Since I've gotten no reply from GreenSaver I pose these questions to the tinkerers on this forum about the (SP27-12) or (SP36-12):
When charging in constant amp mode what is the idea amps GreenSavers like to be charged at?
When charging in constant volt mode what is the ideal voltage formula (temp compensated)?
What is the ideal float voltage?
Can they be floated indefinitely with no detriment if the current is low enough?
-----
Goals (in no particular order): Keep batteries in a string balanced, fully charged, and never over charged for an indefinite time. Keep costs to a minimum. Make it small enough it can be stored inboard. Provide constant desulfation. Provide the longest life for these batteries as possible. Keep project open source, with socketed microcontroller for firmware upgrades. Temperature compensated charging. Negligible battery drain for long term unplugged storage. >5 year service life with daily use consisting of 8 hours of operation. Avoid patent infringement based lawsuits. Upgradable for management or more or less batteries / other chemistry.
----
Plan A: Build a pulsed constant current string charger with individual shunt voltage regulators on each battery.
How it works: A transformer (rewound microwave oven transformer?) that saturates (at a to be determined amperage) provides current limiting. The transformer's output gets full wave rectification then smoothed. Mosfets pulse the charging current at an optimum battery absorption rate (variable rate?). The batteries soak up the amps. The batteries one by one reaches their charging voltage ceiling and the shunts engage limiting the duration of the pulse the charger receives, then disengage as the battery voltage drops below the voltage ceiling (voltage ceiling automatically adjusts with a thermistor circuit, thermistor also stops process if an overheating condition is detected). Once the transformer reaches an upper voltage ceiling for a preset length of time it shuts off and waits for the self discharge of the batteries to fall below their fully charged resting voltage (temp compensated) and begins again with step one. This will keep the batteries topped off without ever over charging them and keeping costs to a minimum. A green light will illuminate saying it's fully charged. I believe this strategy will work with any sort of battery technology too, but haven't done extensive research into it.
Notes: I went with pulsed current, because my online research has shown it speeds charging of lead acid batteries significantly. The space between pulses allow the chemical reactions inside the cells to complete. Another benefit to is to it are thermal. Higher current can be put into the battery without causing the internal temperature and pressure to build (GreenSavers have been charged at 10A without much heating, but it's as of now uncertain if the silicon was insulating the heat to a large degree if only GreenSaver would answer e-mails). The shunts will be cooler too with time to dissipate the heat. If there is ever an small overcharging condition and an air bubble forms, (in AGM batteries at least) they are designed to recombine the gas during the drop from charging voltage to resting. The shunt's switches are controlled by a 14bit ADC monitoring each battery's terminal voltage. This will additionally allow for active monitoring of each battery so to shut down if there are any serious problems during charge, and all kinds of fun things like a cheap LCD to tell you the things PakTrakr can + whatever code people add to it.
Question: If shuts are used it will surely provide overcharge protection, but will it mean that other batteries get undercharged? According to USATracy the answer is yes, but I'm not convinced.
White Papers:
1) http://www.nrel.gov/vehiclesandfuels/energystorage/pdfs/evs_17paper.pdf - Charging Algorithms for Increasing Lead Acid Battery Cycle Life for Electric Vehicles
2) http://www.ipenz.org.nz/ipenz/publications/transactions/Transactions98/emch/2wilkinson.PDF - A new pulse charging methodology for lead acid batteries
----
Plan B:
Use stock string charger and use 1 giant ultra capacitor with a switching algorithm to take power from the strongest battery and put it into the weakest. Monitor each battery voltage with 14 bit ADC to find the strongest and weakest battery. Shut off the charger if an overcharging occurs, turn it back on once batteries are better equalized but not yet charged (voltage compensated). Place capacitor in parallel with controller power input for boost in acceleration. Display battery information output on LCD.
Benefits: More simple design. May extend the range since it's limited by your weakest battery.
Drawbacks: Possible lawsuits due to patent infringement. Loss of 10% of the energy you move from one battery to the next. No desulfation.
----
Plan C:
Build the charger from plan A, and the single capacitor equalizer from B.
Note: Shunts are cheaper than capacitor equalizers ~$8 vs ~$20 per battery respectively
This post took way too long to type. 5hrs. It's the result of my first week of obsessively researching this stuff.
The chip I'll use to avoid lawsuits
The chip I'll use to avoid lawsuits
The chip I'll use to avoid lawsuits
The chip I'll use to avoid lawsuits
The chip I'll use to avoid lawsuits
Plan D: Use five "off the shelf" chargers.
Is there a reason you don't include this option? I use the Vector 2/6/10 Amp "smart charger". By the time you strip off all the extra packaging you can get all five under the seat:
I know that not all five chargers will produce a perfect balance - but how important is it to keep the batteries balanced if they're being charged independently? Isn't it the fact that the initial small differences quickly increase over several cycles into large differences when string charged? However, if bank charged the initial small difference remain the same.
This solution doesn't address desulfation - but I think it addresses everything else and is reasonably cost effective and straightforward.
John H. Founder of Current Motor Company - opinions on this site belong to me; not to my employer
Remember: " 'lectric for local. diesel for distance" - JTH, Amp Bros || "No Gas.
Hi John,
Beautiful job on the charger setup. I was not going with plan D because I'd like to really pamper these batteries, and not spend more then $100 to do it. Once you add a Bank Charger, PakTraker, BattEQ, you're down a significant lump of change. I wanted a charger that would give the greensavers their ideal voltages, amps, and adjust it with the ambient temp to keep them happy. Sulfation is a legitimate problem, so I wanted a charger that addresses that too. Mainly John, these projects are a pure joy for me.
When I began my research I just wanted the BattEQ circuit so I didn't have to pay the monopoly markup. I found it here http://www.electric-lemon.com/?q=node/151 Some people reported longer range with the BattEQ, so I figured it wouldn't be much more work to have it communicate with a charger. Kill the overcharging problem and battery equalization (for range) birds with one stone, but I worried about patent issues. So plan A was formed.
If your batteries are a little out of balance it doesn't matter as you run them down, just good to know where your lowest battery is so you know when to turn around.
XM-3000...
-DC-DC converter replaced with a Dell D220P-01 power supply.
-72V mod
-Expensive bank charger until I come up with something better... Still trying.
-
Hi Sparky.. this stuff can get addictive, can't it? :)
It sounds like you've got some good ideas, & the knowledge to carry them through. It is good to get all the real world info you can.. it sometimes differs from the formulas!
I wouldn't worry too much about patent law infringment. You can make whatever you want.. & if you decide to market a new, improved version of something already out there, hire a lawyer for an hour for counsel. But your creative juices & problem solving will necessarily be built on others' work.. that's how everything is invented or created.
The kind of info pioneers (like you!) in a growing industry come up with causes the product to improve, cost less, & become more useful to the general population. I think battery problems are the main reason electric vehicles aren't more common. Too many things can go wrong. A single cell goes bad. A charger malfunctions, either not charging a battery, or frying it. A bad connection somewhere. The power source for electric vehicles needs to be more 'plug & play'. Us tinkerers don't mind watching over our little monsters, but that's too much to ask for the open market.
I was under the impression that the silicone batteries liked higher amps, or could tolerate it more at charging. I've been bank charging mine at 6, & they have done well.. though i'm only at a few cycles. I'm thinking of varying the charging just to keep the batteries from getting bored.. ;)
The float question is a good one.. i tend to not want any current running after they're charged, so turn everything off. I seldom go more than a day or so without running it, so dormant draining isn't much of an issue.
Good luck with your project, & keep posting your thoughts!
scotty, Sedona, Az
xm-3000 60v, 38ah silicones, since 7/08
quazar 48v, 12x12ah, since '05
5- vector 2/6/10 bank chargers
I hear ya! Kind of thought that might be the answer.
Can someone confirm my thinking - if the batteries are being charged independently then a small voltage difference won't be too bad (because the difference won't get increasingly larger over time). Is that true or not?
sparc5 - definitely keep us posted on your progress - a good DIY alternative is always a great addition.
John H. Founder of Current Motor Company - opinions on this site belong to me; not to my employer
Remember: " 'lectric for local. diesel for distance" - JTH, Amp Bros || "No Gas.
John,
Got your answer. The voltage differences aren't a problem (in my world of theory at least), other then watching out to not deeply discharge one, (or overvolting and gassing the battery) When batteries are placed in series everything follows Kirchoff's laws. Inductance adds, capacitance divides, V+'s add.
I've consulted with an electrical engineer friend who recently made a battery charger, according to him, it's ok to float the batteries indefinitely as long as the current is negligible. This opinion is jives with what the batteryminder (quality charger) people write in their faq http://www.vdcelectronics.com/faq.htm They say the important thing is to make sure the float voltage is appropriate for your battery for floats longer then 2-3 months at a time. It doesn't go into why, but they are a fairly credible source.
To everyone with GreenSaver batteries, I have a request for you. Charge them fully, unplug the charger and let them sit for 8 hours or more then write to me these 4 things:
Model of GreenSaver
Temperature of the room
Voltage of each battery
Today I was thinking about one property of the battery: The quicker the discharge, the lower the capacity due to the Peukert Effect and internal resistance. None of my plans address this. I don't know if the controller takes energy from the battery in pulses, or if it's smoothed. My hypothesis is if you're driving slow, say .10 duty the capacitor could give you ~15% more range. If it's .5 duty more like 4%
First experiment: 10 camera flash capacitors placed in parallel with the battery string = 800 microfarad @ 67V.
Hypothesis: The capacitors will give the controller a boost with hard acceleration.
Results: No noticeable impact positive or negative.
Discussion: Energy=0.5*C*V^2 0.5*0.0008F*67V^2=1.7956J
Conclusion: Need bigger caps for EV but the array makes for a powerful Gauss Gun.
Bike Shop of the future.
XM-3000...
-DC-DC converter replaced with a Dell D220P-01 power supply.
-72V mod
-Expensive bank charger until I come up with something better... Still trying.
-
sparc5 - thanks for the answer and thanks for the picture - I'm glad to see someone's garage is as tidy as mine! ;-)
Is that an oscilloscope next to the bike? A friend of mine just "gave me on permanent loan" his old 'scope (he's the sort of guy who has an old 'scope lying around! He's the "mad scientist" doing the WFC experiments with me). Unfortunately when I powered it up I can't get the beam to focus down - the "dot" is about 5mm wide! Any ideas?
John H. Founder of Current Motor Company - opinions on this site belong to me; not to my employer
Remember: " 'lectric for local. diesel for distance" - JTH, Amp Bros || "No Gas.
Yeah, it's a oscilloscope and next to that is a frequency counter.
Cool friend, maybe he can tell you how to fix the beam. As for the dot, it should only be a dot if you have the freq set to one of it's lowest settings, otherwise it should be a line. Connect it to something like a power supply or battery so the input isn't floating. My scope has a focus knob, which you adjust after turning down the intensity so the dot is dim, that should narrow the line or dot to something tiny. That would be what I'd try first, after that just goof around with it until you find what was wrong. If it has two channels, maybe you try the other, try the various trigger modes, zoom, etc..
Comon people give me some battery info (see two posts up).
XM-3000...
-DC-DC converter replaced with a Dell D220P-01 power supply.
-72V mod
-Expensive bank charger until I come up with something better... Still trying.
-
sparks-
i'm assuming your capacitors aren't overvolting, just providing more amps at startup.. in that case, the controller would handle it & tone it down.. iow, there would be no noticeable difference. That's the controllers job. Even if you had 112 ah batteries, the controller would only see the 60v, & limit the current to the motor.
When you overvolt, then you get more out of the motor, because the controller limits the current, & the higher voltage gives you more watts.
I'm working on a log & will post some battery specs, volts, charging time, etc, on my greensavers. Real life gets in the way of these things sometimes...
scotty, Sedona, Az
xm-3000 60v, 38ah silicones, since 7/08
quazar 48v, 12x12ah, since '05
5- vector 2/6/10 bank chargers
Thanks Scotty you've always been helpful and encouraging.
I was using the capacitors in parallel with the string to provide more amps at start up. I'm not ready to try overvolting the controller yet. I was curious if the batteries max amp output was below the motor controller's max amp input. If true, capacitors could act as a turbocharger for our scooter.
Today I looked at one battery with my scope. I'm making the observations in real time, take everything I say in the video with a handful of salt. :)
I provide evidence that a capacitor array could be helpful at extending range on the scooter by lowering the peak rates energy is drawn from the battery. I'm still not sure the motor controller takes it's energy out in pulses or if it's smooth, what shows up on the scope could just be switching noise.
Note: Each vertical line on the grid represents 2V, each horizontal line is 10uS. I apologize for the blurriness and shakiness.
XM-3000...
-DC-DC converter replaced with a Dell D220P-01 power supply.
-72V mod
-Expensive bank charger until I come up with something better... Still trying.
-