I received my grid charger/power supply from Acopian this week and got to work to intall it to realize the benefits of grid charging. I tapped the battery at the pre-charge resistor so the current can charge the battery with the ignition off and it goes through the SOC current meter with the ignition on. Thanks to retepsnikrep for the great article on tapping the battery (viewtopic.php?f=4&t=7055).
The charger is an isolated 180vdc, 200mA power supply (https://www.acopian.com/single-u-goldbox-m.html) only costs $85 plus $15 for shipping and can be used in series or parallel configurations if you want to add multiple units. I used 200 volt, 6 amp diodes on both ground and positive taps, 10 guage wire and a 20 amp fuse right near the connection. I plan to install a smaller fuse outside of the battery compartment so I don't have to remove the battery cover in case of an oops.
I was experiencing frequent lower end recals and figured the pack could use a ballancing top off charge as a side benefit of the grid charging. I don't plan on topping off to 100% every time, just every few weeks as needed.
After installing the unit I figured I could portably power it with a 300 watt Porta-Watts inverter I've had for a dozen years or so and the Optima Blue-Top battery from my ski-boat. I'm going to call this an I-Boost system (for inverter boost, DC-AC-DC) so it is not confused with Mike's V-Boost which is DC-DC. I plan on upping the charger wattage, inverter size and battery capacity after my 'proof of concept' is completed and bugs worked out.
So far with a week of driving, no negative recals. I was only getting them when using heavy assist and down to 5 or so bars on the SOC. I began using heavy assist after the charger install by climbing hills in 4th and 5th instead of 3rd and getting assist to come on in 5th while accelerating and then carefully backing off on the throttle to get to 55-60 MPG and keep the assist going. I used to do just the opposite, when assist would come on I would quickly back-peddle to get out of assist and ease into the throttle to 50 MPG to stay out of assist. I have only had the car for 6 weeks and I have logged most of my trips to track improvements and refine my hypermiling techniques. I reset trip A to track my lmpg as I bought the car from Billy, a very accomplished hypermiler and he handed me the car with a 86.4 lmpg with 35k on it. Billy spent about four hours teaching me to drive the car 'correctly'. I started out at 75mpg and now have the 'A' counter to 84.3 with about 2500 miles counted.
My typical trip when I'm home is 16 miles uphill to Mt Shasta, 1950 ft to 3600 ft elevation and return. I'm able to FAS for about 9-10 miles on the return trip so I can effectively use up a grid charge on the uphill and not put it back with gasoline if I want. I have achieved 80 to 81 mpg on the round trip all three times I've logged it. My first trip with the grid charge top-off was 86.6 mpg for a solid 6 mpg gain. This was without a +recalI after charging so I was in forced charge for the entire trip up. I think I actually did the uphill faster than before as running at 55 mpg with assist on climbs about 5 mph faster that 50 mpg with no assist. Very encouraging, considering the additional weight in the car (about 55 lbs).
The only problem so far is trying to get a +recal. Some members reported that a +recal could be had by turning on the ignition for a few minutes before starting the engine after a grid charge. Mine doesn't do this. Some have also reported the SOC would update in the first few miles of driving, mine will do this but only if the first few miles includes some significant regen >25 amps. If I leave my house and go south on I-5, I drop about 400 feet in three miles and it will +recal. If I go north, I start climbing right away and I don't get any oppertunity to regen for the 16 miles up. The SOC goes down to 3 bars and stays there for what seems like forever. It still lets me use normal assist at 3 bars for a few miles of climbing then eventually backs off a bit but still at 3 bars SOC. On the return trip, if I regen for about 3 miles of the steep Dunsmuir grade, the SOC slowly climbs to about 1/2 then jumps up to full in a few quick two-bar steps.
It seems the BMS won't report a SOC lower than three bars unless the pack actually starts suffering a low SOC. It also seems that it won't do a +recal unless it realizes that the battery 'acts' full while giving it a decent amount of charge.
I tried pulling both IMA fuses to see if the SOC would update after grid charging, no luck. Before going any further I thought I would get this post up and seek help. My main concern is pumping 50 amps of regen into a full pack to get the +recal, probably not good for the batteries.
I really would like to get a +recal before leaving the house without leaving the ignition on during charging, and with the I-Boost, get the SOC to update after short stops leaving the I-Boost on.
Future plans:
Up the booster battery capacity, 2500w inverter and maybe a manzanita micro charger with control through MIMA or from the drivers seat to avoid the zero crossing current sensing. This would be injected on the motor side of the MPI current sensor with the ignition on, direct to the 144v pack with the ignition off with a simple relay setup.
It seems the BMS won't report a SOC lower than three bars unless the pack actually starts suffering a low SOC. It also seems that it won't do a +recal unless it realizes that the battery 'acts' full while giving it a decent amount of charge.
Nice work but a couple of points. I see the psu you are using is unregulated with a voltage range from 190-162v for the 180v model you selected. 190v seems a bit on high side, I would though a lower V one would be better so that current tapers off more naturally as batteries approach full charge. Just my opinion of course. How is the psu current limited when batteries are low? Is it a linear or switch mode psu?
I have also noticed the 3 bar phenomena you mention above. I use MIMA to push pack to full on occasion, and then under load it will drop steadily until it reaches three bars and then remain static for quite some time before finally dropping the last three bars.
Nice work but a couple of points. I see the psu you are using is unregulated with a voltage range from 190-162v for the 180v model you selected. 190v seems a bit on high side, I would though a lower V one would be better so that current tapers off more naturally as batteries approach full charge. Just my opinion of course. How is the psu current limited when batteries are low? Is it a linear or switch mode psu?
Thanks for the input, the psu is unregulated but I don't think it will be a problem. It is rated at 200 mA at 180 volt, zero at 190, and it does steadily increase as pack voltage drops, there is no current limiting. In practice, with a 500 mA meter on the DC output and a 'kill-a-watt' meter on the AC side of things, it acts a little different than the specs. I have charged the pack from the grid with 152 volts (210 mA) start to about 167 volts read at the psu output (with two diodes the pack must be a few volts lower). After reaching 167 volts the voltage steadily drops down to 163 volts and the current drops as well from about 190 mA to 150 mA. So it seems when the voltage dip is reached the impedence of the pack changes enough to limit the current. At these current levels I don't think any acute damage is likely. I'm guessing that the +recal is triggered by this relationship of the concurent voltage-current drop.?
With the I-Boost, things are a little different. My inverter is a cheap modified sine wave unit and the psu changes it's output with this AC supply. The output goes up to 280 mA from 190 mA and draws a proprtionally equivelent wattage increase reported by the kill-a-watt meter. It also buzzes, where it was totally silent on grid supply. I was concerned about overloading the unit so I emailed Acopian about it. They stated that these units are extremely overbuilt and it will simply blow the fuse long before any damage is done. As I suspected, 'more area under the curve' for the output waveform from my inverter maybe why the output increases with the inverter. The buzzing is due to the 'steep steps' of the AC waveform, no problem as well. I'll just count this as bonus output.
While driving with the Boost on, I'll see from 25 watts on the watt meter during regen, to 50 static, and up to 100 watts with a low pack and full assist. At 100 watts, assuming 85% combined inverter-psu efficiency that's maybe 130 volts at 650 mA. I know there is lots of info about using constant current to charge NiMH, but I think a stock insight pack almost never, if not actually never, sees a CC charge bottom to top. I beleive that at these current levels, there will be no harm.
If and when I step up to a large I-Boost system (10-15 amp), I will charge the 144v exclusively with the BCM on and all safeties in place. It should restore the SOC fast enough from the grid, it will be no hassle at all to just sit in the car and top up.
Still need to figure out how to +recal in the current configuration before starting the engine. Anyone?
As for being concerned about the IMA doing regen with a full SoC pack... I don't think it is an issue as long as the stock BCM and IMA are intact... When I fully charged the 20 individual sub-packs and put them back together into the pack and back into the car... it did a recal yes... but the car refused even with MIMA on full regen or when braking down a steep hill to do more than about 15 to 20 amps of regen as seen from the MIMA display.... after the car corrected for the SoC it seemed over eager to give assist and pull the SoC down and reluctant to give much regen even during braking or MIMA use.... It took a little while before IMA operation returned to normal... Interestingly it recovered to normal operation faster and had less over assist and under regen when on the second time I let the car sit and idle for the SoC correction.
Now over charge could be an issue with the grid charger over a long enough period of time... but it seems while the car is on and the current is controlled by the IMA system and BCM that it is not an issue.
As for being concerned about the IMA doing regen with a full SoC pack... I don't think it is an issue as long as the stock BCM and IMA are intact... When I fully charged the 20 individual sub-packs and put them back together into the pack and back into the car... it did a recal yes... but the car refused even with MIMA on full regen or when braking down a steep hill to do more than about 15 to 20 amps of regen as seen from the MIMA display.... after the car corrected for the SoC it seemed over eager to give assist and pull the SoC down and reluctant to give much regen even during braking or MIMA use.... It took a little while before IMA operation returned to normal... Interestingly it recovered to normal operation faster and had less over assist and under regen when on the second time I let the car sit and idle for the SoC correction.
Now over charge could be an issue with the grid charger over a long enough period of time... but it seems while the car is on and the current is controlled by the IMA system and BCM that it is not an issue.
Thanks IamIan! That helps some of my worries and seems consistent with my car's behavior. This second time you refer to, was that a grid top off or pack removal/pack switch off? How long did the SOC correction take sitting at idle?
I had tapped the wires into my pack a few weeks prior to my psu arriving (no charging) and upon re-powering up there were no SOC bars and It did recal, but I'm trying to avoid this forced charging. I think I will open the pack back up when I get some time and run a third wire out so I can install an ignition powered relay to switch the charge point and try charging with the ignition on through the switched positive output terminal. I'm not thrilled about the idea of leaving the ignition on for hours while charging but as this is the 'proof of concept' trial, I'll put a charger on the 12v battery as well and see how it goes.
I wasn't thinking this far ahead when I tapped the pack, the inverter booster plan I got from the lengthy grid-charging thread from you guys back in 2006, I cannot take credit for the idea. The nice part about it is it is all off-the-shelf stuff, engineered, with safeties and the efficiency is pretty good.
I forgot to give some battery performance information. The Optima blue top is giving me about 4 hours run time at a rough average of 60 watts on the watt meter. The inverter shuts down at 11.25 volts after a few minutes alarming. My abused 6 year old blue top is rated at 55 AH at a 20c rate and taking it down to 10.5 volts.
55 AH x 12 volts = 660 watt hours
60 watts x 4 hrs / 90% inverter efficiency = 266 watt-hours realised, not very good! I think my blue top is tired, ran the ski boat just fine though.
I've got a group 31 Sea-Volt dual purpose marine battery (flooded pba) as well, just checked the date, jan '01. I'll give it a try sometime as it is rated at 80 AH and I've taken better care of this one although she's old.
Thanks IamIan! That helps some of my worries and seems consistent with my car's behavior. This second time you refer to, was that a grid top off or pack removal/pack switch off? How long did the SOC correction take sitting at idle?
Your welcome... that is the nice thing about community forums like this ... different people can share their information and knowledge and everyone benefits.... I know I've benefited from other peoples work on here several times.
The first time I pulled the pack out was the test the taking the pack apart down to the sub-pack level and using the triton to cycle test and charge the batteries... The pack used was one of Mike's .... this revealed the SoC issue between sub-packs... and with some cycle exercising with the Triton all the sub-packs came back up to about 6Ah or so... then with them all near 6 Ah of capacity... and all fully charged I put them into my Insight to see if there would be an issue with rebuilding a aging pack like this and to test to see how the BCM and IMA system handle a 100% SoC on the HV IMA NiMH battery.... While I had the pack out I also pulled one of the PTC strips and rand some primitive tests on it.
While getting some real world driving behavior effects from the rebalanced and rebuilt battery pack I did the same thing with my own stock HV IMA NiMH battery... after a week or two I swapped them so that my original was back... I did the idle SoC correction instead of just taking off like in the first one...
I have not done a grid charge... I live in a apartment and do not have an exterior electrical outlet I could use to do that... One of the limitations on the PHEV Booster Battery I am planning is that I will have to disconnect and carry it into the apartment every time I want to grid charge it.... Eventually several years from now when I live in a house again a grid charge system would be a nice option... but for now I can only move forward with the portable / removable booster HV PHEV battery system.
I did not time it ... but the idle SoC correction time was not long... Definitely under 5 minutes... and it seemed closer to about 2 minutes if even that long.... but I did not time it... probably should have.... Try to remember that for the next time.
My grid charger will probably evolve over time. But for the first attempt I'm planning on using a Variac + bridge rectifier, and a 33 ohm resistor array to limit current. This simple design will drop current as the pack comes up to what the rectifier is feeding it. I like the pricing on the one you found, but 200ma is kind of lacking for 2 packs.
Setting at 170 volts and starting at 140volts (empty), I figure it will take 10 hours to go from 140 to 160 volts, an additional 6 hrs to get to 165 volts, and an additional 10 hours to get to 168 volts. The charge should be slow enough that these should be stabilized voltages. If I need to slow it down I can either lower the Variac voltage, or raise the limiting resistor. For the second version I will probably take Ian's suggestion and go with an LM317HV. I will also use fuses on all mains. I am planning on HV diode isolation, and will install all but possibly the Variac in the electronics bay. The Variac would most likely be in the side storage area. I only plan on having an ac plug mounted near the license plate.
My grid charger will probably evolve over time. But for the first attempt I'm planning on using a Variac + bridge rectifier, and a 33 ohm resistor array to limit current. This simple design will drop current as the pack comes up to what the rectifier is feeding it. I like the pricing on the one you found, but 200ma is kind of lacking for 2 packs.
I admire your progress with the parallel pack, keep up the good work!
I did want to go with something with a little more current, I searched alot of websites and didn't find anything close to the Acopian unit for appropriate voltage and price vs output. They can be paralleled so more is still an option. This current level should keep me out of trouble though.
My work has a basement with tons of old DC tugboat stuff, large boxes of various motor brushes and large resistors, fuses and some very old radars. My boss said take whatever I want, I grabbed a few large (20-100w I think) resisters to add to my 'junk drawer', some have a slideable center tap on a long tubular coil. I'm in Tahoe for the weekend but will be back to work on tuesday, I'll look for something suitable for your application if you want. PM me with a shipping address, I'll gladly donate to the cause if anyone can use this old stuff. Someday they are just going to toss out all that stuff anyway.
I've got my plug above the licence plate, I can tuck it up inside or let it hang out to flaunt it- depending on mood. I first hooked the cord directly to the psu but then realized it would be convenient to be able to plug in an onboard charger for the boost battery, so I added a female plug on the inside. I then bought a standard outdoor AC receptical box that I will mount to the aft bulkhead in the shoebox area for a safe, secure multiple plug outlet. This will tidy up the wiring a bit.
I am happy to see so many people getting over their fear of fooling with the battery packs. This should allow us to make some rapid progress in resolving the many issues involved.
I would only like to add a few things.
1. The end of charge voltage dip will not happen with 120 series batteries, all at slightly different SOC, and with low current, so that is not going to be a reliable end of charge determination.
2. To really be able to use a V-boost to run at highway speeds continuously, and get over 100 MPG, one will need to have 20-25A of boost. My 16.5A and some careful use of MIMA will allow me to get 100MPG for the boost duration at ~55-60 MPH max, but the boost can just barely keep up.
3. The max voltage that the pack will reach is very dependent on the current applied. As the current drops, so does the max voltage. At full 50A regen, I have seen 180V, but with 200ma, I expect that one will never get much over 170.
4. The Acopian supply looks like a great start for a grid charger, but you are dumping 36W into the pack, and without regulation that could all be turned into heat at the end of charge. You may want to run the battery fan, as a minimum.
5. Armin sets a timer to terminate the grid charge based on his ending SOC when he turns of his car. This would also be a nice addition to take care of the times you have a nearly full SOC.
The AutoGuide.com network consists of the largest network of enthusiast-owned enthusiast-operated automotive communities.
AutoGuide.com provides the latest car reviews, auto show coverage, new car prices, and automotive news. The AutoGuide network operates more than 100 automotive forums where our users consult peers for shopping information and advice, and share opinions as a community.
Grid charger and I-Boost installed
I first hooked the cord directly to the psu but then realized it would be convenient to be able to plug in an onboard charger for the boost battery, so I added a female plug on the inside. I then bought a standard outdoor AC receptical box that I will mount to the aft bulkhead in the shoebox area for a safe, secure multiple plug outlet. This will tidy up the wiring a bit.
