PSI -- LiFePo4 -- Testing.. 1 ... 2... Testing..

[Jozzer]

Indeed Miro, as I said before, your wiring is skinny, and measuring voltage anywhere but at the cell will not give you the cell voltage.
Also, I just tried 2 wattsup meters (customers order, he asked for them to be tested) on the same battery and got 0.4v difference between them on a 12v source..

 

[Battboy]

Under the end stickers ( red / black ) you can see bulges where the Vents are i assume, i don't want to break the seal just yet so for now what's under there is gonna be a mystery..

That is definitely a Vent which is there for a reason. You will find out if you overvolt a cell.

Don

 

[miro13car]

My wiring skinny??
Is it 10AWG size not enough for 50A current??
I think it is thick enough, more than enough for 50A.
never really reach, the highes ever was 49.1A , ever.
WatssUp is very accurate comparing to Fluke MULTIMETER, example this morning 1 hour after end of my ride to work WattsUp read 39.54V and Fluke read 39.52V, if you reduce this to one digit after decimal point it is the same.
During writing this message I went and did test.
WatssUp read 40.13 and Fluke measured 40.12V - very accurate, but it is resting voltage , not recorded momentary 1-2 second spike I usually get on hard acceleration.
When you look at VMS board /pictures on this thread/ there are not any Mosfet or any power semiconductor to conduct 48A,even without not knowing much how this board works you can say load current cannot go through this boards.
MC

 

[miro13car]

these WatssUps were definitely defective,
it is clear when you read WatssUp manual about resolution of readings.
MC

 

[Pete]

WatssUp is very accurate comparing to Fluke MULTIMETER, example this morning 1 hour after end of my ride to work WattsUp read 39.54V and Fluke read 39.52V, if you reduce this to one digit after decimal point it is the same.
During writing this message I went and did test.
WatssUp read 40.13 and Fluke measured 40.12V - very accurate, but it is resting voltage , not recorded momentary 1-2 second spike I usually get on hard acceleration.
MC

There is no such thing as too much copper when it comes to high current wiring. But the wire awg is only part of the story. Total wire length, type of connectors , how are they crimped / soldered and how many junctions are also major players. I don't have the lookup table handy, but my gut feeling is 10awg is slightly on the skinny side for 50A continuous, but should be adequate for your profile.

Your voltage measurements are not a valid test, sorry. To be valid they need to be performed under load. Hook up your Fluke directly across the battery and compare voltages when you are pulling 50A.

 

[Jeremy Harris]

I agree. Choice of wire gauge for an EV is more about minimising resistive losses in the wiring than theoretical current carrying capacity. For example, even a high temperature Mil Spec cable will be rated to carry a lot of current even at a relatively thin gauge, but that doesn't mean it won't drop voltage and get warm in use.

As others here have already reported, upping the gauge of the high current wires frequently gives a noticeable performance improvement. My guess is that 10g wires will give measurable voltage drops at 50A. It's easy enough to check, just measure the relative voltage between the battery terminals and the controller terminals under load.

Jeremy

 

[miro13car]

Pete, Jeremy,Ypedal
What I did I compared reading when bike powered not moving from WattssUp to reading from Fluke connected to where WattsUp is connected - that is strait comparison.
If you all read my posts carefully - I wrote my bike draws Just under 50A for COUPLE OF SECONDS, not continous here.
There is hard to maintain any continous on EV .
Why do you think AWG10 is not enough .
Connectors are different story.
I don't have any on my bike, wire terminated with ring connector /rated for size10 wire/to battery,
soldering on another end to controller. Run from batt to controller maybe 90 cm, that's all.
What I am concern is a "bottle neck" with maybe 10 cm size 14 wires of WattsUpp.
Would it be worth to bring my AWG wires directly to WattsUp - that is real question.
MC

 

[Jozzer]

We are all saying that under load the voltage read by watts up may be different. The voltage wont drop over your wiring UNTIL it is loaded. You need a fluke reading from the battery terminals (or better yet the cells directly!) whilst under load..

 

[Doctorbass]

I call that Fine tuning your ebike.. like trying to get lightweight part, having aerodynamic shape, better mosfet.. better efficiency (Wh/km)..etc.. all these parameter AND resistive loss are important!

Often, Resistive loss are most important than the others.. I often see 72V 40A controiller using only 12 gauges... .. for couples of feet long... and people blaming their battery voltage drop that cause the controller to cut the power probably because it reach the LVC too fast...

Well.. about that .. my oppinion is that if people are ready to pay for a 800$+ ebike kit... they should not forget to have a good wire/connector quality and size!.. that will certainly affect the performance and efficiency!..

at 50A, loosing 1V in the entire wire network is 50W.. and during a 2h ride it could result of loosing 5-6km range !.. Wire size cost less than heavier battery pack!

Also... for those who like acceleration, WIRE SIZE IS THE SECRET!

I use 8 gauge and sometimes a parallel pair of 10 gauge on all my ebike electrical network.

The total volt drop from dead stop to max acceleration and 5500W of power is only from 71 to 66V.. so it's only 7% at full power drawing 85A..

so the total R loss is 0.058ohms


Anyway.. i''m sure that if the septic would just taste that difference, they would gladly upgrade to better wire size :wink:

Doc

 

[miro13car]

"Also... for those who like acceleration, WIRE SIZE IS THE SECRET!"
It depends to what controller do you connect your battery.
Tidal Force controller is not your average Chinese ANALOG controller. It is controlled by software. Resistance produce heat everyone agree. But you forget some controllers would NOT allow more current on acceleration, no matter how fully throttle pushed.
Clearly TF controller for Canadian market is limited by sofware, it is obvious if you read theory of TF motor/controller.
My battery is make-shift connected for testing right now and I can try number 8 wire, however even 1000W TF made in the USA model for USA market uses AWG 10 wire.
Restriction is in controller not in AWG 10 wire.
Again I am considering my wires directly inside WatssUp.
Kind of weird to have size 8 on all bike and "bottle neck" WatssUp thrown with like 10cm of AWG 14 wire.
Do you think AWG 14 sections of WatssUp make any difference ?
MC

 

[miro13car]

Andy,
refering to your chart.
From my riding experience obviously I cannot maintain any constant current for longer than stright streach of bicycle path.
During my trials of my 36V10Ah LifveBatt typically my draw was from 5A to48A with everage I would say 10A.
After 10.5Ah used voltage on terminals read 37.5V five minutes after getting off bike.
Your chart can be made only at the bench test.
What voltage can I read on my batt after 45A draw until 10Ah used?
I don`t know hard to say.
Ypedal test will be very useful here with PSI cells if he can draw continous 40-50 A from cells until say 10 Ah used.
I cannot wait to see results.
MC

 

[Patriot]

Wire Size ?!?!?!?!

If I have a 5303 motor, using a Kelly 100a/50a controller, and wanted to increase its efficiency, do I have to replace the entire 12ga wire from the motor?

What I'm wondering, is whether or not I need to break open the motor casing to replace the short 12ga wire that enters the motor. If I have bigger 10ga wires from the hub of the motor to the controller, it would obviously be better than the stock 12ga.

But, will having a few inches of the original 12ga wire in the motor casing create a choking point, thus destroying my entire attempt at gaining efficiency?

I would hate to have to break open the motor if I don't have to.

the same goes for micro13car's question about a Watts Up meter. I plan install one as well.

 

[AndyH]

Andy,
refering to your chart.
From my riding experience obviously I cannot maintain any constant current for longer than stright streach of bicycle path.
During my trials of my 36V10Ah LifveBatt typically my draw was from 5A to48A with everage I would say 10A.
After 10.5Ah used voltage on terminals read 37.5V five minutes after getting off bike.
Your chart can be made only at the bench test.
What voltage can I read on my batt after 45A draw until 10Ah used?
I don`t know hard to say.
Ypedal test will be very useful here with PSI cells if he can draw continous 40-50 A from cells until say 10 Ah used.
I cannot wait to see results.
MC

Hi MC,
I found testing the Yesa cell that I picked up capacity on break-in, and temporarily 'lost' capacity under steady discharge as the load increased. Here's constant discharge on a cell that's broken in:


Pack voltage will vary with current load, wiring used, temperature, and where the measurement is taken (at least). The LiFeBatt cells tested by Sandia Labs only produced 9.61Ah to 9.88Ah (C/2 discharge, 25ºC) when new and 10.14Ah (measured at a 1C discharge rate and 35ºC) once broken in.

I'm looking forward to seeing these results as well.
Andy

 

[Pete]

If I have a 5303 motor, using a Kelly 100a/50a controller, and wanted to increase its efficiency, do I have to replace the entire 12ga wire from the motor?

No, but.

Look at it like this; when your wiring is under load, think of every inch of the wire as a low value resistor. The thinner the wire, the higher the resistance (for that particular inch of wire). The more high resistance pieces, the more voltage you will drop and consequently power you waste as heat.

As long as the thinnest piece of wire in the circuit will take the max current you can draw (without turning into a little blue puff of plasma), then from a purely functional perspective you are good to go. I have a feeling that ALL of the guys here that are considered to be experienced / experts are probably running wire gauges at least one or two sizes bigger than a 'typical' setup of the same power output.

I agree with Doc's comment that spending a few cents extra to save tens of watt/hours is better than spending many dollars extra to carry those wasted watt/hours in a larger battery.

 

[miro13car]

the point is that this resistive/heat losses happen only when current flows throug wire.
You must agree that AWG 8 for 48A at 36V drawn for couple of seconds at the time is overkill UNLESS it is really long .
Cost is not a problem , in my case is free , weight/copper/, connector size and bulk is a problem.
MC

 

[Jozzer]

The point is you are complianing of voltage sag under load..

 

[Ypedal]

Alright, getting back on track here...

Can't discharge until you can recharge, so went by the local Circuit City today and found the best buy to be a roll of Monster Cable speaker wire, looks like 14 gauge..nice thick rubber jacket and flexible.. 8)

I need to hunt down a stock of these ring terminals, 8 per 8 cell pack, and i have 6 of these to build. They don't look at my funny anymore at Home Depot, well,.. you know..

 

[Ypedal]

Crimped and soldered ! ( I don't ever want to have to open up this pack unless i absolutely have to, once it gets built ! )

The tips are the Yellow 12/10 gauge, so i stripped double the lenght of copper and folded it back on itself to fill the crimp better. I removed the yellow plastic covers and will just heat shrink them instead, the plastic tends to look like crap once you crimp them, and they melt when you solder the ends gumming up your iron if you slip.. looks better with shrink anyways.

 

[johnrobholmes]

Waiting for some good tests of these. I love the screw on terminals.

 

[Ypedal]

I'm anxious to get it on the road, but can't risk killing any cells or denting them with improper padding at the very least.. before that happens i need to cycle the cells a few times on the bench to determine usable AH capacity on the CA meter they will be used on..

With that in mind i went parts gathering today.. spend well over 200 $ on shrink tube, crimp rings, more wire, velcro, tape, etc etc etc...

That and a 1/4" drive troque wrench !!!! :wink:

 

[Ypedal]

One down !

 

[johnrobholmes]

Balance plugs?

 

[Ypedal]

Yep, i plan to hook these babies up to Gary/Bob/Fechter's ( Hey.. we need to come up with a name for that thing if it starts to get produced in volume !! ) bms , right now i'm using single cell chargers to insure full charge, without overcharge, and until i get at bare minimum LVC boards on there i'll keep it to 8ah or less per cycle on the bikes..

Voltphreaks changed their single cell chargers, i don't like the new units, they don't alway start if you hook them to the battery before the AC line, and they don't do the orange/change to green stage, they just pump 2 amps and then flip green at 3.7v... if you remove the AC line, the LED's stay on feeding from the battery pack.. and they get WAY hotter than the originals.. i sent Tony an email about it but the originals are no longer available. I"m contacting Soneil shortly for some bulk units.

 

[michaelplogue]

Subscribed.... Definately want to keep my eye on this thread.

 

[Battboy]

When you are finished what will be your final Pack Configuration ? Are you planning a Pannier arrangement or ?

Best,

Don