2859
1981 Jet Electrica with Generator Trailer
4/26/09

    PURPOSE:

    • a hands-on hybrid EV test bed
    • direct operating experience the primary goal
    • minimum $ cost
    • quick results (no component development)
    • not a permanent vehicle
    • not optimized for anyting
    • basis for future permanent vehicle projects

    FEATURES:

      • original GE? motor an 4 speed manual transmission
      • battery pack of 10 Costco Deep Cycle Marine 12 volts
      • auxilary battery same as traction batteries
      • battery pack chager manual variac to bridge rectifier 115 VAC
      • auxillary charger is a 3 amp laptop AC adapter on main pack
      • 15KW gasoline air cooled generator
      • 240 to 120 volt transformer to get to battery pack voltage

    CONDITION: 

    1981 vintage basically original Jet Electrica with minor modifications.  Exterior dents, wrinkles and dings. Original paint and interior.  Seats and exposed surfaces show the age.  Battery boxes seriously corroded.  Original tires?  ~15,000 original miles?  About 5,000 miles since brought back to life in 2007.

    SURPRISES:
    1) Batteries don't like to absorb the capacity of the genset
    1.5) Batteries have high resistance--voltage really drops to get rated current.
    2) It takes about 50% more energy for ~50% increase in load
    3) Road bounce floods a float carb on a genset.


    THINGS TO DO:
    1) Add low side AC voltage and current meters
    2) Check battery discharge voltage curves with independent meters
    3) Characterize trnasformer input voltage versus idle current
    4) Find better batteries
    5) install connectors for data & control lines so trialer can be disconnected
    6) Obtain fuel burn versus KWH data
    7) Install a computer type data collecting system
    8) Make a force versus speed curve for the car (and same for trailer)
    9) Find a way to eliminate the transformer
    10) Get scope images of various point in they system
    11) Do some calibration of the instrumentation
    12) Generate wiring diagrams

Links:
  • under construction

    •  

    This is an experimental test-bed setup for obtaining some real world data.  I would discourage anyone from trying the same thing at this point.  Intial performance data looks mixed.

    I am currently running this as an electric traction only plug-in hybrid.  What that means is I charge from the wall then drive until the weakest battery gets to a resting voltage just above 10.5 volts.  This gets me about half to 2/3rds of one leg of my daily commute. Then I fire up the generator.  I will idle or turn off the generator if we get stuck in stop and go traffic or turn it on if going up hill or needing more acceleration or speed.  The commute is about 35 miles round trip each day.  

    We have gone about 500 miles todate with this configuration. We currently run a driver and 2 passengers.  Speeds range between 30 and 50 (except where limited by traffic).  Current average KWH per mile is .42 (wall and genset power combined).

    The generator puts out nominally 240 volts.  In order to match the 120 volt battery pack, we are running through a 15KVA transformer and then a rectifier and then parallel with the controller across the battery pack.  We can start and stop the generator while driving, as well as disconnect it electrically which drops it into an idle mode (when no current is drawn).   (We don't have a remote choke so we need to start it to warm it up a bit when cold.) We intercepted the voltage regulator loop of the generator and feed it through a small variac on the dash to give us some control over the generator output voltage (the original idea was we would need this to prevent stalling the generator under heavy acceleration--but this has not been needed to date).
    It looks like it takes 50% more energy to move the arrangement down the road at the same speed as it did with just the car.  Having added only two small wells and less than 50% more weight, it seems about right except that the running currents are about twice what they were  (70 to 80 amps at 35 mph now versus 35 to 40 before).  At such slow speed, wind resistance should not be an issue.  Only other thing I can think of right now would be trailer wheel alignment--tires are not obiously scuffing--will have to try to figure out a way to check that.  Having refilled just once, it looks like the generator gets terrible gas mileage.  Even though it is providing just a third of the KWh used, and is running less than half the time, it is burning about a gallon every 20 miles.  (Obviously, more data needs to be gathered and the calculations reviewed.)

    The obvious problem at this point is the battery pack.  The 10 12-volt 115 amp-hour marine deep-cycle and starting batteries can dish out a fair amount of current (albeit with considerable voltage drop) but they simply don't want to charge any faster than about 30 amps.  I have about 120 amps continous available from the generator but when I let go of the throttle coasting to a stop or when stopped, the batteries only draw about 30 amps.  While the generator governor does back the engine off, I am still wasting the potential of 90 amps of recharging.  On accelerating away from a stop, I can see peak current draw from the generator reach 140 amps (about 77 amps on the 220 volt side--so I am loosing some in the transformer).  (At sea level, I should be able to see a peak current of about 90 amps but at 5000 feet, the engine just does not have that much power).



    The trailer and genset add about 50% more tire and weight. The generator wet weighs about 500 pounds and is supposed to be about 30 hp.  The transformer is about 200 pounds.  The trailer and deck is about 300 pounds.  Additional electricals are about 100 pounds.  The 15 KW generator size was picked as it would have provided about the right continous power for traveling 60 to 65 miles per hour with the current draw of the car without the trailer.  The reason for the unexpected current draw with the trailer is not yet determined.  The current suspect is toe-in of the trailer alignment.

    The exhaust of the genset was getting everything around it very black indicating serious over fueling or rich mixture.  We took it to a small engine shop where it was determined that it was just fine sitting  still.  However, if you jostled the trailer simulating what happens driving down the road (even though it does have springs), the engine would over-fuel badly from gasoline sloshing out of the carberator and into the intake manifold.  The current attempted solution is to reset the float level as low as possible without starving it.  This is facilitated by having positive pressure fuel delivery.  It suggests that going to throttle body fuel injection might be desireable.




    The image below is of the contents of the metal box above.  On the right are a pair of bridge rectifiers on heatsinks with fans to the left of the heatsinks.  These retifiers are currently hooked in parallel.  The Fuji rectifiers are rated 800 volts 100 amps each.

    In the center are buss fuses for the signal wires.  both sides of signal lines are generally fused this was.

    Two the left are a pair of 200 amp fuses.  Between the fuses is the DC contactor that is remotely switched in an out to connect the genset to the battery pack in the car.


      

    The image above shows the label on one of the bridge rectifiers.
    The image below is the taped up current shunt on the DC lead to the battery pack that gives the DC current from the genset.

     


    There are 4 basic operating modes:
    1) driving on battery only.
    2) accelerating with genset running--all genset power including surge capability going to the motor and supplemental power as needed coming from the battery.  The acceleration period where genset surge capability is used is very short--a minute or less.  Other high current demand--climbing a hill--needs to be limited to continuous power rating.
    3) running with genset on such that genset is supplying running power and excess continuos rating power is recharging the battery pack.
    4) genset still running when motor demand is zero (coasting and stopped)--maximum continous genset power going into recharging batteries until an "on the road" full charge state (50%?) is reached at which point the generator is shut down.*

    * With the current air-cooled genset, I want to allow a couple minutes of idle cool down before actually turning the genset off based on past experience with air cooled aircraft engines and old VW bugs.  This should not be as necessary on a liquid cooled engine (depending on the coolant system design).



    The metal box above houses the ac to dc rectifier, power contactor and current shunt for the generator.
    The blue octagon in the image below is the 15 KVA single phase 240 to 120 volt transformer.  The gray box slightly above and to the left houses the generator digital KWH meter.



    Below is the EKM KWH meter that keep track of the energy taken out of the genset.  It is on the 240 volt side so its 100 Amp capacity is more than adequate.  It is in the gray electrical box with the weather cover window.  To the right in the box is the 240 volt AC current shunt.





    The image above is the placard of the transformer.



    The images above are the genset labels.
    The images below are diagrams and and specs from the genset manual.  It  looks like it might be possible to change to 120 rather  than 240 volt output if the ends of the power windings can be accessed inside the gnerator case.

     

    The spec. says 1.6 gallons per hour at half load.  Currently, I think we average about half load when the genset is running.  We have it running about 15 minutes of each 18 mile trip.  That implies we are using .4 gallons per 18 mile trip or a gallon per 45 miles.  I don't think we are getting near that good.  Plus, this represents only about 2 of the 8 KWH consumed per 18 mile trip.

        

    The 140 or so max DC amps from the generator trailer connects to the car with standard electric forklift type battery connectors.  The collection of sensor and control wires have not yet been fitted with connectors so are currently hardwired to the car (can't run around without the generator trailer at this time).

    The image below shows the view under the hood.  The DPM showing 137 volts is for monitoring the battery pack voltage on charging.  The white box just to the right of it is the wall chargeing KWH meter.  The large variac is used to adjust the charge current/charging voltage.  The muffin fan to the right and below the variac is for cooling the charger rectifier heatsink.  There are 3 traction batteries and the auxillary battery up front.



    This is an experimental test bed setup.  Most every thing as subject to change.  I try to see that things are tidy enough so as not to be a problem but I don't want to go to the time/cost of making things permanent when I know they are going to change.



    There are currently 8 digital panel meters on the dash.  In the image above, the generator is powered up and connected.  In the image below, the generator is not connected, running at idle.  The square button just to the right of the left-most fuse. connects and disconnects the generator. The upper left meter shows the DC voltage just down stream from the generator rectifier.  The LCD meter below that shows the 240 volt AC output of the generator (the value ranges from about 265 down to 210 depending on the load on the generator).  The meter showning 19 in the image above is the 120 volt current output from the generator.  The meter reading 11 below it is the 240 volt side current (having this meter pair has been important as we found it easy to get too much voltage on the transformer which resulted in saturation and really poor efficiency.  The dial in the middle turns the small variac that runs to the generators voltage regulator that gives us the ability to control the output voltage of the generator over a nominal range of values.  The toggle switch below it is the generator starter and the button to the right of that is the generator kill switch.  The LCD Above and to the right is the battery pack voltage. The meter below that is the motor current draw (if it is smaller than the generator current, then all traction current should be coming from the generator and the battery should be getting charged (hardly ever happens).  The upper right meter is the individual battery voltage--there is a multi-position switch to the right on top a black box that allows selection of any one of the 10 batteries--reading #10 in this case.  The last meter is the voltage of the auxillary battery.  (Note that we have not made any effort to calibrate any of the meters at this time.)  We have a video backup camera in place to aid seeing what is behind as the generator blocks the view of the rearview mirror.



    DATA:

    We did a little battery testing last Friday.  Unfortunately, I have not figured out how to imbed a spreadsheet and chart in this HTML document.  If the link below does not work directly, you should be able to paste the URL below it into your browser.

    Battery Load Test
    http://www.jetelectrica.org/battery%20curves%201.pdf

    First we took the auxillary battery out and stuck it on a battery load tester we have.  This battery is about 18 months old.  It is the same as the Costco batteries we are using in the main pack but about a hear older.  It gets charged from the main pack via a 3 amp laptop AC adapter all the time with a maximum voltage of about 13.5 volts.  It has developed really serious internal resistance based on these results.  We tested as we took it out of the car, and then after we charged it with a microprocessor controlled automotive charger.  There was hardly any difference.  We only got about 140 amps out of it at 8 volts.

    We took amperage and voltage reading on the fly while driving home Friday afternoon.  The results are plotted in yellow.  This was done within a couple miles of having been charged out of the wall.  We got to 300 amps at 110 volts.  This is contrasted to the two reading in yellow that were taken in the morning shortly after charging.  Here we only got 200 amps at 105 volts.  The only obious diffenence being it was around 80 degrees F in the afternoon and 50 degrees in the morning.  (Also note that the testing was done with completely different equipment.)

    Followup will be to do the testing with more matched conditions/equipment.


    BLOGGY TYPE NOTES:
    4/21/09
    -Had the genset looked at by small motor shop--concluded gas splashing from carb due to bouncy ride was causing overfueling.
    -Genset side DC voltage meter misbehaved--showed a fixed 66.  Later went to normal operation with no known cause.
    -Tech made a metal sun shade for the meters--the LCD's went black Saturday sitting in the sun without enough ventilation.
    -Genset would not start on the trip home when the time came.  Did not stop soon enough and get out to use the choke (no remote choke at this point).  Genset battery too low to start then genset when I did stop (was obvious we did not have enough battery to climb the hill to home).  Unfortuately, not carrying any jumper cables.  Took out auxillary battery and used long battery hold down bolts to jump from auxillary battery to genset battery to get it started.  Got ;in and discovered the genset voltage was outrageously high (300 plus volts) and charging current was much higher than normal for no running load.  Genset tripped out within 15 to 20 seconds.  Discovered the genset voltage control variac setting had gotten changed when the sun shade was added.  Could not get the genset voltage control to work for several minutes--then it came back. Spent several minutes trying to relocate the previously used variac setting.  (Need to revisit the setting for minimum transformer loss versus the ability to run at higher charging voltage.  Need to be careful not to zap the Curtis 1231 120 volt controller.)  
    -Rest of the drive home pretty much normal--battery more depleted than normal.

    4/22/09
    Turned the key in the morning and heard the fans come on but no lights on the display and not throttle response.  Spent a good hour trying to find the problem (yes, schematics would have been very helpful).  Finally gave up and wired a switched jumber from the battery to the main battery pack solenoid so we could go.  On arrival at our first stop, discovered the key switch was working again (it did not shut down when I turned my temporary jumper switch off).  Once I got to the shop, my tech noted he had seen the same problem once and that it was the intlock switch in the wall charging plug.  Had one of the guys make me a jumper cable long enough to reach from the auxillary battery to the genset battery in case I had the dead battery problem again.  Checked the gensets 12 volt output--seems to be working but not very strongly.

    4/23/09
    Sure enough, the same thing this morning.  Turned the key and gauges did not come one.  Took a number of attempts with the wall charger plug interlock button to get the switch un-stuck.  Will need to get deeper into it for an actual fix.

    4/24/09
    Did initial battery discharge curve testing as noted above.

    4/26/09
    Started crunching some numbers.  Updated this web page.  Found an error in the initial KWH per mile calculations (its better than I first thought).

    charles@cavlon.com