Some calculations related to battery temperature and cooling
An IC member reported a 92% cycle efficiency while charging and discharging an IMA battery. This applies below the threshold of overcharging, where charging efficiency drops to zero. This prompted me to do a simple calculation:
If you model the IMA battery as 22kg of nickel, it has a specific heat of 0.445kJ/(Kg*°C) = 10 kJ/°C. Adding 1kWh of charge at 96% efficency dissipates 0.04kWh, or 144kJ. If the battery is insulated (or has no fan and the charging happens quickly), its temperature rises 14°C. If the entire pack is uniform, no cells peak early, and the charge terminates at that time, you won't have a problem, and your fan never needs to kick on.
Suppose heat can be conducted very quickly throughout an entire stick of six cells. If one cell peaks early, and you're passing 350mA through it, it's dissipating 0.5W. The temperature of the stick will rise by 3.6°C per hour. Once all six cells in the stick peak, it's 22°C per hour. Obviously, this spells a serious problem if the fan is not running.
What this means: If the pack doesn't cool while you're charging it, you'll be able to start with a pack at >9°C on a -5°C day. It's not terribly impressive, but it's better than nothing.
Also: once the pack peaks, you will begin to do major damage to the cells after about four hours if you're not running a fan.
An IC member reported a 92% cycle efficiency while charging and discharging an IMA battery. This applies below the threshold of overcharging, where charging efficiency drops to zero. This prompted me to do a simple calculation:
If you model the IMA battery as 22kg of nickel, it has a specific heat of 0.445kJ/(Kg*°C) = 10 kJ/°C. Adding 1kWh of charge at 96% efficency dissipates 0.04kWh, or 144kJ. If the battery is insulated (or has no fan and the charging happens quickly), its temperature rises 14°C. If the entire pack is uniform, no cells peak early, and the charge terminates at that time, you won't have a problem, and your fan never needs to kick on.
Suppose heat can be conducted very quickly throughout an entire stick of six cells. If one cell peaks early, and you're passing 350mA through it, it's dissipating 0.5W. The temperature of the stick will rise by 3.6°C per hour. Once all six cells in the stick peak, it's 22°C per hour. Obviously, this spells a serious problem if the fan is not running.
What this means: If the pack doesn't cool while you're charging it, you'll be able to start with a pack at >9°C on a -5°C day. It's not terribly impressive, but it's better than nothing.
Also: once the pack peaks, you will begin to do major damage to the cells after about four hours if you're not running a fan.
I'm not sure about your math. 3.6C * 6 cells is still 3.6C over a larger area. The question is how much of that heat will dissipate naturally.
I think the fan kicks in at 40C.
__________________
2001 MT #284 100K miles salvage vehicle restored
MIMA #70P + FAS + TPS mod
ScanGauge II, Euro seat covers (citrus)
I'm not sure about your math. 3.6C * 6 cells is still 3.6C over a larger area. The question is how much of that heat will dissipate naturally.
I think the fan kicks in at 40C.
A single cell that peaks early will generate enough heat to raise its temperature 22°C/hr. But my assumption here was that there's a good amount of heat conduction throughout a stick on the timescale of interest, so you could model the entire stick as a single thermal mass that rises at one sixth that rate.
Quote:
Originally Posted by JimIsbell
Since running the fan will cost less than $3 and take less than 10 minutes to implement in a charger, why not?
Well, that depends on where you live, how you use the charger, and what your goals are.
If you live in the north, then your battery is colder than you like for 4-6mo out of the year. Charge and discharge performance drops off pretty badly as the temperature falls. Below 0°C out, sometimes regenerative braking is too weak to bother with.
If you're using the Insight as a PHEV in the flatlands, where your usage profile is "charge at home, discharge once" with little regen, you'll be interested to know the battery temperature rise will be minimal and the fan will probably never come on.
There's also the small fact that the battery fan draws 0.6A * 12V = 7W. Leaving that on for eight hours a day unnecessarily is like... leaving a light bulb on for four hours a night. I'm the kind of person who will look for an alternative to a 7W load if it's convenient and cheap.
If you're using the Insight as a PHEV in the flatlands, where your usage profile is "charge at home, discharge once" with little regen, you'll be interested to know the battery temperature rise will be minimal and the fan will probably never come on.
This is my experience, my entire commute is flat enough to where the only regen that is useful is at stop signs and stop lights, I don't have enough stop signs and stop lights in my commute to recover the amount of power I use with my headlights on the way to work. Outside of the 4 month period of the summer where it is light out when I am leaving to drive home from work in the morning, 100% of my regen opportunities are put towards recovering the DC-DC load and I was pretty much always at 18 or 19 bars this winter/spring.
I have only once had the fan come on, that was when I deliberately cycled the battery down with assist and all the way back up again within the same trip. It charged up faster than I thought it would, the difference between 5 bars and 10 is about the same time in regen as it is 18 to 20 for me.
I won't know whether or not I need the fan while grid charging but if I don't have it, you bet there will be some form of temperature disconnect worked in there but I think I'd rather just run the fan as it is much easier to do and won't cost me pack life if I make a mistake somewhere else and induce some overcharging.
I'm sorry, but these calculations are not bourne out by my experience.
A pack left to cycle at 10 amps with the fan accidentally left off will reach an unhealthy 130 degrees farenheit or so (I haven't made that mistake in more than a year), but I've left full packs soaking on a 0.5A trickle for up to six one-hour periods in a row with the fan on and there has been no discernable heat rise in the air being exhausted. My chargers run to delta V and then kick over to trickle for an hour. 20 chargers at once make the air perhaps 1 degree above ambient.
Quote:
Originally Posted by RobertSmalls
A single cell that peaks early will generate enough heat to raise its temperature 22°C/hr. But my assumption here was that there's a good amount of heat conduction throughout a stick on the timescale of interest, so you could model the entire stick as a single thermal mass that rises at one sixth that rate.
Well, that depends on where you live, how you use the charger, and what your goals are.
If you live in the north, then your battery is colder than you like for 4-6mo out of the year. Charge and discharge performance drops off pretty badly as the temperature falls. Below 0°C out, sometimes regenerative braking is too weak to bother with.
If you're using the Insight as a PHEV in the flatlands, where your usage profile is "charge at home, discharge once" with little regen, you'll be interested to know the battery temperature rise will be minimal and the fan will probably never come on.
There's also the small fact that the battery fan draws 0.6A * 12V = 7W. Leaving that on for eight hours a day unnecessarily is like... leaving a light bulb on for four hours a night. I'm the kind of person who will look for an alternative to a 7W load if it's convenient and cheap.
__________________
2001 MT #284 100K miles salvage vehicle restored
MIMA #70P + FAS + TPS mod
ScanGauge II, Euro seat covers (citrus)
I dont think battery efficiency is anywhere near that personally.
Next issue is that NiMH uses around 6kg of Nickel per kWh so thats alot less, I guess its the film and electrolyte along with the metal housing that form your main heat sinks.
Uniform temperature distribution is not going to happen either.
Natural convection relying on the bouyoncy of air is a good cooling method on many things, gues it doesnt help a big deal on the insight though as the cells are all higher than any of the air ducts to the IPU!
Personally I dont see why fan control with a grid charge is discussed I think it should be included?
I'm sorry, but these calculations are not bourne out by my experience.
A pack left to cycle at 10 amps with the fan accidentally left off will reach an unhealthy 130 degrees farenheit or so (I haven't made that mistake in more than a year), but I've left full packs soaking on a 0.5A trickle for up to six one-hour periods in a row with the fan on and there has been no discernable heat rise in the air being exhausted. My chargers run to delta V and then kick over to trickle for an hour. 20 chargers at once make the air perhaps 1 degree above ambient.
Fair enough, and thanks for the input. The biggest limitation of the above calculation is that it guesses at a lot of the numbers, and I have no ability to verify it experimentally.
Packs soaking on a 0.5A trickle charge: >90% of the energy going in is converted to heat, right? They should be giving off >80W each at equillibrium, which is a moderate amount. My computer (built for efficiency) gives off 60W, and I can sometimes tell the difference in ambient temperature with it on vs. off.
Quote:
Originally Posted by jonnyvtec
I dont think battery efficiency is anywhere near that personally.
Next issue is that NiMH uses around 6kg of Nickel per kWh so thats alot less, I guess its the film and electrolyte along with the metal housing that form your main heat sinks.
Uniform temperature distribution is not going to happen either.
Natural convection relying on the bouyoncy of air is a good cooling method on many things, gues it doesnt help a big deal on the insight though as the cells are all higher than any of the air ducts to the IPU!
Personally I dont see why fan control with a grid charge is discussed I think it should be included?
There are only two fan control schemes worth considering. Fan always on is one. It's easy, foolproof, safe, and cheap. The other is temperature-based fan control. It's less of all the above things. But it allows you to start with a warmer pack, which I expect will be nice come wintertime. I'm doing it, but I wouldn't recommend it to a friend.
Regarding the uniform temperature distribution assumption: a "lumped capacitance" thermal analysis is valid if heat is generated and removed much slower than it's conducted through the body. I think that's fair.
Anyway, don't place too much stake in the analysis; it's got its flaws.
E27006, Most NiMh chargers implement a good deal of overcharge though, I've personally watched mine and monitored each of the 6 cells both for negative delta voltage and temperature gain. My charger put in nearly 1Ah extra after all of the cells peaked and were dropping in voltage and temperature was gaining 2 degrees C every minute at 1C with the charge rate determined by the measured discharge capacity. If I let those same cells that just finished the charge at about 45C rest until they are cooled to room temperature 4 hours later and restart another charge it will burn off over 1Ah before terminating. Granted the delta V is adjustable, I hadn't adjusted it because I knew that my packs weren't balanced and the extra power going in balances those sticks for me and I didn't mind as the sticks that I have aren't suitable for use in the car anyway, each for their own different reasons.
It's hard to measure efficiency when this sort of equipment is designed to overcharge to correct balance issues and ensure a full charge every time. When I've stopped the charge at the point where I thought it was full and then discharged, I got better than 92% efficiency between the charge and discharge, in one specific example from a discharge to 5.4volts, I put in 6504 at 7amps and took out 6186, which is just a hair over 95%. The car very specifically stays away from the top and the bottom for a good reason. The top 5% or so leaves you with a fair amount of energy spent with heat as the result. The bottom and you risk reversing cells and when there is less energy leftover in the cells and you continue to pull high amperage from them, they tend to shed more heat as they just don't have the same availability of power leftover so efficiency takes a hit there, especially when the voltage drops down and you do the math of volts*amps=watts The current you continue to pull gives you less watts but still pulls down your state of charge as your voltage continues to sag lower.
Efficiency should match what the car is doing but that is hard to measure. It would take some sort of test like taking the known capacity, emptying the cell, charging it with 80% of its measured discharge capacity, discharging it 80% and repeating the process until the battery is measured at empty. This would take time and probably has enough flaws to make it not worth the effort though.
The AutoGuide.com network consists of the largest network of enthusiast-owned enthusiast-operated automotive communities.
AutoGuide.com provides the latest car reviews, auto show coverage, new car prices, and automotive news. The AutoGuide network operates more than 100 automotive forums where our users consult peers for shopping information and advice, and share opinions as a community.
Some calculations related to battery temperature and cooling
