methods 26kwh Emoli bug build

[steveo]

So then I get some batteries. . .

So Luke calls me up (you know - the mentally deranged guy we call Liveforphysics) ranting about some batteries that I have to buy. Babbling about how they are $0.18/watt and they would be perfect for my build. Another long story short that includes a frantic trip the the bank 3 minutes before closing an a wad of cash so big it only fit in a coat pocket and we have the following:

Batteries - lots and lots of batteries

2,650 to be exact -> or about 26KWH. To put it in perspective, that is 500lbs of A123 sized batteries. About half a Tessla pack

I went thru several testing rigs - first testing at 20A, then at 100A. I started by trying to bin up the Ah of each bank but got board of that quick. What I am doing now is testing the behavior or the cells under a 3C load

Bank = 2S 12P
Rated Ah = 34.8 AH
Test current = ~100A
Apx C rating = 2.9C

I use 14 x 1ohm 30W batteries in parallel sandwiched between 3 aluminum Infineon cases cooled via a computer fan.
Datalogging at the cell level happens with the iCharger
Datalogging of the current happens with the CA


View attachment 10

This is the GUI for my BMS - here I am only using it to display/record Amps and Ah coming from the CA

View attachment 9

I have started collecting parts on Ebay - this here is a 50mV 100uOhm 500A shunt.
QUALITY! After creating blue lightning by smoking out my first shunt (a story I wont repeat) I learned my lesson -> this ain't no ebike

View attachment 8

Ignore the voltage - that is just a pack I use to power the CA. The current is all we are really interested in as we are testing 2S. Unfortunately 6V is not enough to power the CA

View attachment 7


This is how I am going to configure them: Each little bank will be 2S 48P. Rated that would be 140Ah but it is actually more like 120Ah. If I were to run the packs hard - say at 400A continuous it would be more like a 100Ah pack followed by another 20Ah at "normal driving" 150A range.

View attachment 6

I went thru a few variations of this laod - currently it is running at something over 2KW. Anyone who has ever tried to dissipate that kind of heat knows.... It is no joke. I am blasting the crap out of that stack of resistors and it is still hotter than hell. At one point it was reaching 230C Now I have it under control. A 12V 2A PC fan sounds like a jet engine but does the trick.

View attachment 5
View attachment 4

I dont know if you have ever tried to open a circuit that is drawing 100A DC -> errrr..... it can be a bit testy and connectors dont last long. For that reason you use a pre-charge resistor. This particular resistor is 1.5ohms 30W and it draws about 5A. First I clip this on then I make the main connections. This keeps the sparks down to a dull roar, especially when disconnecting.

View attachment 3

I am starting to prepare the wiring and collect an entire new set of "big boy" tools. This is ghetto - but it will work for what I want to do.

[ATTACH type="full" alt="115_Isolator.jpg[/attachment

Yea - that is right. My very own 4/0 cable.

attachment=0]116_Isolator.jpg"]1[/ATTACH]


-methods

Hey Methods,

would you mind explaining how you built that load tester?

I'm looking to build a nice load testor to test my a123 cells... each row is configured in 11p... would be nice to see what i get out of them at 100amps continous...

how did you cut the power when the test was done... just disconnected the load manually?...

thanks in advance ...

pics attached of the cells i've tested at 5amps discharge..

-steveo

 

[methods]

Dude - what is that rig you are using to hold the batteries?
Will that work with 26700 cells? (as opposed to 26650)
I need 10 rigs that will do 1S 48P


DISCHARGE LOAD:

The problem with a resistive rig like mine is that is produces an ass-load of heat and moves around with temperature. If I were to do it over I would use Light bulbs - lots and lots of light bulbs - but you need to consider your voltage.

I = V/R
P = V*I

So I was doing only 2S (8V) @ 100A so call it a kilowatt.

If you want to do only 1S @ 100A you are going to need a load that is like 0.040 ohms. You would need about 5,000,000 lightbulbs in parallel to get you that low so....

I did the monitoring with a CA.
Just take a 6S battery and power the CA
Hook the shunt lines up to a big 500A shunt
Now the CA can give you your current (but not your voltage)

If you cant get the resistance down low enough you can always start stacking banks in series - but WATCH OUT -> your power is going to go up really quick - and the more power you develop the more you have to get rid of.

1S, 4V, 100A, 400W
3S, 12V, 100A, 1.2KW

Lot easier to burn off 400W than 1.2KW

Anyhow - I am clearly babbling.

Search around the forum for constant loads - I have seen a few designs in the 500W - 1KW range.
If you go resistive you WILL need fan cooling
If you go with lightbulbs you will need quite a few in parallel to get the resistance down

Disconnecting is no problem for 1S or 2S -> Just pull apart.
Disconnecting for 3S or more can start to pull a good size arch - be careful!

TRICK
*** Put a 20W resistor in parallel with the connector you use to connect and disconnect the load. This will eliminate the spark on contact and help eliminate the spark on removal. Just move fast as the 20W resistor can heat up very quickly.


Ok - now tell me where you got that big rack for the cylindrical cells?

-methods

 

[liveforphysics]

It's from a failed Prius lithium conversion pack Methy



I used a bar of 3/4" wide 1/8" thick 304 stainless steel for my load bank. With a welding electrode at the far end of the 48" length of it, it was like 60amps at 3.3v. As you slide the welding electrode clamp up the bar, it would increase to about 800-900amps as you got in close, the bar would get hot enough to smoke the oils from your skin off, but it stayed in safe temp limits of 304 stainless, and the welding clamp stayed a safe temp as well. (using 0000 cable here)

Cost a whopping $30 in materials for the bar, welding terminal and electrode clamp. Enables testing of multi-hundred amps at single cell voltages.


You don't need to discharge each section at a zillion amps, just compare relative Ri between the groups. It's not like you're evaluating A123 cells as a product, you just want to find where/if you have bad ones.

 

[methods]

...just compare relative Ri between the groups. ....

Thats assuming he can get an accurate Ri reading.

-methods

 

[steveo]

Dude - what is that rig you are using to hold the batteries?
Will that work with 26700 cells? (as opposed to 26650)
I need 10 rigs that will do 1S 48P


DISCHARGE LOAD:

The problem with a resistive rig like mine is that is produces an ass-load of heat and moves around with temperature. If I were to do it over I would use Light bulbs - lots and lots of light bulbs - but you need to consider your voltage.

I = V/R
P = V*I

So I was doing only 2S (8V) @ 100A so call it a kilowatt.

If you want to do only 1S @ 100A you are going to need a load that is like 0.040 ohms. You would need about 5,000,000 lightbulbs in parallel to get you that low so....

I did the monitoring with a CA.
Just take a 6S battery and power the CA
Hook the shunt lines up to a big 500A shunt
Now the CA can give you your current (but not your voltage)

If you cant get the resistance down low enough you can always start stacking banks in series - but WATCH OUT -> your power is going to go up really quick - and the more power you develop the more you have to get rid of.

1S, 4V, 100A, 400W
3S, 12V, 100A, 1.2KW

Lot easier to burn off 400W than 1.2KW

Anyhow - I am clearly babbling.

Search around the forum for constant loads - I have seen a few designs in the 500W - 1KW range.
If you go resistive you WILL need fan cooling
If you go with lightbulbs you will need quite a few in parallel to get the resistance down

Disconnecting is no problem for 1S or 2S -> Just pull apart.
Disconnecting for 3S or more can start to pull a good size arch - be careful!

TRICK
*** Put a 20W resistor in parallel with the connector you use to connect and disconnect the load. This will eliminate the spark on contact and help eliminate the spark on removal. Just move fast as the 20W resistor can heat up very quickly.


Ok - now tell me where you got that big rack for the cylindrical cells?

-methods

These are hymotion packs version 1

they are from toyota prius basically the upgraded battery pack or extra range packs ...

the plastic ends came with the cells i purchased... i wish i had some better pics of the plastic covers on hand .. but they secure the cells in place..... they were for a123 cells .. i'm assuming the a123 cells are fater then the emoli cells you have by the look of the photos i've seen of your pack..

oh yea....

wanted to ask you... whats your range like on your ev at 120ah with your 72v pack?

It's from a failed Prius lithium conversion pack Methy



I used a bar of 3/4" wide 1/8" thick 304 stainless steel for my load bank. With a welding electrode at the far end of the 48" length of it, it was like 60amps at 3.3v. As you slide the welding electrode clamp up the bar, it would increase to about 800-900amps as you got in close, the bar would get hot enough to smoke the oils from your skin off, but it stayed in safe temp limits of 304 stainless, and the welding clamp stayed a safe temp as well. (using 0000 cable here)

Cost a whopping $30 in materials for the bar, welding terminal and electrode clamp. Enables testing of multi-hundred amps at single cell voltages.


You don't need to discharge each section at a zillion amps, just compare relative Ri between the groups. It's not like you're evaluating A123 cells as a product, you just want to find where/if you have bad ones.

i might go check out the stainless steel bar idea... sounds powerful....

can't compare resistances... my megapower sr 960... only goes as low as 0 ohms .. and thats what i'm getting heeheheh..

so far the 11p rows range from 22ah-- 23ah+...

-steve

 

[liveforphysics]

Methy.
I have a CNC bro.
If you want fancy end plates, CAD it, and i will cut it for you.

 

[methods]

Methy.
I have a CNC bro.
If you want fancy end plates, CAD it, and i will cut it for you.

pfffttt.... yea, I will hold my breath waiting for you to get around to that

-methods

 

[methods]

wanted to ask you... whats your range like on your ev at 120ah with your 72v pack?

-steve

Actually it is 20S48P so 48 * 2.9Ah = 139Ah
But many of the cells are thrashed so I never discharge past 100Ah

So the best way to look at that is as a 7.4kwh pack and what you want to ask me is what my wh/mile is.
I consume anywhere from 250wh/mile to 350wh/mile depending on how I drive.
That is a 20 to 30 mile range

Considering that my pack weighs only 250lbs that is reasonable.
The car weighs less than it did stock, has a better weight distribution, and has respectable acceleration.

-methods

 

[methods]

Ok - back on track.

I have put a few hundred miles on the the Emoli Bug going back and forth to work and around town. The little 4hp motor has served me well - only getting hot enough to stink on a few occasions. The forced air cooling really did the trick.

You may recall that I have the 120V / 1,200A Kelly waiting in the wings (courtesy of Steve - thanks!)
Well... I found a mate for it.

Behold the Warp 9

View attachment 1
View attachment Warp9_First_Day008.jpg


I was going to sell off my Emoli cells when I found a guy selling this motor (obviously used) only a few miles from my house. He actually has a bug nearly Identical to the build I am doing so.... I can take a hint. Looks like the MX-5 will have to wait - I am going to put this in the bug and see how it runs first.

Now.... To make my crappy 6" adapter plate work with this monster 9" motor.

-methods

 

[katou]

So this would be an upgrade from the 4 hp motor?

Katou

 

[amberwolf]

Unless it also forces him to replace the springs on that end of the car for the extra weight. But for the power, it's probably worth it. 8)

 

[methods]

I wish to perform copious smokey burnouts for the pleasure of my friends and colleges. I wish for them to appreciate the electric revolution. I dream of spraying molten rubber into their hair and faces.

I cant wait to smoke the first guy who pull sup with one of those 1994 Accords with a 4" tail pipe and 3-tier spoiler

-methods

 

[Arlo1]

I wish to perform copious smokey burnouts for the pleasure of my friends and colleges. I wish for them to appreciate the electric revolution. I dream of spraying molten rubber into their hair and faces.

I cant wait to smoke the first guy who pull sup with one of those 1994 Accords with a 4" tail pipe and 3-tier spoiler

-methods

If you get luke worried enough he will panic and just use to much laughing gas and wont even finish :wink:

 

[liveforphysics]

I wish to perform copious smokey burnouts for the pleasure of my friends and colleges. I wish for them to appreciate the electric revolution. I dream of spraying molten rubber into their hair and faces.

I cant wait to smoke the first guy who pull sup with one of those 1994 Accords with a 4" tail pipe and 3-tier spoiler

-methods

If you get luke worried enough he will panic and just use to much laughing gas and wont even finish :wink:

ohhh damn!! lol!

This is actually what's going to happen.

Option A: He sends it a reasonable amount of current, never takes advantage of the power potential of the motor, and it's just another semi-gay EV.

Option B: He uses the potential of the motor, and on the 3-4th transmission/axle failure, he get's frustrated and either quit's EVs, or go back to option A.

 

[steveo]

Ok - back on track.

I have put a few hundred miles on the the Emoli Bug going back and forth to work and around town. The little 4hp motor has served me well - only getting hot enough to stink on a few occasions. The forced air cooling really did the trick.

You may recall that I have the 120V / 1,200A Kelly waiting in the wings (courtesy of Steve - thanks!)
Well... I found a mate for it.

Behold the Warp 9

View attachment 1



I was going to sell off my Emoli cells when I found a guy selling this motor (obviously used) only a few miles from my house. He actually has a bug nearly Identical to the build I am doing so.... I can take a hint. Looks like the MX-5 will have to wait - I am going to put this in the bug and see how it runs first.

Now.... To make my crappy 6" adapter plate work with this monster 9" motor.

-methods

you bastard LOL...

how you find a warp 9... your so lucky!!!

you must of got a killer deal compared to a new one ..

-steveo

 

[katou]

Option A sucks (except that I've seen some semi-gay dudes downtown that look like they could accidentally beat the tar out of me while sneezing)
Option B also sucks.

You're the man with the high-strength power tranny plan LFP, so what's the way to go here?

And how powerful is that thing anyway? I couldn't really get a feel from looking up the stats.

Katou

 

[AussieJester]

And how powerful is that thing anyway? I couldn't really get a feel from looking up the stats.

Katou

You know WhiteZombie...worlds fastest street legal lecky car, that uses/used 2 of the a Warp 8 albeit a lil tricked out i believe...

KiM

 

[methods]

And how powerful is that thing anyway? I couldn't really get a feel from looking up the stats.

If the tranny is going to break it will most likely be caused by stepping on the gas too hard while there is still "slop" in the gears. This thing has enough instantaneous torque to shatter the transmission. I have to be very careful even with the little 4hp motor.... Any time I take off I have to "wind up" the gears before I step on it.

This motor is by no means a true tire smoker. You would really need 2 of these. It will make for a great driving car - but this wont be any sort of drag racer.

Luke is joking but he is right.... I have already done the "lets spend $20k building something awesome just so that it can be broken 95% of the time" thing. That is total BS - and I am not going to do it again. I am going to crank the current limit as high as it will go (because to me, that is the only setting) but after the first thing breaks I am dumping this thing in the gutter.

-methods

 

[liveforphysics]

Luke is joking but he is right.... I have already done the "lets spend $20k building something awesome just so that it can be broken 95% of the time" thing. That is total BS - and I am not going to do it again. I am going to crank the current limit as high as it will go (because to me, that is the only setting) but after the first thing breaks I am dumping this thing in the gutter.

I expect nothing less from a quitter with a broken spirit.

I've been thinking about how I'm going to do an EV. I'm going to base the whole build around the rearend/tranny layout. Start from something that is proven to handle 4,000hp launches on slicks thousands of times without breaking. The custom fit it to whatever chassis seems best, and just cross-off drive-train failures from the list of things to go wrong.

 

[liveforphysics]

BTW- For folks looking to build an electric car chassis that can handle every bit of torque you ever though of throwing at it, that beautiful rear-end, complete with the brake setup, hubs, linkage plates etc, is only $3050 brand new (free shipping at Jegs!), and generally $1500-2000 used off a dragster/racecar (if you can find one).

It's also shockingly light weight for how bulky it looks.

 

[jag]

BTW- For folks looking to build an electric car chassis that can handle every bit of torque you ever though of throwing at it, that beautiful rear-end, complete with the brake setup, hubs, linkage plates etc, is only $3050 brand new (free shipping at Jegs!), and generally $1500-2000 used off a dragster/racecar (if you can find one).

It's also shockingly light weight for how bulky it looks.

What does lightweight mean? 100lbs?

So omitting the gears and with some fair amount of engineering I imagine one could fit the motor to just the differential on the rear axle.

A more realistic alternative would be to find a used VW bus transmission. The internal gears are stronger. Some models have the same mounts as the bug tranny, so you can use the same adapter plate.

 

[methods]

BTW- For folks looking to build an electric car chassis.....

Yea... I think anyone looking to build an EV should start with one of these




And then they should crack open their treasure chest

View attachment 1


and buy them selves a face-paint job like this





-methods

 

[liveforphysics]

I

BTW- For folks looking to build an electric car chassis.....

Yea... I think anyone looking to build an EV should start with one of these

View attachment 2


And then they should crack open their treasure chest

View attachment 1


and buy them selves a face-paint job like this





-methods

All solid advice.

I was thinking more for when you cry about breaking miata diffs daily, cause I predict the scrapping of the bug inside of 2 weeks after getting the warp9 hooked up.

 

[Arlo1]

And how powerful is that thing anyway? I couldn't really get a feel from looking up the stats.

If the tranny is going to break it will most likely be caused by stepping on the gas too hard while there is still "slop" in the gears. This thing has enough instantaneous torque to shatter the transmission. I have to be very careful even with the little 4hp motor.... Any time I take off I have to "wind up" the gears before I step on it.

This motor is by no means a true tire smoker. You would really need 2 of these. It will make for a great driving car - but this wont be any sort of drag racer.

Luke is joking but he is right.... I have already done the "lets spend $20k building something awesome just so that it can be broken 95% of the time" thing. That is total BS - and I am not going to do it again. I am going to crank the current limit as high as it will go (because to me, that is the only setting) but after the first thing breaks I am dumping this thing in the gutter.

-methods

Yup then you will just be even more pissed off at the world Oh man.... I need some workers to come finish all my stuff so I can show you how to do it!

 

[Arlo1]

Luke is joking but he is right.... I have already done the "lets spend $20k building something awesome just so that it can be broken 95% of the time" thing. That is total BS - and I am not going to do it again. I am going to crank the current limit as high as it will go (because to me, that is the only setting) but after the first thing breaks I am dumping this thing in the gutter.

I expect nothing less from a quitter with a broken spirit.

I've been thinking about how I'm going to do an EV. I'm going to base the whole build around the rearend/tranny layout. Start from something that is proven to handle 4,000hp launches on slicks thousands of times without breaking. The custom fit it to whatever chassis seems best, and just cross-off drive-train failures from the list of things to go wrong.

First off hows your other 100 projects coming along? And secondly Why have a diff? Just wind the motors to wheel rpm and either have hub motors or a good cv shaft strait from the motor to the wheel! I thought you knew this already?