Toshi
10 kW
Ok, I'll check the voltage on each cell before shipping it.
Denver commuter e-bike
- Thread starter Toshi
- Start date
Toshi
10 kW
So none are 0V but it's not particularly well balanced. Hot off the charger with a total pack voltage of 56V some cells read as low as 3.51V while others read 3.8+V. Then again, I was reading them off the terminals while they're all wired together so there's probably some inaccuracy produced there.Toshi said:
I'll let the BMS circuitry do its thing overnight, recheck in the morning, and possibly stick it on the charger again.
dnmun
1 PW
yep, even though you don't say which ones, i bet the low ones are #1,2,3,4 and the rest are all bumping up against the HVC and shutting down the charger. you will see that if you have the voltmeter on it while it is charging. there is usually one that is the high one that shuts down the charge and if you drain some charge off of it then the other cells can charge up more to catch up.
Toshi
10 kW
Yeah, the first 3 were the lowest. The last several were all pretty uniformly high, though. Won't the BMS drain down the high ones per its own action?
dnmun
1 PW
nope, the BMS only just shunts the current around the fully charged cell when the charger is running. when the charger if off, nothing happens. the reason the first 4 cells are low is because they supply the current to run all the electronics there. after a while, the circuit current consumes all the stored charge in #1,2,3,4 and the voltage on those cells drops down to the point where the output is disabled from low voltage cutoff LVC caused by one of those 4 cells. that was why you could not measure any voltage on the output when you got to the pack finally. maybe it was so low that the charge you were adding did not turn the output FETs back on.
so the low cells may have less than .2Ah stored and the others at the top still have the full 10Ah. the shunt current for the v2.5 signalab is about 70mA. so to add the extra 8Ah back, you would need to charge for 9.8/.07 or 140 hours to add the charge back to match the other cells if the BMS was not constantly shutting off the charge FET as the voltage on the high cell climbed to 3.9V. so even if you hold the high cells below the 3.9V cutoff, it will take a long time to balance.
if you make a jumper with alligator clips, you can clip your load onto the pack between the top of #4 and the top of #16 and discharge the top of the pack down to the level of the low cells, and you can do this while it is charging on the charger too. when the voltages get down to 3.5 on the high ones like the low ones, then you can take the load off and let it charge all the way back up and they would be close for first pass, then leave the charger on and eventually it will balance, but it may take a while.
i guess i assumed this was a signalab BMS.
if you have the older v1 signalab, the shunt current is about 180mA so it takes less time to balance.
so the low cells may have less than .2Ah stored and the others at the top still have the full 10Ah. the shunt current for the v2.5 signalab is about 70mA. so to add the extra 8Ah back, you would need to charge for 9.8/.07 or 140 hours to add the charge back to match the other cells if the BMS was not constantly shutting off the charge FET as the voltage on the high cell climbed to 3.9V. so even if you hold the high cells below the 3.9V cutoff, it will take a long time to balance.
if you make a jumper with alligator clips, you can clip your load onto the pack between the top of #4 and the top of #16 and discharge the top of the pack down to the level of the low cells, and you can do this while it is charging on the charger too. when the voltages get down to 3.5 on the high ones like the low ones, then you can take the load off and let it charge all the way back up and they would be close for first pass, then leave the charger on and eventually it will balance, but it may take a while.
i guess i assumed this was a signalab BMS.
if you have the older v1 signalab, the shunt current is about 180mA so it takes less time to balance.
Toshi
10 kW
I don't know which BMS it is. It's whatever Justin at ebikes.ca specced, and it's the second revision: it was shipped with one then I soldered this one in after they sent it, saying it's better somehow. 
What load would you recommend using to discharge the cells? I don't have any resistors or anything, really, handy, but could certainly buy something. Our space heater is encased in a plastic shell (that it doesn't touch--no, it doesn't melt each time we use it) so using its elements may not be such a hot idea (cue rimshot).
What load would you recommend using to discharge the cells? I don't have any resistors or anything, really, handy, but could certainly buy something. Our space heater is encased in a plastic shell (that it doesn't touch--no, it doesn't melt each time we use it) so using its elements may not be such a hot idea (cue rimshot).
vanilla ice
1 MW
Got incandescent light bulbs? Or halogen.
oatnet
1 MW
Toshi, I see you are sporting the Tourmaster Transitions 2 jacket - I got the same one (in silver) and I love the ventilation, not to mention the left-wrist key pocket. I started using it on my ebike rides too, so I can have armour for those rides.
I had the same problem, and resolved it by attending an MSF BRC a few months ago. The instructor said everyone turns better in one direction. I discovered that I was sticking out my inside knee as I went through the turn, preparing to put down a foot to keep the bike from falling. I watched one of my classmates doing the exact same thing, and saw the bike swerve wildly as he extended his knee. Moving your knee changes the balance and makes you feel like you are falling. I finally took the instructor's message to heart - grip the tank with your knees - and the problem was fixed. He said that we should be steering the bike with our knees, and that the handlebars were just there as a place to rest our hands, and mount controls. It took a while to master, but by the end of the course I was riding the bike with my knees. Then I got back to my Vectrix, and realized that it doesn't have a gastank for my knees to grip.
I also noticed that you are shifted off the seat towards the inside - I see racers doing that on high speed turns, but for slow speed I think our butt was supposed to be square on the seat, and then we would angle from the waist.
-JD
I had the same problem, and resolved it by attending an MSF BRC a few months ago. The instructor said everyone turns better in one direction. I discovered that I was sticking out my inside knee as I went through the turn, preparing to put down a foot to keep the bike from falling. I watched one of my classmates doing the exact same thing, and saw the bike swerve wildly as he extended his knee. Moving your knee changes the balance and makes you feel like you are falling. I finally took the instructor's message to heart - grip the tank with your knees - and the problem was fixed. He said that we should be steering the bike with our knees, and that the handlebars were just there as a place to rest our hands, and mount controls. It took a while to master, but by the end of the course I was riding the bike with my knees. Then I got back to my Vectrix, and realized that it doesn't have a gastank for my knees to grip.
I also noticed that you are shifted off the seat towards the inside - I see racers doing that on high speed turns, but for slow speed I think our butt was supposed to be square on the seat, and then we would angle from the waist.
-JD
dnmun
1 PW
if the space heater has a fan inside then it can't be used for the load. but if you can disconnect the fan you can use it for the load across the top of the pack. measure all the cells while it is on the charger using jumpers with alligator clips to make the connections to the top of the cells.
to measure, put the voltmeter across the terminals of each cell or across the spot on the BMS where the sense wires plug in. then measure after you take the charger off to confirm the difference. but the discharge curve for lifepo4 is very flat and just a few tenths of as volt separate charged and partially charged so keep that in mind too.
post up the charging voltages on the first few cells. maybe if you can put in a picture of the BMS we can evaluate it or ask justin but if it will take a charge then keep charging all the time until it pushes the low cells up to the same level as the others.
can you measure current in line with the charger or discharge? that will help and the load just has to be a few hundred watts or even a hundred watt lightbulb will do. i use 2 oil filled radiators and a radiant heater all on full blast and that gives me about 3-4A discharge current. but no heat. almost heat, but not really noticeable. BOL, dennis
to measure, put the voltmeter across the terminals of each cell or across the spot on the BMS where the sense wires plug in. then measure after you take the charger off to confirm the difference. but the discharge curve for lifepo4 is very flat and just a few tenths of as volt separate charged and partially charged so keep that in mind too.
post up the charging voltages on the first few cells. maybe if you can put in a picture of the BMS we can evaluate it or ask justin but if it will take a charge then keep charging all the time until it pushes the low cells up to the same level as the others.
can you measure current in line with the charger or discharge? that will help and the load just has to be a few hundred watts or even a hundred watt lightbulb will do. i use 2 oil filled radiators and a radiant heater all on full blast and that gives me about 3-4A discharge current. but no heat. almost heat, but not really noticeable. BOL, dennis
Toshi
10 kW
Update: With no manual cell discharging, only several day-long series of leaving it off the charger or on the charger (through the BMS) all cells are reading 3.74 or 3.75V as measured pairwise from the copper bus bars between each cell. 8) This was hot off the charger so things might settle a bit, but that's a good showing for the BMS at face value, no?
Toshi
10 kW
A further update: after cells have lost their initial hot-off-charger extra charge, the cells are showing between 3.55V (low #ed cells) to 3.57V. That's close enough for me to be satisfied. I'll re-shrinkwrap the pack and ship it off to my parents soon.
Toshi
10 kW
So it looks like I'll not be buying that 3 ton SUV anytime soon. I found out earlier this week that my career path will be taking me and the wife back to Seattle from Long Island in July 2013. Yes, 2013: In my field we plan quite far out in terms of appointments… 
8)
We'll be living on the cheap at her mother's house, saving up for a down payment on a house of our own. Her mother lives about 12 miles from my future workplace, and there's a notorious traffic bottleneck on the way there. Oh, and parking at work would be $15/day if I did somehow slog through the traffic in the car.
All this, in turn, means that I might come full-circle back to where I was in 2008: I might end up buying the e-bike back from my parents, who are barely using it if at all out in Wyoming.
Downsides would be that I've already been down that road, that the 75 lb beast is a pain in the ass to lift onto the bus bike racks (negated in part because I'd use the bus as a range-extending/laziness-enabling "crutch" less), and that a 24 mile round trip commute with a major hill in it is nearing the battery's range limit. I used about 30-33 Wh/mile with it typically, and the battery only theoretically holds 576 Wh. On the other hand, I did buy an extra battery for the parents earlier this year when this whole dead-pack-revived affair cropped up, so could definitely work out a system either with a spare battery or with recharging at work. Actually, it'd likely involve both: recharging the 48V 12Ah LiFePO4 pack at work (and at home, of course), with the second pack (48V 8Ah, I believe) as a get-me-home reserve.
Upsides would be that it really did and does fit my Seattle commuting needs pretty well: gratis to lock up at the bike racks at work, bike-like enough to ride on bike paths without guilt or fear of prosecution, full fenders for the slop, waterproof panniers for carrying work-related stuff, wide enough tires to not need to run 120 psi, and enough extra oomph to really speed things up. (I haven't even ridden it with the current 9 Continents/Infineon setup that it has, btw!) Not having to wait for a potentially late or full bus would be a bonus, too.
Hmm…
8)
We'll be living on the cheap at her mother's house, saving up for a down payment on a house of our own. Her mother lives about 12 miles from my future workplace, and there's a notorious traffic bottleneck on the way there. Oh, and parking at work would be $15/day if I did somehow slog through the traffic in the car.
All this, in turn, means that I might come full-circle back to where I was in 2008: I might end up buying the e-bike back from my parents, who are barely using it if at all out in Wyoming.
Downsides would be that I've already been down that road, that the 75 lb beast is a pain in the ass to lift onto the bus bike racks (negated in part because I'd use the bus as a range-extending/laziness-enabling "crutch" less), and that a 24 mile round trip commute with a major hill in it is nearing the battery's range limit. I used about 30-33 Wh/mile with it typically, and the battery only theoretically holds 576 Wh. On the other hand, I did buy an extra battery for the parents earlier this year when this whole dead-pack-revived affair cropped up, so could definitely work out a system either with a spare battery or with recharging at work. Actually, it'd likely involve both: recharging the 48V 12Ah LiFePO4 pack at work (and at home, of course), with the second pack (48V 8Ah, I believe) as a get-me-home reserve.
Upsides would be that it really did and does fit my Seattle commuting needs pretty well: gratis to lock up at the bike racks at work, bike-like enough to ride on bike paths without guilt or fear of prosecution, full fenders for the slop, waterproof panniers for carrying work-related stuff, wide enough tires to not need to run 120 psi, and enough extra oomph to really speed things up. (I haven't even ridden it with the current 9 Continents/Infineon setup that it has, btw!) Not having to wait for a potentially late or full bus would be a bonus, too.
Hmm…
Toshi
10 kW
The wife and I haven't moved back to Seattle yet--June!--yet the e-bike is back in my hands already. This came about unexpectedly: My parents in Wyoming had it but are moving back to Oregon sooner than planned, selling their Wyoming house while at it. Instead of moving it to Oregon and waiting for me to retrieve it my mom had it shipped to me.
Incidentally, shipping big lithium batteries is nearly impossible. As it turns out, the original 48V 12Ah LiFePO4 pack could not be shipped, so I'll have to retrieve it in person next year. The 48V 10Ah LiMn pack that I bought last year was shippable, though, but only because I was mistaken and told them it was NiMH! Amusing how that works.
Anyway, the bike is back in my garage, now with an Infineon controller, LiMn pack, 9C hub motor, and completely redone cable routing and controller positioning. I have the controller affixed atop the rack now, secured by multiple zip ties--hell, I must have 30 or 40 zip ties on the bike in total, securing my rear light, the controller, the Airzounds horn, the crank cover, and routing and coiling all the various and sundry cable runs.
First results were positive in that the Cycle Analyst powers up and the hub motor spins up from rest albeit with radically incorrect mph readouts. I haven't ridden it yet since its return. Tomorrow I'll fix the speedo: 23 poles on the 9C vs. 8 on the old Crystalyte, but since my v2.0 CA can only do 1-14 poles iirc I'll go with 1 pole in the setup and a wheel circumference of 1/23 that of reality. I'll also get to sample what 1-1.3 kW feels like vs. the old 700W setup and my 60 HP gas motorcycle...
Incidentally, shipping big lithium batteries is nearly impossible. As it turns out, the original 48V 12Ah LiFePO4 pack could not be shipped, so I'll have to retrieve it in person next year. The 48V 10Ah LiMn pack that I bought last year was shippable, though, but only because I was mistaken and told them it was NiMH! Amusing how that works.
Anyway, the bike is back in my garage, now with an Infineon controller, LiMn pack, 9C hub motor, and completely redone cable routing and controller positioning. I have the controller affixed atop the rack now, secured by multiple zip ties--hell, I must have 30 or 40 zip ties on the bike in total, securing my rear light, the controller, the Airzounds horn, the crank cover, and routing and coiling all the various and sundry cable runs.
First results were positive in that the Cycle Analyst powers up and the hub motor spins up from rest albeit with radically incorrect mph readouts. I haven't ridden it yet since its return. Tomorrow I'll fix the speedo: 23 poles on the 9C vs. 8 on the old Crystalyte, but since my v2.0 CA can only do 1-14 poles iirc I'll go with 1 pole in the setup and a wheel circumference of 1/23 that of reality. I'll also get to sample what 1-1.3 kW feels like vs. the old 700W setup and my 60 HP gas motorcycle...
Toshi
10 kW
Speedo is fixed. Controls have been adjusted. Maiden ride has been accomplished in grand fashion: 8.2 miles, 4.6 Ah, 27.8 Wh/mi. Max speed without pedaling, wide open throttle was around 27 mph on flat ground. With me pedaling along I could get it up to 30 mph. Comfortable cruising speed was more in the range of 23-25 mph while pedaling along, throttle part of the way open. Max indicated power from the battery to the controller (so less at the wheel due to motor inefficiency) was about 1600W at 27 mph. Max regen coming down from 25 mph was around 550W, decreasing as speed decreases. Overall I regened 6% of power input on my errand run today.
The only real bugs were a loose CA mounting screw, easily addressed, and a throttle that both likes to creak on the handlebar and sometimes doesn't reset its signal to 0V immediately on being released. I had to toggle it back to "off" a time or two to get the power to stop, which is mildly concerning but not a huge issue provided toggling it always works.
I'm pretty psyched with its current setup. 25 mph with a rack, panniers, and lots of rear visibility aids (see photos below) is worlds better than 11-15 mph with a backpack in terms of cars paying heed. Assuming this ~30 Wh/mi energy usage holds true for my commute then the current 48V 10Ah LiMn pack on it should get me to work and back without a recharge there, since it's a flat journey just shy of 7 miles each way. I can always slow down a bit and pedal more, too, if I need to get down to more like 25 Wh/mi.
Anyway, in celebration of it being back together and working I snapped some photos of it post-first-ride:
9 Continent 2806.
OG V2.0 CycleAnalyst firmware.
Note both the Nexus 7 twist shifter on the left as well as the e-brake cable emanating from the lever for regen.
Airzound air horn. It's LOUD.
48V 35A Infineon controller, now relocated to the rack from within the frame as on previous iterations of this bike. The left pannier holds a 48V 10Ah LiMn battery, with another 48V 12Ah LiFePO4 battery hanging out with my parents since UPS refuses to ship it.
Part of my rear lighting/reflector setup. There's another flasher on the seatpost. Yes, I'm paranoid.
The only real bugs were a loose CA mounting screw, easily addressed, and a throttle that both likes to creak on the handlebar and sometimes doesn't reset its signal to 0V immediately on being released. I had to toggle it back to "off" a time or two to get the power to stop, which is mildly concerning but not a huge issue provided toggling it always works.
I'm pretty psyched with its current setup. 25 mph with a rack, panniers, and lots of rear visibility aids (see photos below) is worlds better than 11-15 mph with a backpack in terms of cars paying heed. Assuming this ~30 Wh/mi energy usage holds true for my commute then the current 48V 10Ah LiMn pack on it should get me to work and back without a recharge there, since it's a flat journey just shy of 7 miles each way. I can always slow down a bit and pedal more, too, if I need to get down to more like 25 Wh/mi.
Anyway, in celebration of it being back together and working I snapped some photos of it post-first-ride:
9 Continent 2806.
OG V2.0 CycleAnalyst firmware.
Note both the Nexus 7 twist shifter on the left as well as the e-brake cable emanating from the lever for regen.
Airzound air horn. It's LOUD.
48V 35A Infineon controller, now relocated to the rack from within the frame as on previous iterations of this bike. The left pannier holds a 48V 10Ah LiMn battery, with another 48V 12Ah LiFePO4 battery hanging out with my parents since UPS refuses to ship it.
Part of my rear lighting/reflector setup. There's another flasher on the seatpost. Yes, I'm paranoid.
I like the chrome finish on the 9c, that would help with keeping mine cooler in this TX summer sun.
Don't feel weird about having multiple rear flashers/lights. I once saw a guy with about 5, and you can bet you are not missing that.
If you drop your speed to around 20mph when possible, it will really help your range. I get around 20wh/mile with lots of 25-30mph stretches, but also a lot of slow city coasting and 23mph cruising. I have the same winding 9c and a 50v a123 battery.
You should try out a Mac geared hub though. If you really like pedaling with the motor off and have hills, it is a good option. It is significantly more 'punchy' and does everything below 20mph better than the 9c. The 9c does getting better wh/mile with cruising though. I am going to have two bikes, one with each motor. I'll prob use the Mac more in the winter since I will pedal more to keep warm.
Don't feel weird about having multiple rear flashers/lights. I once saw a guy with about 5, and you can bet you are not missing that.
If you drop your speed to around 20mph when possible, it will really help your range. I get around 20wh/mile with lots of 25-30mph stretches, but also a lot of slow city coasting and 23mph cruising. I have the same winding 9c and a 50v a123 battery.
You should try out a Mac geared hub though. If you really like pedaling with the motor off and have hills, it is a good option. It is significantly more 'punchy' and does everything below 20mph better than the 9c. The 9c does getting better wh/mile with cruising though. I am going to have two bikes, one with each motor. I'll prob use the Mac more in the winter since I will pedal more to keep warm.
Toshi
10 kW
In the winter I'll be on my motorcycle with my heated jacket liner and gloves, or my wife will be driving me in her Prius! (I actually have a battery for off-motorcycle use of the heated liner that could come in handy on the e-bike...)
Anyway, the rear light all zip tied up in Hannibal Lecter style is my own creation, and is ridiculously bright. It rocks:
[youtube]dXSZkQNI9Ok[/youtube]
Monday will be my first real commute with the reborn e-bike. Maybe I'll take the charger with me, top up at work, and then see how close to 10 Ah I get over the two legs without the risk of actually running out of electrons.
Anyway, the rear light all zip tied up in Hannibal Lecter style is my own creation, and is ridiculously bright. It rocks:
[youtube]dXSZkQNI9Ok[/youtube]
Monday will be my first real commute with the reborn e-bike. Maybe I'll take the charger with me, top up at work, and then see how close to 10 Ah I get over the two legs without the risk of actually running out of electrons.
BikeFanatic
1 kW
I remember that bike.
I like your rear light that's a cool setup especially for daytime riding. I may try that setup for winter.
At 29 watts a mile, I would say you are going a little fast, nothing wrong with that, but your battery may not last as you say.
One thing I do now if I need more electrons for the trip home , I add a small Lipo in series with my pack for the ride , like a 3series. Charging at work is risky as there are rules and regulation. I leave my battery on the bike.
The 3 series or 4 series hard packs are very cheap at hobbyking USA, 25$ including shipping, and a 50 watt RC charger is a good thing to have anyway, can use it as a single cell charger as well. Just a thought.
If that controller is like my infineon from Ebike kit, then they are good to 62 volts.
I like your rear light that's a cool setup especially for daytime riding. I may try that setup for winter.
At 29 watts a mile, I would say you are going a little fast, nothing wrong with that, but your battery may not last as you say.
One thing I do now if I need more electrons for the trip home , I add a small Lipo in series with my pack for the ride , like a 3series. Charging at work is risky as there are rules and regulation. I leave my battery on the bike.
The 3 series or 4 series hard packs are very cheap at hobbyking USA, 25$ including shipping, and a 50 watt RC charger is a good thing to have anyway, can use it as a single cell charger as well. Just a thought.
If that controller is like my infineon from Ebike kit, then they are good to 62 volts.
Toshi
10 kW
Funny, I just tossed my aging collection of 3s RC Lipos since I wasn't using them for anything. I'll just charge at work if it's an issue, in the locker room or in the reading room where I'm actually working, perhaps.
Toshi
10 kW
I jetted to the library today. 5.6 miles round trip, 18-22 mph cruising speed while pedaling at "12 mph" non-e-bike effort, gear 6 of 7 on my Nexus hub... and that yielded 19.8 Wh/mi.veloman said:
One thing I still want to address is wind noise around my ears. Even with my normally-for-motorcycle-use -35 dB Westone molded earplugs in place there's still a lot of it, no doubt due to the aerodynamics of my head as it goes away with my head turned.
Something Iike this would work but would be ridiculous:
I think a few years back I considered half faced scooter type helmets in this very thread, but then at that point why not just wear my Shoei QWEST full face? Hell, I could wear my hi-viz yellow mesh moto textile armored jacket while at it. Hmm. That seems even more ridiculous...
Toshi
10 kW
Just slightly less ridiculous: Giro Advantage 2. $123 at REI, which may not be the cheapest but has a great return policy and a local store...Toshi said:One thing I still want to address is wind noise around my ears. Even with my normally-for-motorcycle-use -35 dB Westone molded earplugs in place there's still a lot of it, no doubt due to the aerodynamics of my head as it goes away with my head turned.
Something Iike this would work but would be ridiculous:
http://www.rei.com/product/809816/giro-advantage-2-bike-helmet
My god, I'm a nerd...
Toshi
10 kW
First real commuting day. 12.64 mi, 5.26 Ah, 21.4 Wh/mi. Looks like I won't have to opportunity charge at work. (Current battery is 10 Ah, the original still with my parents is 12 Ah.) Score.
You don't want to be draining your battery 100% every day. Plus there will be windy days where you need more power. I still think you can get below 20wh/mile if you don't have to go above 25mph too much. Maybe grab onto a passing truck and put your regen on for a few miles....
Toshi
10 kW
Those were my round trip stats, not for a single direction. Gear 6/7, pedaling along constantly, 21-23 mph.veloman said:You don't want to be draining your battery 100% every day. Plus there will be windy days where you need more power. I still think you can get below 20wh/mile if you don't have to go above 25mph too much. Maybe grab onto a passing truck and put your regen on for a few miles....
In Seattle next year I'll have a 12 mile each way trek (and a large hill!) but not here in NY. I'll probably run the two batteries in parallel and opportunity charge at work for that, longer commute. (or take the bus...)
Toshi
10 kW
Toshi said:Just slightly less ridiculous: Giro Advantage 2. $123 at REI, which may not be the cheapest but has a great return policy and a local store...
http://www.rei.com/product/809816/giro-advantage-2-bike-helmet
My god, I'm a nerd...
Well, it didn't make things worse, but didn't make a perceptible difference in noise, either. Plus it's quite hot on the head, with greatly reduced airflow despite the few, token vents on it. All that and it didn't increase my overall efficiency (18 Wh/mi, with recent pre-aero helmet trips over the past week hovering between 15-20 Wh/mi), which isn't surprising given my upright riding position, big frontal area from shoulders and torso squared to the wind, and wide panniers without a tailbox.
Therefore it'll be going back to REI for a refund, and I can close out this embarrassing mini-phase of my life.
