Unusual problems found with "normal" use patterns.

Here's what I've found so far as the BugE is used for it's intended use - a pure electric commuter vehicle that operates in cool rainy conditions while keeping the occupant free of windchill and rain.

First, in terms of BugE specific issues, I've noticed more tire wear with the front two tires with little wear on the drive tire. This suggests my 'toe' is out of adjustment. I'm guessing too much toe-in. So, an adjustment of the steering rods will be needed. Fortunately, this is a simple task. Visit the express links on the right to see how this is done.

Another unusual incident happened. It has to do with the BMS system. This type of BMS is no longer sold and has been replaced with a more reliable system. Unfortunately, I have the older setup. So, I came home to find the meter showing my batteries were rather low . I knew this, only due to noticing the BMS indicator lights were completely off. What apparently happened was that the DC-DC converter was doing a steady draw down of the batteries to power the BMS. However, the BMS wasn't triggering the AC charger to replenish the main pack. Finally, the DC-DC converter had the sense to cut off. However, the lead-acid battery discharged completely (turning off the BMS and it's indicator lights.) I'm sure the lead-acid battery took some damage but it's duty is normally rather light. So, it may have survived well enough to remain in service.

Fortunately, the main pack was not completely drawn down, just low. The cause was apparently due to a faulty LIN HUB. It failed such that it would only work if I bent the connector. When bent, all components seemed to work. When not, nothing worked. Unfortunately, when I tried solving the connector problem by trying to solder on a more robust replacement connector, I managed to loose what little function that circuit board still had. Unfortunately, it's a multi-layered board so it's beyond my ability to repair.

So, I removed the board, did an unbalanced charge of the pack to 50V and eliminated all parasite loads. Although 50V was not a full charge, it hopefully kept the lithium batteries in a safe state. Meanwhile, my vendor shipped me a replacement hub. Once replaced, the BMS returned to normal operation. Hopefully, the batteries are OK. We'll see if this is true in the next range test.

The last failure was severe enough to get my BugE trucked home. At times, if I accelerated even slightly, the vehicle stopped. After powering the vehicle on/off several times, the problem would go away. When I first noticed this, I thought the fault might be in the spade connectors I used for the throttle cable. So, I soldered the throttle wires to the speed controller inputs. This helped for a while, then the failure returned more often. So, this caused me to think the throttle potentiometer was at fault since it seemed to have similar symptoms to my previous throttle failure.

But no, these electric cars can be sneaky. Testing out the various components with a multimeter showed what the real cause was. After putting the rear tire up on blocks, I measured various voltages along the system. I found the potentiometer in the throttle was working just fine. However, the contactor (also known as the solenoid) was giving a voltage difference across it's terminals when engaged. This was not good since it's supposed to simply act as a hunk of wire between it's terminals when active. In fact, that's how I tested my hypothesis that the solenoid was at fault. I put a hunk of wire across the terminals. Sure enough, the motor driven by the speed controller worked just fine. So, I've ordered a new solenoid, at nearly $100 bucks. Hopefully, after installing the new one, I can take apart the old one to salvage it for a spare. It did cause me to question why these solenoid's are needed at all. For an explanation, look HERE.

September 30, 2011 - My first commute!

At last. It was time for my first real commute from Auburn NY to Wells College in Aurora NY. Russel Watson's "Faith of the Heart" song came to mind when I took off on this historic journey.

Although the BugE will do 50MPH on flat ground, this test wasn't a speed trial. Rather, it was to figure out the speeds that will be fast enough to be useful yet be slow enough to not loose too much energy to wind resistance.

Distance was just over 20 miles. Morning temperature was 60 degrees and sunny. Canopy practically eliminated windchill. Speed was kept at a minimum of 30mph, up the gentile slope to the ridge line. Along the top, maintained 35mph. The journey was so pleasant that I overshot my planned turn. Fortunately, I had an excess of charge so I could take the next turnoff which made the journey slightly longer than planned. The last 2 miles downhill were done at 45. Arrived with 40% of charge still left. Commute time was 45 minutes which was two minutes longer than I had planned.

The BugE recharged at the Stratton Science Building where it was put on exhibit for the day. Charging was done before noon. At the end of the day it was time for the return journey. Rolling the BugE out of the building, the outside temperature was 80 degrees & humid. This caused a temporary fog on the canopy that quickly dissipated in the sun. I did the long climb up to the ridge at 25mph to see if a lower speed would help conserve the battery. It did. Once half way up the hill, I maintained speeds of 30-35mph. Since I had more charge than expected, I took the last 2 miles home at 45mph. Pulling into my driveway, I still had 1/3 charge left.

The trip had one noteworthy incident.
The throttle didn't work! So, I exercised the interlock relay several times. Then, the throttle worked. When I returned home, I tried some contact cleaner compound on the throttle spade connectors. That seemed to fix the immediate symptom. However, this likely isn't the real cause. I suspect the interlock arrangement I have may be at fault. So, I'll be doing a wiring change to test this hypothisis.

BugE Q & A

This blog posting is a partial response to some questions fermi_of_borg had in his BugE build.

The first photo shows how I oriented my direction switch. I found it was easiest if I had the forward direction pointing backward. However, this is probably opposite of what people normally would think when it comes to direction. I had to remove quite a bit of material with a Dremil tool to get a sufficient sized notch for the switch to go into. I decided to go with a notch instead of a hole since I open and close the battery compartment quite often for small adjustments. Now, no need to remove the handle!

The second photo shows the switch orientation without the cover. The direction switch is held in place with a couple of "L" brackets attached to a platform I made from a cutting board I bought at Wal-Mart. It's cheap, easy to cut, non conductive and needs no painting. Anyway, In the lower left is a little photo of a wiring change I did. In the original 48V wire diagram two wires were to be attached to a contactor terminal. Problem is, it's quite a tight space. So, I decided to attach the two wires together instead at the top terminal of the reversing switch. Electrons don't care which end of a wire they are on. To them, one terminal of a wire is identical to the other terminal of the same wire.

When it comes to connecting the lug rings, the type of nuts should connect very TIGHT! Using Nylock nuts will help. Also, double check the little screws on the lithium battery pack. They should be tight too. One way I happened to find a couple of a loose screws that I had forgotten to fully tighten was when my BMS said it had a "volt diff" error when I accelerated. Inspecting the pack, then tightening the loose terminal screws got rid of that error.

Last is a photo of my BMS wiring harness which is an example of perfection being given up for progress. It's ugly, but it works. Making a better harness will be a winter project. Meanwhile, this mess seems to work for me right now.

Another range test

This range test was a 22 mile test to simulate a commute in terms of distance and hill climbing. This would be to the half way point, then do a return on one charge. First part of journey was at going to half-way point on relatively flat ground, then downhill from 860ft to 420 ft. Speed as before was kept between 30-35mph. At bottom of hill, I briefly tested battery recovery. Unlike a lead-acid pack, parking for 5 minutes showed no significant battery voltage recovery from the batteries in the main pack. Then back up the long slope at 15-25mph. The remainder leg home was done at 40-45 over the flat and downhill sections of road.

However, as I feared, my charge controller could not charge the accessory battery fast enough to keep up with usage. So, near the last mile of the journey, the 12V battery became weak enough to make the headlight interlock relay de-energize. So, the main pack still had 25% charge left but I couldn't go! No problem. I pushed the BugE to the side of the road. Turned off all navigation lights. Then waited 5 minutes for the charge controller to catch up. Then quickly proceeded home. Still had 20% left.

What this says is either that I'll either need to increase the accessory battery size, reduce loads or use a faster method using the DC-DC converter to re-charge the accessory battery while under-way. Too bad I can't find a reasonably priced DOT approved LED headlight for a motorcycle! Hopefully, lower cost LED headlights will appear soon.

Lithium charger mounting

When using a lead-acid pack, I found an external lead-acid charger arrangement worked for me. However, the new lithium battery charger is larger, more expensive and has a really wimpy data cable that would quickly be destroyed through repeated connects and disconnects. So, I've decided to find a way to mount the charger so it does not need to be removed for each journey. To the left, you can see the end result. The charger has sufficient ventilation to keep cool. It also allows inspection of it's indicator LEDs on the end of it.

To the right is an image of how the holder just before it was installed. To mount, it's bolted in with nylocks and washers to the fiberglass battery pan structure. If assembled correctly, the charger should just "snap" into place. The only cable modification I needed to do was to lengthen the battery charger data cable so it could reach the "LIN hub". I suspect a "new" BMS system would only provide connectors & cable without being assembled, providing an opportunity to make cable lengths correct the first time.

Parts & tools required:
6ft piece of angle iron (you'll have some left over)
an angle iron bender (or use a vise & hammer)
hacksaw
drill
file
scratch awl
tape measure or ruler
(10) 6-32 flat head screws with matching Nylock nuts
(4) washers - used for mounting to the Plexiglas battery tray
spray paint and primer
zipties (to run data and power cable to battery tray)

Range tests

One of the most frequent questions I get asked usually includes the phrases, "How far and at what speed?" Due to these factors being dependent on conditions of the day, It's an impossible question to accurately answer. However, I can say this.

I did my first extended range test today. My route to work is 20.3 miles. I decided to go to the halfway point on the route I plan to take to work. According to the GPS, I start out going to work at 620ft above sea level. Then took several small hills for an average climb height to 860ft. I figured if I had 75% of the pack left at that point, it's likely I could make it to work with a comfortable reserve since the remaining part of the journey is either flat or down hill. Once at work, the plan would be to charge up during the workday for the return trip.

For this test, temperature was around 80 degrees, overcast, with occasional sprinkles of rain. Travel through town was stop and go which limited speed to around 25mph. Once out of town, roadway turned to tar & stone. The roadway outside town had enough bumpy areas that I limited my speed to between 30 and 35mph. Visibility was very good. I was followed briefly by one car. Otherwise, I saw only three cars going in the other direction. For the return part of the journey, I decided to take a faster route home. Speed was kept to about 40mph for most of the journey. However, once I encountered the smooth road of the arterial highway, I decided to speed up to 50mph for about a mile. No problem. I returned home with 60% of charge left.

First test drive on the lithium pack

Woha! The new battery pack makes this BugE perform much better! I decided my performance test would be at night. I drove in a mix of conditions with stop-and-go traffic along with hills. Total mileage was just over 10miles. I tried to keep speeds at least 30mph and go 40mph where I could. Passing cars was fun and I could even do so going up hills! At the end of the test, I decided to do a speed test on flat straight road. For the speed test, I sustained a speed of 50mph over 2 miles! At the end of the journey, pack still showed 60% of capacity left.

I decided to keep the retro analog meter to see how acceleration affected the pack. I noticed that the pack was either in white (full) or upper green at all times even during acceleration. I was worried that the 12V accessory battery would be depleted but it apparently has enough capacity to keep the lights relatively bright. Although I didn't have a digital meter, I did observe that headlight brightness seemed to be unchanged through the journey. The lithium pack also seems to have other benefits too. For example, braking and acceleration seem to both be better. Also, the BugE seems to deal with bumps better too.

The drive was not problem free. The BMS electronics have problems. For example, the LED readout has two LEDs that have already failed. Fortunately, I have the larger readout I can use instead. Also, the wiring seems to be of all the wrong lengths. I'm guessing this is because in it's original installation, the LIN HUB and all instrumentation was probably installed in the dashboard rather than in the battery compartment.

Still to do. The BMS system works but shows an error when the battery charger isn't present. So, next step is to mount the battery charger in the cargo area.

Charging up....

At last, the batteries are installed and the wiring is connected! The pack is doing it's first charge cycle. Along the way, there were some interesting incidents.

Upon connecting the LIN HUB for the first time, the LED indicator panel showed a red LED "wiring fault" and the main display showed that the "battery capacity" was 0%. Yet no alarms were showing and going through the screens, it appeared as though each battery circuit board was being polled successfully. Also, when the charger was plugged in, the individual battery voltages seemed to be charging correctly. The fix was simple. Leave the charger plugged in. Then go through the settings. I found the unit was set as a zero AH pack. So, entering 60AH for my 16 cell pack size turned off the fault LED and updated the bar graph to show 100% capacity. The main pack, fully charged, shows 53.8V.

Another interesting incident. When I powered up the DC-DC converter, the charge controller for the accessory 12V battery was flashing that it was having trouble charging. I found the problem rather quickly. I had wired it in backwards! The battery was trying to charge the DC-DC converter, obviously failing to do so. Fortunately, no damage was done. After reversing, the unit appears to be charging the 12v battery properly. However, the charge controller, being only a 7A unit, seems to be on continuously even though no load other than the BMS is drawing from the battery. So, it's hard to establish that this circuit is charge limiting. At least I know the 12V pack, initially charged, was at 12.9V. I'll be measuring that voltage again later.

For now, the pack charger is removed when not in use. A mount will be constructed and the charger installed into it. After, it will be time to take the BugE out on it's first drive with the new pack.

On my BugE, all lighting plus the BMS is directly powered by the 12V battery. I did this since the battery safety cutoff circuit should stop all power being drawn from the pack including the DC-DC converter. That means in the event of a battery cutoff during a journey, not only would the BugE have no power, but it also would not be able to signal to other traffic that it had a problem!

I built my own wiring harness. The 12V system was built to be completely independent of the bottom of the BugE. That way, during assembly, I could just plug the + and - into the cowl lighting to test all lighting plus speedometer indicator lights without requiring the rest of the BugE to be present. Both cowl lighting and BMS run directly off a12V accessory battery (autocraft part#5L-BS). The size of the battery was picked due to it being on sale rather than being an optimal size. However, that battery size seems to work just fine. To charge the battery, I use a Sevcon DC-DC converter to feed into an ICP 7Amp solar charge controller. Not that either are optimal components. Both are much more expensive than are needed. However, to preserve the ability to go back to a lead-acid pack with no need for a supplemental accessory battery, I decided to keep the DC-DC converter arrangement as-is for now.

Post construction note:
If I were starting from scratch again, I would not muck about with a DC-DC converter at all. The weight savings isn't a big enough benefit. Instead, I would simply use a motorcycle battery with a separate 12v battery charger, mounted on the tail shelf where my DC-DC converter now sits.

Deciding pack orientation for lithium battery upgrade

When deciding how to install the batteries, it's useful to study photos. In the first photo is my setup for a lead-acid pack. Lifting up the rear, you'll see 4 Optima batteries plus Powercheq battery balancing modules on the shelf on the right. Note, the addition of a spatter shield wall that I made of fiber cloth with resin. I later added a rubber skirt for additional length plus a chain spatter shield. Since doing that, water has not been a problem for my BugE.

The next photo is from Allen Clark's installation. He decided to mount the lithium batteries to one side. This frees up some space for battery management electronics. In addition, I'm hoping I can fit a small 12V battery for emergency 12V accessory power. This would be needed in case the main pack goes into an unexpected shutdown while on a journey. If I was to use this configuration, the center hole in the front of the battery tray where the 12V harness goes through can not be used. Instead, a second hole would need to be drilled to the right. Fortunately, I left enough slack in the 12V wire harness that moving the pass-through location will not be a problem.

So, to check the arrangement of battery electronics and to avoid doing too much in a confined area, I decided to set up the lithium pack and electronics outside the BugE first. One issue I ran into was the inability of arranging a battery pack in series just by using the provided jumpers. I finally just made a cable to complete the series circuit (the brown cable on the bottom). The arrows trace how the jumpers make the cells into one big series circuit. The arrow path shows how the circuit begins at the (-) terminal and ends at the (+) terminal. Also, note the small stubby screwdriver on the lower left. When it comes to working on batteries, it's safety first! Either use a small stubby tool or at least wrap the metal stem of a regular screw driver with electric tape or heat-shrink so accidentally dropping it on the pack can't cause a short!

An alternate way of arranging batteries was suggested by Baka Nihao although his batteries are different in shape. The batteries in his tray are arranged in parallel and held in place for the season with "great stuff" insulating foam. I asked about battery management. He uses none. He also brought up an interesting point on having a separate 12V battery for lighting. If I was going to use a 12V battery for lighting and have it charge from from the main pack via a DC-DC converter, the battery will be in danger of being over charged.

Unlike Baka's setup, I have a 12V battery management system that will need continuous 12V power to balance cells in the large pack even if the pack needs to be shut down. So, I'm using an independent 12V motorcycle battery for powering the BMS. I'm going to try using a solar charge controller that I have left over from a previous project.

The LiFeMnPO4 pack and BMS arrives. What next?

Note: This BMS system is no longer sold by Elite Power.  However, the batteries appear to still be available from electricmotorsport.com. As for my experience below, the batteries, charger and BMS came in two boxes. The batteries have removable purple covers. There is extra room in them. So, it's likely a different type of BMS boards could be installed under the covers to make the wiring look nice. Each set of four cells came pre-connected with metal bands on them (as shown in yellow box). Measuring the 4-pack gave a voltage of 13.1V . Also the BMS kit had a small LED bar graph display (middle black thing). Also, the vendor provided a larger LCD display as well. The reason it's "free" is that it has a display defect. It works, but it looks kinda ugly. Should I want to have a better looking one, I can purchase another one later. The BMS circuit boards came pre-wired, probably because I bought a "demo" system rather than a new one. On a new system, it's likely I would need to wire each board as I found it in the string. So, this is the "big picture" of how the battery management system should go together. Rick Suiter, my sales rep, emailed me instructions for a 24 cell model instead of a 16 cell model but I got the general idea. Like most things, these representative images are simpler than reality. Since I need to make some decisions on mounting locations, I decided to lay out the components. The LED bar graph looks like it can be press-mounted into a rectangular space, perhaps somewhere on my polycarbonate dashboard. The "Control Unit" also has a mounting hole for a screw. However, the "hub" has no native way to mount it. So, I can mound it somewhere with double-back tape. Wire length may be the deciding factor on where things get mounted. There were also some pieces that puzzled me. I got an assortment of additional metal pieces that I take to be conducting strips that work with the circuit boards. I'm also thinking having these strips will allow me to change the battery arrangement to have a wider or thinner pack. There were also some flat to round adapters. They puzzled me. Why so many? They seem rather wimpy for a mechanical connection. However, wimpy they appear to be, Rick assured me that the parts would be sufficient for the job. Plus, he sent a photo too. He states below:

For #4 wire you can fold the edges around the cable and crimp it with a die size one or two larger. I've done this with 4ga wire and it seems to work well. See the attached picture. Soldering is an option, but there seems to be issues in a vibration intensive environment where the solder joints can fatigue and micro crack. If you did want to solder I would crimp first then back flow solder in to the wire, that way you are not relying on the solderto hold the connection, it is just there for extra conductivity.

The other mysteries had to do with recommended interlocks to prevent the pack from being damaged from too much discharge. Looking over my parts, I found I had a mysterious small heat-shrink covered component which was the reset switch. Since it was too small to be captured by my camera, I used Sketchup to doodle out an approximate image of what the component looks like. On each end are the letters "L" and "D". So, from the response below, the "D" was supposed to be an "O". The little black piece with L and O on it is the low voltage cut off switch. ... The L side connects to the LIN hub, the O side is your output signal. You must provide a 12 volt and ground signal to the output side per the diagram, the middle wire is your switched signal for low voltage. I attached a document for the alarm cut out, it is for another version of the hardware, but the circuit on the last page will work to amplify the signal to drive a small relay directly if it is useful to you. Our warranty requires that the low voltage cut off be connected such that it disables the vehicle if a low voltage cell occurs. Speaking of loads, in the diagram, there are two DC-DC converters. One small one seems to be for only powering the LIN hub with "control unit". This converter apparently remains on 24/7. The other is the high capacity 12V DC/DC converter for everything else. The larger converter appears to be able to talk to the LIN hub as well. If a lithium cell either exceeds it's temperature or falls below a safe voltage, the BMS should "disable the vehicle" automatically. To do this, Rick offered the following suggestion. In most cases the alarm switch signal will be used to drive a relay which will either break an enable signal to the vehicle main controller or break the power to a main contactor coil. Since the 5 milliamp signal is typically insufficient to directly drive a relay the below circuit can be used to amplify the output signal enough to drive a relay coil.

This circuit can be built with Radio Shack parts:

MOSFET IRF510 Transistor Catalog #: 276-2072 0.5-Amp SPST Reed Relay at 12VDC Catalog #: 275-233 1N4001 Micro 1A Diodes Catalog #: 276-1101 1M Ohm 1/4-Watt Carbon Film Resistor Catalog #: 271-1356 However, warranty or not, it's safety first. I don't have a problem with the traction motor loosing power if the pack reaches it's discharge state. However, if I'm gliding to the side of the road, my BugE needs hazard lights to warn the other drivers behind me that this BugE is goin' down! So, how can I provide power to lights while the main pack recovers? Well, I need to re-think the decision of having a single DC-DC converter versus having a small 12V battery, charged with a 12v DC/DC converter. With Optima batteries, I think having a converter with no separate battery is the best approach. It's proven to be simple and effective. However, with a fussy lithium pack, a supplemental 12V battery is needed. So, this will require some changes to the wire harness to add a battery, charger, charge regulator and interlock. Then, there will be the challenge of finding a place on the BugE to mount a small motorcycle or gel cell battery. As they say, nothing is simple. As for where to mount the other components, it looks like the LED bar graph could be mounted on the transparent dashboard where the existing 48V meter is. I could mount the larger LCD display on the dashboard as well. However, given the condition of my "free" display, it's likely I'll only use it for testing but not for everyday driving. So, I'll probably Velcro that display to the glove box area so it can be easily removed when not needed.

From Lead to Lithium - trading simplicity for performance

I've decided to trade simplicity for performance. The lead-acid pack with battery management circuitry was very easy to maintain. Just plug in & charge. Lots of warning if the pack was being overdrawn and there was no need for temperature monitoring. They were durable batteries but they just don't give me the range I want. So, I ordered a set of higher-performance batteries plus a battery management system from Elite Power.

16 GBS-LFMP60AH LiFeMnPO4 - (4 banks, 60AH each)
1 BMS-16S battery manage system (no display)
1 EMC48-15 charger
With shipping, it all came to $2083.88 As a bonus, they did throw in a slightly damaged display. It's missing a couple LCD lines but it still has some use to it.

Even still, the whole system is quite a chunk of change! However, it will be worth it if it finally allows me to be able to use the BugE as a commuter vehicle. Over time, I hope to save enough gasoline to justify the cost of the pack. If gasoline prices stay where they are, that should not be too hard to do.

As a performance comparison, my real-world test range on a slightly damaged 2- year old lead pack was 12 miles with varying speeds & hills. Comfortable around-town range seems to be 10 miles one way. So, from the anecdotal evidence of lithium performance, I'm expecting an improvement in range due to...

1) not being damaged, unlike my present lead-acid pack
2) being somewhat lighter than the lead-acid pack
3) being able to use more of the capacity without damaging the pack
4) having more charge/discharge cycles before performance deteriorates

So, it's not beyond imagination to think I might be able to make the commute distance of 20 miles depending on how much energy I loose to wind resistance and hills. In order to do it, I'll need a re-charge at work for the return journey.

My journey will consist of roads that have rather easy slopes and allow me to drive at low speeds most of the time. All roads I've picked typically have good visibility and people are used to looking for roadway hazards such as deer or slow moving farm equipment. I have no doubt I could complete the journey at 20mph but be out of energy if entirely done at 50mph. So, the task is to find an acceptable mix of faster and slower speeds to allow the journey to happen in a reasonable amount of time.

The BugE returns to the roadways!

This time of year it's hard to predict the weather. A day might start out nice, then be rainy and cold. When this happens and I'm riding on my petrol motorbike, I go through an extensive 15 minute "dress up" and "dress down" procedure. Options include using chaps, rain suit, boots, additional layers of clothing, gloves with liners, earplugs, sunglasses (more for eye protection due to bugs than sun protection) and applications of RainX and NeverFog to my helmet visor. This is all needed to deal with rain, wind and noise that comes with riding a conventional motorbike. Today, thanks to an inaccurate weather report, I needed every bit of that protection to get home from work today. After the biker "fun" in dealing with all the wet gear after the trip, I realized I still needed to go to the monthly Green Drinks event I host in the area. However, it was still raining!

So, instead of taking my car, I decided to drive the BugE. The contrast between my earlier motorcycle ride could not have been greater. I just unplugged the BugE, put on a light jacket and was immediately on the road. No protective motorcycle accessories required (other than a helmet). Even with a steady 36 degree drizzle, I arrived at my destination neither cold nor wet. Nice!

Winterizing the BugE

Roadway salt is harsh on all vehicles. So, once the first salt carrying snowplow rumbled by my house, I decided BugE season was over. No special steps for winterizing were needed. Just park it in the corner of a garage with a battery maintainer plugged in and cancel the insurance. Cover it, check it once a month or so. Once spring arrives, I'm hoping all I need to do is re-insure & go!