I Split my LTO pack today into a 6 cell block. I decided to setup my cellphone camera and make a quick video on how I did it. Its fairly simple. You have to make sure you don't squeeze the pouch cells as you can easily damage them.
You can probably use a drill and drill bit to remove the tab. That may actually be better than the Dremel and ball bit I used.
The car worked fine with the A123 12V battery today on a 40 mile test trip sitting at ~13.8V avg 3.45V/Cell
The 12V backup starter does work albeit with a low voltage alarm signal from one of the cells when actually cranking.
I'm not bothered about the alarm which is at a conservative 2.7V and any actual under voltage sag would be tiny duration.
I have a little 3.6V balancer board I might fit as well as the voltage display just to burn off a bit if anything goes over the limit.
Installed my 6 cell LTO pack today. This took way too long to put together, didn't help that I had to figure a lot of stuff out as I went along. I wrote this up if anyone wants to actually try and do this install for the ultimate insight lead acid battery replacement.
This thing is a dream battery, plus it starts the car stronger than I have ever felt before with the 12 volt starter.
This needs the WHT/GRN wire cut for it to stay charged at 14 volts.
I wanted this battery to be the last battery this car will ever have.
I built this to be bullet proof. No chance of loose connections, battery very stable and supported properly. I come from a background of building ebikes so I have learned a lot about the importance of having perfect solder joints connections, and wire placement.
I also designed it so that I can easily put back a lead acid battery.
The first thing you want to do before installing the battery is balance it and test it carefully for any weak cells. This can be done easily by balancing the cells to .001 volts, you need a quality volt meter that tests to .001 volts, then run a 10-15 amp charge and make sure the cells are moving together exactly at the same voltage while charging. Then check after charging a few amp hours that the cells all stay exactly the same voltages.
Any weak cells must be bypassed as they will go downhill quickly.
I first wanted to point out to not bend the battery tabs or they will crack. I bent this tab up when I was cutting the battery in half (as you can see in the video I made) and it immediately made a crease in the battery tab. This later broke with just a little wiggling. You can see the crease in the picture. This forced me to solder directly on the battery tab right over the battery cell which isn't good.
You can see the Positive and Negative battery wires soldered. I also soldered balance lead wires to be able to easily check the cell balance.
This soldering was much harder than it looks. First, I'm not really sure if the tabs are nickel or aluminum. The solder doesn't stick easily at all, you have to really get a lot of flux on it and eventually the solder will stick with enough heat. The reason I don't know if its aluminum or nickel plated is because regular solder shouldn't stick to aluminum but it eventually does. If it was nickel tabs, the solder should stick a lot easier. I just don't know, but it does stick when it finally sticks and it does make for a very strong solder connection. I would probably rip the tab off the battery if I pulled on those wires, before the soldering joint broke.
If you re going to solder to the battery, you do not want to keep the soldering iron on the battery tabs too long, people have said that this could damage the cell and it may affect cell balancing in the long run. You really need a very good high quality soldering iron and don't hold the soldering iron any longer than you have to. Optimally, you want to solder as far away from the battery as possible, that is why I soldered away from the battery where I could.
If you don't know what you are doing, and will do something dumb like holding a cheap soldering iron on the tabs for a very long time as someone inexperienced would do, then you may damage the cells, but I'm not sure how easily heat damages these robust LTO cells. Even a little damage will throw the cells out of balance with enough cycles, since I don't plan on balancing these cells ever, this is not something I want to deal with.
I have a spot welder but I didn't test to see if it would actually work on these tabs. My feeling is they probably will not as the tabs are aluminum
In the picture below, I covered the top of the battery with kapton tape to seal it up, but you could use clear packing tape. You can see the 6s JST/XH balance lead, typical for hobby lithium batteries.
To make a very nice and reliable connection to the insights battery terminal, I decided to solder the 10 gauge wire directly to the terminal. It takes a bit of solder to do it effectively and you have to solder directly to the copper wire, but it does make for a very nice and strong connection.
I did not want to clamp in a ring terminal as I wanted all these connections to be extremely sturdy with no chance of loosening. This will be the last 12 volt this car sees and I don't want to check connections. The battery also will see as high as 300 amps through these connections.
I placed the the original bottom covers with the gel pad or what ever that stuff is, I believe its a thermal pad, back on the bottom of the cells after cutting it in half. This is probably used for dampening or something as the cells stay elevated on this gel material. I would rather the cells bump around on top of this stuff then on the corners. It is very easy to damage these pouch cells, and the corners are the first things that usually get damaged.
You can see in the picture how the cells ride on top of this material and not on the white plastic, and the corners of the cells are not touching anything. This is actually a really smart design and another reason why these cells stay in perfect balance, because they are not getting damaged anywhere.
You can see the battery installed and surrounded by foam. I used much softer foam on the two smaller sides of the battery to help compress it a little. I used xt150 connectors on both the positive and negative wires. This makes for a quick disconnect without tools, great for resetting those IMA engine errors.
I ziptied the old positive terminal after duct taping it and placed it on top of a piece of wood. I'm paranoid about rubbing wires that will eventually short the battery.
I didn't compress the top too much with foam, just enough to hold the battery from moving too much.
There you go, this thing took way too much time to do and is built for the rigors of an off-road ebike.
Here is one of the best purchases I have made. A hakko FX-951. I did my research at the time and this is by far the best purchase you will make for your money in the USA. This heats the soldering iron up in seconds, and the important thing is it maintains that temperature. This is important for jobs like this battery soldering, where you need to heat something up fast so the heat doesn't travel too far into the battery. A normal soldering iron will drop in temperature when you touch the surface and it will not heat up the component properly or take too long causing damage to the component or surrounding components.
Unfortunately, unless things have changed since I bought this, you need to spend the $250 dollars to get the heating element near the actual soldering tip.
For my 12 volt LTO battery, I was thinking about how to prevent it from over discharge. If you forget to shut off just a single dome light in the car, like just a single interior dome light that draws .400 amps. This means it would drain my LTO pack, which is usually around 50-60% or 10-12 amp hours, charge, in about 24 hours.
This tell me that some kind of protection would make sense as its fairly easy to drain your battery to 0 volts by mistake.
I could put a low voltage cutoff switch directly on the positive line of the battery. I believe the starter has a separate wire and I would NOT put the switch on that line, just the other line powering the car.
I could use a switch like this which has a cut-off voltage of 11.5 volts and is rated for 50 amp continuous:
This would cutoff at 11.5 volts or 1.91 volts per LTO cell. Which is an OK cutoff as I don't plan on the battery ever going that low unless I leave the lights on. I had an LTO battery that lost only about 3-5% after falling to 1.5 volts and sitting at that voltage a long time, like 1-2 years.
Then what I could do as a backup is have the meanwell DC/DC HRP-100-15 , 7 amp, 12 volt power supply powered from the IMA battery that will charge the 12 volt battery with a throw of a switch. This would just be used for a backup.
This would mean you would never have a dead 12 volt battery again, probably for the rest of the life of the car with this setup. You may even be able to drive the insight with everything powered off, with the 12 volt dc/dc converter alone if the car uses less than 7 amps at 14 volts. If the cars DC/DC converter ever gave out, you would also be able to use this backup DC/DC converter to buy you extra time by charging the battery.
It is an interesting design. I think the idea of having a backup for your Li 12V might be a bit of overkill. The find on the low voltage safety switch should provide enough protection. I really like that and will probably copy you yet again
There seems to be a bit of a trade-off between using 5 or 6 LTO cells. After some deliberation, driven by some of eq1 analysis, I chose to go with 5 cells. I also have a 6 cell sitting on the bench. The 5 cell spends a bit of time at 14.1V which is slightly over the cell max spec. So far it is operating well, but running at that voltage may shorten the battery life. At least, the battery is usually fully charged with lots of energy on tap.
The 6 cell, if it settles around 14V, will be at 2.3V/cell, so a good operating point. I think the entire battery will be storing a bit less energy than the 5 cell variation.
Jime, I'm really surprised you are charging those cells to 2.8 volts per cell. It takes very little to really spike the voltage after 2.6 volts, this means unless the cells are perfectly balanced you will have cells probably hitting over 3.00 volts. I just personally would not do that. I would do the 6 cell and charge it to 14.1 volts, that would give about 2.35 volts per cell which is more than enough capacity for the insight.
The DC/DC power supply is overkill, but I'm going to do some cross country driving, I'm considering doing a trip to Las Vegas and may stay there for a couple of months. I also got stranded one time in the desert of New Mexico by falling asleep in my car and forgetting to turn the lights off and it wasn't fun finding out that almost nobody carries jumper cables anymore. For $45 dollars, this DC/DC will be a cheap piece of security.
I wonder if the battery voltage falls to 11.5 volts, if that would be high enough for the IMA to start normally and charge the battery. I bet the LTO would even be able to start the car at 1.9 volts per cell with the 12 volt starter motor.
Jime, do you know if the starter motor has its own wire run directly to it from the battery? I know there are two wires coming off the battery but I am not sure where they go.
Yeah, the 2.8V/cell is out of spec slightly. Most of the time, the voltage is a bit lower - because of the charging controls I suppose. My cells are well matched and I'm considering installing a balance harness and using the PL8 occasionally as needed. I'm not too concerned about the 5 cell longevity since I have the 6 cell waiting. We'll see
Yes the starter has its own dedicated high current wire directly from the battery, so no worry there.
Hey @jime, on your 2.8V cutoff, if you are using automation to hard stop charging when the first cell reaches 2.8V you should be OK but you might want ot stay below the knee anyway. See the chart below. If you are going by total battery voltage it's too easy for one cell that's charged more than the others to reach 3.6 volts which was the voltage used in a study to find out what damaged occurred with a single excursion to that voltage (a permanent 25%+). If you extrapolate the slope below, it looks a cell leading the the pack by being charged only, say, 2.5% more than the rest will reach 3.5 volts when the others reach 2.8 volts. For reference, the balancing circuit that Peter is testing in another thread was found to end balancing with 0.4 and 0.55 volt delta between highest and lowest at 2.45 volts. You can look at the charge and see that this works out to be about a 5% difference in cell capacity.
Page 1034 of the following document (there are only a few pages) documents the overcharge test and has a chart showing the effect of the overcharge event:
I bought and received my 12 volt cutoff switch. After looking online, I think this one is probably the best for my needs. I modified and raised the cut-off voltage from 11.3 volts to 12.75 volts, and now its pretty much perfect for my 6 cell LTO pack.
What I liked about this device is:
1) Uses only 8 ma of power when idle. I believe I will remove the LED to drop that further.
2) Turns off 100% when it hits voltage cut off, to not further drain your battery
3) handles up to 50 Amps of continuous load with a low resistance
You need to go below the cut-off voltage for 10 seconds for it to trigger, this means starting the car will not turn it off.
However, the one thing I didn't like about it so much is that it cuts off voltage at 11.5 volts, and when I tested it was actually 11.3 volts cutoff. That is far too low for an LTO 6 cell pack as that would mean the battery would cut-off at 1.88 volts per cell. I wouldn't want to start the car with the 12 volt starter at this low voltage, especially in the winter.
I decided to modify the unit for a higher cut-off voltage. It has a voltage divider that determines how it reads the voltage. I removed the 100K resister and replaced it with a 118K resistor. This changed my cutoff voltage to 12.75 volts. This means my LTO pack will cutoff at 2.12 volts a cell.
Since the insight always keeps the LTO pack at 14.00 volts, there should be no reason why I would ever drop that low to 12.75 volts. If I did drop that low then I want my battery to cutoff because then I have some serious issues I need to look into.
The 12.75 volts will also keep the pack at a healthy 2.12 volts a cell, which is more than enough to start the car in the winter.
I plan on replacing the external resistor with a SMD resistor.
Trying to figure out the resistance value I needed to cutoff at 12.75 volts.
Replaced with two resistors I had on hand. Yes You can use a POT but I didn't want to risk it changing the voltage on me.
This is how it will have to stay until I replace it with a SMD resistor.
It cost me $75 US dollars and I ordered it directly from Australia to USA. They do have cheaper shipping so could be like $65, but its very slow.
While I agree it would be nice if it was adjustable, and especially if you can program the value. But thankfully all it takes is a simple resistor change. The resistor is also quite large you need to remove, not one of those tiny SMD chips which are a nightmare to work with. None of these low voltage cutoff devices have adjustability in them, you pretty much always need to modify them.
To change the value, best to hook up a POT and see exactly what voltage you need, then buy that SMD resistor.
You could also add a pot to it easily. I just don't know if the pot slowly changes values with the vibration of the car.