Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Doctorbass]

The demand is so high for FAST charging  that i thought that for those who can't affort a 1500-2000$ solution like the Elcon,  but that have some DIY experience in electronic like me could have an alternative.

BE AWARE THAT IF YOU BUILD ONE YOU ASSUME ALL RISK AND MIGHT AFFECT THE WARRANTY OF YOUR ZERO.

IT IS ALSO RECOMMENDED TO HAVE A  BACKGROUND IN ELECTRONIC.

Also be aware that this cheap compact solution IS NOT WATERPROOF so you would have to cary them with you in a side case or a box in the gas tank.

The 2013+ Zero have a battery that when it is fully charged, it is at about 116V.

For FAST charging you need a powerfull DC power source that can do that voltage and that is also  regulated in constant current and constant voltage.

Fortunatly it exist ALOT of solution for a powerfull multi kW power supply  that have very great price per watt.  One of the most popular solution in teh ebike community is the use of the server power supply. These exist in a wide  power range from 400 to 3000W at up to 58Vdc

What is nice is that these  are VERY COMPACT!! sometime about 22W per cu inch !! and cost less than 100$ for 2000W.

As i said the 2013+ Zero need up to 116V and these great server psu are just availlable at up to 58V and mostly 56V max. But what is nice is that  you can connect two in serie!! so two of 56V would give you 112V and would charge at up to 90% and the 58V would give 116V when in serie and give 100% charge.

As well, charging at 2, 3, 4 or even 7kW can rarely be done on a standard 120V ac outlet and would require 240V outlet.  For that it exist a company that sell these great J1772 Adapter that can connect to the most standard electric car charge stations.  these cost between 80 and 200$ from TucsonEV.

If you search on ebay for SERVER POWER SUPPLY and look for the 1000 2000 or 2500 or 3000W by searching with 2000, 2500 or 3000 you will find alot of these that are compact and cheap. To operate these you might need to connect some jumper on the rear panel multipin connector but most of the time you can find the datasheet on the internet and find the activation pins to ground or connect to the 5V.

Also you might need to isolate the case of the power supply from the negative pin or ground  to avoid a short circuit when connecting them in serie.

If you are lucky you could get two power supply that fit in teh "gas" tank of the Zero as many are about 10" long per 5" wide..

Here is an exemple of these great power supply i have found:

Tyco NP2500 48-V, 2600W Power Supply or Rectifier
http://www.ebay.com/itm/New-Tyco-NP2500-ROHS5-Series-1-2A-48-V-2600W-Power-Supply-or-Rectifier-/390939207985?pt=LH_DefaultDomain_0&hash=item5b05cae931

Teh popular but VERY LONG ESP120 from HP.. a 3000W 52V unit :
2 for 69$.. that's 6kW !
http://www.cafr.ebay.ca/itm/HP-ESP120-Series-2950W-Power-Supplies-Lot-of-2-USED-/351364875011?pt=LH_DefaultDomain_0&hash=item51cefa8303

This one is 6kW split in 3 unit in a rackmount that can be refolded in a stack of 3 for about 270$
http://www.ebay.com/itm/Artesyn-Emerson-Network-Power-UFR6000-00J-with-3x-UFE2000-96S48PDJ-48V-Modules-/301213471460?pt=LH_DefaultDomain_0&hash=item4621b8d2e4



Some are 12V or 24V and most of them are 48V ( 48 to 58V generally)

Also if you find a great deal for two that can just go let say 52V max   you can add a 12V third unit in serie that will  raise the voltage to 116V but it will have to support teh same current as the two 52V unit.

IMPORTANT: you will have to add a shotky diode polarity protection in parallel to the output of every unit connected in serie. This will protect each unit when these start and will avoid reverse current going to the last unit that turn ON.

 As you see.. this require perticular attention can care and knowledge ans skill but it represent a solution that might cost less than 500$ for a full compatible J1772 charging solution of up to 6kW!

I have personally did that last year for a total of 7.6kW total !! but i have used $$$ very tough Meanwell power supply from the RSP serie ( cost about 1200$ of material total)  but it charge my Zero very well and is as compact as a box of shoes! I transport it on a bag on the top of my "gas" tank.
http://electricmotorcycleforum.com/boards/index.php?topic=3085.0

2000km trip with 13.4kWh total and 7.6kW portable charging last year: http://roulezelectrique.com/quebec-la-sarre-quebec-en-moto-electrique-2000km-pour-0/

These server power supply are very similar solution than my meanwell but these cost 3-4 time less and require some little mod.

I tought that sharing that idea will certainly help some DIY guys here!





For those who dont have the skill to make their own , Electric Cowboy is also releasing a charge tank solution.

Doc

[trikester]

Here's an irony for you Doc. Before we sold the company, and I retired, I was a principle owner of a power supply company that made server supplies as well as a lot of other switchmode power supplies and DC power systems. I was also VP of engineering and later in research. Our biggest business at the time we sold the company was from tell-com, cell site power and large server installations. We built a big power system that supplied 48 v at 10,000 amps (used in a Cisco server installation)! The individual supplies in those systems were hot swap-able so power was never interrupted when a supply failed. Our company was also the first to mass produce switchmode power above 1 KW using power mosfets (my early designs).

The irony is that now that I'm into electric bikes I no longer have any contact with the power converter industry. However, I'll take retirement any day!

Thanks for all of your information, insight, and inspiration, Doc.

Trikester

[Doug S]

Doc, I applaud your resourcefulness and DIY spirit, but I'm not sure I'd recommend that sort of approach unless you REALLY know what you're doing.

Batteries can be fairly sensitive devices, especially when you start talking about charging them at high rates (rates approaching or exceeding 1C). If you want to do that sort of thing, you need to be very careful to obey the charging envelope the manufacturer recommends, or you risk damaging or prematurely aging your batteries. Particularly toward the end of their charging cycles, batteries can't take a huge amount of current -- that's why you often see the "80% charge in 30 minutes, 95% in one hour" sort of spec given for fast-charging technologies. Charging current must be tapered toward the end of the cycle to preserve the battery's health. The same is true if a battery is very depleted....it needs to be charged very slowly until it reaches a certain threshold, then it can take the full "bulk charge" rate.

A fixed-voltage power supply isn't going to do any of that. It's going to simply attempt to supply its preset voltage, subject to its internal current limit. It will supply full short-circuit current right from start of charge, right up through its programmed voltage, and then will abruptly change to "constant voltage" mode, without regard for whether it's violating the safe operating area of the battery.

I'd also be a little bit concerned about operating the power supply in current-limiting mode for the great majority of its operating life. Power supplies generally aren't designed to operate in current limit long-term, and I'm not sure I'd be confident it wouldn't be harmed by that sort of operation. In fact, some supplies use "fold-back" current limiting, so they wouldn't give you nearly as much current as you think you're going to get during most of the charging cycle.

My guess is, you'd usually get away with that sort of thing, maybe shortening the life of the battery a bit but not disastrously. But you might just find yourself having to replace your bike's battery very prematurely.

[Doctorbass]

Thanks for your comment Doug.

I am aware of all concern about that cahrging way and i have indicated it in BOLD in the first line of the thread

If it would not have been succesfull for me i would not have posted that idea. That idea is now confirmed to work when done properly.

I know very well the cells in the 2009 to 2014 battery from Molicells , EIG and Farasis. I have my own lab of battery testing that i developped over the 8 last years.

Luke a great friend  also know VERY VERY VERY WELL these battery in the Zero and he have no dount that if done properly, these battery will remain balanced very well.

Bulk charging at up to 1C can be done if the battery is in great shape. All cells keep balanced with in 1or 2 mV witch is excellent.  In fact the QC that Zero get of these cells is according to their very specific and severe specification.  So bulk charging ant 2012 and up cells at 1C is not a problem. The cell temp remain under 35 degree C with ambiant temp of 25 degree C at the end of cahrge ( tested personally)

For the power supply i've been using more than a dozen of these 1kW+ power supply in current limiting mode and they never failed. If properly air cooled these will last longer than the life of the motorcycle. These are used only when charging witch is 1 or 2 hour per 100km...Especially the RSP serie fro meanwell witch are really tough and the Meanwell Sales rep told me alot of impressive story about how tough conditions these are used in and these continu to operate.

Even tough their MTBF would be reduced by a factor of 10  that is still 1000 hours of operation!

I also tought about using the Zero BMS feedback signal (like throttle enable) to cut power to protect the battery against overvoltage of one or more cells.. but these remain very well balanced.

Think about Terry!! he is the best exemple! he charge at more than 1C and also bulk charge and he sometime overvolt them to 4.20V instead of the standard 4.15V per cell  and it work perfect for him... most of his charger was not using any feedback signal from what he told me when i have meet him last year in California.
My guess is that he was overvolting them to extend the constant current mode to charge faster...

On my 2012 S ZF 9 + 2x ZF3  i just checked my battery in storage since 5 month now and they are all at 3.700 or 3.699V!! and was bulk charged alot

So... personally i think that if people are aware of what it involve and that they have the knowledge to understand what they do, i woudl say go ahead and try it!

Making long trip  i Zero become possible and become a real charm !!

Doc

[firepower]

Hi Doc, I going to try using these rectifiers as they were free due to being obsolete for use at work. They normally are in a sub rack x3 rectifiers. I need 4 so will use them without subrack.

http://www.eevblog.com/forum/reviews/invensys-r524-rectifier-module-24v-20a/msg569931/#msg569931

[Burton]

I have purchased another tank ($99) to see if I can modify it to hold the RSP-2000 series chargers better as well as the wires. No clue how long it will take to get to me given there is no guarantee delivery time with Zero. I will post up when I start to modify it for my purposes.

The RSP's are only rated to 90% humidity and should not run where condensation might be a threat. So even enclosing it into a tank like system with cleverly placed intakes to prevent water from getting into the chargers wont limit you to when / where you can charge.

I did a full test with my chargers yesterday at home (via a 14-50 outlet), without the voltage sensing circuit since I need to get some more 22ga wire which isn't solid copper to make it. They slowly tapered off as the pack reached 90% which is where I set them to (114vdc). The max charging rate I got from them was about 52 amps with the stock chargers, or about 111 minutes to charge a 11.4 bike to 90% from 0% or if you charge from 20% then 87 minutes.

I might end up with a couple of elcons in the future as a backup / tandem charging method for super fast .9C charging when touring but we will see.

[Doug S]

Again, Doc, I'm not trying to say it won't work, or that it will definitely harm your equipment. It does seem likely that the manufacturers, primarily motivated by liability limitation, probably do put out very conservative specs.

But people need to be very aware, before they do something like this, that they're clearly violating the manufacturer's recommended operating area, maybe by a substantial and significant margin. That's all I'm saying. The likelihood of causing a failure may be low, but the consequences of a failure (a dead battery pack) are harsh.

[RNM]

Other Power supply:
http://www.tdk-lambda.com/products/sps/catalog/eng/fps.pdf

http://www.ebay.co.uk/itm/NEW-TDK-Lambda-FPS1000-48-S-1000W-Dual-Output-Power-supply-48V-21A-12V-0-25A-/161599285590?pt=LH_DefaultDomain_0&hash=item25a0116956

[Burton]

Doug,

Is charging to 90% with a CC-CV PSU at a max of .5C not good for our battery chemistry?

I have noticed as my RSP-2000-48's get close to 114 vdc (they are set to 57 each) the amps seriously drops down and the output is below the rated 2000w at this voltage.

I have seen some endless sphere wiki posts stating a CC-CV is the ideal charger for LiPo cells but idk about our chemistry.

EDIT: Reference to post I am thinking of http://endless-sphere.com/w/index.php?title=Charging_Lithium_Polymer_batteries

[Doug S]

Is charging to 90% with a CC-CV PSU at a max of .5C not good for our battery chemistry?

I have noticed as my RSP-2000-48's get close to 114 vdc (they are set to 57 each) the amps seriously drops down and the output is below the rated 2000w at this voltage.

I have seen some endless sphere wiki posts stating a CC-CV is the ideal charger for LiPo cells but idk about our chemistry.

I'd have to spend some time looking at the full charger spec as well as Faresis's recommendations for the battery to give the definitive answer, but it sounds pretty appropriate. Generally speaking, a four-state charger is best -- a small trickle charge if the battery is highly depleted (ours may be prevented from ever getting into that state by the BMS), then a fast "bulk charging" state at a constant current ( "CC" -- 1C is traditional but 0.5C is allowed by many manufacturers nowadays), followed by a fixed-voltage (CV) stage as the current ramps down, and finally a "trickle" stage at slightly reduced voltage to maintain the battery at full charge. I like hearing that the current drops as the voltage approaches the final value -- that indicates that the most important thing, the reduction of charge current near full charge, is being performed.

Also generally speaking, Li-NMC is a pretty rugged chemistry and can stand a lot of abuse. I'm sure that's one of the reasons Zero picked it. I don't want to sound like an alarmist, I just think people need to be aware that a fixed-voltage supply really isn't the best solution for battery charging.

[Doctorbass]

I agree with you that a fixed voltage power supply is not appropriate..

That's why i recommend these kid of power supply. These are all CC-CV.. constant current and constant voltage .. just like lithium ion battery require.

Otherwise a simple constant voltage power suply not regulated in current would just bow or trip the breaker or goes in protect.

NMC, NCA, LiFePO4, LiMn, LiCo etc.. all these chemistry just need CC-CV.

BUt as you said there is an important thing: not bulk charge a fully deplated battery. This would cause gassing or accelerate the corrosion of the electrode inside of the cells.


That's why i always start by charging few minutes with the OEM zero charger.. Then i connect the DIY bulk charger.

Oh.. btw the 4x  320W Meanwell unit that charge the Zero to 1300W are  in fact  CC-CV normal power supply.... commonly used for LED application. The server power supply work the same way in CC-CV. But their CC is the constant current  limiting feature.

Doc

[DynoMutt]

Even if the bulk charger can only work in two states, if the high-voltage mark is set lower than the pack can handle, couldn't the onboard charger just fill in during that time to top it off because it has to be on for the contactor to be engaged anyway?

Simply put, if the packs are in parallel, which I am assuming they are, why would there be a threat to them if the bulk charger never goes above 114VDC?

Please correct me if I'm wrong, I'm trying to understand this for my own benefit as well.

[Doug S]

Simply put, if the packs are in parallel, which I am assuming they are, why would there be a threat to them if the bulk charger never goes above 114VDC?

On further thought, maybe I'm worried over nothing. If Doc's right (and he usually is), and those types of supplies are designed to run indefinitely in CC mode (which would make sense in a rack-mount situation where you might have several of them operating in parallel to power the bus), and if the BMS never lets the batteries get to a badly depleted condition, maybe it's perfectly acceptable, and within manufacturer's specs, to do what Doc's talking about. If all of that's true, the only quibble I can come up with is that when the CV portion of the charge is complete, a "smart" battery charger drops the terminal voltage a small amount, terminating the charging cycle and beginning the trickle "keep-charged" phase. But if the terminal voltage is set to the trickle value, I can't even make that argument.

Doc, I stand corrected. If done right, what you're suggesting should not be a problem.

[Burton]

What is the lowest SOC we should be charging from at a rate lower than 1C to prevent scaling ? (Yeah I watched the battery charging video posted here a little bit ago)

The rsp 2000 won't even let you charge if the voltage is 80%+\-6% of the set voltage

This would mean if set to 114 out they wouldn't let you charge less the voltage of the pack was at least 84 or 98 correct? (Not sure which it causes but I have had it happen before when set to 116.5 output)

[Patrick Truchon]

On further thought, maybe I'm worried over nothing. [...] Doc, I stand corrected. If done right, what you're suggesting should not be a problem.

Please excuse the slight tangent; I just wanted to mention that I love the level of discussion on this forum: civil, polite, helpful.  Thanks you all!