Tuesday, August 3, 2010

500 miles at 216V

This weekend, I passed 500 miles in the Volt914 with its new 216V pack.  It is staying very well balanced - the Rudman Regulators are doing their jobs well.  And, with quite a few trips in it, I'm able to assess its efficiency at the higher voltage and with the new charger.  Take a look at the miles vs. kWh plot:
As you can see, the least-squares fit line shows that a good approximation of wall-to-wheels electricity use is a 910 Wh fixed overhead per charge, and about 280 Wh per mile.  This is wall-to-wheels, so it accounts for all the efficiency losses of the AC-DC conversion process as well as the reverse DC-AC conversion for driving.  Not too shabby!  And, much better than the old 144V floodies with the Zivan NG-3 charger, which were giving me 1,230 Wh fixed overhead per charge and 380 Wh per mile wall-to-weels efficiency (!).

Edit: Forgot to mention, at my current cost of 7.5 cents per kWh, I spent  about $14 for the electricity for those 500 miles, at an average cost of 2.8 cents per mile.  Which compares to 3.7 cents per mile with the old 144V system - a 25% reduction in cost.

Of course, I will soon be paying essentially *nothing* for the electricity my cars use:

This is a 5 kW array, which will generate about 7,000 kWh per year.  Between them, the Volt914 and the Electrojeep use about  2,000 kWh per year, if I drive both of them every work day.  Which I don't.  But I could :-)

Sunday, April 25, 2010

Out with the old, in with the new

I finally got around to replacing the ammeter that came with the kit. For some reason, it was a -50 to 50 amp meter. The controller generally draws more than that except in economy mode. So, I got a -150 to 150 amp meter, along with a 150A 50mv shunt. Removing the old shunt was a little bit of a pain, but fortunately I had mounted it high enough that I could access it without removing the battery boxes.

Here are the new shunt (on the left) and the old shunt next to each other:

Notice the discoloration and warping on the old shunt? Here is the same area from below:

Definite heat damage. It's not clear to me whether there had been a loose connection at some point (generating extra resistance == heat), or whether the 50A rating of this shunt was just exceeded too often, but it is clearly damaged. I'll monitor the new shunt for damage frequently for a while.

Driving with an operational ammeter is extremely useful. For one thing, I can see that my "normal" and "economy" modes indeed are not different - which I had *felt* but not proven. Both pull 50A max. I'll tweak that at some point. It would be useful to have something in between 50A (economy) and 150A+ (performance)...

Addendum: this is very similar to damage that Randy Pollock suffered (although not as extreme):

We have the same shunt, in the same system, and had the same failure. I don't think this is a coincidence.

Tuesday, February 23, 2010

Of Mice and Men

As in, "the best laid plans of." Or, while I'm on film allusions, how about "The Remains of the Day:"

But I'm getting ahead of myself. The car has been running very well - except for the nagging fact that the battery warmers have not been drawing any amps except for the first day they were plugged in. And the batteries have not been kept warm - instead, I've been keeping the car in a heated garage ($$$$). So, tonight I decided to start with the middle battery rack - the one with 3 batteries. It's the easiest to access. So, I pulled the batteries out and pulled the warmer out. Another allusion, "What Lies Beneath:"

And what a mess it was that lay beneath. The clear plastic around the thermocouple was discolored and clearly burnt:

Dissecting the clear plastic, it is obvious that the nicrome wire got too hot and simply vaporized:

Seeing this, I decided to pull all the warmers out. Each one had similar damage. And, in fact, some of them had burn-throughs in the middle of the cable:

Even when the cable had not burnt through, it is clear that the cable produces too much heat for the foam insulation. Here is one of the dual-sided pieces, melted all the way through:

Which brings us back to the pile of stuff at the beginning of the post. I'm very fortunate that no other damage occurred. So, it is back to the drawing board for battery warmers. I'm considering some of the battery heating pads, or other options. Stay tuned.

Saturday, February 20, 2010

Misc Saturday

It is snowing today. Grr. I won't drive my car in the snow - too much of a risk of someone crashing into it. So, I decided to get the heater working again. As mentioned here, the Crydom solid state relay was stuck closed. I happened to have a Tyco EV200AAANA contactor lying around. It handles 500A of current, so it is way overkill for the 3A or so max needed by the DC-DC converter. Oh, well. Here is the new contactor in the place of the Crydom relay:

I installed it, hooked up the cables, and turned the heater on - voila! Heat! So now, when it is cold but not snowy, I can drive around. I put a clamp-on ammeter around the HV cable, and measured 10A when the heater starts up, which decreases to around 5A as the heater warms up and the resistance increases (V = IR, after all, and V is constant). The updated PDF of the wiring diagram is in the usual place (nothing major changed, just replacing the Crydom with the contactor).

I also installed a goody I got from Electro Automotive. This was supposed to have been part of the original kit, but got overlooked somehow. They were gracious enough to send it to me gratis. It is a combo regen enable/disable switch along with a 3-position rotary "ECONOMY / NORMAL / POWER" switch. It hooks up directly to the 4 wires that have been dangling in my cabin for 2 years. I attached it where the radio would normally go:

Finally, I got around to attaching the door sills and speaker grills - these are what holds the carpet in place. I'd been frequently pulling the carpet back into position, and got tired of it. It took me about 10 minutes, I don't know why I put it off so long:

Also, earlier this week, I put a new rear-view mirror on. The original Porsche mirror would not stay stuck to the glass, so I went with a traditional after-market metal "button" and rear-view mirror combo. Works very well:

P.S. - this is post #100 to this blog. I never thought it would go on so long when I started...

Monday, February 15, 2010

On the road again!

The big news of the day is that the Volt914 is an automobile again! I drove it a total of about 12 miles today. But I'm getting a little ahead of myself...

The morning started with a large, immobile lump of steel, lead, glass, and plastic sitting in my garage. Turning the key did not result in the proper effects. So, I went after it with my digital multimeter. I found that I had swapped two wires - the keyed +12V and the "lower interface terminal block" wire. Fixing that resulted in all the right things - the oil pressure light came on, etc. The 12V system was alive!

However, all was not ready for driving. The Crydom relay seems to be stuck closed - it is always transmitting full HV regardless of the control inputs. As a temporary workaround, I used the heater contactor to control the DC-DC converter and voltmeter. I had to extend the wires to reach the contactor lugs:

With that set up properly, I made sure the DC-DC converter was properly wired. That yellow block is a 60A 12V fuse (the DC-DC converter is rated for 55A). Before hooking it up to the fuse, I verified that it is putting out the 13.3V it is supposed to (it is a battery-charging DC-DC converter):

With that taken care of, it was time to seal everything up. Here is the front compartment with all the wiring in place - it will get neatened up a little more some time later, but for now, it is operational:

And here, for the first time in months, is the front hood closed:

I then hooked up a computer to the controller and tweaked the necessary parameters to convert it from 144V nominal to 216V nominal. For the record, here is what was tweaked:
  1. EE2AccelBatRamp - 16 (was 12)
  2. EE2NoAccelBat - 189 (was 126)
  3. EE2RegenBatRamp - 18 (was 34)
  4. EE2NoRegenBat - 240 (was 160)
  5. EE2BatVLoMem - 0.018 (was 0.0122)
  6. EE2BatVHiMem - 0.018 (was 0.0122)
  7. EEXMinAccelPower - 10800 (was 9609)
  8. EEXNormAccelPower - 21600 (was 24000)
  9. EEXMaxAccelPower - 32400 (was 36000)
I had previously tweaked the EEXNormAccelPower and EEXMaxAccelPower variables in an attempt to get more power out of the 144V system. It did not really work. It will be interesting to see how the power selector switch from ElectroAuto works - it should arrive later this week.

I took two drives today. The first was just over 4 miles "around the block" (never more than a mile away from home in case of breakdowns). It feels peppier than it used to - losing 180 pounds of lead probably helps. The regen brakes have the "pulsing" effect again - I will need to tweak the RegenBatRamp variables some more. The very good news is that the combo of the new DC-DC converter (rated at 55A) plus the massive 4-gauge cable I used for a main 12V bus have made it so there is never any voltage drop, no matter how much 12V stuff is happening.

The second drive was to my local AutoZone, where I snagged stick-on letters to put a "VOLT914" sign on the back:

I charged in between the drives, and at the end of the second drive. Mostly to try to get the timer adjusted properly. It was reading 6A when it cut off the first time, so I increased its time by 30 minutes (from 45 to 75 minutes). It was reading 3A the next time - I probably need to tweak it up another 30 minutes to 105 minutes.

After playing with the charger some more, I hooked up the rear fans. This involves a relay box and stringing various wires:

Here is a diagram of the rear fan control. The DPDT relay is there so that when the key is not turned on, the fans are being driven by the AC-DC power supply when plugged in (this keeps the charger cool). When the key is turned on, they are instead driven by the car's +12V:

The PDF may be found here. Regarding range, it is a little early to get a good read, but the second drive was 7.6 miles, and consumed 3.1 kWh, for a net 408 watt-hours per mile (from the wall). This correlates very favorably with the 144V / Zivan system, which consumed about 4.0 kWh for the same distance - so at first blush, the Manzanita charger and new batteries are about 25% more efficient.

I also made a little chart based on Concorde's own data, showing the SOC versus voltage for my "216V" pack:

And that's all for today. I'm exhausted. This week - mostly just test drives, but I may have the opportunity to show the car off for some government big-wigs on Friday - stay tuned.

Sunday, February 14, 2010

Pushing toward completion

I've been very busy the past few days, doing instead of blogging. So, here is a day-by-day summary. Not quite road-enabled yet, but getting close...

Feb 10, 2010 - Regulator Covers

A while ago, I cut plexiglass covers for the regulator boxes. Today, I finally got around to drilling the holes and installing them. Here is an example. The covers will keep rainwater out, as well as curious fingers. They also clamp the 10 gauge input lines in place:

Feb 11, 2010 - Regulator Boxes Attached

Attaching the regulators to the battery box covers was pretty straightforward. First, I found the ideal placement - close enough to attach the wires, spaced well enough to allow access to plug in the RJ11 RegBus. With the regulators attached, simply mark and drill holes. I then put bolts through and clamped them in with nuts, leaving a forest of bolts. This makes it easy to do final attachment of the regulator boxes.

And here they are, all attached. I brought them inside to work on the RegBus RJ11 interconnects.

Here is the front box with RegBux interconnects. If you have the proper crimping tool, building these interconnects is a breeze. The Rudman Regulator manual is very clear on how to build the cables.

Here is the rear box with the RegBus cables attached and the Metric Pak 480 connectors connected:

Feb 13 2010 - Push Toward 216V Completion

Today I focused on getting the 216V system ready to charge. I did some modifications to the original VoltsPorsche instructions. First, I wrapped most of the HV cable with FlexGuard, in red, to highlight that they are high-voltage cables. I also put an Anderson 350A connector in so the controller would not be plugged in while working on the high voltage system:

Here is more FlexGuard. This will not only alert hypothetical emergency responders to the presence of high voltage, but it will also protect the cables from abrasion from whatever might abrade them:

Still more FlexGard. In addition, I moved the CamLok connector from the middle compartment to here. Much easier to reach, and plenty of room. This makes it very easy to disconnect the 216V circuit:

Since I removed the CamLok, I redid the fuse mounting. I cut a notch in a conduit box so the fuse would fit, and bolted cable to it:

Moving back to the rear trunk, I added vent holes so that later fans could bring cooling air in - both the controller and the charger are happier when they are cool:

More vent holes, on the right:

A 12V fan installed on the left. This faces up, so it pulls cool air in from outside:

And the other 12V fan. This faces down, so it blows warm air out:

The fans are not yet hooked up, that will wait for another day.

14 Feb 2010 - Charger Attached, Configured

After getting all the 216V system hooked up, it was time to focus on the charger. I moved my kWh meter to the rear compartment, and attached twist-lok AC connectors to the charger input:

With the charger hooked up, and the RegBus hooked up, I flipped a dip switch on the charger. This lit up the yellow LEDs on the regulators to verify the RegBus was correctly attached. And it is. Here are the yellow LEDs glowing in the front compartment:

I also kept an eye on the voltage levels - the voltmeter is reading the voltage from the Anderson 350 connectors. The voltage I'm aiming for is 14.30 volts per battery, or 257 volts:

And also on the amperage - I got a new clamp-on DC ammeter just for this purpose. Later, after a test drive, I'll use the clamp-on DC ammeter to make sure the constant voltage timer is set correctly:

Finally, I worked on the 12V system in the front compartment. I drilled a new hole to accept the HV input and heater output lines:

Not shown - I crimped WeatherPak connectors on a variety of wires and hooked things up. This is a picture of the middle of the process as I tried to debug what all the wires did on the original AC VoltsPorsche:

Unfortunately, something is wrong with the 12V system. In the summer, when I started this project, the exact same something was wrong - the speedometer does not work, and the green oil pressure light only comes on when the heater is turned on (!?!?!?). The 12V voltmeter reads zero. Back then, I thought all of those problems were from the isolation failure caused by leaking battery acid. Now, it looks like there is something deeper.

Tomorrow, more debugging, and, hopefully, a road test!

Wednesday, February 10, 2010

Batteries Interconnected

A few things have come together over the past few days. First, I mounted the DC-DC converter, the relay box, and the AC-DC 12V power supply on the board where the old Zivan charger had previously been mounted:

Next, I reinstalled it in the front compartment:

Next up was attaching the interconnects. But I discovered that my BMS wiring was not quite the right length - so I modified several of them. I made one side longer by about 4", and shortened the other side by a similar amount. Here are a few of them, rebuilt:

And here is the front compartment with interconnects and BMS wiring installed:

Middle compartment:

Rear compartment:

It is getting so close to done I can almost taste it. I predict a late-night final push some time this week...

Sunday, February 7, 2010

Revised Front Wiring

This weekend ended up cold and snowy, so mostly I worked on finishing up the front wiring. Photos will come later, for now, here is the revised circuit diagram. This includes the details of which connectors go to what - this will make it much easier to attach things when in the car:

You can find the PDF here. It is sized to print on legal 8.5 x 14 paper.

Sunday, January 31, 2010

Battery Interconnects

After procrastinating for quite some time (not to mention healing my injured arm...) it was time to figure out battery interconnects. I had originally been planning on using 3/4" wide 1/16" copper bar stock, stacked two high, as the interconnects. But putting the bend in it was a pain. And I hate drilling through thin copper. I got some hole punches, but the 1/16" copper laughed at them.

So, on to Plan B. The local metal supply store had sheet copper 0.020" thick. This turned out to be perfect - I can cut some of the weird shapes right out of the sheet. Cutting straps 3/4" wide, 7 of them would stack up to form the same cross-sectional area as 2/0 cable (0.105 in^2). After much careful measuring (and a first failed prototype - good thing I tried cutting the paper out and fitting it first!) I created a layout that fit on a 8.5" by 14" legal piece of paper:

You can find the PDF here. I printed out 7 of them, and then cut out around the outside edge and used spray adhesive to stick them to the copper sheet:

I cut around each one to make for easier handling:

Next up - the holes. This hydraulic punch did not do much with 0.0625" copper, but it sails through 0.020" copper like butter. I call it Mr. Punchy:

Here are the holes punched in one section. To reach the interior holes, I cut each piece in half. Mr. Punchy can reach pretty far in, though. It is *much* easier to line up the hole punch than it is to drill through thin sheet:

My assistant, Naiche, came by and together we used aircraft shears to cut out all 105 pieces:

Now for the tricky part. To make sure that the straps don't pry the battery posts out of the batteries, I designed a bend in each one - basically, a 3/8" by 3/8" triangle. I tried just using pliers to bend it, but it was very difficult to get a precise bend. So, time to design a tool. This is the bending jig. It is basically just a piece of plywood with holes drilled to align with each different piece. A 1/8" deep inset is routed into the surface, and a welded metal piece is screwed to the base. This piece has a small angle welded to it, on top of which a welded stamp fits:

You can find the sketchup file here. I extracted a PDF of the base and rotated it so the holes would fit on 8.5" by 11" paper (did not want to make another trip to the copy shop):

This PDF can be found here. After a pleasant afternoon of metal cutting and welding, plus wood routing and drilling, I had my bend jig:

You can see the stamp on the upper right. To bend a strap, all you do is put screws through its end holes into the holes on the jig. This keeps it aligned. Then you place the stamp over the ridge in the metal base and pound it down with a hammer:

Remove it all, and you have a nice bent piece of copper strap:

And, finally, after bending all of the pieces and stacking them up, I put heat-shrink tubing over all of them. All done:

Next up, attaching the straps, hooking up the PakTrakr and the regulator harnesses, and then finishing the wiring. So close I can almost taste it...