I have been eager to add some sort of charge control to my 1985 VW Cabrio conversion (a Mike Brown kit, 8" ADC motor, Curtis controller and many other improvements over the years). I took this car in the Tour de Sol in 1995 and 1997. the first time I was running flooded Trojans, and the second time I had two paralleled strings of Optima YT's in the car and a small trailer with flooded 12 volt Trojans. I confess here and now that I must be a serial battery murderer.
When I first installed the Optima YT's I put in an early vintage of Rudman regulators. These seemed to do the trick but were very temperature sensitive and I was constantly having to adjust the trim pots at the change of every season with our vastly different temperatures here in the Northeast. I was always curious about the state of charge at different points and trying to figure out which was the weakest batt. Shy of crawling around the batt boxes with a meter, this was tough. I wanted a management system I had a lot of control over. Hopefully some advanced battery chemistry will be in my future, and so something adaptable would be required. Perhaps higher voltages than my 120vdc are in my future, so something extensible would be nice. I was eager to tie the charge management system into my two transformer based chargers. For instance, starting with both at 220vac input and ending with one at 110vac input to taper the charge current. Being able to turn on heaters in the winter and battery box fans in the summer would also be a plus. In looking at the Smoother and other commercially available systems, they didn't seem to have this flexibility. They also cost a heck of a lot more. From reading the EVDL I was aware of Lee's system and decided to buy the bare boards from him and try my hand at assembling and installing them. While I have done some electronics work in the past I am most certainly not an EE. Unfortunately I proved to be the poster boy of what Lee was fearfully predicting when he sold a bunch of Balancer boards and threw this nifty design into the public realm. Enter Roger Stockton (another Balancer purchaser) who along with Lee painstakingly took me thru a correspondence couse in electronics as I got this thing up and running. Hopefully some of my pain will spare future builders some grief. Roger has done an excellent job of presenting the Balancer on a website: > http://www.geocities.com/sorefeets/balancerland/ My installation has yet to make it into enclosures and into the car; and presently lives in the garage and is only used on (and after) charge. Lee apparently uses his on both charge and discharge. I plug the car into the balancer with a multiconductor heavy guage connector when the charger gets plugged in. At present, my system is using a fraction of the available Balancer flexibility. Each batteries voltage is measured and written into scratchpad RAM. The lowest battery gets charged from the on-balancer-Vicor for two minutes, and then the cycle is repeated. The cumulative amount of time each battery spends "being balanced", i.e. getting a sip from the Vicor, is also tracked. Note that the Balancer can be on in standalone mode or doing its thing which the main charger is going. The balancer can also do its thing fed by the main pack voltage and with no AC connection, allowing balancing-while-driving or balancing-while-parked-with-no-AC in sight. At present my display and actual data after a couple of hours of balancer activity looks like the following: Batt # Voltage Cumulative Balancer time (multiply by 2 to get minutes) 01 12.64 2 02 12.71 0 03 12.72 0 04 12.69 1 05 12.60 3 06 12.64 3 07 12.61 20 08 12.51 2 09 12.51 14 10 12.55 23 house batt 12.26 9 On my near term "to do list", are the following. Use one of the onboard relays to turn off the charger(s) when a trip voltage is hit. Wire up the two thermistors provided for on the board. Wire up the battery heaters I have ordered. Figure out a good cooling fan configuration for front and rear batt boxes. Develop an LCD keypad and screen to plug into the onboard console port (presently my laptop is talking to the onboard Basic STAMP in Debug mode). Implement Lee's magic number. This is detailed on Rogers website and is an algorithm for determining how long to charge the lowest battery. I am calculating and tracking it but my near term goal has been to match the batts in the pack, weed out the dying soldiers and get the thing balanced in the first place. Wish List: While Lee's design is excellent and very versatile, I hope we can add some things to future spins of the boards: - ability to handle longer strings. 15 batts is the present max although Roger is grafting some magic onto his board to handle longer strings. - socket the board for the larger Basic Stamp, the 2P40. I am using the Basic Stamp 2p which has more memory and a bit more flexibility. The 2p40 would allow even more versatility. - add current limiting resistors for the LED's, so their brightness can be decided by the builder. I found that an "LED dongle" which plugged into the DB9 connecting up the relay boards, was vital for diagnostics and installation. I'd like to see LED's added to the relay boards and these could be selected or disabled by jumpers (no need to have 'em blinking after things are installed and debugged. Folks balancing while driving or sitting with no AC connection won't want the drain.) Impressions: This has been a much longer and more painful project than I thought but has been well worth the effort. Roger and Lee have been superb help and always right there with a tip, a suggestion, a workaround, and lots of encouragement. (I think I startled Lee early on by emailing him to ask which end of the soldering iron I should hold; the end I had been hanging onto was getting really hot.) This gizmo is really going to improve the state of discussion and control over battery management. Perhaps something like this could be combined with Rudman Regulators: the Balancer could control regs, disabling them for equalization charges, report on battery voltages and nurse along stragglers with sips of current from the Vicor, while the regs could clamp voltages on batts getting charged first. There was some discussion on the EV tech list about how the balancer would perform with a battery that went into a runaway condition. There are lots of built in safetys. The Balancer uses an opto-isolated DMM which it speaks to thru a DB9 connection. There are fuses on each relay board, and fuses on the batt. Lee has rigged up a cap and resistor which only allows one relay to be energized at a time. Thermistors can look at overtemp conditions. I suppose some unforeseen condition could arise, but I feel a lot better about charging with this device than without. Kudos to Lee for designing this and putting it in the public realm. Kudos to Roger and Lee for supporting boneheads like me to get this up and running. For those of you who are technically inclined and/or concerned about what's going on inside your pack, I heartily commend the Lee Hart Balancer!
