Big voltage drop!

[Electricdirtbiker]

I'm charging my a123, 1s 10p packs with a Web-Tronics 3.3v, 20A powersupply. So far so good. But I notice a huge voltage drop across the charging leads. The leads are about 1.5 ft, about 16 or 14 ga, they do get warm at first. The charge current tapers off long before the 3.6v battery voltage is reached.
I think heavier cables are the solution, but how heavy? Is 12 ga enough? Is it worth hacking the powersupply for the sense leads? This would make for a huge wiring mess plus connectors if I use many chargers for my pack.
Any ideas?
Thanks
Wes

 

[rkosiorek]

so you are taking a discharged cell, probably reading 2.7 or 2.8V with a current of 20A? and you are surprised that 14awg or 16awg wire get a bit hot?

when discharged the 10p cells will charge at whatever current your supply will allow. the voltage will rise from the discarged value as the cell is charged. at the same time the current will decrease. the voltage drop is not in the wire it is from the discharged cell. the wire gets hot because you a pushing too many amps through it.

for 20A charge current maybe you should try 10awg or maybe even 8awg. for a short length that should remain relatively cool. the voltage though will still sag to match the cells empty voltage. this will be at the constant current your charger can deliver. it will stay at this constant current until the voltage reaches the max voltage your charger is set set for. is that 3.3V or have you set it for something higher? once it reaches the max voltage the voltage will hold but the current will decrease.

that is what is meant by CC/CV charging. a charger is a powersupply so i use the term interchangeably here.

rick

 

[Electricdirtbiker]

The powersupply is set for 3.6v. The voltage sag, or voltage drop is from the light weight leads, checked with a meter.
I could use some ideas on how to deal with this when the powersupply to the battery pack (in the bike) is about 2 1/2 ft away. I know shorter is better, but in use, this is the best trade off. I will try 8ga leads, but becomes a handfull with 16 batteries. Turning up the powersupply voltage can correct the voltage drop, but the battey may become overcharged as the current tapers down.
Thanks for the reply,

Wes

 

[tailwind]

Hi Wes,

I have one of these 3.3v 100w power supplies.
I only use it once per week for my balance charge.

When I first got the PS I ran a test to see what the charging profile was like.
The voltage doesn't hit 3.6v to well into the charging process.

http://endless-sphere.com/forums/viewtopic.php?f=14&t=4057&p=61227

You can see that it puts out 83w as a maximum.
My setup is a 4p 13s A123 pack.
It was going to be too much of a hassle to connect up 13x PS. Plus I would not be able to tell at a glance when the charge was finished.

During the week I do a bulk charge of the pack with a 350w 48v dumb power supply.
This allows me to charge at a higher rate.
I monitor the voltage on each of the 4p packs to make sure I dont over charge any of them.
The pack stays in good balance for the week.

Once per week I will charge the pack up to 90% capacity, I then remove the batteries from the bike, disconnect the serial connections and then connect my parallel wiring harness.
This make one 52p battery. I connect this up to the 3.6v PS with a WattsUp inline.
When its finished, put the batteries back in the bike and I am ready to go for the week.

Of coarse it is easier to use a BMS to limit the voltage of each cell when charging.

Greg

 

[Electricdirtbiker]

Greg,
Thanks for the charge plot and explanation on your system.
I have 4 of those 3.3v, 20a powersupplies and eventually 16 of them so no balancing needed. But I didn't fore see the size of cables needed.
Your bulk charging requires only 2 wires and a way to monitor each battery. Also easy to top them off inbetween rides. How do you monitor your batteries during charge and lvc?
Your way of balancing them is about as simple as you can get, just a little work. The way I will install my batteries, I'll have to balance them on the bike.
How are your cells holding up?
This was a question for a separate post. Where do you keep your batteries? In the house, garage, corner of the yard. I know the a123s are the safest, but with no actual induced short, is there any history of them cooking off on their own? This is of course with no over v, under v or shorts. The only thing different we are doing here compared to the tool packs, is paralleling the cells. Everyone is doing that with good results, even the Killacycle people.

Wes

 

[tailwind]

If you are just planning to use the 3.3v PS then your wiring will need to be able to handle
the 24 amps during the current limiting phase.

How do you monitor your batteries during charge and lvc?

Very manual method im afraid to say.
I think every one should have a BMS to protect their batteries, but I dont.

I typically use between 4 and 5Ah for my 24km commute. So lvc is not something I have to deal with on a daily basis.
One of the 4p batteries (#6) has slightly less capacity than the others. This is always the one that has the lowest voltage after use. As such it will be the one that reaches 2.0v first.
I did buy a small digital meter that I was going to use to monitor the voltage of #6 but I have not installed it. Instead I just use the cycle analyst to limit my AH used to 8Ah or less.
The bike is perfectly rideable up to 20 km/hr without any motor drag so I have never been stuck.


For charging I use a WattsUp inline with the PS and a DMM to probe each of the 4p.
The side of my battery box comes off and I am able probe and access each of the 4p seperately. Each of the 4p batteries has 4 powerpoles. Two for the serial connection and two for probing the voltage.

I typically charge in one hour. It is only in the last ten minutes that I really have to baby
the charger to keep it to around 3.6v per cell. Most are at 3.6 but some will be just over and under.
Because it is so easy for me to probe the voltages I have a very good knowledge of the health of the pack.

I did buy a couple of the "MKS 3in1 Balance/Discharger" to help with displaying the voltages
but I have not connected them up.

How are your cells holding up?

Great batteries. I only have a 20amp controller and I have only seen 25amp on the cycleanalyst so these cells are not even stressed.

Yes LiFePO4 is a great chemistry but I never charge the batteries unattended.
You need to test each of the A123 cells for capacity and if you can IR before assembling them into a pack. That way you can weed out the dud cells.
The metal can is live. The cardboard sleave acts as an insluation layer. I put an extra layer over the end where the + and - come into close contact. I also taped a rubber strip over the end of each 4p to act as another insulation layer and to stop the possibility of one group of cells rubbing through the cardboard layer of its neighbor.
Once per week I take the batteries out of the box to balance charge. This gives me the opportunity to inspect the cells for damage.


Greg

 

[Electricdirtbiker]

Greg,
Your right about all the cable handling. Sounds like you are very diligent about your battery maintance. I could make a list, but seems there is no way around all the manual cable handling. This is of course without the use automatic electronics.

a) Bulk charge, 2 med cables, careful manual monitor with a meter. Balance once a week, 2 light/med cables / battery, no real v monitoring. Easy to top off charge, requires mannual monitoring.

b) Balance charge per battery, no real mannual monitoring, 2 heavy cables / battery. Somewhat messy to top off charge. Need 1 charger per battery.

c) dc-dc converter charger / balancer / battery. Little v monitoring. Since each converter stays attached to battery, requires only 2 heavy cables to charge. Requires large bulk charger. Adds converter weight and space. Unknown current back flow and never fully disconnected. Easy to top off charge.

For me top off charging is useful. Any of the methods of couse depends on needs too. A system of both a and b is the most flexable. a, could be much improved with o.v. sensors that shuts off the bulk charger when a battery reaches 3.6v. The o.v. and lvc could be combined.

I'm considering the Chargery lvc monitors, but they don't think they have a o.v. for 3.6v.

I'd like to post a pic of my 1s 10p battery, but I don't know how to reduce it to fit, or post.
I solder all ten cell polarities to a copper strip on each end and reuse the black end caps. Its very tight, compact and good for 300-400amps. Stackable and phyically sturdy. No need to for reinspection.

Wes

 

[tailwind]

My lack of BMS is just an interim measure until I find one that I am happy with.

I am looking to have only the lvc on the bike and to have a seperate bms to handle the hvc when charging at home.

The Chargery unit that methods suggested with the selectable lvc and buzzer looks very good. Tiberius has a small lvc board and GGoodrum designed some as well.

With regard to the wiring up 16x PS. Have you considered just using the 4 PS that you have.
If you are able to disconnect the serial connections between the batteries then you can combine 4 packs in parallel and make up a 4 into 1 parallel harness. Each wire would only need to carry 1/4 of the current (6 amps).
The down side is that it might take up to 8 hours to charge.

Greg

 

[dnmun]

voltage sag on charging is irrelevant. let him put 40A through the 14G no problem.

when the cell finally collects some charge and the potential builds up the current will drop and eventually the charging current will drop.

the guy with the 48V power supply could still use a regular 12S BMS instead of 13 like he intends. he can adjust the power supply down to 44V or 3.66V per cell for 12S and he can use a normal 12S BMS

 

[Electricdirtbiker]

Greg,
My battery packs each have 2 copper strips for bolt through posts. So unbolting to perform balancing is a lot of work. The other option is solder on short leads with deans connectors for daisy chaining, simuliar to what you did.

Dnmun,
You right about voltage sag. At the end of the charge, as the current drops, so does the voltage sag, down to nil. This does stretch the charge time, but the alternitive is huge charge cables.

Does anyone know how to add o.v. cut off to Gary's lvc?

Thanks all.