jonescg's NEW electric racebike BUILD thread!

[harrisonpatm]

I just appreciate this addition to the copper-nickel sandwich discussion, because according to his tests, you can get reliable copper welds at low-ish power, this copper tape is quite cheap and easy to manipulate, and although it is quite thin, his tests are showing that there actually is an increase in current carrying capability. Much appreciated

 

[j bjork]

This sure is interesting
The nickel copper sandwich is nothing new on the forum, but I dont think I have seen copper tape used before.
(I am not sure, I havent followed the discussion closely)

Have I got this right, you just take a nickel strip and tape the copper tape to it?
Then weld it to the cell, but not on the other end? And you still get good connection with the glue between the nickel and copper?

 

[jonescg]

This sure is interesting
The nickel copper sandwich is nothing new on the forum, but I dont think I have seen copper tape used before.
(I am not sure, I havent followed the discussion closely)

Have I got this right, you just take a nickel strip and tape the copper tape to it?
Then weld it to the cell, but not on the other end? And you still get good connection with the glue between the nickel and copper?

So far, yes, that's what I've observed.

The 0.07 mm copper adhesive tape is stuck to the 0.2 mm thick nickel. The copper side is welded to the cell top through the nickel above. I used a Sunkko transformer-based spot welder set to about half of maximum power. Whatever the "8 ps" setting means, anyway. Just two bzzts (four contact spots per end of the cell and it's solid. I loaded it up with 60 A for a good 7 or 8 seconds and I could leave my finger on the tab with no burns. Barely got warm.

In my case, the other end of the tab is soldered generously to a 0.9 mm copper busplate designed to handle the 200+ amps a typical 5p battery will be passing, as this soaks any conductor-derived heat pretty effectively.

In effect, we have a parallel resistor where one resistor is 5 times lower than the other. The adhesive is no impediment to the spot-weld, but it doesn't seem to come off the nickel either.

 

[jonescg]

Figured a video was the best way to explain it:

 

[j bjork]

Thanks, so you solder both nickel and copper on the other end.
The conductivity though the adhesive is not really an issue then.
Good idea for an easy nickel copper sandwich

 

[amberwolf]

People *have* tried copper tape over various conductive surfaces to improve resistance, but I don't think any of them welded it to them; I don't recall the results in most cases. The few I recall weren't very successful, probably because the adhesives weren't nearly conductive enough. I think some of them were using them to conduct heat vs electricity, and those might have been more successful.

Some of the previous posts using "copper tape" as the words, there are probably others with different keywords.

Search results for query: "copper tape"

endless-sphere.com

 

[jonescg]

Well I'm satisfied I have the termination system sorted.
I was disappointed to measure the DCIR is 16 mOhm, which is more or less what the spec sheet says. Even with 5 in parallel and 168 in series, the effective pack DCIR is about 0.53 Ohm. At my preferred 250 A discharge, we're talking about a 33 kW heater in the shape of a battery pack Well, this will be suited to short, stop-start tracks. I reckon it would do some damage at a power track...

 

[DogDipstick]

It will actually be less of a beast than before, but should be able to manage twice as many laps.
What real world C-rates are folks seeing from these Molicells? Because at 45 amps per cell, I'm looking at about 140 kW. We got about 180 kW from the pouch cells, so outside of Philip Is or Queensland Raceway, we won't really notice the difference.

Lol They suck. Half that 45A and they are still loiving a short life.

Like... 0.7kW for every kG vs... like 3kW for every kG of a good highway lipo ( such as the Chevrolet cells I use... 40K cycles to 50% DOD baby! )

. Last 10kWh I bought cost 1400$ brand new lol.

I would think it would be significantly less powerful given you can only stuff 0.7kW for every kG. RaTHER the 180kW you claim for the pouch cell pack. Plus the lifespan detriment.

Padja data. Plus some from others. Comparing the densities of the Model3 2170.

How much ( kG ) will this pack weigh?

 

[calab]

I reckon it would do some damage at a power track...

Damage to tires and marking asphalt you mean or damage to the spread of points scores or the damaged ego's?

 

[DogDipstick]

Well I'm satisfied I have the termination system sorted.
I was disappointed to measure the DCIR is 16 mOhm, which is more or less what the spec sheet says. Even with 5 in parallel and 168 in series, the effective pack DCIR is about 0.53 Ohm. At my preferred 250 A discharge, we're talking about a 33 kW heater in the shape of a battery pack Well, this will be suited to short, stop-start tracks. I reckon it would do some damage at a power track...

How do you reach that?

based on 16mOh per and 4p?

For 168s. Lol.

I could do the math as I know it if you like.

You will probably disregardful of my numbers tho. Usually are.

This is the math as I know it.

Vd=xS/xP*Rc*I

You will get your Voltage drop by number of series cells divide by number of parallel cells times average cell resistance in ohms time current draw in amps but you know this.

Oh no.. ... its half a godamn whole OHM! 536 milliohm. 5p 16mOh cell 168s. Lol 33kW? heater on a 250A discharge. Toasty. Bahahah. Whole Chevrolet pack is ... 48mOh. 96s 2p of 1mOh cell. For its 120kW.. AND its OEM water cooled for that lifespan and output. Good luck removing 33kW of heat ( if those round cells dont comply first)

5p of 16mOh cells means a 3.2 mOhms. Per string.

3.2 mOh ( per string) ( x) 168 ( strings) = ?

How much ( kG ) will this pack weigh?

It is 12.1kWh nominal?

6,000$ in cells?

 

[jonescg]

Err, yes, you have done the maths correctly. Thanks for reinforcing the arithmetic.
5p is 0.016 Ohms / 5 = 0.0032 ohms per cell group. 168 of these in series = 0.5376 Ohms. So yeah, if this pack was even capable of 250 A like the LiPo pack was, that's I2R = 33,600 W. Your Chevy cells sound great. I assume you can fit 12 kWh inside a motorcycle designed to operate at 168s?

Look, I knew I'd be taking a power hit by going for a more energy dense cell. But our local track sees bikes exceeding 220 km/h for about 3 seconds at best, and the rest of the time around half that, so even 100 kW peak will be enough to have fun. My main goal is to complete an 8 lap club race. Something the previous pack couldn't do.

 

[jonescg]

Damage to tires and marking asphalt you mean or damage to the spread of points scores or the damaged ego's?

Damage to itself... Depends on the duty cycle we give it, but there will be time for it to cool between sessions.

 

[DogDipstick]

Nah that's just a thing it's I mean if you want to fit 20 kW hours in a bike you got to use a good energy cell. I certainly understand the motivation. Seven or eight kilowatt hours just won't cut it.

180 kW must have been insane though.

 

[jonescg]

180 kW must have been insane though.

On my first proper trackday with it, I took until the third session just to muster the balls to wind the throttle to the stop. Turn one comes at you fast :O

 

[rivvs]

Hey, just found this thread. I am building a battery pack using p42a and got my busbar made on alibaba. This copper with nickel tab laser welded to the copper. And the nickel is copper plated.

crf450r 45kw race bike build

Here's an article I wrote with this build in it. "Two Builds using a 3shul VESC Controller"

endless-sphere.com

 

[jonescg]

Very neat! That's an awesome bulk production solution. What's the application?
[Edit - followed your link, duh]

 

[jonescg]

Well my busbar design is complete, and I've sent it off to the jet cutters for a quote. It's probably going to be around $1500.

The final termination will be like the module I made for the eBike:

I'm hoping to be able to just use M8 stainless screws to hold 35 mm cable links between the four modules, or copper bus bars if it's a bit tight for space.

 

[riba2233]

Hey, just found this thread. I am building a battery pack using p42a and got my busbar made on alibaba. This copper with nickel tab laser welded to the copper. And the nickel is copper plated.

crf450r 45kw race bike build

Here's an article I wrote with this build in it. "Two Builds using a 3shul VESC Controller"

endless-sphere.com

View attachment 337191

Ok this looks sick!

 

[jonescg]

My copper arrived. After the initial jigsaw puzzle, the next step is to countersink the fixing holes, then flip the underside busbars over and do the same.

Then it's time to make up 1700 18 mm long, 8 mm wide copper nickel sandwiches, and hole-punch the end which is soldered to the busplate. This will ensure good contact to both the nickel and the copper. Otherwise the risk is only the thin copper bonds to the busplate, and the nickel just held on with adhesive.
I've not yet started the plastic CNC fabrication job. That's about $2000 worth of work. I think polycarbonate is still the best option, as most other plastics are impossible to stick to, or are harder to machine.

 

[jonescg]

Can't say I'm enamoured with this approach... If only there were enough folks in Perth/Oz who were interested in a pulse arc welder...

 

[CONSIDERABLE SHOUTING]

Can't say I'm enamoured with this approach... If only there were enough folks in Perth/Oz who were interested in a pulse arc welder...

How are you bonding the nickel strip to the copper? I thought the sandwich method was using the resistance in the steel to hold the copper directly to the cell.
EDIT: Nevermind! I re-read the portion about the copper tape. Very interesting...

 

[jonescg]

It's not the best, but I have found that hole-punching the soldered end of the layered tab results in a really solid nugget of solder between the copper busplate and the nickel.

 

[hallkbrdz]

Hey, just found this thread. I am building a battery pack using p42a and got my busbar made on alibaba. This copper with nickel tab laser welded to the copper. And the nickel is copper plated.

crf450r 45kw race bike build

Here's an article I wrote with this build in it. "Two Builds using a 3shul VESC Controller"

endless-sphere.com

View attachment 337191

Did you buy NxN slot sheets and have them cut locally, or can you actually send them the patterns you need?

Also, what is the copper thickness?

 

[jonescg]

I reckon it will take four solid days to put the nickel-copper strips together, cut them to length, hole-punch one end (massively improves solder joint quality) and solder all ten of them to each busplate. I could find a faster way to do it, but that would take about 4 days to research... I'm trying to get onto U-sheen for some more thermal epoxy but they aren't replying to emails.
Should also get the polycarbonate capture plates machined in the next three weeks.

 

[jonescg]

Okay, so who wants some thermally conductive, electrically insulating epoxy potting/adhesive compound?
Cause the minimum purchase is 70 kg! That's AUD$3550. I only need about 10 kg, but will buy 20 kg if I have to. But 70 kg!?