XM-3500

Matt wrote:

im actually quite interested in how xtreme got their charger to detect when a single cell hits 4.25v and shuts down.

I suspect it was just coincidence - a cell or cells reaching 4.25 volts coincided with the pack reaching the shutoff point of whatever charging protocol the charger used. The pictures of the charger connector show that it is just a plain bulk charger, so the charger couldn't possibly "know" what the voltage of any single cell is?

As far as a simple zener and resister setup, unless you charged at an impractical low rate. I don't think this would work, having messed around with manual balancing of my pack with automotive bulbs, you need to shunt up to a couple amps current past the highest cell if charging at about 10 amps. They don't make zeners that big!

All the balancing schemes (analog balancers anyway) developed so far in the Endless Sphere Forum involve using a using a LM431 "adjustable" zener which provides the base signal to a PMP transistor and resistor which actually does the shunting. But unless you use large transistors and heat sinks - resulting in a bulky BMS, the shunt current is still limited to less than a half amp. So they also include a circuit to "throttle" the charge current back when the first cell reaches the target 3.7 volts.

My apologies, I was not considering the odometer was in kilometers instead of miles.

The actual distance is 10.0 miles; the odometer read 19.6 miles traveled.

19.7 kilometers is 12.24 miles. Figuring the odometer is 18 percent off the miles would be 10.04, which is the distance I travel to work.

Tomorrow I plan on riding it to work (the reason I purchased it), which is exactly 10 miles

The distance obtained per charge seems to be around 55.1 km, 34.23755 miles (times .82 = 28.074791 miles per charge) The last 3.2 kilometers (1.98839 miles) the voltage gauge goes from the "H" to the half mark. Maybe the one bad cell (#11) is having an impact on the distance (more than 5 percent).

Anyway, if the speedometer is also 18 percent off, that would make 50 mph then be 41 mph. I wonder if the original design was for 16 inch tires. That might explain the 18 percent difference.

I rechecked the speedometer. Using 0.62137 as a conversion factor (1 kilometer = 0.06137 miles), I observed the actual speed (by car and GPS) was 82 percent of the speedometer reading (within one mph.) i.e. The speedometer read 90 kph, the actual was 45 mph, the conversion is 45.85711. (The GPS used is in increments of whole mph.)

so the charger isnt recieving any feed back as to when a single cell reaches its upper limit then. bummer.
back to plan A.

the zener balancing is only good for packs that have cells that are within about 5% of each other.
they arent meant to shunt all the current.
the other side of the bms is to either throttle or shut down the charger when one of the cells goes out of bounds, as has been suggested.

remember, a BMS is to keep good cells in order, not to cover for bad cells.
if the cells need balancing of more than 0.25AH per charge (for cells up to 90AH) then something is wrong.
even thunderskys arent *that* bad.

a BMS that can shunt the full charge current should (imho at least) be only reserved for packs where manual balancing isnt possible when the cells fall out of the 5% range the zeners can deal with.
such an application is a setup for balancing brand new cells en mass (otherwise you need a power supply per cell, which believe me is time consuming)

Matt

TanWare,
There are no secondary reactions in LiFePO4 batteries aside from the lithium plating experienced at high voltage (which is to be avoided at all costs).
secondary reactions are where most of the coulomb efficiency is lost in other chemistries.

light cycling wont help balance them as the shunt current loss is so small.
with the thundersky packs in particular, its the shunt current loss that pulls the pack out of balance.
though until the pack is balanced, it is still best to only ride short distances.

xm-3500 owners,
until the pack is balanced you wont experience anywhere near full capacity. if you have no BMS, and dont intend to install one to keep the warranty, charge each cell individually to full charge to start with.
at least this way the cells will start out balanced.

after adding a bms or even just a balancing setup of some kind, i suggest getting an AH meter.
using a volt meter for SOC measurement is incredibly inaccurate with LiFePO4.

Matt

If you have no BMS, and dont intend to install one to keep the warranty, charge each cell individually to full charge to start with. At least this way the cells will start out balanced.

I start looking into electric vehicle not too long ago. As a beginner myself, would keep track with 19 cells be too much of work?

antiscab
LiFePo, as with all lithium chemistry batteries, suffer higher inefficiency charging at 90% capacity and up, as this we hit this mark the batteries that are under 90% still have their greater charging efficiency. This is a simple algorythm based on that property. It is what they are depending on and this is why they even stated it costs about 5% of the total charge to 100%!

What I would love to see instead of one large single bank charger like the one used is a 20 bank 10 amp charger so each cell would have its own charger. Agreed it would be somewhat more expensive, but it still would be nice.........

3000 Watt Electric scooter, lithium ion LiFePO4 battery
This scooter comes with an individual cell charger, which some people refer to as a battery management system (BMS). Also known as a bank charger.

I found this information about this 19 cell lithium iron phosphate powered scooter. It seem to indicate it has individual cell chargers. Here is the link.

www.fun-ev.com

Eric Fisher
www.SiliconeBatteriesUSA.com

TANware,

You seem to be giving a lot of new information regarding LiFePO4 cells, which may or may not be the same as the other lithium ion chemistries. I think it is too early to tell myself. Can you cite some sources for you information?

Matt,

I have heard of this plating damage to cells once in the past. Can you explain under what conditions it occurs, and what the symptom of it occurring are?

Anyone for moving this discussion to the batteries and chargers forum?

I have two 3500li we have received in and are testing with similar results to what is posted here. WE have them and they each have been tested for a while. So we are selling our two demos for the first two who want them for $3985. Call me at 800-806-7109, ask for Kurt.

Kurt

ive started a new thread for LiFePO4 battery discussion here:
http://visforvoltage.org/forum/4541-lifepo4-batteries

Matt

Well, folks, I've reluctantly canceled my order for an XM-3500Li. It's taken me weeks to decide. As the the deficiencies reported here began to accumulate, I was in a quandary, but still felt this was likely the best electric transportation I could afford. The real top speed is so much lower than I'd expected that I realized I would never be able to even briefly go onto a freeway. That I reluctantly accepted, though in the Bay Area, never being able to cross bridges is a serious problem with one's primary mode of transport. Yet I was going to be willing to sacrifice range and speed for the sake of doing the right thing ecologically.

What really blew the XM-3500Li for me, though, was the advertised, but nonexistent, ABS and battery management system. Then it appeared there was something fishy about the batteries themselves. I wanted to go to lithium chemistry as a long-term money-saver in comparison to my old Expresso S's SLA batteries. I didn't want to find that lithiums themselves might become a new money pit of enormous depth.

My salute to any of you who can afford to go ahead and surf along this particular bleeding edge. For me, I need reliable, affordable transportation. That means, sadly, that I will have to consider a gas scooter with good fuel economy in the mix of my options.

HalfMooner,

Sorry you decided to cancel - but I think you likely made a good choice for your circumstances. You also make a good point - we're on the bleeding edge. Unfortunately we're taking a few cuts and bumps along the way.

However, if you can wait three to six months before purchasing I think you'll have a viable solution That's based on the fact that it appears that BMS is the main culprit and there are several folks (see below)

Oh, BTW, I think you mean non-existent regen? It does have ABS.

So, having been out the loop for a long weekend...

1) It appears that the batteries aren't being conditioned and bought into balance initially. Thus, a DIY approach for initial balancing would help? If that's the case - please put on your thinking caps and come up with a good DIY approach for us!
2) There are several different BMS options in the offing:
- PJD mentions a system from Gary Goodrun at http://www.tppacks.com/products.asp?cat=26 (currently showing out of stock) - but the price is attractive - you'd need two boards at $90 per board (I believe you have to assemble the board yourself).
- sparc5 (of this forum) is working on an open source modular BMS called the GreenBMS at http://visforvoltage.org/blog/sparc5/4147. I believe he hopes to have a prototype functional in the next week or so. Erik Kauppi of Corsa Instruments has also offered to help Jeff. Corsa makes data acquisition to go on race cars (amongst other things) so he knows automotive grade electronics. Erik is also one of my partners in our nascent "rEV" branded bikes (watch for an announcement in the fall). Jeff's target is sub $200.
- Tracy Ingram over at Nova Scooters has sourced what he describes as a "cost effective and feature rich" BMS: http://www.voltsrider.net/vr/forum_posts.asp?TID=143. He mentions a couple of weeks and less than $500.

I'm currently backing the GreenBMS horse. Go sparc5 go! ;-)

Thanks again, Henry, for your reports. That reply from Thunder Sky is a shocker, to put it mildly!

Yup. So much for my XB-3500 lust.

Halfmooner,
I too have made the same decision as you. My electric progression went from an XB-600 to an Ecruiser220 (1500w) and my next purchase would have been an XM-3500 (or DFUN Model D as one was for sale in San Jose). I can charge at work and it is only a 9 mile commute so I badly wanted to stay electric but as I spent more time riding I realized I wanted to take pleasure trips into Napa and out to Bodega Bay. This required a 20+ mile range. I made the reluctant decision to go gas. I bought a used Yamaha Majesty and do not regret it at all. The downside is routine maintenance and not the gas expense as I got 53mpg on my first tank. I suggest you seriously look into the Majesty, Silverwing, Burgman, and Kymco Xciting scooters for you bridge crossing needs. There is definitely a big difference between these maxi-scoots and the electrics. I feel like I have much more presence on the road with the Maj than with my Ecruiser. I can ride comfortably at 70mph knowing full well I can go faster if desired. The weight of the scooter makes freeway speeds comfortable and I wonder how the folks on smaller bikes can handle the wind at those speeds.
When I want to scoot around town I just use my Ecruiser but I feel much less safe on it now so it stays in the bike lane. I don't want to wear my full armor all the time. When a scoot like the Vectrix can sell for $7000 and have a highway range of 180 miles I will buy one but we are a very long way from that. I do appreciate everyone that makes the financial sacrifice to help push this technology. I would be glad to buy a used xm-3500 to replace my ecruiser.

I decided to test the theory "Battery #11 is bad", so I bypassed it. The result of the bypass is the max speed was 80 kph (actual 40 mph), I had obtained 45 mph (actual) with 20 batteries. I rode 31.4 actual miles. With 20 batteries I rode 28.3 actual miles. The voltage gauge was at the 1/2 mark as I accelerated up "The Hill". The voltage once I stopped was 60.5. The gauge responded normally. (i.e. dropping a little at a time.) Before (with all 20 batteries) the gauge went from full to half in about 2 miles. I kept the throttle wide open almost the entire time.

A few days ago I purchased a Triton Jr. battery charger from Hobby Town for $79.99. It is designed to charge many types of batteries, It can be programmed to charge one to four Li-Ion batteries. The max is five amps and it is designed to use a 12 volt car battery. Anyway I was able to get the voltages in all the batteries to be the same (except #11). It took several hours to do that, however, all the batteries are now staying at the same voltage (again except #11)

Does anyone know what the max voltage rating is for the 3500 watt motor? My rough idea is that 19 batteries obtains a speed of 40 mph, 20 batteries obtains a speed of 45 mph, would 21 batteries obtain 50 mph? (This is ignoring wind resistance)

Everyone - the XM-3500LI has left me concerned, but I have it and plan on moving forward. I have purchased the Paktrakr Lithium 800 model to track and monitor voltages of cells as well as using it to alert me when a cell is low. I am also going to build a simple BMS using a TL431 driving the base of a MJE210 PNP transistor with a 2ohm/10W resistor as the bypass shunt connected to the collector (and set the base current to <=100mA assuming the TL431 output is about 2V when "on"). I initially will only plug this in when charging and then monitor the state of charge using the Paktrakr with an LED on each BMS to easily alert me when a cell has reached max before the others. I am going to set the TL431 to trigger at 3.75V and will use 0.1% resistors with the 0.4% tolerance of the TL431 to get a ~0.5% set point accuracy (and will add 50mV of hysteresis so it does not oscillate).

Does anyone know what the max voltage rating is for the 3500 watt motor?

The motor can probably take 3.5 extra volts just fine. The bigger issue is the controller, but I don't think the components in the traction power circuit (MOSFETS, capacitors) would be specified with maximum ratings so close to the 20-cell voltage that it couldn't take an extra 3.5 volts as well.

So I think you would be fine adding an extra cell.

Agree on the motor and controller being fine with an additional cell, but the key is getting the charger to put out the extra voltage or you will just be charging to the same as the 20-cell string. Also need to determine if the controller is limiting max voltage/current to the motor as adding another cell would not help top end RPM if it is volt limiting.

Once my warranty is up on the controller and charger, I might take them apart and do some reverse engineering to see if the controller is limiting the max voltage to the motor by PWM limiting and extending the top end if it is to get more RPM. Then look at adding an additional cell tweaking the charger max voltage limit. However, I first plan to get the Paktrakr on to monitor cell status and then run the 3500LI as is to see if there is any infant mortality issues that would be covered under warranty so I want the warranty to still be valid (the BMS circuit I am developing will only be used if there is a cell way out and only to bring it back in so there is no premature stress and damage; I would expect X-treme not to be too concerned doing this as it will only save them money by not having to replace a cell damaged by initial unbalanced charge). I will then consider a full-time BMS once the warranty is up on the batteries.

With all the impressive engineering going on among you clever, driven, and intrepid XM-3500Li owners, it seems to me you are well on your way to having the collective experience and know-how required to build your own electric scooter! Why not just get the rolling stock chassis from the Chinese manufacturer (no batteries, no controller, no hub motor, etc...) and add them yourselves? That way you'll get your BMS, your regen, and your "real" ThunderSky batteries instead of the sub-par components currently causing so much trouble with the vehicle. If everybody works together - like they do in open-source software communities, the result could be good. Perhaps a company such as Revolution EV could import and distribute the rolling stock, and the community could collectively publish the design and the parts list and sources for the EV components.

Battery #11 had 0.25 volts at the end of today, the others were 3.15 volts.

I suppose it is time for an over all opinion (Sorry for the lengthy post.)

I like the way it handles, and it feels stable while cornering. It has enough acceleration to meet my needs. (It is about the same as how fast I accelerate in the car (no “popping” the clutch etc.) Also, remember that I am a beginner to the world of riding. A 1200 cc engine would be too much power for my abilities.

I think the speedometer being off by 18% make it easy to determine mph. Instead of having to multiply by 0.6 to get mph, I just divide by 2 (e.g. 90 kph is an actual speed of 45 mph.) I understand that the odometer (in kilometers) will show 18% more kilometers that the actual kilometers, however, I do not consider this an issue. (i.e. I’m not going to be advertising it for sale with low miles any time soon.)

There is not a BMS, if I remember correctly one of the dealers said something about problems with the BMS so it did not ship. Maybe the manufacturer will ship it later. Anyway, the batteries are under warranty for two years and I am able to bypass the one bad cell and still get decent performance (40 mph), at least for me. I am going to order another battery (which I was going to anyway) to make 21 batteries in the pack. This was I will have 20 good batteries until the warranty battery arrives. I think that either the extra battery or tuning the controller (or both) for a higher capacity will result in at least 50 mph if not 55 mph. I was figuring on around 50 miles per charge. If my computations are correct, I will get around 50 miles per charge.

There is the irritation factor in all of this, however, I think that it is to be expected with a version one device and I think the norm for early adopters. As far as the one bad battery goes, every manufacturing process has a percentage of defects even with high quality controls. There are several issues/defects with the XM3500Li, which are either minor (wrong decals) or easily corrected (charger connector dangerous close to the bolt that attaches the storage area to the frame. The alterations I have made so far are totally reversible. (The reason for mounting the Barrier Strip on plywood.)

There was a $500.00 discount for pre-orders, which I think will compensate for the purchase of a BMS, extra battery, etc. Is it right, well no, but given a choice of sending the XM3500Li back for about $250.00 in shipping charges or buying miscellaneous devices/parts, I think I will buy the parts. I have never trusted voltage gauges anyway. So I was already planning of buying an amp meter.

To sum it all up, for me, I'll keep it, even with some acceptable irritants, but I expected there to be some, due to it is a new product (version one).

I canceled my order for the xm-3500Li yesterday for the very same reasons Halfmooner did and the fact I may not be able to operate the hand brakes of this scooter do to the severe rheumatoid arthritis in my hands I really wanted to be a green zero EV owner.

It seems very quiet here all of a sudden. As an interested observer (not yet an XM-3500Li owner), I had hoped for more user reviews by now, but the talk so far has been mostly about batteries and BMS systems. Very little "real world" riding experiences have been posted, other than a few back-handed comments ("handles good", etc...).

What I'd like to know is, what's this thing like as a daily rider on city/town streets, highways, up and down hills, around curves, in the rain, etc...? Is it really the equivalent of a 125cc scooter in terms of performance? And PLEASE, let's have some more info about the day-to-day range! How can I buy one of the September allotment without this information? I am not like the first-adopters: I can't afford a "close but no cigar" experience for that much money. My wife is a wonderful person and very much with me regarding zero-emissions transportation, but she would kill me if I got six months use out of this $4K expense and then basically had to put it in the back of the garage. So, if I am not to just give this a pass and get a gas scoot at 1/2 the price, I need more basic information about this vehicle.

I understand the current fixation on batteries and BMS systems, but it seems to me if the market for the XM-3500Li is going to ever move beyond the early-adopter phase, we (the non-early-adopters) need to know if this thing is a "go" or "no go" as daily transportation! Is it a vehicle that can be used to commute 15 or 20 miles to work every day and charged overnight for month after month without hassles, or is it a continual "project" that requires hours of attention and is off-line for stretches at a time?

Perhaps it's too early for such a verdict. If so, where are we along the way?

Thanks!

njsupercub - for 15-20 mile daily use and recharge everynight, the XM3500LI definitely fits this. As fas as 125cc equivalent, I would say it is closer to a 80cc on average. I do not have much experience with extensive riding to give more details.

As a note, jdh2550 has some QA on his website under the XM3500LI that gives some performance info that might answer some of your questions.

P.S. One note to everyone; if you want to stop fast, this thing will do it. I know there was a couple of questions on the "passive" ABS brake system but it works fine.

As fas as 125cc equivalent, I would say it is closer to a 80cc on average. I do not have much experience with extensive riding to give more details.

That's why I'm staying quiet. I want to wait until I can give a good side-by-side comparison with a modern day 125cc. I'll be comparing against a 125cc Yamaha Vino and producing real world numbers. However, my XM-3500Li has one dead cell which X-Treme and/or Elite Power are working on replacing. I might "steal" a cell out of my EFun-D so that I can get the show on the road (because the darn batteries are back-ordered too - there was a snafu and there weren't any spare batteries in the first container - there are spares in the second container).

As far as not wanting to be an early adopter - you might want to wait until next riding season (if you're in a seasonal place like I am). Remember that there were only 11 to 18 bikes in that first shipment (I've heard two different numbers). So, that's one reason you're not hearing much - there aren't many of us out there!

BTW, my two least favorite words are "back" and "ordered".

Just FYI -

Link removed.

Auction ended.

They are selling a 2007 model. Weird, you would think they had sold as many as they could make.

Bernie

Hi njsupercub I do believe that it would be a (no go) for you if you are expecting the Xm3500Li to perform like your super cub does, my brother had a super cub in N.Y.C.in early 60's so I know it's perfomance. The first week he had it he kept it in the foyer of our apt. until the landlord gave him a storage space.I had a order in for the XM3500Li and canceled it after I had more time to investigate the misleading spec.s and info. It is not what people are led to believe it is. Do you home work, if you get an(A) you will give up. I call it the big XM-3500Lie

Seems like we are doing China a favor by perfecting their deficient bike, just like one of the reader said, we should import only the frame and add all the electrical component ourselves ;)

Initially I got excited about the Li3500 and did place an order which I canceled 4 weeks ago, I do ride Honda Helix 250cc and I am intending on keeping it for longer trips. After ordering the Li3500, I did more research on electric bikes and I found 3 others but with higher price tags ( the Vectric, Enertia and the GPR-S). I did extensive research about the parts and the quality of these bikes including the Li 3500 and there's no comparison at all. Truly you get what you pay for.

Two weeks ago I decided to go with the Vectrix and I have been riding it to work daily with over 200 miles on it now.
The quality and the performance are something else, it is a joy riding it everyday, it perform like 300 or 350cc bike, top speed fix at 62 miles per hour (real speed) speed is not affected by the terrain or even having an extra passenger behind you :) Remember it runs on 125 volt system, so you have lots of torque. If you go for the Vectrix you will feel bad only one time and one time only, :) that's when you pay for it. But if you have 40 miles of commute a day, think about it, if you commute to work 30 miles each way, the bike will pay for itself in 2 short years only. I battery will last over 10 years. The Range is 44 miles @ 55 mph speed, and longer if you drive slower up to 65 miles per charge.

After reading your reports and opinion about the Li3500, I feel very happy that I canceled my order, the last thing I need is something I am not going to use occupying a space in my garage knowing that I paid 4 thousand dollars for it.

China is known for cheap and low quality product, if you are looking for high quality commuter, I think you should consider the Vectrix (still way cheaper than a car), after all it is made in 3 countries (Germany for the electronics and batteries, Italy for suspensions, brakes frame and wheels, and the motor is made in Ohio). After all, don't we all want to keep the jobs in the USA?

If anyone has question about the Vectrix don't hesitate to ask.

Have a great day everyone ;)

I battery will last over 10 years. The Range is 44 miles @ 55 mph speed, and longer if you drive slower up to 65 miles per charge.

You are dreaming!

What is your actual ever longest range?

Mr. Mik

Guys, this thread is very long, and keeping track of it is getting awkward. I can no longer click on 'x new' to get to the end of the thread, since it brings me to the 1st page (and then the 2nd page no longer shows anything as new).

How about using one of the various continuation threads, or starting up a new thread? Please??