Quantifying Battery condition

[birmingham]

Also i saw a few posts about lithium ion batteries from a leaf, im guessing the cost of this must be astronomical because from what i can gather it must not be a popular conversion. guess there just aren't enough leafs in the wrecking yards yet.

[JohnnieB]

All sorts of red flags waving.

First of all, there are several different makes and models of LED Bargraph SoC meters and all of them tell more lies than politicians do.

If you know what the voltage is when the various number of bars are lite, it can be used as a "Fuel" gauge, but it doesn't help to know if the "tank" is 3/4 full when you don't know how many gallons the tank holds. (Amp-Hour capacity of your batteries equates to tank capacity)

You need a DVM to measure the battery voltage and with an accurate voltage, you can calculate the SoC (State of Charger), but you cannot determine the AH capacity of the batteries.

SoC simply means percent full, so a 5AH battery, a 50AH battery and a 500AH battery will all have the same voltage when are at 75% SoC, or any other SoC.

To determine the battery AH capacity, they have to be load tested.

If the cart runs okay to the store, there probably isn't anything wrong with the controller or motor. Once a good set of batteries are installed, it will very likely do okay on the way back from the store.

How often do you charge the batteries? They ought to be charged after each use.

The charger's output has to match the battery pack voltage.

Yes you can charge two batteries of a 8 x 6V 48V pack separately, but not advisable.

Same goes for tapping the main battery pack for 12V accessories. It is better to use a separate 12V battery for accessories, or use a DC-DC voltage converter some the entire pack is loaded equally by the accessories.


It is normal for the motor to get hot, but the cables shouldn't.

[birmingham]

Originally Posted by JohnnieB

First of all, there are several different makes and models of LED Bargraph SoC meters and all of them tell more lies than politicians do.
.

So the bars dont directly correspond to the battery pack voltage? assuming i need to add wire extensions to the battery pack attach to my volt meter and monitor while operating the cart.

Why not advisable to charge last 2 batteries separately? this lead to uneven wear/chargers not charging to same level?

The cart is charged every time it is used now, not always been the case in its life the batteries have not been taken as good of care as possible.

[JohnnieB]

Originally Posted by birmingham

1. So the bars dont directly correspond to the battery pack voltage? assuming i need to add wire extensions to the battery pack attach to my volt meter and monitor while operating the cart.

2. Why not advisable to charge last 2 batteries separately? this lead to uneven wear/chargers not charging to same level?

3. The cart is charged every time it is used now, not always been the case in its life the batteries have not been taken as good of care as possible.

1. There are several flavors of them and they don't all work the same. Some are true bar graph voltmeters, while others adjust to the voltage that existed when they are connected as full-scale and others "recalibrate" to full-scale every time the battery pack is charged. Some have linear scales and others don't. All of the look about the same.

Once you figure out what type you have and verify what the voltage is when each bar is lite, they might fill the role of a rough estimate type fuel gauge, but that is all.

The attached chart shows the voltages vs bars lit for someone's 48V bar graph meter. It is for a specific meter, so you need to do something similar to find out what the various bars on your bar graph meter actually mean in voltage terms. Of course, if you have the type meter that recalibrates every time the batteries are charged, the voltage corresponding to the bars change.

There are DVM panel meters available for monitoring actual battery voltage. (Second attachment is what I use)

2. They are discharged in series, so they should be charged in series unless you charge them equally in parallel, which isn't likely using a 36V charger designed to charge deep cycle batteries and a 12V charger designed to charge car batteries.

If the batteries in the pack aren't kept reasonable well balanced, their useful lifespan is reduced and they have to be replaced sooner.

For maximum battery lifespan, use a charger that matches the pack voltage, uses dV/dT (change in voltage over change in time) technology and float charges after regular charge cycle finishes.

The are a few chargers that meet that criteria out there, but the best bang for the buck is the DPI Accusense. http://www.cartsunlimited.net/48v_Battery_Charger.html

3.

[remac]

You don't need to monitor the voltage while operating.
It can be useful but it can also be misleading.

There are different types of voltage readings on the batteries.

1. At Rest - voltage of battery(ies) while not under load.
Most accurately measured 1 hour after use or 12 hours after charge.
This is what gives you the SoC of the battery(ies). Keep SoC above 50%.
If you measure the pack (all of the batteries) you should get 36-38 volts.
If you measure the individual batteries, you should get 6-6.5 volts.
The individual battery voltages should be within one or two tenths
or else you have a potentially (pardon the pun) bad battery.

2. Under Load - voltage of the battery(ies) while under load.
This will vary depending on the load, whether you are coasting,
going uphill or down, etc., so it can be misleading.
This can be as low as 1.5vpc (volts per cell) or 27 volts total
up to over 2 vpc (36 volts total).
1.5 VPC IS VERY BAD AND WILL KILL YOUR BATTERY(IES).
1.75 vpc (31.5 volts) is about the lowest you want to go UNDER LOAD.

3. On Charge - voltage of the battery(ies) while being charged.
This varies depending on your charger and battery pack
but can go as high as 47 volts on a 36 volt pack.

Practically - I check my voltage at the charging port before and after I ride.
Then I check the individual battery voltages to look for a bad battery.
(They should all be about the same, around 6.3 volts when charged).
After I am done riding I check the fluid levels before I charge the pack.
If no plates are exposed I charge the battery pack, checking voltage as I go.
Once I am done, I check the fluid again and add water if needed.
12 hours later I check the voltage again for a final 'at rest' reading.

The other thing I do is read/read/read all the posts here.
There is a ton of good info, a lot on stuff you don't know you need to ask, yet.

[rlw]

The MooshiMeter would be an awesome tool for logging this kind of testing. It's a recording V-O-M that logs data and displays it on your smartphone via BlueTooth (apps available for i-things and android-things)... It can also record two of its inputs simultaneously (Amps/Volts, Volts/Ohms, etc.). It's available for pre-order right now at $119 and the website says it's in manufacturing and will be shipping in a few weeks.

It could be vaporware, or it could be real - YMMV.

I've been looking for something like this for a long time...

<full disclosure> I don't have any relationship with the MooshiMeter folks, just interested in a device like this </full disclosure>. I saw an article about it on Hackaday.com.

I still have my dad's old paper chart voltage recorder (but no paper!), and one of his polar paper chart recorders (I think that one recorded pressure on a gas line). Both have been sitting in an unheated garage for 25 years, so neither likely works.

RLW

[cgtech]

MooshiMeter? If it is what says it is, AND not Vaporware... Take My Money! I have been trying to do the same type of thing for a while with RaspberryPI. No, wait, i still want to build it wit RaspberryPI.

[rlw]

Originally Posted by cgtech

MooshiMeter? If it is what says it is, AND not Vaporware... Take My Money! I have been trying to do the same type of thing for a while with RaspberryPI. No, wait, i still want to build it wit RaspberryPI.

If you *do* build it with a RasPi, publish the specs here, pleeeeeeeeze!

Throw a 1/4" x 3" shunt on the S1 wire to the motor and monitor the current AND voltage at the same time.

I bet JohnnieB'd have a field day with something like that...

I've been looking for something like that ever since I got a $12 Chinese OBD2 BlueTooth device and the Torque app for my Galaxy S4. I figured someone would eventually build a multimeter like that.

RLW

[birmingham]

Ok, So finally got all of the stuff together and made my new high amp wires for the cart, this made a difference (guessing that means that one or more of my cables were toast) Also the charging wires looked terrible, cant believe this was working.

My findings, Ill skip right to what i think is wrong, after about 2 miles on mostly flat ground i was still going along well, so i did some uphill, found my voltage to be dropping drastically under load, Using volt meter. So i hooked up to each battery individually This is what i found:
At Rest Under Load
Bat 1 6.1v 4.9v
Bat 2 6.09v 5.08v
Bat 3 6.07v 5.1v
Bat 4 6.07V 5.57v
Bat 5 6.03v 2.3v
Bat 6 6.2v 5.71v

Winner winner #5. Next question
1 im recharging them all now, is this safe to recharge with a "bad Battery"?
2 is it ok to replace just that battery with a cheaper new one?

i know this isnt best but i really dont want to replace them all now because i wanted to stage in my upgrades and go 48V so i want to replace the bad one get it running again then around xmas do the controller and solenoid, then when the batteries really die do 48v and the charger. At some point I want to make this thing look better to.

[birmingham]

Sort of off subject but im assuming that the 4 wires hooked to the motor,

2 big ones feed it juice
2 smaller ones work with the "regen?"

Is this correct?