Mark Farver's Journal
10th January, 2005. 1:21 pm. Daily driving
Been driving the MR2 a few times a week. Alyssa is back in school so I need it as a second car to get to work. I replaced the DC/DC fuse and connected it back up. With it mounted to the trunk floor it has worked without incident for a few weeks. Still not idea if the fuse blowing problem was heat, or a loose 12v ground wire.
Driving to work this morning the roads were damp from rain. I made a right turn and got on the throttle as I was exiting the turn. The rear end broke loose and with the short wheelbase the car really swings out fast. I managed to correct quickly with only a slight overcorrection and lifting the throttle and got it under control.
Stock MR2 are pretty stable. The only major danger area is lifting the throttle suddenly when cornering. The weight will transition forward, the rears break free and basically there is no way out. A lot of MR2 have been wrecked this way. The MR2 EV does not seem to handle like stock, with the higher rear wheel weight, and with most of the weight close to the rear axle the weight transfer is a lot less aggressive. Most of my skids have been power related, the high torque of the electric drive breaks the rears loose and the rear end stops cornering. Lifting the throttle immediately stops them from skidding and give back some control.
The battery are sagging a lot more under load, even a fully charged pack can dip into the 10.5V per battery range under moderately to heavy accelleration. This might be the cooler weather, or I might have another faulty battery in the pack.
8th November, 2004. 6:46 pm. Old hardware
The MR2 is working well electrically.. but a couple of other problems are beginning to surface. Its quite disheartening.
The DC/DC was returned from Alltrax, and installed. Everything seemed to work. The stalling issue was finally resolved after an afternoon of driving back and forth in the driveway. The Cat5 cable between the hairball and controller was intermittantly shorting internally causing the hairball to reboot. This is likely the cause of the hairball's programming getting wiped out as well.
Last week I went on a week business trip, leaving the car with Chris to take to a clean air show at the mall over the weekend. Chris drove the car, um, aggressively and returned it with a dead DC/DC and a destroyed third gear syncro. It is now very difficult to shift in and out of third. Third gear has never been eashy to shift into, and Chris was using second gear for burnouts. I think shifting into third under load with only a partially effective clutch was the final straw.
The u-joint in the steering rack has been making noise for some time (I think it is the cause of the banging whenever the front tires go over a bump) and it is heading downhill fast. The steering rack is no longer available rebuilt, so I will have to take my chances with a junkyard rack. The alignment is way off, and the front passenger lugnuts loosen over time.
So the car is drivable, but a number of expensive repairs are on the horizon and my enjoyment level is way down. $200 for a new steering box, $600+ to rebuild the tranny, $50 alignment, new tires etc.. The DC/DC went back to Alltrax for more repair work, under warrenty, thankfully.
A newer car would have less of these issues... and parts would be easier to obtain. I strongly recommend against converting a car older than 5 years unless you have a really good reason.
18th October, 2004. 2:39 pm. Life in the slow lane...
The weekend I showed the MR2 at the Festival de las Plantas at Guerrero Park. One way distance was about 13.76 miles. I had never done a range test on the MR2, but I've been working on 25 miles as the approx based on other EVs. The organizers offered outlets so I set out expecting to get enough power for the return trip doing a slow 15amp charge during the 8 hour long Festival.
At the event power was in short supply, the organizers plugged AustinEV and all the food vendors into the local power utility's portable solar/battery trailer. The trailer had been oriented badly by the setup crew, with its 16 panel solar array pointing North-NE and under heavy shade. By 10am the inverters on the trailer began shutting down every few minutes. Initially we assumed it was due to overloading and disconnected appliances and the EVs. 4 hours later Aaron was charging the Henny Kilowatt from a hidden outlet near the ballfield, and the MR2 was still waiting its turn. By 2pm it became apparent the trailer was shutting down due to low battery voltage, and its panels were still in shade (it didn't begin generating until almost 4pm). The ballpark outlet breaker had tripped, and the groundspeople were unable to find it to reset it.
By closing the MR2 had received less than 15 minutes of charging. The 180VDC pack was showing about 186V. I left the parking lot with some uneasiness. A quick acceleration told me the pack was still pretty strong. I followed the Henney at 45mph to the IH35 feeder. Feeling a little more confident I hopped onto the freeway at 1st St. By 15th street I could tell things were weaker, the pedal had to be held down a little farther to cruise at 55mph. At 50th street I could have stopped an Aaron's house for a few hours of charging, but power is limited and I feared the oncoming darkness. The trip to the park was done in the early morning, requiring headlights. The DC/DC is with DCP/Altrax being repaired and I had no portable charger for the 12v system.
Still cruising I skipped the 50th off ramp and continued on. passed 290, started the gentle climb to the 183 flyover. Crested the hill at 183 and relaxed, the next 2 miles is a long downhill to my Braker Ln exit. Approaching the exit speed dropped rapidly and the low voltage light on the dash came on. I was forced to pull over just before the ramp. At this point I was feeling pretty good, I was less than 1.5miles from home and the rest of the trip was on 35mph surface streets. I gave the pack a 10 minute rest and drove home, timing the lithts weel wnough to keep momentum up.
Arriving home I checked the pack... all the batteries were dead, floating at about 11.5. All except #13
which was at disheartening 10.5VDC and #9
that was destroyed at 6.5VDC. I put #9
on a 12V 10amp charge for about 1 hour, and it rapidly returned to 11.6VDC. Odds are it needs to be replaced and probably sags heavily under load. #13
took a 10amp charge for about 2 hours to come up to 11.5, there is some hope for it yet. Once all the batteries were vaugly in balance I fired up the PFC-20 for a 20amp charge.
So.. like every EV newbie, I am well on my way to destroying my first pack of batteries.
18th September, 2004. 11:12 pm.
This morning I pulled the driver's seat for easy access to the Hairball. Carefully (becuase a few feet in either direction lay the wall or the garage door) I moved the car forward and back in the garage. The problem reappeared a few times. Following our theory on wiring I probed the power terminals on the Hairball. Nothing, it would not fail when it was being probed. Checked the wires which seemed tight ut following a hunch I grabed a jewelers screwdriver from Sean and tightened all the terminals. Previously I had been using my smallest Craftsman screwdriver with the 1/8" blade. It fit tightly but still tightened the screws.
Using the jewelers scredriver allowed me to add another turn to the screw terminals. A quick ride around the block failed to reproduce the cutting out. The surging problem in 2nd gear has also disappeared.
The cutting out while backing into the driveway usually happened while going over the curb. The sudden restoration while coming home happened after bumping over the curb onto the sidewalk. The slight pressure from the meter probes was enough to make it work in the garage.
Learn from my mistake. Intermittent connections are a major problem in the high vibration environment of cars. Solder permenent connections and the wire to the pin of any connectors. Use automotive rated connectors when disconnect is needed. Use locking washers on any screw fitting. Double check every connection.
17th September, 2004. 11:12 pm.
I've actually made enough progress on the car that I haven't posted squat.
I hooked up the Lee Hart sytem balancers. I'm not actually using them for balancing, but it makes it quick and easy to measure the voltage of all 15 batteries. I drove the car to work once, starting out with all batteries at about 13v and within about .1v of each other. Checked again after a little driving and found battery #10
was down to 10.5 volts... its either dead or has a reversed cell. Swapped it for the spare (I originally planned on 16 in the car). Alyssa drove the car to school a few times(about 1.5 miles) and had no problem, but it cut out on me a few times while backing into the driveway.
The DCP/Altrax DC/DC is dead, no output power. This happened once before, and it was a blown input fuse. So I opened the cover to check the fuse and found a nice scorched area on the circuit board. Some component had vaporized, leaving only two wires protruding from the circuit board. The nearby filter cap had a large black mark. CAlled DCP, them were nice and didn't seem to surprised, asked me to RMA it. Boxed it up and sent it to DCP... couldn't afford 2 day service, so it didn't reach them until Thursday, no word yet on the repairs.
I had been attributing the cutting out problem to a low 12v battery. Without the DC/DC the 12v has to run all the lights, vacumm pump, water pump and contactors all by itself. The problem was compounded by accidentially leaving my 12v battery connected to a 40amp dumb charger, returned hours later to the smell of sulfer. The hairball has been logging 1223 and 1224 errors (the first a low bat warning, the second is low battery shutdown.)
Thursday I replaced the bad SLI battery with another from Batteries Plus. $90, ouch. Took it to work Friday. The car died when I was giving my co-worker his first ride. He was unimpressed and mocking during the entire walk back. Such a nice fella. During lunch I took Alyssa's car and retrieved the big marine deep cycle I purchased for the RV. Wired it to an anderson SB175 connector and plugged it into the jumpstart port I installed in the rear trunk. Drove home. Made it to the traffic light about 1/4" from the house before it died in a left hand turn lane. Sat dead for several minutes annoying taffic before I got a break large enough to push the car onto the sidewalk. Pulled the fuse for the water and vacuum pumps and waited 5 minutes for the SLI battery to recover, assuming the marine battery was never fully charged. Drove another 1/8 mile before dying, once again at a traffic light. Walked home.
Hours later I returned with my roommate Sean and another battery. Still didn't work and the palm pilot I use for debugging the hairball was dead. Went home again, fetching a multimeter and the laptop and Derrik, another AustinEV member. Connected hairball to laptop. The car would startup normally, closing contactors and clearing the Check Engine light. The instant the accelerator was pressed and the car started moving forward it would die. A few seconds later the contactors would open. Running the motor with the car in nuetral and it would work fine. Sean suggested monitoring the SLI voltage while trying to move the car. The voltage stayed good, 12.41 volts and more interestingly the laptop displayed the hairball's bootup screen the moment the car died. For some reaason the hairball was resetting each time the car started moving. Theories abounded on the problem, mostly centered around bad wiring between the battery and Hairball, or some noise issue. Tired and working in the dark we decided to rope tow the MR2 home behind Sean's Jetta.
6th September, 2004. 11:44 pm. Clutch piping
I finally found the correct adapters for my clutch problem.
The MR2 uses 12mm inverted flare fittings on each end of the clutch pipe. The counterperson at Austin High Performance found a 12mm inverted flare to AN6 fitting in his box of adapters. AN6 is too big for a clutch line, so another adapter converts to an AN3. From the AN3 a compression fitting converts to 1/4" steel line. 1/4" line would have been perfect, but they didn't have long enough lengths. So two 12" length were used with a 1/4" threaded fitting to 3/16" adapter. Then 3/16" pipe was used to connect the endpoints together.
The slave cylinder end is the same with the addition of an AN flex line and female to female adapter:
Ugly, but with plenty of Teflon tape, leak free. We spent a few hours trying to bleed the system and found the clutch would not work. The throwout arm moved to its stop without encountering the pressure plate springs.
Rented a transmission jack on a holiday weekend special: Saturday thru Tuesday morning for only 1.5 days cost or about $38.
Dropped the Kostov, leaving the tranny in the car. Pretty easy to do, but the quarters are tight. Made several measurements inside the bell housing and consulted some of my original adapter drawings. Turns out we mounted the adapter hub on the motor with the motor shaft flush with the flywheel alingment ring, it should have been mounted farther out. Remounted the hub a 1/4" farther out, moving the pressure plate that much closer to the thrust bearing.
Took the opportunity to turn the motor 90" on the adapter plate, correcting a mistake made eariler that placed the motor terminals at the top of the motor, and under the battery rack. A 90" turn makes them accessible in the rear of the battery compartment.
Several hours of effort was required to remount the motor, including redrilled the motor end engine mount. The Kostov only has two threaded holes on the commutator side, and they are not simply oriented opposite each other. Reinstalled the axles and was ready to run.
18th July, 2004. 9:01 pm. AC
Selling the neon that has been my primary gasser to free up some space in the driveway. I've been working on home improvements for the last few months, but I need to finish up the MR2.
Battery balancing.. the curse of my existance.. its harder to build a BMS than it seems, and certainly time consuming. Might just break down and buy some MK2s from Rudman to tide me over.
Air conditioning. Its summer in Austin, and without AC the T-Tops and large glass to volume ratio make the MR2 a sweatbox on the trip home. Vintageair.com sells AC components and kits for retrofitting AC onto custom and vintage cars. The grill area of the MR2 is 13.5" tall by 27" wide. With R134a systems the bigger the condensor the better. VintageAir's Superflow condensor's are available in 12 and 14" heights (Doh!) and up to 24" widths. They have standardized on the Sanden 508 rotary compressor ($~199) and have a flexible hose kit for $99. The hose kit comes with crimp on ends. I don't have the crimp tool, but hopefully Napa or a local AC shop can be convinced to do the work. I can reuse the MR2's evaporator, which is nice since it is custom fit into the dash. The expansion valve needs to be changed, but VintageAir claims to have a bolt in replacement. I already have a 3phase 1HP motor and small 1/2HP 3 phase inverter to spin the compressor.
Driveshaft seals. The tranny leaks ever so slightly around the half shafts. I have the replacement seals, but removing the shafts to replace them is a pain. I've been avoiding it, but once the car is on jack stands its an afternoon of work.
Clutch piping. Have all the parts, after my trip to the junk yard, just have to bend and route the pipes. A weekend under the car...
Radio... the Infinity speakers are in the doors, but I have no radio or amplifier to drive them.
11th March, 2004. 4:11 pm. Oops...
Reassembled the car and drove it around the block. Everything felt good so I took it out onto Braker and floored it.
The good news.. Otmar's fixes to the controller, plus warmer temps makes for much more amperage. Bad news.. much more amperage means having good tight battery connections is critical.
I wouldn't be so bad if all the flaming debris had ended up somewhere other than the lid of the Zilla. Now I have to meekly ask Otmar what a new cover might cost.
Live and learn...
9th March, 2004. 8:32 pm. Spring has sprung
As usual I figured out that the replacement strut cartridges do fit the car when I went to return them and an additional bag of parts fell out of the box. In fact you are expected to remove all the oil and piston assembly from the stock struts, and install the aftermarket cartridge inside. Of course, this is after I paid $200 for the air springs.
The air springs are not going to be easy to fit onto the car, since they have studs sticking out of them with too little room for the strut cylinder to fit within. My original idea was a flat plate welded to the end of the cylinder with bolt holes which won't work. So instead of screwing around with the air stuff, I went back and got the aftermarket strut cartridges and reassembled everything. At least now the car might be driveable when my sister arrives late this week.
The new springs made quite a difference in the ride height.. probably too much (the after side of the pic was taken at a higher angle, the gap above the wheel is even larger than the picture suggests.)
At least I manage to do everything without the spring compressor failing...
27th February, 2004. 10:05 pm. Solenoid release
Back A Page
Alyssa thinks I'm too amused, but after about an hours worth of fooling with it I got keyless remote to open the fuel door.
It was actually challenging since the solenoid came with a short 1 foot cable sleave, and a 14 inch steel cable pre-attached. The door is held shut by a small plastic pin that is attached to the cable. Pulling the release lever pulled the pin back, a spring pushed it back into position. The existing cable sleave had a odd snap fitting that locked into the hole in the sheet metal and held the pin assembly in place. The most obvious solution had me just connecting the existing cable to the solenoid cable, but that would have required creating a bracket to hold the two cable sheathes in place.
My solution was to cut the existing sheath about 13" from the pin and discard the existing cable. The new solenoid cable was then threaded into the existing sheath and a new stop was crimped onto the end. I had some initial trouble with the new stop shifting positions and jamming the mechanism. This was solved by surrounding it in my new favorite, hot melt adhesive.
I still need to mount the solenoid permanently, probably to a cross brace that is near by. Will be tough to get screws into it. The welder would be ideal, but it is in San Antionio for Nick's Jeep project.