Split pack battery charging...

How do you store and manage your electricity?
User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sat, 15 Dec 2012, 05:43

I think that will fit! It means a busy pack, but there is room for it at least. I might be going crazy with that insulating fabric though.

Image

Life size demonstration of the top of the pack - 320 mm by 220 mm, including the Lexan box.
AEVA National Secretary, WA branch chair.

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Fri, 21 Dec 2012, 06:44

OK everyone, can you check my circuit?
This is what I came up with using MS Paint:
relaylogic_zps12686dc0-1_zps78508d39.JPG
relaylogic_zps12686dc0-1_zps78508d39.JPG (114.12 KiB) Viewed 1033 times
Both circuits are the same, it's just that one is more PCB correct. The LEDs are illuminated by completing their circuits from the auxiliary contacts in each of the Gigavacs. This way if a contactor sticks the LED will remain lit. The charge LED is wired in parallel to the charge enable relay coil. There are some shared 0V Gnd headers to reduce the number of leads I have to stuff into one hole.

The switched circuit is the positive side of the ignition switch, and unswitched is straight from the battery positive terminal.

The 555 timer and associated cap and resistor are designed to give a 5 second delay before the main contactor closes. Can you all check it to see that it's wired up right?

Finally, Dmitrii was very kind to draw this up as a circuit board gerber for me so I can get it made up and have it look all nice like. Pretty nice eh?

RelaylogicPCBFT_zps47a8df06.JPG
RelaylogicPCBFT_zps47a8df06.JPG (99.5 KiB) Viewed 1033 times
Last edited by jonescg on Thu, 20 Dec 2012, 19:50, edited 1 time in total.
AEVA National Secretary, WA branch chair.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Split pack battery charging...

Post by weber » Fri, 21 Dec 2012, 18:40

jonescg wrote: OK everyone, can you check my circuit?

My brain just keeps sliding off it I'm afraid. The usual way would be to use schematic/PCB software where you start by drawing the schematic. The human checks the schematic (which is drawn so it is easy to follow) and the software checks that the PCB matches the schematic. DesignSpark (from RS Components) is free and easy to use.

I've mentioned these things before but they got no response: (a) Those relays are "SPDT", not "DPST". (b) You only have 560 uF of capacitor to charge right? So 5 seconds would be right for a 1200R 50W precharge resistor (500 W peak). But you could do it in 2 seconds with a 560R 100W resistor (peak power 1 kW) if the physical size is not an issue.

Does a 555 really have enough output current capability to drive an automotive relay? We use a TC4227 MOSFET driver. And in fact we use a MOSFET instead of the automotive relay. You can see our precharge schematic here.
viewtopic.php?title=weber-and-coulombs- ... 449#p35449
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Fri, 21 Dec 2012, 18:52

My apologies, schematic is here:
schematic_zps9d600f0e.JPG
schematic_zps9d600f0e.JPG (42.05 KiB) Viewed 1032 times
On the relays, what is a pole and what is a throw. I thought the thing bridging two poles was a 'throw'. Anyway, they have a common and an NO and an NC contact.

On the duration, I have a couple of 2kR 20 W resistors which are huge, and presumably rated for high voltages. They look like this:
resistor_zps365b52e1.JPG
resistor_zps365b52e1.JPG (44.56 KiB) Viewed 1032 times
So I have sort of based it on these being used, but I can change the value of R to something smaller so it's over quicker.

555s are good for a maximum of 200 mA, which is plenty to operate a 400R coil.
AEVA National Secretary, WA branch chair.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Split pack battery charging...

Post by weber » Sat, 22 Dec 2012, 17:31

Neither a "pole" nor a "throw" is an actual component of a switch or relay. The number of poles is the number of independent circuits that can be controlled by the switch or relay, or the number of independent contact sets. The number of throws is the number of positions in which a connection is made from one terminal to a different other terminal.

"Double throw (DT)" is synonymous with "changeover (CO)", while "single throw (ST)" can be either "normally open (NO)" or "normally closed (NC)". A relay with two independent contact sets, one of which is normally open and the other normally closed, doesn't really fit the "pole/throw" naming scheme, which is why I prefer the "xNOyNCzCO" scheme where x,y,z are small integers, usually 1 or 2. But note that in the case of switches (but not relays) "CO" can also stand for "centre off".

SPST = 1NO or 1NC (4 pins including coil)
DPST = 2NO or 2NC or 1NO1NC (6 pins including coil)
SPDT = 1CO (5 pins including coil)
DPDT = 2CO (8 pins including coil)

From http://en.wikipedia.org/wiki/Relay#Pole_and_throw

Image

Those 2k 20W resistors should be fine for voltage and just barely OK for peak power. However, as well as needing a long delay, there is another problem that can occur with larger value precharge resistors, and that is not being able to fully charge the caps due to quiescent loads inside the controller. These quiescent loads may be resistors connected across the caps to discharge them over 5 minutes or so, or they may be a DC-DC converter deriving gate-drive power from the DC bus. They may only amount to tens of milliamps but 20 milliamps times 2kR is 40 volts, plus maybe 8 volts remaining because we've only waited 5 RC time constants. This 48 volts, remaining across the main contacts when they close, can be enough to do serious damage to these contacts given that they may carry 1000 amps or more briefly as they are closing. See the curves and table of "Electrical Load Life Ratings for Typical EV Applications" in this EV200 datsheet.

http://relays.te.com/datasheets/ev200.pdf

With 32 volts it's rated for 50,000 operations. With 64 volts it's only rated for 50! Your Gigagvacs are probably similar.

I suggest temporarily connecting the inverter to the battery (or similar voltage source) via your precharge resistor and precharge contactor (no need for the main contactor) and measuring what voltage remains across the resistor after 5 to 10 seconds.

You're right about the 555 timer. The automotive relays I have on hand have a much lower coil resistance (80R), but even so they would still be OK with the 555.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Thu, 10 Jan 2013, 06:24

It works!

Well, there is one minor problem. You might notice that the relays have the NO and NC positions mixed up. So when I first made the circuit board up, everything was working in reverse Image

The solution is to solder everything except the 555 IC on the reverse side of the board. There is a downside; you can't see the text to guide you as to where you need to attach what. This caused a short circuit, lifting up some trace Image Still, I managed to salvage it, and once coated in epoxy it will work well. I won't use this board on the bike though; I'll make a fresh one. Good thing I got 6 boards made.

With everything off:
relaylogicworks001_zps312c606c.JPG
relaylogicworks001_zps312c606c.JPG (64.63 KiB) Viewed 1031 times
Turning the key ON:
relaylogicworks002_zps2e2b5711.JPG
relaylogicworks002_zps2e2b5711.JPG (65.75 KiB) Viewed 1031 times
Notice that the half pack LED isn't illuminated - this is because I don't have a contactor with auxiliary leads. I'll pick one up tomorrow.

Then about 5.5 seconds later:
relaylogicworks003_zpse1517fb7.JPG
relaylogicworks003_zpse1517fb7.JPG (64.28 KiB) Viewed 1031 times
All LEDs except the charge LED are lit (just pretend the half one is lit).

Then when it comes time to charge, you plug in, completing the charge enable circuit:
relaylogicworks004_zps9e545bec.JPG
relaylogicworks004_zps9e545bec.JPG (66.21 KiB) Viewed 1031 times
The main and half contactors are open, quarter contactors are closed, and the LED is lit saying you need to be patient Image.

I didn't photograph it, but if the key is on and the main is closed, completing the charge enable circuit causes the main and half to open, quarters to close. Like it was supposed to Image

Only thing I need to do is decide on a precharge duration. If I go for some big resistors of 1kR, I should have it over in 2 seconds.

Thanks Dmitrii - the boards look the goods!
AEVA National Secretary, WA branch chair.

celectric
Groupie
Posts: 56
Joined: Tue, 29 May 2012, 05:14
Real Name: Tim Bates
Location: Adelaide SA

Split pack battery charging...

Post by celectric » Thu, 10 Jan 2013, 16:26

Chris, where did you get your PCB fabbed, how much did it cost and how long did it take?

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Thu, 10 Jan 2013, 18:09

Hey Tim!

I got a friend in Melbourne to interpret my MS Paint sketch and re-present it as a Gerber file. I then forward the Gerbers to OurPCB in China and once paid, I receive the boards in the mail 2 weeks later. PCBFast does the same thing in Melbourne, but unfortunately they are about 4 times more expensive. Since I'm trying stuff that may not work, I will go for the cheaper option. Despite this one not working as planned (it is backwards) it works well enough that I can go with what I have already.

Tooling cost is $80 one-off, boards are about $1.50 each and shipping is about $60, but only because I ordered more battery PCBs too. So probably something like $20 for a half-dozen boards like this.

I am soldering the final version today, and will begin putting the main control box for the battery together. The good thing is, this board can be used in the car as well, or for anyone who is running a 300 V+ system which needs to be split in half and quarters. So if you want a board let me know and I'll send one your way.
Last edited by jonescg on Thu, 10 Jan 2013, 07:12, edited 1 time in total.
AEVA National Secretary, WA branch chair.

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sun, 13 Jan 2013, 04:15

I should probably start my own build thread for the race bike, but this is applicable to the split pack charging idea.

Today I decided to finish the main control box. This box contains the final 700 V cables, main contactor, precharge and discharge resistors, high voltage relay to drive said resistors, the 500 A shunt, charge leads, and of course the relay logic board to drive it all.

I found Altronics made a 220 mm by 165 mm box, which was 90 mm deep. Perfect fit, however it was going to be difficult to fit everything in and maintain isolation. So the trick was to use three sheets of 0.8 mm G10-FR4 board glued together to mount the shunt, main contactor and high voltage relay to. The nuts easily foul the base of the box, so I had to cut the bolts as short as possible to ensure they fit snugly.

There wasn't enough room for the 12 V relay logic, so I had a brain wave - double stack! Jaycar sells 32 mm nylon spacers for M3 screws which were exactly the right distance to mount a finger-proof G10-FR4 board to. It clears the main contactor and shunt but leaves enough room to mount the relay logic board. I will mount the HV relay and precharge resistors between the shunt and the contactor.
Maincontrolbox001_zps3b47848a.JPG
Maincontrolbox001_zps3b47848a.JPG (61.67 KiB) Viewed 1030 times
I had to epoxy the nuts to the mounting board so I can unscrew everything to bolt down the high voltage lugs etc, then screw it all back again. This box will be covered by the tank cover, but I want to ensure it's water proof. So I bought some honking great big cable glands and mounted them through 25 mm holes. Dremmels are awesome tools! The cover is clear, so you can see sh*t just as it happens.
Maincontrolbox003_zps1c96b799.JPG
Maincontrolbox003_zps1c96b799.JPG (48.79 KiB) Viewed 1030 times
This will mount vertically on the back of the battery box, between the rider's balls and the pack. There will be a fibreglass tank cover in between though ;)
Maincontrolbox004_zps20b75f01.JPG
Maincontrolbox004_zps20b75f01.JPG (47.69 KiB) Viewed 1030 times
Two more glands need to go into the top (charge leads and the 12 V coil wiring) and one more at the bottom (charge leads and charge enable). I need to mount a waterproof multi-pin plug for the 12 V power supply, ammeter leads, voltmeter leads (hopefully I can get a panel meter to do this at at a much safer voltage) and all of the LEDs for indicating the status of each contactor. These could plausibly be soldered directly to the board since the thing is clear, but you can't see it through the tank cover :). I think they will look good on the dash ;)

Getting there. Randy says the CNC'd billets shoud be ready for the frame next week, so work can start on putting it all together. Hopefully shipping a frame over here won't kill the budget, because the front end, fairings and wheels sure will Image
AEVA National Secretary, WA branch chair.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Split pack battery charging...

Post by weber » Sun, 13 Jan 2013, 06:14

Good work. Looks good.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sun, 13 Jan 2013, 06:51

Thanks Dave!

I have a query about the power dissipation of a wire-wound enamelled resistor. I have thought about wiring my 20 W 2kR resistor to a 14 W 1kR in parallel, so I effectively have a 666 ohm precharge resistor which is now considerably more powerful. It also means the precharge is over in 1.5 seconds, so I can have the main contactor close after 2.5 seconds. I could stick with the 1kR resistors, but they are only 14 W, which is probably well underrated. Alternatively I could parallel the 14 W resistors, but the resistance is lower still (500R).

Unfortunately I don't have room for the big 50W resistors...
AEVA National Secretary, WA branch chair.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Split pack battery charging...

Post by weber » Sun, 13 Jan 2013, 21:51

jonescg wrote:I have a query about the power dissipation of a wire-wound enamelled resistor. I have thought about wiring my 20 W 2kR resistor to a 14 W 1kR in parallel, so I effectively have a 666 ohm precharge resistor which is now considerably more powerful. It also means the precharge is over in 1.5 seconds, so I can have the main contactor close after 2.5 seconds. I could stick with the 1kR resistors, but they are only 14 W, which is probably well underrated. Alternatively I could parallel the 14 W resistors, but the resistance is lower still (500R).

Unfortunately I don't have room for the big 50W resistors...

Can you point us to datasheets for these resistors (20W and 14W)? Or tell us make and model?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sun, 13 Jan 2013, 22:20

These are the 14 W resistors:
http://docs-asia.electrocomponents.com/ ... ce7a55.pdf

In the RS catalogue it rates them at 750 V, but the spec sheet says 650 V. 10 times overload for 5 seconds is pretty good, but whether that means 100 times for 0.5 seconds, I'm not so sure.

The 20 W resistors are a random brand that came with a Kelly controller. Brand RX21. Some details are here:

http://jingdacheng.en.alibaba.com/produ ... istor.html
AEVA National Secretary, WA branch chair.

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sun, 13 Jan 2013, 22:56

Wait, I just realised I can get away with one resistor which serves as both a precharge and a discharge - The resistor can go between the open side of the contactor (positive) and the high voltage relay common. Then I just need a wire joining the negative to the NC, and a wire joining the other side of the contactor to the NO.

This means I can stack up to four large resistors in the box to serve as both a precharge and discharge. And I can save a bit of money on the HV relay!

AEVA National Secretary, WA branch chair.

User avatar
weber
Site Admin
Posts: 2601
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Split pack battery charging...

Post by weber » Sun, 13 Jan 2013, 22:58

I suggest ignoring the 20W resistors because we have no pulse or short term overload data for them.

The 650 V will be a working voltage, i.e. continuous for those with a high enough resistance to be within their power rating. Based on data sheets for other similar resistors, they will probably take at least 2 kV for up to a minute. Also based on similar datasheets they should take at least 25 times power for 1 second.

So I'd use a value of 750V*750V/(14W*25) ~= 1600 ohms and parallel as many of those as you can fit. (1500 ohms probably ok)

One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sun, 13 Jan 2013, 23:26

OK I have ordered a packet of 5 of the 14 W 2.2 kR resistors. If I parallel all of these I should have a power rating close to what I need, and a resistance of 550 ohm. I can't believe I didn't see the HV relay problem staring me in the face... Probably because I was entertaining using the half contactor as a main contactor.

Must get Gigavac on the phone Image
AEVA National Secretary, WA branch chair.

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Mon, 14 Jan 2013, 01:46

Maincontrolbox_zps27a37ffd.JPG
Maincontrolbox_zps27a37ffd.JPG (95.76 KiB) Viewed 1029 times
I think I need to put a physical divider beside the shunt so should any wire go stray it can't short. I should also put a fuse on the ammeter circuit... The SPDT high voltage relay from Gigavac is also much smaller, so I can still mount it through the top board with the 12 V coil wires exposed.

http://www.gigavac.com/products/relays/ ... /index.htm

The charge leads are going through next, and I will have to put some kind of conduit on the inside to stop them from going stray.
AEVA National Secretary, WA branch chair.

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Fri, 25 Jan 2013, 04:12

Two 2 kW TC chargers tuned to 350 V will be heading my way shortly! They are good for 8 amp charging which will have my pack fully charged in about 90 minutes.

I will be able to run them at the same time from a 32 A plug at full power, or if I build the voltage divider for the Enable pin I can dial them back to 1 kW each (Wakefield Park's power supplies are woeful).
AEVA National Secretary, WA branch chair.

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sat, 26 Jan 2013, 04:17

Finally got the HV relay. It's the size of a spark plug with the ceramic bit chopped off. $275 but hopefully it will be worth it.

I soldered the HV wires on and hooked them up to the contactor / shunt.
Maincontrolboxnearlydone001_zpsb33174fd.JPG
Maincontrolboxnearlydone001_zpsb33174fd.JPG (93.39 KiB) Viewed 1028 times
Unfortunately the charge wires don't fit with the relay in the way, so I have conceded defeat and will run them straight out of the main pack.

Now as for wires coming out of the box, there are the two main HV wires in and out top and bottom. Then there are the three 12 V supply wires (unswitched positive, switched positive and negative). I also need to make a voltage divider using my 14 W resistors and fit it to the bottom of the box, then run the leads from this out to a panel mounted socket. This will also have the ammeter leads going to it. As these are all attached to the HV system they are orange and rated for it, as well as insulated in that woven tube stuff. Finally the LEDs need to plug into a multi-pin panel mounted plug. Hard to find a weather proof one which would work well, but a D-sub 9 should suffice.
AEVA National Secretary, WA branch chair.

User avatar
coulomb
Site Admin
Posts: 3735
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Split pack battery charging...

Post by coulomb » Sat, 26 Jan 2013, 13:51

jonescg wrote: ... as well as insulated in that woven tube stuff.

If you mean this:

Image

the technical name is "spaghetti"   Image

From http://www.jaycar.com.au/productView.asp?ID=WS5502 .

Also known as varnished cambric tubing, I believe. But spaghetti seems easier.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sat, 26 Jan 2013, 17:11

Just thinking - if I am running the ammeter sense wires and the voltage divider output wires up to the dash, should I be shielding these from EM? I imagine high currents from the battery would generate a sufficient field to interfere with a millivolt signal...
AEVA National Secretary, WA branch chair.

User avatar
evric
Site Admin
Posts: 500
Joined: Sun, 20 Jul 2008, 01:57
Real Name: Eric
Location: Adelaide SA
Contact:

Split pack battery charging...

Post by evric » Sat, 26 Jan 2013, 17:52

You can run each pair to the dash as a twisted pair. (Place the end of each pair in the vice and the other end in a (hand) drill and turn.) Voila. This is what I have done for the XPert Pro shunt sense wires in a few cars.
Prius Plug-in Conversion: http://www.evplus.com.au ...Holden Barina EV: http://www.evric.kestar.com.au

User avatar
jonescg
Senior Member
Posts: 2949
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Split pack battery charging...

Post by jonescg » Sat, 26 Jan 2013, 18:16

Cool! And this works as well as shielding cable? Good to know!
AEVA National Secretary, WA branch chair.

User avatar
evric
Site Admin
Posts: 500
Joined: Sun, 20 Jul 2008, 01:57
Real Name: Eric
Location: Adelaide SA
Contact:

Split pack battery charging...

Post by evric » Sat, 26 Jan 2013, 18:21

Shielded may be better, but I have never experienced any interference.
Shielded cables often have the pair twisted inside as well.
Prius Plug-in Conversion: http://www.evplus.com.au ...Holden Barina EV: http://www.evric.kestar.com.au

Post Reply