Need more capacity

[azisme]

I have a Crystalyte 408 motor with a Crystalyte 4825 controller driven by a Ping 48v 10Ah battery. I use my bike for a daily commute of 14 miles each way. Toward the end of each commute I have to be very careful with my throttle on startups. If I hit it too hard I will trip the BMS due to too much amperage draw. If I unplug the battery and plug it back in, it works fine again until I draw too much from the battery again.

I initially had the motor on a mountain bike and did not have any issues. I then purchased a new comfort bike and the issue popped up due to the increased wind resistance of sitting upright. Incidentally the resistance also cost me about 1.5 to 2 mph in electric only speed.

Here are the ratings for my battery:

• Rated Discharging Current: 10 Amps
• Max Continuous Discharging Current: 20 Amps
• Max Discharging Current: 40 Amps
• Cut-off Current (Max Current): 35 Amps

My question is this. What is the cheapest way for me to increase the total discharge current of my battery? The way I see it I could add a small booster battery in one of the following manners:

a. Add a booster lifepo4 battery in parallel with the same voltage at a very low Ah rating (~1.2Ah).
b. Add a booster lifepo4 battery in series with the same amperage rating and low volts (~3v)
c. Add a 12v 7Ah SLA battery in series.

I know mixing battery technologies is not recommended, but choice c. would be my preferred choice as I have 5 extra SLA’s lying around that I could use as well as a 12v SLA charger. I really don’t mind risking the SLA’s as long as I don’t risk damaging my Ping battery. My second choice is b., so that I could hopefully recoup the 1.5 – 2 mph that I lost. I would also be more willing to spend addition money if I would see an increased top speed. I believe my controller will handle at least 72 volts. I looked up the components, but I am unable to locate the specs at the moment.

The last bit of information that I can think to offer is that at the end of my commute my battery voltage is somewhere between 51.5v and 52.7v. Which brings to mind, what is the minimum voltage I should let the battery get to?

 

[Rassy]

Welcome to the forum AZ. I don't have any highly technical answers to your questions, but: If you try to mix SLA in parallel with your LiFePO4, the SLA will try to pull the LiFePO4 down, since it rests and runs at a lower voltage. I had the same problem with a 48V 20AH pack running a WE BD36. My solution was to add some solder to the little shunt inside the BMS, which was limiting the battery output to about 30 amps. (I think this was Fechter's idea). Since your battery is a little small to start with, this might shorten it's life a bit, but since you are almost ok as is it should solve the immediate problem (I hated being forced to stop and reset). The LVC should still work the same.

 

[azisme]

Thanks Rassy. I found a some posts from Fechter along with a couple of others about soldering or otherwise bridging the shunt. I would rather take an approach of adding more power rather than adding more strain on my battery. Although I only hinted at it, I would really like to increase my speed. I am thinking 6 -12 volts would be the most I would want to add to stay within my controller tolerances but the additional speed though minimal would be a real plus.

 

[dogman dan]

I think the real problem is you are reaching the lvc cutoff point on the pack, not the discharge amp limit. The reason I think this, is twofold. One, it happens at the end of the ride, and two, it's cooler now, even in AZ. Over here in NM I am having problems with the range of my 20 ah ping getting shorter in the cool weather. To confirm this, I rode a full discharge range test on the last 80 degree afternoon we had, and viola, range was back to normal at 80. In the morning, riding just above freezing, I am using so much power that the thing won't recharge by quitting time anymore. I wouldn't hesitate to do the shunt mod, to the bms you have though, just to see if it helps. But what you describe is just like what happens to me if I get close to the lvc. A few miles before the final cutoff, I have problems if I haven't finished climbing the big hill and need big amps. Your best bet at this point is possibly to pack the lead for the winter, or untill you have the cash to get another ping to parallel with the one you have. I'd just run the lead for part of the ride, and then switch to lifepo4 for most of it. Slowing down would also work, but ugh

At some point, it's clear you want more volts to go faster. Me too, but it's not in the budget to go for some headways, or lifebatts this year.

 

[praskal]

I have a 48V 10ah LiFePo4 pack as well, and a front wheel road runner motor. I tried it for my long commute the first time today. The commute is about 16.3 miles one way. The problem is that there is a huge elevation change; I have to go up a bad hill for several miles. The damn thing cuttoff about 12 miles into it. I had to bike the olf fashioned way for the remainder (over 4 miles miles up and down). I'm pretty sure the sustained hill drained the juice, and then the low voltage cuttoff kicked in. Not sure if it was the BMS lvc or the 48v 25a controller lvc.

Anyway, I'm frightened of the ride home. But at least most of it is downhill.

 

[pwbset]

I have a 48V 10ah LiFePo4 pack as well, and a front wheel road runner motor. The commute is about 16.3 miles one way.

480wh goes quickly when you involve hills. I average 400-440wh going just shy of 8mi one way and that's about 2.5mi flat and 5.5mi steep climbing. If you cut out at 12mi sounds like you're averaging around 40wh/mi... 16.3*40 = 664wh. There is nothing worse than cutting out and having another 500 vertical to climb with a 70lb rig resisting you the whole way. Ugh! :lol:

 

[azisme]

Hi dogman. I wouldn't have suspected the cause was low votage rather than low amperage considering the lowest I have measured on my battery was 51.5v. But I really don't know much about these batteries. I did ask for a Cycle Analyst for Christmas so I will eventually learn more about them.

So back to my original question, is there a safe way to add a small booster battery for added speed and/or range? I am not willing to carry around 4 or more SLA's along with my Life battery.

 

[dogman dan]

I'm not big on electical engineering, more of a here's what I learned the hard way guy. But I would think if you wanted to series connect another lifepo4 or one sla for more voltage, a diode could be used to protect the lifepo4's bms. You'll have to hunt down the specifics on the diode connection in the forum though. Untill I do something, I have a hard time getting the details correct. As for what to use, mabye ping can make you a 12v 10 ah ? Or you could get some tool packs to series connect. A single sla is likely to run down before the lifepo4 does, so you have a big likelyhood of killing it if you don't disconnect it halfway, so you might have to carry two, in parallel, and then series connect that to the lifepo4. I have contemplated that idea several times, but packing another 20 pounds turns me off too much. A big front sprocket so I could peadle like hell is how I got 3 mph back on my bike.

You may still not get there in cold weather though, just get to the point where you run out quicker. I'd like more speed and range too, but I don't see any real practical way to do it short of big bucks for new lifepo4. I solved my range problem by shortening the ride, using the car to go just far enough to make the ride fit my capacity in the cold weather. So I drive 10 miles a day and ride the ebike 20. Not a perfect solution, but a bike rack was much cheaper than a new ping. My main issue is just that I don't work a long enough day to fully recharge a big pack.

Once you scrape up the cash though, two 36v 15 ah pings in series would get you flying. But to really go fast, have you even considerd a gas scooter or small motorcycle? My wife bought a used 150 cc scooter for half what I spent on a 25 mph ebike. 70 mpg is still pretty cheap and it can go 60 mph.

 

[azisme]

Thanks dogman. I had searched the forums for a longer time than I care to admit while only picking up fragments of information on how to connect different battery technologies/voltages together. Your post prompted me to do one more search and I finally came up with enough info to go ahead and bite the bullet.

The way I read it is the bms on the ping should protect itself so the main risk is to the lower voltage SLA's. Regardless, I will be using two diodes to protect the batteries per this thread: http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=4897&p=72809&hilit=+diode#p72809
I went ahead and ordered 3 diodes that were referenced in the thread. I have four 7.2Ah SLA's that were pulled from a UPS at work. They seem to be in good condition, but I haven't done any load tests on them mainly due to lack of knowledge. I may parallel two of them or just run one for testing, not sure yet. I will post results once I test it out. By the way two 12V 7.2Ah batteries weigh in at around 11 pounds total, almost identical to my 48v 10Ah ping.

If your biggest issue is charging time, you should be able to find a higher amperage charger. I haven't researched this much but the big thing with life batteries it they can take on a charge real quick. Ping just chooses to supply a lower amperage charger, I assume for cost reasons.

 

[dogman dan]

I allready use a 4 amp charger for the at work charge. I just really work a short day normally, and It got even shorter after I broke both collarbones. 30 years in construction has pretty much trashed my body above the waist. Including the brain, too much fun with solvents. Luckily I have a job where a 4 hour day is acceptable. In warm weather I used little power on the 15 mile ride in, and could charge in 4 hours or less, but now in the winter, I'm needing about 6 hours to charge, leaving me riding home, 15 miles, 70% uphill, maybe with a headwind, and a partial charge. So I drive to the bottom of the huge hill, and ride from there. I guess I really need about 24 ah pack size in the winter. The morning ride is at just above freezing, so It is using up the range in the morning.

 

[azisme]

I received my diodes, but I decided against adding any sla’s to my Life battery after doing more research on my controller components. My Clyte 4825 has P75NF75 mosfets in it. I looked them up and they are rated for a maximum of 75 volts. I also read somewhere that you need to leave something like a 12 volt margins due to surges or whatever. So basically my controller won’t handle the additional voltage.

In the mean time, my battery issue has become worse. On my way home from work last Monday, the battery (or controller?) kicked out on me after only about 6 miles. I tried cycling the controller off and on with no results. I then unplugged the battery momentarily and the motor would work again; until I gave it too much throttle. This continued with me resetting the battery about 15-20 times with ever decreasing throttling before the battery would kick out again. I finally found an electrical outlet within 2 miles of my house so I decided to try charging the battery for a few minutes to see if there was any difference. After 15 minutes max of charging I was able to complete my journey home at full throttle. I even tested it at the end by coming to a complete stop and hitting full throttle several times without any problems.

I have a front hub motor and before this episode I had the battery located in the front. I moved the battery to the rear for better weight distribution the weekend before this happened. I did have the battery in the rear on a different bike earlier without any issues whatsoever. The only difference is (a) I patched in an additional length of wire this time so there was one additional connection (I am using 10 gauge wire and Anderson connectors, unsoldered) and (b) the battery has gone through more stress since it was on the other bike.

I have not been able to trust taking it to work since then and have not had the time to go through any further testing or troubleshooting. I am hoping I can do some testing tomorrow.

Any ideas?

 

[dogman dan]

Sounds to me like that last ride simply used up all the power in the battery. That was the low voltage cutoff tripping, not the high amp cutoff. You're gonna have to ride real slow to stretch your range till you can afford a bigger ping. The extra wire will case some resistance and the connector too, but it doesn't sound like you did it too skinny to me. Are you sure you are getting a full charge? A loose connection on a plug or the wires on the circuit board on the 12v side of the charger will look the same as a finished charge on the led. I had some myseteriously short rides till I found the loose wire in my charger.

 

[azisme]

You were right about me using up the power. I took the bike out for a test ride yesterday and it kicked out on me again at around the 6-mile mark. I kept resetting and babying the throttle as I did last time I rode it. By the time I got home, I had about 11 miles on it and had reset the battery 6 times and did some pretty hard pedaling as a result (I decided to have some fun while testing and took it out on some dirt bike trails in the desert near my house).

So anyway, when I got home my voltage was 51.7 which is about the same voltage I have been getting after my 14 mile commutes with much less intensive pedaling. I suspected a failing cell and confirmed it with my meter. Coincidently, I had recently removed the duct tape from my battery and replaced it with gaffers tape. I took that opportunity to take readings of the individual cells so I have a benchmark to go by.

As you can see cell 7 is well over .2 volts lower than the other cells. It also had the lowest voltage on 11/28.

Cell 11/28 12/15
001 03.50 03.59
002 03.49 03.58
003 03.48 03.58
004 03.49 03.58
005 03.49 03.59
006 03.49 03.58
007 03.45 03.32
008 03.48 03.57
009 03.49 03.58
010 03.49 03.58
011 03.49 03.59
012 03.49 03.58
013 03.48 03.57
014 03.49 03.59
015 03.46 03.57
016 03.49 03.58
Total 55.75 57.03

I am sure it is on the forum somewhere, but do you know where I can find info on how to revive a failing cell? Also, how do I test the BMS? If I have no luck after checking those two options, I will check to see if Ping will sell me one or two cells as replacements.

This is really getting to be a maintenance intensive endeavor.

 

[azisme]

I found a couple of posts regarding reviving cells. One mentioned draining the remaining cells by using an automotive light bulb so the charger won't shut down before the lower performing cell reaches full capacity. I tried this last night, put it on the charger and checked the voltages this morning. As soon as I took the battery off the charger the voltage of the suspect cell started dropping significantly faster than the other cells. Sorry, I didn't write down the specifics.

I am now trying the second suggested option of charging only the lower voltage cell. I am doing it in a sort of makeshift way as I don't have a bench charger with selectable voltages. I am using a cell phone charger that states it is 3.7 volts, but testing it with my meter I see it's putting out over 9.5 volts with no load. I brought the entire battery down to about 53 volts (~3.3 volts per cell) by using a regular 60 watt 120v ac light bulb. I then brought up the weak cell to 3.9 volts momentarily. I have read voltages of 3.8 from some of the good cells immedialty after taking it off the regular charger, so I figured this to be safe. The charge took 15-20 minutes to go from 3.3 to 3.9 volts. In the time it has taken me to write this, the voltage of the cell has dropped to 3.45. The rest of the cells are holding at their previous state of ~3.33 to 3.34 volts. I am going to bring the low cell back up to around 3.8 volts, let it settle for a short time and then plug in the pack to the charger and see what the bms does.

 

[dogman dan]

Ugg, sounds like time to email Ping and get out the soldering iorn. I have always felt that the minimum pack size for a long lasting, 1000 cycle battery is at least 20 ah. Your pack is very likely to be getting discharged pretty close to it's maximum rate, which has to wear it more than a 1 c discharge. These batteries are great, at a good price, but I think they need a discharge rate that is moderate to really last.

 

[azisme]

Thanks for your replies dogman, I just want to be clear that at this point I am hoping to just help out someone out by documenting what I am trying. I am thankful for your comments as well as others though.

For those following along, I forgot to mention in my previous post that I removed the wires going to the bms from my bad cell before attempting to charge the bad cell individually.

Continuing … I brought the bad cell back up to 3.8 volts. The second charging only took 8 minutes this time. I definitely have too much charging current. I let it settle for about 15 minutes during which time I resoldered the bms back to the bad cell. Immediately before charging the bad cell was at 3.74 volts and the remaining cells were around 3.33 volts. When I plugged in the charger it cycled off within 15 seconds and then continued in the normal cell balancing cycling that it goes through.

Eight hours later, immediately off the charger the bad cell read 3.74, remaining cells 3.74-3.77. It’s getting hopeful.

Here are my readings after 40 minute rest:

Total battery: 59.4
Cells 1-16
3.73
3.71
3.71
3.71
3.72
3.70
3.70
3.71
3.72
3.71
3.71
3.72
3.71
3.72
3.71
3.72

Performed a very minor discharge, then recharged and let rest for 3.5 hours:

Total battery 59.2
Cells 1-16
3.71
3.70
3.69
3.70
3.70
3.69
3.69
3.69
3.70
3.70
3.70
3.70
3.69
3.71
3.70
3.70

The bad cell is not outperforming, but it does seem to be holding its own. I just need to perform a slower discharge on the battery to see how the cell holds up. I am still thinking of the best way to do that.

 

[dogman dan]

With the wheel off the ground, the motor will not draw near as much at no load, as it does when you ride. Not bad, it looks hopefull for you to revive the cell a bit! But the original problem remains, that the whole pack is being used a bit harder than a bigger pack would be. The weak cell may be real sensitive to high rate discharging now. And thanks for documenting, it's what the forum is all about. Many of the technical posts allways will go right over my head, but I have learned so much in the last year by reading everything.