Print

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

[Burton]

Figured I would start my own thread here as I am ready to start building out my charging solution.

The build will start with three RSP-2000's ... two 48's and one 12 then later expand with another set of 48's in the future. (or I might just say screw it and buy all 5 at once to get a price break)

I am making this thread to both document the build and confirm parts before ordering them. I will adjust the build as required. There will be both images and video in this series so others can learn from what I make.

Chargers in question:
http://www.meanwell.com/search/RSP-2000/RSP-2000-spec.pdf

Configuration (as shown below)
Two 48's in series with another 12 when the NEG switch is at poll A and the POS switch is at poll A
Two 48's in series parallel to two 48's in series, in series with a 12v when the NEG switch is at poll B and the POS switch is at poll B
[note IDK if it would be cheaper to do two separate stacks of 3 chargers in series or to do it with the proposed switches and 5 total chargers. I can see where having two separate stacks would have it's benefits. ]


If I don't have access to two J1772's I can simply switch to a single "stack" solution. This first stack is what I will build out first.
I assume I also need to make switches for the current sharing configuration so it can switch from 3 in series to (2s p 2s) 1s (this should be "fun" to figure out from the limited information about parallel current sharing at the bottom of the spec sheet)

First I will start off with the new images I made based off of DoctorBass's suggestions http://electricmotorcycleforum.com/boards/index.php?topic=3503.msg21935#msg21935

Wire diagram less wiring required for voltage sharing.


Same as above but just output wiring


I assume I would be safe with a 20% buffer for voltage and current limit on the Schottky Diodes and I found these on the mouser page http://www.mouser.com/ds/2/205/L271-79525.pdf I will need confirmation these are ok before ordering but they seem like overkill to me They are in a pair configuration so I would need 5 of them unless DrBass suggest otherwise.

Some things I have yet to finalize:

• wire gauges and connectors for all wires in circuit
• Receptacles and plugs (do I want to go with a 14-50p and have a conversion box(s) from J1772, or just go straight J1772)
• Need to confirm diodes and placement
• Need to confirm trimmed voltage of each RSP module
• Need to confirm settings of each RSP module
• Need to figure out current sharing settings / wiring for total stack and add switching array as required.
• Mounting and enclosure(s)
• Monitoring equipment?

I have a feeling this month is going to be an expensive month between all the metal I have to order for the subframe and the charging solution >_<

As always Input welcome.

[BSDThw]

Are the center diodes at the "Battery Neg relays" not in the wrong direction?

The hole system looks very complicate to me. I do also use one diode in series when doing my charger-project it prevent the return current and saves a lot of other problems, but one at a string is enough. Don't underestimate the powerloss =>heat at a diode with your current! 

You need diodes parallel at each output - it will protect the charger if it is not on/fails. (using a series Diode to decouple the battery revers current allows me to leave this protection diodes) 

I never saw the schematic you use for a diode so it makes it harder to tell if it is OK or not.

   

[Mr. Fusion]

Why are you using a 12V RSP in series? I'm also thinking about building such kind of a charger and I think with the 2x56V of two RSP-2000-48 you're already at a pretty high SOC where the charging current is anyway limited. So i figured at this SOC I'd just let the onboard charger finish the job?!

[Burton]

Why are you using a 12V RSP in series? I'm also thinking about building such kind of a charger and I think with the 2x56V of two RSP-2000-48 you're already at a pretty high SOC where the charging current is anyway limited. So i figured at this SOC I'd just let the onboard charger finish the job?!

I believe you needed 117 volts to get a full charge on the 2013 S. Two 56's would be 112 volts which is why I included the extra 12 volt power supply in series.
http://electricmotorcycleforum.com/boards/index.php?topic=3503.msg21952#msg21952

Are the center diodes at the "Battery Neg relays" not in the wrong direction?

You need diodes parallel at each output - it will protect the charger if it is not on/fails. (using a series Diode to decouple the battery revers current allows me to leave this protection diodes) 

Sorry you are right the symbol is wrong -_- wasn't to awake when making this lol

There updated with questions bellow:


When you say parallel at each output do you mean like the diodes at position #2 or #1?

If I am reading this wrong can you point me to a link online showing what you are recommending?

Thanks!

[Mr. Fusion]

Yeah, but do you want to charge with full power until 100% SOC? I thought it would be enough to charge full power until about 90% SOC and then let the onboard charger finish...

[BSDThw]

I tried to explain with the diagram below. It is a standard scenario when driving things in a daisy chain. I think with a battery as load it will not happen so easy!

The upper left is what you normally see, if the top RSP is off it will be forced to run the current trough its inside in reverse! Therefore it could happen like the top right diagram shows that you have a high negative voltage at your supply that will likely kill it.

The bottom diagram explain how the parallel diode save your device because it will only see 0.4-0,7V depend on your diode material!

I agree with Mr. Fusion if you don't need to charge your last 10% so fast it will be enough to use the 112V. Also if you don't reduce the current when you reach the battery end voltage your have to shut off the chargers and you will see the battery has ~80-90% because the high current has raised the voltage!

Have you thought how to protect the system?

Where do you feed the current to your system? I fear quick charger input is fused to low and the wires are to small but I don't know I never saw the 2013S from inside.

If you feed at the point the quick charger do; it will be like feeding to the Sevcon +/- Connection. This is what the 2013+ do but you have to take care, you first have to "wake up" the Sevcon before you feed current at the +/- because if the Sevcon is not slowly "precharged" it will nearly shortcut your chargers- the charger are current limited but the Sevcon can be destroyed.
If you start your onboard charger first it will do the wake up...
But you should use the control wire from your quick charger port to see if your bike is ready and more important if your battery pack is not in need to switch off. I believe since 2013 the BMS can cut off the battery with its main relay but the correct way would be to see if the control wire is released from ~0V back to the value you pull it.
Trikester has explaint how he use this wire -
http://electricmotorcycleforum.com/boards/index.php?topic=2698.msg24006#msg24006
-Personally I have watched how my 2012 Deltaq was doing it, it only connect 3.3V to the control wire (I used 10-20KOhm in series) and if the Bike will charge it pulls the line to Bat - (~0V) if eq the Battery is out of balance it my happen one cell reach the 4.15V but the pack is lower 116.2V the control wire is released and you should shut off your chargers now.

[Burton]

Ok so you are using the diodes to route the voltage around the individual chargers. I will update the diagram bellow to make sure I understand what you're talking about.

I get what you two are saying now >_< Since 117 would be a 100% SOC coming up short at 112 wouldn't be too bad. I think DrBass pointed out the 125% current you get when putting the RSP's in current limiting mode would drop some when you trim them from the stock 48v to 56v. I will get two 48's and see how they work out and move on from there.




What voltage is a 0% state of charge on the 2013 S ?


I planed on hooking up my output voltage from the RPS's to the controller on the B+/- terminals as they hook directly through the contactor. When the bike goes into charge mode (by plugging in stock charger first) it closes the contactor and allows more amps through than the stock anderson connector will because it is fused for only 30 amps. I was informed by EC this is likely the easiest method to do safely.

How long do you have to wait after starting the stock chargers before turning on the aux chargers?

[Burton]

Ordered two RPS-2000-45's ($930 shipped)

Haven't got any other parts yet. I might just wire each to a Nema 5-20p (?) and then run them through a 14-50p to 3 x 5-20r (so I can plug up my stock chargers as well) and then make or buy one of the J1772 to 14-50r boxes.

http://cordtec.com/products.asp?id=715&classId=234
http://modularevpower.com/Sales_J1772_to_NEMA_14-50_Adapters.htm

This way I can charge at a J1772, 14-50r or individual 5-20p house circuits (on their one fuses each)

I need to get my wire gauge chart out so I can figure out what wire I need to buy and connectors / terminals.

Going to ride Turbo into work today, 52miles round trip. Will likely take it the speed limit to see what battery power I have left when done. It is 65 out right now but it will be low 50's this week.

[Burton]

The power supplies are due in mid next week. :/ gives me time to get other components in the interim.

Slightly off topic. Went into work today and started with 10.58 battery capacity and ended with 5.512 meaning I used 5.068 kW for the 29 miles traveled but the zero app says I only use 4 with an average of 137watts per mile when it was really 175.

Going to have to get a cycle analyst to see my watts real time going to have to dig up docs post and see what I need to get this done.

[Burton]

I have been giving this some thought and need "real" electrical input.

My objective is pretty simple. I want to carry the least amount of adapters / wires required to charge from a 120v nema 5-15/20, 240v 14-50, and J1772

My thoughts are to take a 14-50p and split it three ways. The split ends would be where the chargers plug in if all interfaces are 5-15. Then have a J1772 to 14-50r for use when using J1772 sources. (example but with 6-20 http://visforvoltage.org/forum/12613-charging-multiple-nissan-leafs-same-time-one-charging-station-evse )

Another problem is the nema 5-15 is rated for 15amps and the RSP-2000-48's pull 16A/115VAC and 10A/230VAC. Does this just mean I increase the cable size to 12awg from 14awg?

The existing extension cord on the bike is 14awg C13 to 5-15p. However the C13 receptacle attached to the power supply has 12awg cable running to it which makes me think I can upgrade the wire. But will a 5-15 survive 240v @ 10 amps?

Any other suggestions more than welcome, I want to order parts ASAP!

EDIT:
Should I just wire up the two RSP-2000's with a 6-15 or 6-20 then make all three splitting off the 14-50 the same type. I would then have to wire up a C13 to 6-15 or 20 or just require the outlet to also be 6-15/20 an then make a adapter cord for it. I would also need at least one, if not 3, 6-15/20 to 5-15 for the rare case I am at someones house and they have at least 1 and up to 2 circuits I can plug into at 120v and 20amps.

[Doctorbass]

Hey Burton, you might be interested by the setup i made.

I am using the popular Powerpole 30 ( model PP30) connectors that when used with gauge 10 they can take 45A continuous but are really small.

I added them to all NEMA 6-50, NEMA 14-50, NEMA 6-30P and my J1772 adapter.

This is very compact and lightweight and reliable!

take a look:

[Burton]

I wish you responded a half day earlier >_<

I like your setup and will likely adopt it. I have 3 x L6-20P, 4 x L6-20R inbound as I was going to do a 14-50 to 3 x L6-20R and then wire the chargers with the L6-20P (including stock) and then make a J1772 to 14-50 and some others ... But your solution beats them all in size and weight.

Going to have extra plugs and receptacles laying around now -_-

Now that I am thinking about it I can charge through the stock Anderson connector with a single stack as it is under 30amps

I will likely wire a PP30 inline with the stock C13 outlet as well so I can use it on top of the meanwells ... which came in today

[Burton]

Are you using the size 45 connector pins in the PP30's?
I would love to see more detail on the circuit your using for your J1772 with display

What gauge wire did you use for your meanwell stack input and output?

Crap just remembered I haven't ordered my diodes yet >_< need to hunt them down again :/

[Doctorbass]

Hello Burton,

I could have posted these pictures sooner but my crapy Samsung galaxy ACE II X is in problem with the usb and SD card and my computer is also not reconizing it...  Anyway... I hate these poor processor low budget cellphone.. that's the last cheap cellphone i'll have!

In fact i have used the PP30 contact that i used a long nose cutter to enlarge the hole and insert the AWG 10 Silicon ultra flexible turnigy wire ( RC stuff again...)

I could have use dthe 45A contact but i hathe these and dont have the right tool for them. I only have the tool for the pp30 contact.

All my probe, aligator connections, parallel and serie adaptor are made of Anderson connections. The advantage is that you dont have any MALE or FEMALE.. these are all working together and remain protected.

The J1772 circuit is pretty simple and i could have just built it myself but i bought the entire J1772 plug with the communicating circuit because i tought it was complex... than i received it and discovered that it's just a kind of resistor voltage divider. It's simple as this:



I have bought the J1772 connector here at Tucson EV: http://www.tucsonev.com/index.html

I have bought the grey plastic box at Home Depot nd the current/voltage display on ebay for about 10$  http://www.cafr.ebay.ca/itm/Hot-Dual-Display-AC100-300V100A-LCD-LED-Panel-Meter-Current-Voltage-Transformer-/191192554719?pt=LH_DefaultDomain_0&hash=item2c83f6dcdf

I have used 14 gauge wire to parallel all the AC input and used silicon ultraflexible 6 gauge with SB-50 Anderson connector.
The 6 gauge is becoming a little hot ( 110F) but it is built  for that =) I did not want to use 4 gauge these are too heavy... but in my case i'm running 90A into them... if you run lower this will be ok too.

two parallel RSP-2000-48 give about 90A DC but it depend on how do you tweak the voltage from the nominal value.

btw if you parallel them ( in 2s2p config) YOU MUST CONNECT THE CURRENT SHARE PINS LIKE SHOWN IN THE TECHNICAL DOCUMENT. This will equalize and share the current equally between all paralleled meanwell.

When connecting in serie it is not required.. you only need to ensure that all serie  section have a parallel diode to their output in normally non conducting mode.

Right now i am reverse engineering one of my RSP power supply  to understand howe the current regulation circuit work and i'll add a potentiometer to adjust the current output . This will be usefull in situations where  itS, only 208V and 30A availlable... in that case at 208V each psu draw 12A wich represent 36A AC and trigger OFF most of the 30A rated charge stations... so when i have to use one that is 208V i'll be able to reduce the current to 30A.

3x RSP 2000 draw 30.5A at 240Vac and 36A at 208Vac.

These are the data you might find interesting that come from the original thread:


TESTED results:
Fast Charger efficiency measured: 92.5% ( 6980W AC 6460W DC out)
Fast Charger Current consumption ( at max output power): 30.5A AC
Fast Charger max power output ( mean at end of charge) 6.60kW
Fast Charger Voltage output: 74.0V ( adjustable from 29 to 76.8VDC)
Fast Charger Current output: 91A (0.75C charging rate)
Fast Charger weight ( including connectors and harness): 13.4lbs
Fast charger dimensions: L 13.2" x W 5" x D 5"
Fast charger max temp ( case): 43 celsius
ZF9 battery temp at end of charge ( internal OEM temp sensor): 32 celsius
ZF9 Battery measured capacity (with  2 bar flashing on the gauge): 6600Wh


 and original thread: http://electricmotorcycleforum.com/boards/index.php?topic=3085.15


Good luck and dont forget to cvontinu posting pics !!

Doc

[Burton]

I actually have the same image you posted in my zero folder on my computer I was more interested in how you wired your display and LED (if you have time to share)

diode question for you. Is 150V reverse, with Forward Continuous Current 30Amp good enough for this application? Forward voltage drop would be around 1v if I recall with something like that and 200Amp peak current. I assume with a Dual Common Cathode diode you would simply use pins 1 and 2 ignoring pin 3? Do they need mini heat sinks? (I know when I use to assemble laboratory equipment ours had heat sinks on them)

Our 2013 Anderson connectors have silicon wires on them. I love that stuff but man it is expensive >_< I think the cheapest I can get it is 1.75ft and I doubt I am going to care what color it comes in. I can use the PP45's to indicate that. I will need to track down some sleeve material for the wires as well.

Do you find your small extension cord to be long enough and if so how long is it?

I wonder what the connector is on the stock Aux charging circuit and if I just rip it out :/