able to assemble the last four BMS modules. I had destroyed the microcontrollers on these four, and had ordered replacements from RS Components about 4 weeks ago. I got tired of waiting for them, so I ordered some from Element 14 as well. Then finally I found them all at the Post Office! It turns out that the courier who used to deliver to our door has now decided that we are too far off his route, and he dropped them off at the PO. Would have been fine if he had told me! So the RS chips sat at the PO for a few weeks, and were then joined by the Element 14 ones...

In the meantime I moved the BMS modules around so that the four lowest voltage cells had no BMS. I also altered the code so that the charger turned off at quite a low level (a voltage low enough so that no modules went into balancing). I also only drove short trips, and measured the voltage on those four cells frequently with a multimeter.

So I've had BMS modules on all cells now for about a week. I set the charging cut-out back up to a cell voltage of 3.7V, with a balancing voltage of 3.5V. On my first charge I had something like 2 modules balancing when the charger cut out. Next time it was about 12. Within a few days of short trips to work I had 47 of the 48 balancing, and the charger was reducing the current rather than the BMS.
But 47 was where it sat. Cell 17 was quite a lot lower voltage than all the others. While they were all 3.50V, cell 17 was more like 3.31. At this stage the charger was running all day at about 1A or less, so cell 17 was getting a trickle charge while every other cell was shunting.
On my day-off I decided to speed up the process by attaching my (cheap, small) laboratory power supply to charge that one cell. The power supply boasts "2.5A" output. I set it up to 3.5V with nothing attached, and then hooked it up to the cell (with all vehicle isolation switches switched, and the car un-plugged from the mains). Once I got my connections correct it did indeed draw about 2.5A according to the read-out on the power supply. These are 130Ah cells, so I thought it wouldn't take long to top up this one. I also wasn't confident about the power supply - would it cut out when the voltage got high enough? So I sat out there with it to watch, doing some bug-tracking on the laptop in the tray of the ute.
It took about 20 hours! If I believe the current read-out on the power supply, this adds up to about 50Ah in a 130Ah cell! So cell 17 was seriously depleted compared to the other cells.

The power supply finally reported that it was finished at about 20:00 last night. Mrs 4Springs was out, so I decided to take it four a range test. I needed to set the "empty" point on my Zeva fuel gauge driver, and I also wanted to know how far it would go!

This is where I was with the lead-acid pack:
Max: 13kWh (measured), 60km max range, 25kWh/100km (max) 35kWh/100km (typical)
My target for this new lithium battery was 20kWh, which should get me 1.5 times the range, or about 90km when driven carefully (max speed 60km/h).

So last night I drove to work and back (30km round trip), then up and down our 2.5km road. I had previously tested the 30km, but I didn't want to stray too far from home for the last bit of the range test. My BMS is set to flash a light at me when the lowest cell gets to 2.50V, so I decided that that would be my crawl home point. I started this test at 8:20pm. I wouldn't take long would I? I could fit this in before bed time surely!
2 hours later (way past my bedtime), the lowest cell voltage finally started to move. At the far end of the road (of course) the warning light came on. I slowed down and it went off again, but then came back on again soon after. I crawled home in 1st, and by the time I got home that lowest cell was 2.2V. After about 30mins of me fiddling with the fuel gauge driver that cell was up to 2.5V. So I think I need to set that lowest cell alarm to a bit higher - perhaps 3.0V? At least that would give a bit more than 2km to get home. For the wife I have called that light the "stop and call husband" light.
I set the fuel gauge driver to a bit below empty, so when driving first the fuel gauge goes to zero, then the light will come on after that, and then we'll have enough reserve to crawl home, or to drive to the nearest house to beg a power point, or onto a tow truck...

And the verdict? 108km! And I was not limiting my speed. The 30km first bit included some stretches at 100km/h, and even for "our" road I went to 80km/h for most of it (and it includes a gentle hill). Along with all the stopping, turning around and accelerating back up again, this was not a very conservative drive.
As I am writing this the charger is just finishing. My watt meter on the 240VAC side counted 24.4kWh, so this calculates out to 22.6kWh/100km. Our electricity currently costs $0.26807/kWh, so here is the comparison:

Brumby 1.0 (petrol) 10L/100km x $1.63 = $16.30 per 100km
Brumby 2.0 (pbacid) $9.38 per 100km
Brumby 2.1 (LFP)    $6.06 per 100km
Our small diesel Hyundai i30 is $7.94 per 100km

So I have finally broken the diesel barrier!
The drive is much better than the lead-acid car as well. More power available, rarely reaches maximum current drawn. Also in Brumby 2.0 the lights would start to dim once you had used a portion of your pack capacity. The lights would dim when you accelerated and brighten when you let the foot off, presumably because the DC-DC converter was struggling to keep up the 12V load when the supply voltage sagged. None of that on this drive, even when at the dregs.

User avatar
offgridQLD
Senior Member
Posts: 1785
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

4Springs' Brumby

Post by offgridQLD » Sun, 16 Feb 2014, 18:01

Nice Image

I'm sure when you do a constant run you will consume way less than stop start driving around the block. I notice the difference in consumption when I have a good run and get all the green lights and have a constant smooth run vs stop start red lights and accelerating the weight back to speed all the time chews a lot of power.

All you need is some PV and you can drop that electricity cost to zero.

Kurt
Last edited by offgridQLD on Sun, 16 Feb 2014, 07:02, edited 1 time in total.

User avatar
jonescg
Senior Member
Posts: 1996
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

4Springs' Brumby

Post by jonescg » Sun, 16 Feb 2014, 19:05

offgridQLD wrote:

All you need is some generously donated PV and you can drop that electricity cost to zero.

Kurt


Fixed Image
AEVA National Secretary, WA branch vice-chair

User avatar
offgridQLD
Senior Member
Posts: 1785
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

4Springs' Brumby

Post by offgridQLD » Sun, 16 Feb 2014, 19:24

True,but if we are going to get into that discussion just keeping the old diesel is most likely the best financially.

If your on the grid though the the prices being quoted for grid connect systems fully install are quite reasonable. You can't even buy the components for the same price so are worth considering.

That said without battery's (offgrid) and or a generous feed in tariff you need to be able to charge your during the day for it to work in your favor. Not always easy with peoples 9 to 5 schedules.



Kurt
Last edited by offgridQLD on Sun, 16 Feb 2014, 08:28, edited 1 time in total.

User avatar
jonescg
Senior Member
Posts: 1996
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

4Springs' Brumby

Post by jonescg » Sun, 16 Feb 2014, 19:31

For sure. When my parents installed an off-grid system for their place in Emu Vale, 80 W BP solar panels were $900 each. Now you can buy a 250 W panel for $280. As a fully installed system, you're still looking at a very cheap investment. I reckon an Elcon/TC charger could run straight off a 250 V DC solar array (if you were home all day).
AEVA National Secretary, WA branch vice-chair

User avatar
offgridQLD
Senior Member
Posts: 1785
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

4Springs' Brumby

Post by offgridQLD » Sun, 16 Feb 2014, 20:31

The midnite classic range of mppt DC charge controllers can do over 250dc then you can skip the AC charger all together and have fully programmable DC charging directly from PV with all the safty features to boot along with logging / remote online controlled and a lot more they are probably more programmable than many dedicated ev charges. good for over 5000w each and can be daisy chained for more output. I don't know why more home Built ev,s don't use them as a charger for around $ 600 us.

PV is $1 watt sometimes under that.

Kurt

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Sun, 16 Feb 2014, 20:41

While I was sitting in the back of the Brumby waiting for cell 17 to charge I added a PC data out function to the BMS. I had to go to Hagley that afternoon (to pick up a parcel from the Post Office as it happens!), so I logged the cell voltage data for the trip:
Image
Scale along the bottom is minutes, trip is about 30km. There are 48 cells on this graph, so it is quite busy.

Quite happy with this, all traces are on top of each other, meaning that no cell is much different from any other cell.
You can see the recalcitrant cell 17 is a lower voltage at the start (the blue line), but it very quickly became just another cell doing its job in the magic range of 3.1 to 3.3V.

I have the BMS transmitting voltage data now every 2 seconds or so. That was hooked into a laptop sitting in the passenger footwell. Connected using CuteCom, which logs incoming data from the RS-232 port to a file. Pasted this into a spreadsheet with the calculations to convert to voltage. This is interesting stuff - I need to log a "performance" run next.

I can swap over to temperature data if I want, not so much to see how they change over time, but will be interesting to see once the weather gets cold, which cells benefit the most from the warming pads.
I did have a bug that I found when setting up temperature logging. The old "forgot to put the break after the case" trick. Meant that the array of temperature data was overwritten 9 times out of 10 by balancing data. I hadn't noticed up until then, since the function that uses that data will ignore it if it doesn't look right. It looked right 1 time in 10, so that was good enough.

User avatar
Johny
Senior Member
Posts: 3647
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

4Springs' Brumby

Post by Johny » Mon, 17 Feb 2014, 14:59

Nice work Chris. I think your low voltage alarm at 3.0 volts might be a bit high (might sound during acceleration at high revs) but time will tell.
My pack took a long time to get balanced when I started driving daily but has settled down to the point where I just leave it up to "the car" to take care of itself.

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Tue, 18 Feb 2014, 01:58

Johny wrote: I think your low voltage alarm at 3.0 volts might be a bit high (might sound during acceleration at high revs) but time will tell.
Didn't take much time! Image
Drove to work today and the light came on. Stopped, checked the voltages, everything looked ok, drove slowly the rest of the way. Had completely forgotten that I had increased the threshold until I got home and read Johny's reply! Looking at my graph above it is pretty apparent that 3.0 is too high. Might try 2.8 next.

User avatar
Johny
Senior Member
Posts: 3647
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

4Springs' Brumby

Post by Johny » Tue, 18 Feb 2014, 14:13

4Springs wrote:Looking at my graph above it is pretty apparent that 3.0 is too high. Might try 2.8 next.
Wow - I'll frame that reply and put it over my desk! The problem will be in the winter months when your cell voltages drop even lower. My alarms go off at 2.0 volts but I don't have any control over the voltage setpoint. In the early days (before I realized that the modules were WAY out of balance) I nanny crept home about 2km twice with the alarm sounding without cell damage but I knew it was pretty close.

2.8V might be OK for summer and I'd try it but come winter I think you will be lowering it a bit more - just be ready for it and don't get too much of a surprize when it drops below 5 degrees.
Last edited by Johny on Tue, 18 Feb 2014, 03:41, edited 1 time in total.

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Tue, 18 Feb 2014, 22:19

Johny wrote: The problem will be in the winter months when your cell voltages drop even lower.
Ah, good to know. I've set it to 2.8 at the moment, and I'll do some hooning with the laptop connected to see how much they sag under acceleration.
Will be interesting to see how low the temperature goes in the winter, seeing how I have the cells sitting on battery warmers. I never quite knew how well they were working with the old lead-acid, but now I have temperature data at 48 points throughout the battery!

User avatar
Johny
Senior Member
Posts: 3647
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

4Springs' Brumby

Post by Johny » Wed, 19 Feb 2014, 14:23

4Springs wrote:Will be interesting to see how low the temperature goes in the winter, seeing how I have the cells sitting on battery warmers. I never quite knew how well they were working with the old lead-acid, but now I have temperature data at 48 points throughout the battery!
Good planning! Here in Melbourne we usually don't make it down to zero for enough time for the cells to get down that low. The lowest I saw during the winter months was 4 degrees C directly on the modules (can't measure cells).
It will be very interesting to run a couple of similar temp. nights (when it gets colder) with and without the warmers so we get an idea of how much they are contributing. I really noticed the reported range reduction during winter.

User avatar
4Springs
Senior Member
Posts: 449
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

4Springs' Brumby

Post by 4Springs » Sun, 30 Mar 2014, 22:16

Ok, I know there has been a lot said about range extenders in the trays of utes, so I had a go myself.
I built a rack for two range extenders. The rack can be added fairly quickly (two bolts and a pin), and can also be fitted with a padlock. You never know what people will try to steal!
Image

The range extenders themselves once deployed will last for several kilometres, although they are not good in wet weather. There are enough positions on the rack to fit range extenders for all passengers. Ute can be charged while range extenders are in use.
Image

I challenge - oh, hang on...

User avatar
jonescg
Senior Member
Posts: 1996
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

4Springs' Brumby

Post by jonescg » Sun, 30 Mar 2014, 22:31

Image Nice one Image q4qber also has a folding bike in the back of his iMiEV for those 6 hour range extension periods.
AEVA National Secretary, WA branch vice-chair

User avatar
offgridQLD
Senior Member
Posts: 1785
Joined: Tue, 23 Jul 2013, 16:05
Real Name: Kurt
Location: Fleurieu Peninsula, SA

4Springs' Brumby

Post by offgridQLD » Sun, 30 Mar 2014, 23:58

My wife's e folder uses about 10wh -km with almost 1000whr of battery it's got almost the same range as the imiev.

Kurt

Post Reply