Budget Bottom Balancing

Using a LightObject 5740 Volt Meter to build a $100 LiFePO4 bottom balancer

You will need a wall wart power supply, a power resistor, a high amperage continuous duty relay, a LightObject 5740 volt meter, a pair of 30A alligator clips, a pair of smaller alligator clips, some 8 or 10 gauge wire, and some small gauge wire.

Major components:
Wall wart power supply 16vdc 5w $10
http://www.digikey.com/product-detail/en/DA4-165US/454-1400-ND
.1 Ohm 200W resistor $28
http://www.digikey.com/product-detail/en/2-1630019-7/A102386-ND/
Continuous Duty Solenoid 80AMP 12V $16.50
http://www.amazon.com/Continuous-Duty-Solenoid-80AMP-12V/dp/B0050I94XG
Programmable 4 Digit Red LED AC/DC Volt Meter with dual control. $38
http://www.lightobject.com/Programmable-4-Digit-Red-LED-ACDC-Volt-Meter-with-dual-control-Good-for-HHO-System-P408.aspx

I built mine for considerably less than $100, I had everything except for the 5740 meter. I used a wall wart power supply left over from a dead router, several 0.1 Ohm 20W power resistors that I ganged together, and a contactor that will eventually go on my motorcycle pack.

It isn't very pretty, but it works. The resistors that I had, when connected 5P2S across a LiFePO4 cell, draw about 40A. The power resistor listed above from Digikey draws about 20A.

Below I have a JLD404 hooked up with a shunt to show the 40.5A draw at 3.24v with my resistors. I'm going to have to get some heavier wire to connect my resistors... 40A was making them get warm. A fan blowing across the resistors and wires helped a lot. I probably should add a fuse or two in there somewhere.

Here you can see where I rebuilt the balancer with the power resistor, better wires, alligator clips, and a fuse on the power lead. The JLD404 shows that it is drawing 22.7A at 3.27v. This power draw is closer to the recommended 0.3C discharge for the cell. 

Wiring the balancer:

1. Connect your dc power supply to terminals 1 and 2.

2. Jumper terminal 1 and 15 to connect the positive DC power to the common terminal (15) of J1.

3. Connect one side of the contactor to the normal closed terminal (14) of J1.

4. Connect the other side of the contactor to negative DC power (terminal 2).

5. Connect the negative voltage sense lead to the COM terminal (8).

6. Connect the positive voltage sense lead to the 10V terminal (6).

7. Connect one side of the contactor power terminal to the power resistor with heavy wire.

8. Connect the other side of the contactor to a heavy alligator clip with heavy wire.

9. Connect the other side of the power resistor to a heavy alligator clip with heavy wire and a 12V/30A fuse.

Programming the balancer:

1. Press Manual/Enter and enter 0001 to get the relay menu.

2. Set AH1, the high voltage trigger, to your lower voltage (2.60v).

3. Set AL1, the low voltage trigger, to your higher voltage (2.75v).

4. Select End and press Manual/Enter to exit the relay menu.

5. Press Manual/Enter and enter 0036 to get to the configuration menu.

6. Set PuL to 0.000 for no offset
7. Set PuH to 1.000 for displaying 0.000 to 9.999 vdc.

8. Set dot to 3 for 3 decimal points.

9. Set FILt to 0 for no noise filtering.

10. Select End and press Manual/Enter to exit the configuration menu.

When the voltage is higher than 2.75v the contractor will energize, drawing power from the cell. When the voltage gets down to 2.60v, the contractor will open, removing the load. If the voltage rises above 2.75v, it will close again. After several cycles, the voltage should settle just below the 2.75v.

Homebrew EVSE

One of the guys at the Dallas Makerspace has a Tesla Roadster so he decided that he needs an EVSE at DMS as well as at home. It uses an EVSE controller from the Open EVSE Project, a pair of solid state relays, terminal strips, 12v psu, panel box, and a J1772 cable/connector. He did an Instructable on it.

Below is a shot of it mounted next to the warehouse door. It normally has the lid to the box in place for safety.

CALB CA60FI spec sheet

Testing update

I've tested about one third of my cells, and all of them so far are in excess of 60Ah capacity. The cells come from CALB at about 50% charge, the PowerLab 6 takes about 6 hours to charge from half to full, discharge and recharge.

Below are some graphs that the PowerLab spits out, the raw data is also available, so I plan to look at the stats when I'm done testing the entire set.

Here you can see that the PowerLab puts in a little over 30Ahs, rests for 15 minutes, removes a little over 62Ahs, waits for 15 minutes, then puts back in a little over 63Ahs.

Here you can see the steep "knee" as the cell gets full, the plateau as the charger changes from constant current to constant voltage mode at 3.60v, the drop as the surface charge dissipates during the 15 minute cool down. Then we see the sharp drop as a 30A load is applied and the slow drop until the steep knee as the cell approaches empty, and a quick recovery as the load is removed at 2.80v for another 15 minute cool down period. In the final phase, we see a steep curve as charging starts at 30A constant current, then a long fairly flat charging curve and another steep knee as the cell gets full again, followed by the constant voltage plateau to finish.

The CALB paperwork lists this cells internal resistance as 0.44 mOhm. The PowerLab measured it at 0.5 mOhm. The other cells have been similarly off, and similarly flat on the graph. I'm not sure what that means, if anything.

More Power Parts

The hardware for the cells showed up, the bolts seem to be fairly good stainless steel, as are the flat and split washers. The battery interconnects are several thin sheets of copper stacked and heat shrinked together.

I don't like split washers in things that vibrate... like motorcycles, so I ordered some Nord-Lock washers.

I also ordered a JLD404 and shunt from LightObject. I'm not sure where I'm going to mount it yet.

Testing

I ordered a Revolectrix Powerlab 6 to test my cells.

Finally...

I've been busy, and not on fun stuff like the motorcycle. I've finally got some time to get back to work on the bike and ordered some batteries.

I really, really wanted to get Sinopoly LFP60AHA(B) cells, and got some good quotes from Sinopoly, but just couldn't get a reasonable solution to importing and shipping them to Dallas. I ended up getting the new CALB CA60FI cells from Keegan at http://www.calibpower.com/

They arrived UPS Freight in two small crates shrink-wrapped on a pallet. I removed all the extra packing material and consolidated them down to one small crate for storage while I'm working.

Unfortunately, they neglected to ship the bus bars and bolts with the batteries, so I'm waiting on those to show up in the next day or two.

I ordered 48 cells, and plan on installing the best matched 36 of those in the bike. The other 12 will be used on various other projects. I'll update on the testing soon.

According to the CALB paperwork they all tested 65Ah to 67Ah. Two were 65Ah, 25 were 66Ah and 21 were 67Ah. The "Open Circuit Voltage" listed was 3.304v or 3.305v for 45 cells with two more at 3.306v and one at 3.309v.

Round and round she goes...

I bought a ST1100 transmission off of ebay for less that I could get just the transmission output shaft by itself.

It took about 30 minutes to get it all apart and remove the shaft. This leaves my original motor and transmission intact, which should make it easier to sell. Here is the pile of leftover gears.

Now I need to take some measurements of the shaft and get it machined for some bearings and a pulley.

I put a piece of tape on the shaft, and on the tire, and spun the shaft while watching the wheel spin. I measured a 2.83:1 ratio for the rear gears and a 6'9" circumference on the tire.

More power...

My charger kit from EMW came in today. I was really excited to see it arrive... 

until I opened it and discovered that they had accidentally sent the wrong version. I was expecting the water cooled version and as you can see it came with the air cooled heatsink. It is also missing at least one part, I'll have to pull the BOM off their website to see what all is missing.

I've got an email out to Valery, and I'm confident that he will get this straightened out quickly.

Fill 'er up!

I decided to use the gas tank to mount some of the electronics, and to keep the existing front mount point for the seat intact. The first step was to remove the fuel pump/filter and the float, then completely empty the gas residue in the tank. I then put a few cups of water and a little liquid soap in there and sloshed it around to make sure no gas was left behind... then to make really sure I put a heat gun on low and ran it into the tank for about 15 minutes to completely dry the tank. A Sawzall then made quick work of cutting the bottom half of the tank off.

I then took a grinder and cut the filler connector off and ground it flat to the top of the tank. This will let me mount the J1772 connector where the fuel inlet used to be. I needed the 1/2 inch of room so that the J1772 would drop low enough for the fuel door to be able to close.

It is mocking me...

I wasn't sure if I could get 36 of the 100Ah cells in the frame and still be able to get the skin back on the bike (My goal is for the bike to look as stock as possible.) As seen earlier, one of the NTEAA members made some wood mockups for the 60Ah(b) Sinopoly cells for me, but I needed to see if the 100Ah cells would fit. With the help of a couple of the Dallas Makerspace members, I built 40 of the 100Ah and 36 of the 60Ah size mockups out of construction paper. We used Pepakura software to model the cells and used a laser cutter to cut them out, then folded and hot glued the 60's... on the 100's I used clear tape, I was tired and gluing took too long.

I laid out the 100Ah cells and it quickly became obvious that I wouldn't be able to get them to fit, even if I used all of the space up front in the frame and both of the pannier cases at the back of the bike (leaving me no storage at all.) I then laid out the 60Ah mockups and discovered that I would have plenty of room with them.

36 of the 60Ah cells will give me a 115.2 vdc, 6.91kWh pack, which should yield at least a 45 mile range with 80% DOD, assuming 120 wh/mi. A 36 cell pack will use just under 38 liters of volume, and weigh just over 68 kilos (150 lbs). I'm estimating that the cost for the pack will be under $3000, plus shipping.

The 40 100Ah mockups have been donated to the NTEAA for members to use for making their own battery boxes, etc.

Programmer

Travis at EMF-Power.com got me an excellent deal on a Curtis 1311-4401 programmer for my 1231 controller. This will let me tweak the controller.

Technical Difficulties...

I'm sorry for the delay in updating this blog. My laptop croaked and all I had to access the net while it was being repaired was my phone. I'll be doing several updates over the next day or two.

High or Low?

There are two places that the AC-20 can go. Down low where the transmission was, or up high where the gas tank used to be under the seat. I'll have to do some figuring to see which one will work out to be the best position.

The first set of pictures show the motor mounted low. The bottom of the motor would be at the same level where the exhaust pipes used to be. The face of the motor is on a fairly close plane to the face of the drive shaft. 

The second set of pictures show the motor mounted high, under the seat. The face of the motor is a few inches forward of the face of the drive shaft.

This last set of pictures shows the relative size of the motor and batteries in the bike frame. The "tank" is actually just a plastic air cleaner cover. That entire area is open, I will probably be putting the Curtis 1238 controller and the battery charger under there.

The battery mockups are of the Sinopoly 60(b) cells, as you can see there is room for a lot of them. I'm going to also mock up some 100's to see how they fit.

And it is outta there...

Thanks to one of the local members of ST-Owners.com, Ken H, the motor is now out of the bike. I have two different manuals for the ST1100, the factory one and one from Haynes. Both listed the same steps for removing the engine, with the Haynes one being slightly more detailed and having pictures. Neither listed what I actually had to do to get the engine out...

At Ken's suggestion, I removed the right timing belt cover. That gave just enough clearance to lever the engine out with a pry bar. With the engine out of the way, I can start measuring for motor placement and batteries.

Busted knuckles...

Last week was very frustrating. I'm still fighting with the engine, and it is still winning. The manual says it should come out, other people say they have gotten their engine out, but I can't seem to get the right incantation to get it to come out.

Several people on the ST-Owners forum recommended that I try taking the Swingarm off. This requires a custom tool from Honda and a 17mm Hex Allen key. I didn't want to spend the $80 that Honda wants for the tool and then wait for it to ship, and there are several examples of people taking 1-1/8" sockets and cutting them down to make a tool that will fit. The local pawn shop had just the socket I needed for only a few bucks, so I went to measuring and grinding, grinding and measuring, then grinding some more until I got the socket to fit (mostly). It turns out I should have gotten the 1-1/16" socket... this cheap Chinese socket had very thick walls. I tried to remove the swingarm lock nut and ended up banging my knuckles when the socket slipped... twice.

I decided that I really should have the right tool for the job, especially when I needed to reassemble the bike, so I searched on ebay for the tools and ended up getting both an third party swingarm locknut tool and an axle tool for less than Honda wants for just the swingarm locknut tool.

After I finish with this project, I'm going to donate both of these to Mellow, the Admin of ST-Owners.com. This will let any of the local DFW Honda DIYers borrow them when they need them. 

Hopefully I'll have the engine out this weekend.

Juice... lots of juice.

I've been looking at all sorts of chargers for this project and haven't liked any of the commercial options available. They are either too expensive... or too cheap. Too big, too heavy, too poorly designed or cost more than the rest of the bike. Blech.

I've been reading about a DIY, open source charger that has been refined over the last couple of years and is now available in a variety of forms from Electric Motor Works. I've been chatting with Valery Miftakhov, PhD (Physics),  founder of EMW, and I'll be ordering a water cooled version of his 10kW/60A PFC charger kit.

My current plan is to mount the EMW charger on one side of a cold water plate, and the Curtis 1238 on the other side. This will sandwich the cold plate between them, and since the charger and controller will never be operating at the same time I should have no problem of ever overloading the cooling capacity of the plate.

EMW offers a fully assembled version of their charger, and I was considering ordering that. My soldering skills are frankly pretty poor. I was discussing this with some of the Dallas Makerspace members, and one of them has offered to take the time to help me with the charger build, and teach me how to properly solder. By ordering the kit, rather than the completed charger, I can put the savings (~$1000) toward batteries.

More Bits and Pieces.

I'm still fighting that blasted engine. According to the manual, it should just come out to the left now that all the mounting bolts are out and the sub-frame is off. It looks like it lacks about 1/8th of an inch to just fall out onto the floor, but I can't make it budge. One of the members of ST-Owners.com suggested that I may need to remove the swing arm to get the engine out. I was hoping to avoid that. If I don't get it out by the weekend, "Mellow" the admin of ST-Owners has volunteered to come give me a hand on Saturday. It turns out he is a local, and a really nice guy.

More stuff came in this week. I wasn't expecting it to show up so soon, but the AC-20 kit arrived at the Dallas Makerspace on Monday. Luckily one of the members happened to be there and was able to sign for it... I was still waiting for a tracking number from HPEVS and had no idea it had even shipped! Travis with Electro Motive Force was proactive on chasing down the package, HPEVS just dropped the ball on sending out the tracking info when they drop-shipped the motor.

That motor is a lot heavier than it looks. I knew it was over 50 pounds, but I was still surprised at how heavy it felt when I picked it up. The controller has a few scrapes on the plastic at the corners, and has a small chip on the plastic lip of the cable connector. It looks like it was maybe dropped before packing, the packing showed no signs of damage.

The Mean Well DC-DC converter also arrived. I'll need to fabricate some sort of cover for the terminal end.

The $38 Programmable Volt Meter from LightObject.com is an interesting little gadget that I heard about from Jack on EVTV. It has two programmable relays built in, and you can set a high/low voltage for each. With a suitable relay and resistor, it will make a dandy little cell balancer. They also have a Programmable Current Meter for the same price.

Bits and Pieces

One of the DMS members brought up his impact driver and helped me remove the engine bolts. I haven't wrestled the engine out of the frame yet. He showed up just before the DMS meeting started, so we just had time to remove the bolts.

I was able to get enough measurements to decide that the larger DC motor I was looking at would not be a good fit, but the HPEVS AC-20 kit should work nicely. The AC-20 with a Curtis 1238-7601 650A controller retails for about $4250, but can be found for considerably less at emf-power.com. It has been ordered and should be here soon. The AC-20 has a peak HP of 50 and peak torque of 75 ft/lbs. It weighs 53 lbs.

I've also ordered a Mean Well SD-1000H-12 DC-DC converter and expect it by the end of January. It is $300 from JameCo, 72-144vdc in, 11-15vdc, 60A out. It has a couple of interesting features; remote on/off, and 12V 0.25A auxiliary output. It has a 3 yr warranty, and lists a MTBF of 32k hours (3.65 yrs). The Aux 12v out is always on, which is ideal for running the EV Works ZEVA Fuel Gauge Driver Plus, which requires a constant 12v, 35mA source.

In the mean time I've ordered and received several of the small parts I'll be needing to build the bike. I already had the Stebel Compact Nautilus 115dB air horn, it pulls ~18 amps at 13.8v so you MUST use a relay (included) to power it. It can be found online for $40-$50.

The ZEVA Fuel Gauge Driver Plus from EVWorks.com.au will let me use my stock fuel gauge as a SOC meter and my RPM gauge can double as an ahmeter. It cost $225 shipped from Australia.

The Prius inverter cooling pump (OEM # G9020-47031) moves a little more than 20 lpm while using about 2 amps at 13.8v. It cost me $89 on ebay.

I also have a Chennic Hall Effect PB-6 type throttle. It cost $60 shipped from China.

Bit by bit.

More parts are coming off the bike. It took some work, but the carburetors are off.

The coolant has been drained and the radiator removed.

The next step will be removing the engine. I've got the top motor mounts removed,  but I'll have to find an impact wrench to get the lower mounts to break free.