Sunday, November 21, 2010
The original BMS didn't have any wire protection, if for some reason a wire got shorted out and burned up, you'd have to rerun that wire. I added small inline fuses to each cell at the battery. It's a little more cost and time, but something that's worth it in my opinion.
Here is a picture of the completed front end. Six batteries were removed from the original design to all for the AC compressor and pulley assembly. The good news is I finally have AC, but the news is it's winter time and I have to wait awhile to use it. However, part of this was to add the heater and on the cold mornings it really heats up the small cab space of this car quickly.
The motor controller was modified in multiple ways. First the Curtis "whine" was removed and the car is now completely silent, as an electric car should be. The maximum voltage was increased to allow for my fully charged voltage on 54 cells (I charge up to about 184v). Finally the amps was increased from 500 to 750. This made a huge difference in acceleration and overall power. My 0-30 went from (I think it was 7 seconds) down to less than 4 seconds. Overall the total power of the controller went from 72kW to just over 130kW.
I removed that large, and unreliable BMS screen from the dash and replaced it with the E-xpert Pro. This has turned out to be a great little display. On top of showing pack voltage, amps, Ah, and remaining runtime, it has some other extras and is highly configurable. You can customize this meter to your battery, driving needs and how hard you want to push your battery by setting what is considered full, empty, when to raise alarms, etc. So for example I setup mine to alert me when the pack is 30% SOC and to consider 20% SOC empty. This affects how the "fuel" gauge displays its bars. I also turned on a feature that will automatically add the back light as long as >1 amp is going through the shunt. As I drive the back light is on, and 10 seconds or so after stopping it turns off. I also configured the meter to consider the battery full when 184v is reached and the charger is at about 2 amps for at least 1 minute. This then resets the Ah counters and gauge and avoids any calibration errors that might allow the meter to slowly drift over time.
Here is the trunk now. I wouldn't call it finished because things are still a little messy and I'd like to get carpet back in there. Trunk space is still reduce, but I can at least fit something in there now if needed. The little black things on top of the charger are the new Cell Log 8 modules that I'm using to monitor high and low voltage conditions. These are about $13 a piece and will monitor 8 cells each. I build a little circuit board for each to simplify the wiring. They have an internal relay that allows you to provide signalling to the charger, a buzzer, light, etc. They are not isolated, however, so expect to use an additional relay on each module to isolate it. These modules will turn on when the charger is connected to AC power or the key is on.
Another thing I modified this round was the suspension. I added another 9 cells (over 100 lbs) and with so much additional weight in the rear I knew I couldn't avoid it this time. Online I found the spring rates saw that the rear springs were much stiffer than the front. The front springs were 245 lbs/in while the rear springs were 311 lbs/in. I moved the rear springs to the front of the car and replaced the rear with 10" (original springs are 11" unbound) 400 lbs/in springs. The put the rear exactly back to stock ride height and my front is still 1/4" lower but I think within range for alignment. I still need to take the car back down for an alignment to see if this will do.
It's great finally being back on the road. The gas to drive my truck 70 miles to work each day was really adding up. That and I just missed my car, too much fun to drive. I'll keep you posted on miles. I think around 16-17k miles total at this point.
Wednesday, September 8, 2010
I plan to ensure adequate ventilation this time and have installed two 120mm fans. They will connect directly to the 240v charger source and run continuously during the charge cycle. The fans are quiet (30 dB), low speed (1900 RPM), move up to 67 CFM and take 7.5 watts of power each.
Here is a shot of the mounted fan. The trunk is big ugly at the moment but I plan to reinstall the carpeting after the modifications are complete.
Here is a shot from the bottom of the trunk. I've installed a vent which will deflect water spray while driving. They are angled to take the incoming air which should close the vents as well during speed.
Thursday, September 2, 2010
My plan is to install four small fans directly above each set of brushes that turn on when the motor heats up using a thermostat. The thermostat is basically a temperature driven switch that will trigger a relay to turn on the fans. I selected 40C or 105F as the turn on. This will keep it from turning on during short trips but will ensure it starts cooling as soon as possible.
Here are the fans I used. They are 2" fans that do 20 CFM and draw .25 amps.
Here is one of the mounted fans and the thermostat. I made a bracket that attaches to the old temperature sensor and the thermostat bolts to the bracket.
When I manually enable the system I can feel a good breeze coming out the back the motor so that's a good sign. We'll see what happens after I get it back on the road.
Monday, August 30, 2010
When I got the original cells I was told the factory capacity tests were all above 160, if I recall they were mostly around 165Ah. I charged a group of four new cells the same way I did the old cells. That is, not long after switching from CC stage into CV stage I remove charge and begin testing. The old cells were drained at ~130 amps until 2.5v and the worst cells yielded 145Ah, most around 150Ah. The new cells I only drained down to 2.8v and the reason is that at 2.8v they had produced 183.5Ah!
This raised a few thoughts of course and along with the fact they have improved the cells, I'm lead to believe that Sky Energy (CALB) who sells an equal size and weight 180Ah cell for more money is most likely selling the exact same battery. Unfortunately I don't have the 180Ah cells to test, perhaps they are getting 190-200Ah. I believe the batteries might be coming with more capacity also so that after time, they are still considered to be at their rated capacity or higher.
Friday, August 27, 2010
The first and probably hardest step determining how power the compressor. My first option was to modify the compressor to use a standard V belt. In this case it was easy to create a main pulley to attach to the motor. However I determined, after dismantling the compressor, it was going to be very difficult. My second option was to find/make a pulley that could use the same style of serpentine belt the compressor used. I opted for this method. I used the original pulley which accepted a spline shaft. The ADC 9" motor has a much smaller, round keyed shaft. I bought a 3/4" hub from tractor supply. I was able to get my uncle to turn down the hub and we then pressed it inside the original pulley with a hydraulic press. We ran a weld around one end to make sure it could turn inside. With this done it was just a matter of fabricating a frame that mounted the compressor and tensioner to the motor frame.
Here you can see completed setup. Reusing the tensioner instead of trying to make the compressor adjustable is the best route. The tensioner keeps the belt at the ideal tension as it stretches with age. This setup should be maintenance free for years.
Here is a shot from the other side.
I ordered from Jegs and got a Gates belt K060345. This is actually a bit over 35" and is the six v serpentine style belt used originally, just much shorter. This was stocked and I got a "free" hat. Considering their "free" shipping had a $5 handling fee, I'm not sure which one of the two I actually got free.
Here is a 12v functional test video. It's not nearly as noisy as this little camera picks up.
After everything was functional I needed to charge the AC system. The system had been opened for months, and even if it had only been opened for a day or so you need to do a lengthy evac process. An oil based vacuum pump is required which I bought one from Harbor Freight. You'll also need an AC manifold set which I also got from Harbor Freight.
The process is fairly simple. First connect the high(red) and low(blue) side hoses to the A/C system. The sizes are different so you can't accidentally connect the wrong one. This manifold set has a yellow hose and two spots where the yellow hose can connect, one is open and the other is a pressure fitting. Connect the yellow hose to the open side and then to the vacuum pump. Turn on the vacuum pump and then open the high and low valves. Let this run for several hours, the longer the better. The pump will get the system down fairly quickly, but it won't remove all the moisture this fast. You must leave it running to remove the moisture (3-5 hrs will do it). Once complete, first close the high and low valves, then turn off the vacuum pump.
You'll need to check the service manual to determine how much refrigerant should go into the system. You'll also need to replace any oil that was lost. There are refrigerants that contain oil, but it's not enough for a complete recharge so you'll need to do the math and figure out what you need.
The recharge process is simple. Read and follow the directions on the can first and foremost. Disconnect the vacuum pump and connect your refrigerant. The can should be shaken during the entire process. Open the valve on the can, then open the LOW side ONLY. Never open the high side valve while recharging. You'll hold the can upright and rotate 90 degrees every few seconds again given the can a good shake frequently. You'll want to make sure there is enough refrigerant and oil in the system before turning on the A/C system (again consult your service manual). Most systems should have a low pressure safety switch to prevent this. With the A/C system on max you'll notice the low side gauge drop as it compresses the gas to the high side. As it pulls from the low side more refrigerant from the can will pass into the system. As the can empties it slows down so be patient 5-15 minutes per can. Repeat this process to add oil and more refrigerant as needed. When complete close the low side valve, then close the refrigerant valve. The high and low side connectors can be popped off easily at this point, but when you disassemble the manifold some gases will escape so do this in a well ventilated area so you don't breath it in.
That was it, now I have some really cold air coming out of my EV!
Thursday, August 19, 2010
- Greater range. More cells means I have more energy and can travel further before recharging.
- Distributed load. The more cells I have, the less they have to work for my daily commute which will increase the life of the cells.
- Performance. Currently the 144v system allows the motor to hit about 4k RPM before the controller switches to VMax. Any additional increase in RPM greatly reduces torque requiring you to shift to a higher gear. Increasing the voltage will allow the torque to remain steady for higher RPM meaning the motor will want to rev out higher.
Here is a shot of the original trunk space. You can see there isn't much usable space currently above the fuel tank, about 3" and it opens up to only about 8-10" on the far right.
The old spot I use to have the charger has a curve to it and greatly reduces the usable space.
Sunday, August 15, 2010
Below is my battery testing solution.
- Costco a 2300 watt power inverter. This allows me to test four batteries at a time to get my 12v source.
- 1500 watt space heater. This gives me just over 130 amp load on the batteries which is close to 1C and will give a decent enough load to test.
- E-Xpert pro battery monitor from TBS Electronics with a 500 amp shunt. This will track the Ah used for me and give accurate results compared to trying to calculate this myself based off of ever changing voltage/amperage as the batteries drain.
- CellLog8 allowed me to easily monitor each cells voltage to determine when a cell was too low during the load test or too high during recharge.
- For charging I used my original battery charger, a 12v charger I had, and my bench power supply. This allowed me to charge up to 45 amps and helped speed up the testing greatly. I only did quick charges and once the amperage needed was between 5-10 amps I stopped charging and begin the test. This means the cells were NOT fully charged (probably 90 - 95% is my guesstimate) which is important when reviewing the results below.
My original testing plan was going to be to drain the batteries until the first cell reached low voltage, then rotate in another cell and continue testing to get results for each cell. After doing this only once I realized that the capacity of the cells was extremely close and it wasn't worth the extra time. I won't post all the results but basically my lowest cell produced 145Ah. I had other groups producing 148, 150, etc. They were all extremely close. Even the two cells that have terrible voltage sag still produced over 145Ah like the others. Now 145Ah is just over 90% of original capacity, however, remember my charging method was manual and I didn't give it the time to do much constant voltage charging. So to my surprise, I think I'm still near original capacity. The only thing is I didn't do this testing when the cells were new and I'm told they will have at least 160 but usually more, so it's possible I've lost more than I know. Either way this is good news and I'm quite happy with the results.
Monday, June 28, 2010
I've been in touch with a great guy who works at Curtis. This weekend I had the opportunity to test a modified Curtis 1231C. Unfortunately I only had a 500 amp shunt so we weren't able to test the motor side amps. I did see my battery amps peaking somewhere around 650-700 though as the controller reached Vmax. This controller reduced my 0-30 to 5 seconds. I will add the shunt later and get the real output from it soon.
He also has an even higher end modified 1231C, but we were having some issues with it. It should be 200 amps more or so. It would be nice to see a 3 second 0-30 :)
Unfortunately I can't give out any of the secret sauce that is being used to upgrade/modify these controllers since I didn't obtain the information myself and don't want to get anybody in trouble.
We should be doing some more tests with the 2nd upgraded controller later and I'll be sure to have the shunt ready for some numbers. I realize most people won't be able to make these modifications, but it should give you a good idea of expected acceleration for a given amount of continuous amps.
Tuesday, June 22, 2010
The increase in temperature has made a little improvement to the increasing voltage sag. I think the bottom line is I'm pushing these cells harder than I should for longevity. I believe my commute to work is using >80% capacity and this is taking a toll on capacity. Some mornings I have cell that is very close to fully discharged. I don't know how many miles I have left before at least one cell can't make it and possibly gets ruined to get me here.
I keep going in circles, but I'm leaning towards not selling it now. I'm thinking of making some modifications though now that I see how reliable the conversion itself is. Nothing official yet but here are some ideas I'm toying around with.
Motor controller: The Curtis 1231C has been reliable but I want to get the acceleration back in this car. I'm leaning towards the Soliton1 again for a few reasons.
- It will increase my motor side amps from 500 to 1000.
- It's much larger and has great cooling with the option of water cooling if needed. In the hot summer days on my long commute doing 70mph the Curtis is kicking down the amps sometimes.
- It has many features that will make protecting the batteries, and motor a breeze. I'd like to add some features, but reduce complexity of the build and I think this will be a good step in that direction.
- I'll have the ability to run a pack voltage that is much higher than the voltage applied to the motor. This allows you to run more smaller batteries allowing you to still get the capacity you want, but make it easier to fit them. It also will remove volt sag during hard acceleration. The pack itself will sag, but if your pack voltage is high enough the motor will receive a steady voltage increasing performance even on those cold days.
I think I'll end up cutting out the current trunk and expanding it. There is a LOT of room that is not being utilized under there. This will allow me to add more batteries and even move some of the front batteries to the back giving me room to finally mount that AC that I've been missing.
Currently I'll just add some more cells (I'd like to add another 15 or so) so I can get as much from the existing cells. This should ease up on them and extend there life some on my commute. This will give me not only more capacity but the ability to crank up the motor voltage a bit also to increase high end performance along side of the low end performance I'll gain from the controller change.
I'd like to get the car to a point where I don't have to look at anything and worry about what's going on. I want to simplify this side of things most of all. I want to be able to get in the car, look at my "fuel" gauge and know if I'm good to go. If there are any issues, then be able to connect something and diagnose the problem. I'm looking into some options for a low cost over haul here. The biggest problem with adding more and more cells is the complexity of protecting them.
As always, anything I do I'll document and share with you.
Tuesday, April 20, 2010
I decided to do some quick math to see what 12k EV miles really means. The car originally got 26mpg on my commute. 12,000 miles/26 mpg = 461.5 gallons that I didn't burn this year. Gas in my area has been more or less $3 over the last year so that's $1384 dollars I would have spent on gas.
Although it's great to not have given my money to the oil companies or burned the gas, the actual conversion cost vs returns would never pay off before something needs to be replaced (i.e. batteries). EVs of some form are definitely the future, now we just need to wait for all the auto manufactures to get those production models out which will reduce costs. 2011/2012 will be a huge turn for the auto industry...I can't wait!
Wednesday, March 10, 2010
You may recall I reported two cells which sagged considerably more than the rest. I figured these would have went out by now but they are still working fine. On the down side, I now have two more cells that seem to sag more than before (about 0.4v under 2C load lower than the rest). I don't plan to replace any of them until they go out completely so I can really test them out. Speaking of replacing them, I noticed that Elite Power Solutions has finally dropped their prices to be more competitive with some of the other suppliers. We'll hopefully see a continuing trend making this option more affordable for all.