Thursday, January 28, 2016

Stuffing is the hardest part

After rebuilding the Grumman battery tray rollers with technical help from the capable countermen at Motion Industries, we discovered that though the tray could hold 14 batteries with a few rows set on end, two of those rows would just barely interfere with a floor beam, and the cabling gets difficult any time you set batteries on end.  Also, the manufacturer may allow it, but they're never quite as comfortable with any installation where the batteries don't sit on their base as designed.
That and other factors is pushing us towards a simpler pack layout with ten in the main tray, and ten in the forward body cavities.  Those two parallel strings at 120V nominal will still provide over 30 kWh of capacity, yet be a very safe voltage for student mechanics to work with.

BUT, when adult entertainment is desired, a simple switch could unite the two strings in a series circuit of 240V for higher performance.

Also in the interest of speed and simplicity, I'm leaning towards getting the van rolling with a single 10-battery pack to start off with, and then add more once the bugs are worked out of EVerything else.  placing batteries is tough work, especially when safety is job one!

The charging inlet has been changed from a jury-rigged industrial plug that was placing 12 VDC on two pins and 144 VDC on the other two, all in close-proximity exposed blades, to the modern J1772 standard.  The original inlet mounting plate was reused, so of course a billet aluminum inlet from Modular EV Power was used to match!

Tuesday, January 19, 2016

Stuffing the KurbWatt

The original Grumman battery tray will hold 11 Group 27 batteries after removing some of the center partitions Yes, that weakens it somewhat, but in the long and strong direction, but we reinforced the floor and have reduced battery weight by over one third.  We've also found another battery donor, one of the few local users of Valence batteries, for a total of 24 U-Charge XP27 batteries available, which adds up to 39.9 kWh!   This would give the truck a 39.9 kWh pack weighing only 1056 lbs.

 Three batteries will fit in an unused body cavities ahead of the rear wheels on the passenger side, but only two behind the driver due to cable routing.  The seat platforms are nice boxes that will also hold two batteries apiece.

 Add up to five batteries in the motor compartment, and the count stands at 24.   That should be good for up to 200 miles at low speed around town, or no less than 114 miles on the highway!

Battery placement is still being decided, and we may use fewer batteries to KISS, but we also have two Valence U-Charge BMS systems, each originally set up for 12 batteries.  Running two packs in parallel would extract the maximum energy at minimum stress, while operating at safer voltages (144V nominal rather than 288V).  Power delivered by the controller would be greater at 288V, because then the motor could be fed with 170V, which the maximum voltage for decent brush life, and not sag below that 170V during high current (1000 amp) accelerations.  That 170 kW would be 240 instantaneous horsepower, probably more than the Rabbit clutch could tolerate!

144V or EVen 132V is our likely course since students are involved, but it leaves an option open for putting the sub-packs in series when daddy goes to the track!

Tuesday, January 05, 2016

The Grumman Gets Gutted...

Though we were able to get the Grumman Kurbwatt running in very short order, it was due for an all-around update, so as soon as all systems were go and the brakes were road-tested, it was time to take the electric parts apart.

The battery pack is being replaced with donated set of used first generation Valence U-Charge Lithium Ion modules, which match the outgoing Group 27 lead acid batteries.  Our plans are to replace the controller with a Zilla 1k, and the Lester "boat anchor" battery charger with a Manzanita Micro PFC20, complete with full J1772 interoperability for public charging stations.

The Valence cells are probably a half-measure, being of uncertain capacity and remaining life, but will enable extensive road testing and tuning of all the other upgrades before building the ultimate battery pack.

Sure enough, once we managed to pull the control board (with the controller, main contactor and auxiliary relays mounted) it became apparent that there were about twice as many wires as necessary, and every time we opened up a bundle of electrical tape, we found either an ugly splice or abandoned, dead-end wires.  Plan A is to use a multi-pin Amphenol connector to have a single disconnect point on the new board for all small-gauge wiring, enabling easy removal of the board, along with a couple of Anderson connectors for the big stuff.

Charge On!