jonescg's NEW electric racebike BUILD thread!

[c_a]

ok, good luck!

[full-throttle]

A battery pack with 168s3p in a motorcycle is really crazy

Thanks for your concerns

Chris, you should take his advise seriously

Being zapped by mains 240VAC is nothing compared to 450VDC - a tiny leak, even 10mA is potentially lethal

Whatever you do get it properly inspected

[jonescg]

Yes dad....

Nah, I realise high voltage DC is risky, but it seems to be the only way we can get these big motors to run.

I've put a lot of thought into this pack, and the bulk of that thought was how to make sure there's no way even the most retarded of operators would find a way of zapping themselves. The remainder was how to get it to fit...

Seen this video?

http://www.youtube.com/watch?v=6Piu1tjb2Jo

[c_a]

It is more complexe as you can imagine.

You have to buy very specialand expensive wires with big isolation, special isolation for the battery pack. A contactor that can handle this voltage and currents for sure. Keep the voltage near to the packs and transfer only data...

To make it safe and reliable you have to invest 10 times the money as for a 120V/80Ah package.

[jonescg]

Yep, onto all of that. I have 50 mm2 cable which is double insulated and rated to 1000 V DC.

Contactor rated to 750 V, 600 A DC
http://www.gigavac.com/products/contact ... /index.htm

Fuses from Bussman: http://www.cooperindustries.com/content ... -500a.html

A 2 mm thick layer of rubber to line the insides and termination fronts, with foam backing to prevent rubbing... Yeah, I;ve thought of this stuff.

[jateureka]

SUbscribed!

CAn't wait to see this thing in action.

[jonescg]

OK I think I have settled on a trellis frame and battery cradle I am happy with. A friend on Trev's forum (mcnews.com.au) who races a lot and has done lots of modifications to frames reminded me that subframes are disposable and I shouldn't mount the shock anywhere near it. So in heeding this advice, I have come up with a simple motor mount, trellis fame and shock mount.


Open in a new tab to see it fully. I have highlighted the bits I want FrameCrafters to build for me.

The overall length of the bike is a bit longer than I'd like, but I think I may have been a bit generous with the front wheel clearance. I can easily claim another 25 mm there. The battery cradle is important, not because it provides rigidity (I believe the frame as it's drawn should be plenty rigid) but should the bike crash there will be considerable force exerted on the battery pack in the forwards and downwards direction. The cradle can bolt in like that of an RGV.

The other thing to note is the Ducati SS900 swingarm is a bit short at the pivot end. This means a longer than desired runout from the pivot point to the drive sprocket. As long as the wheelbase isn't being extended, this shouldn't affect the handling of the bike in turns.

Lock it in Eddie

[jonescg]

OK I have settled on the pack configuration and voltage - and my eCRX project will use the same charge voltage, just with a few more cells (due to EiG's lower nominal voltage and charge voltage).

So when it comes to charging, most companies don't make 705 volt chargers. Elcon/TC chargers, which I swear are the most bulletproof charger going, only go up to 417 V. But I can run two of their 2 kW chargers set to 352.5 V on half the pack at once. This means a more sophisticated plug system (4 main charge terminals and two charge enable plugs - not impossible) but can anyone see any issues with this? I would do the same for the eCRX, except it would be hard-wired into the pack.

Any thoughts welcome

[jonescg]

Exciting update - the Evo motor is ready to ship next week!

Unexciting update - I don't have the cash to pay the difference I wasn't expecting them to finish the motor so quickly, so I started spending money on other things. Oh well. I should be able to get it delivered in October I would think.

YAY!

[Jay64]

I don't know if I missed it, but where are you getting your EIG cells from?

[jonescg]

Hi Jay,

I won't be using the EiG cells for the race bike, just for the car conversion. I will be sourcing them from Korea directly, but only when the reduce their prices. We're currently trying to haggle a good deal.

Turnigy LiPo for the race bike

[jonescg]

Ever tried to design a substation? That's kind of what the top of my battery pack is looking like .



I have a controller capacitor precharge resistor which is connected to a DPST relay rated for high voltages. With a 2 k ohm resistor (rated for 20 W) I can discharge the caps in under 5 seconds. It also takes 5 seconds to precharge them. This satisfies the TTXGP rule that the battery needs to be completely isolated from the controller (both the main contactor and the precharge resistor) AND it must allow the caps to be discharged to under 50 V in 5 seconds.

The key switch is there to allow precharge to occur, before switching to the ON position where the main contactor coil is energised, closing the HV circuit. There is a 555 timer circuit (not drawn) which allows a 7 second delay between going to ON and actually closing the contactor. It's to save me from belting right through to ON when the caps aren't fully charged up.

Now, safety - all of the HV terminals will be on top of the cell pack separated by a 6 mm piece of Delrin. Then, a cover goes over the terminals, and finally, a cover does over the lot. Between the top and second layer is where the balance taps will be. Not sure whether to go for D-subs or centronix plug. It only needs to be good for 2 amps, but that's still quite an ask for most plugs and the tiny wires that service them.

All up it should weigh about 85 kg, give or take a kg. And it won't have any HV output until 12 V is fed in, so that's a good thing.

Any ideas or advice?

[c_a]

you need to add another swtich in line with the 2k resistor, otherwise you will discharge your battery with 0.35A (700V/2k=0.3525A)

[jonescg]

Well spotted! I will put another resistor in there but on the other side of the contactor.

[c_a]

you need another switch in the line, with another resistor you will increase the precharge time, not more, not less, but your pack will be discharged continually and if you are back from the one month trip to europe, you can buy a new pack

[jonescg]

Sorry I didn't really explain that in my simple post - I will create two switched circuits for precharge and discharge respectively, by using another resistor.


Edit to add: This: http://www.gigavac.com/products/relays/ ... /index.htm would be perfect for the job. Fully isolated DPDT relay. I will update my wiring diagram in a bit.

[jonescg]

OK, here is the corrected relay circuit. I'll use a Gigvac 16 contactor for the main contactor, and one of their high-voltage DPDT relays for switching the precharge and discharge circuits.



As before,
Off - discharge circuit is closed, precharge open, contactor open.
Acc - discharge circuit is opened, precharge circuit closed, contactor open.
ON - discharge circuit still open, precharge still closed, contactor closed.

There remains the overcurrent protection and isolation measures, but I will get to these in due course. The giant Bussman fuse is on the positive, but I can't see how a short which generates an epic plasma ball is going to be somehow less catastrophic thanks to the fuse.

[jonescg]

OK, I took a photo of some gear on top of an outline of the battery pack dimensions. The inside of the pack will be 300 mm by 200 mm, with about 10 mm of acetal plastic all around (not exactly, as the sides will be 6 mm and the front and back will be 8 mm, but I have estimated about 10 mm either side for simplicity).



The switch gear will sit on top of an 8 mm false floor, and below this is the LiPo pack. The maximum height is about 60 mm, over which another final cover will sit.

Please excuse the MS Paint mock-up, but I have drawn in a pack-splitting contactor, as well as the main output contactor, a shunt, two fuses for each half pack, and a control box which will contain the high voltage DPDT relay, power on delay and so forth.



As you can see, you run out of room pretty quickly. Two 352 V packs is surely safer than one 705 V pack, so taking charge leads from either side of the contactor and pairing these with the main pack terminals will give two floating packs which are fused.

There will also be physical separation of the fuses and contactor at the back from the shunt and contactor at the front via a small wall as high as the walls, so it also gives a bit of strength should the bike to a tumble.

All look reasonable so far?

[jonescg]

Hi everyone,

Another update for Voltron Evo

I have made another payment on the motor, so now I'm only about $4000 away from owning it

I have been working on the battery pack and with the help of weber over on the AEVA forum, have come up with a much safer layout for the 700 V pack. It involves more isolation than I had previously considered. I figured that 350 V would be good enough, but since I have the room (just) I can split the pack into 4 quarters of ~175 V each. This means three pack-splitting contactors and a main contactor, and at least two fuses. Ideally a fuse on each end of the full pack would happen too, but I'll see how the real estate works out. The precharge and discharge circuits are good, I just need to put more contactors in series with the packs. My soldered packs are really coming into their own here, as I can slip them in like magazines and bolt the few main terminations on top.

A very basic drawing of how the pack will come together:


The contactor between the two half-packs will serve as a place to attach the output leads of two 350 V chargers. This way the only way 700 V DC is accessible is when the key is ON and the bike is ready to ride. This also comes from the fact that I need to charge with two smaller chargers, as 700 V chargers don't come cheap.

Pictures to come

CHRIS

[jonescg]

Slightly more sophisticated drawing







I might get Danny Sailer to machine these slabs up for me. Unless there is a less flammable plastic, acetal (Delrin) would be best. Can you machine polycarbonate?

[jonescg]

OK this is a more complete version of the box, still missing the side covers. I will shave the corners of the box off so it doesn't protrude too far.



I might be able to make use of the bottom of the pack, but it could be a bit much. And I'm not sure what I'd use it for anyway. Everything else should fit in the top or the front/back.

[megacycle]

Why is the fuse so big?

[jonescg]

It's just the only fuse I had lying around. I reckon a 200 A fuse would work, even if I will be tickling 450 A peaks. The higher the voltage rating the longer it is, so the 200 A 700 V fuses are all rather large

[Ratking]

OK, here is the corrected relay circuit. I'll use a Gigvac 16 contactor for the main contactor, and one of their high-voltage DPDT relays for switching the precharge and discharge circuits.



As before,
Off - discharge circuit is closed, precharge open, contactor open.
Acc - discharge circuit is opened, precharge circuit closed, contactor open.
ON - discharge circuit still open, precharge still closed, contactor closed.

There remains the overcurrent protection and isolation measures, but I will get to these in due course. The giant Bussman fuse is on the positive, but I can't see how a short which generates an epic plasma ball is going to be somehow less catastrophic thanks to the fuse.

Very nice build, I follow all your builds closely to learn and just look at all the bling
I have one question about the quoted drawing. I can't understand any other outcome than a big short when the contactor closes? You have plus and minus at the bottom and they goes into each of the contactors side? I guess there are some major things about the drawing I don't see?

[jonescg]

Thanks - just wait till I start work

The diagram is confusing I'll admit - see how the orange line goes around the top and back down the other side? That is to simulate the location of the batteries. The big + and - refers to the controller B+ and B-.

I have since updated this again, so you will soon have to erase it from your memory I have added more isolation contactors so the pack will be broken into 175 V segments. Charging is done by some kind of proximity sensor which activates only the quarter-pack contactors, so I can use two separate 350 V chargers.