DC charging spec

[schwankl]

Anyone point me to info on DC-DC charging a redshift? Losing ~20% going to AC and back sucks.

 

[snydes]

Anyone point me to info on DC-DC charging a redshift? Losing ~20% going to AC and back sucks.

This is why we need @MVetter and DigiNow to develop a more versatile charger for the Alta.

 

[bluefxstc]

I am not aware of any DC charging for the Alta. Could probably be done. The current charger's output is DC so no reason you couldn't duplicate that with a different system. You would need to work on figuring out the Alta/charger communication so you could duplicate that. For DC fast charging there is a lot of communication that goes on between the vehicle and charger so all of that would have to be worked out. You would also need to figure out what the bike charging circuit (wiring, relays, batteries...) can handle. I don't know what it is, but I do know there is some limit and if you exceed it something bad will happen.

On a bigger note, not sure if there are any efficiency gains to be had are. Unless you have solar/wind/hydro DC power generation source and a battery bank, you are going to have to take AC wall power to DC, so not sure of the advantage. Most EV car chargers that I am aware of have an efficiency of 93% or better, so we are only talking pennies per charge. I have relatively cheap power at a little over 8 cents per Kw so assuming a loss of 10% (higher than I think it is), for a full charge of 5.8Kw, I only have about 5 cents of inefficiency.

 

[SuburbanDad]

I was wanting to be able to charge the bike with lower power sources, and make a more portable charger I could throw in the backpack. Top up at work, pull juice from DC in a pinch, etc. I opened up my charger to hook up a logic analyzer. I found Elcon controller guts, and Alta confirmed that the (non R) charger is an Elcon PFC 2500. e.g. PFC 2500 Battery Charger.

I imagine Elcon could provied a lot of detail and likely some off the shelf products to interface with the redshift. I got sidetracked and didn't go much further with my charger project, but this little box inside the charger is doing all the comms:

 

[Speedkills]

I wonder if the new 3300w rapid charger is just a variation of Elcon TC HK-J 3300W Charger ?

 

[bluefxstc]

The picture is of the CAN controller/adapter for the FPC charger. I have an account with Elcon, I will write them and ask if they know how it is programmed and if we could use one of their 5KW chargers. That would give one hour charging, providing the bike wiring and systems can support it. 5kw at 350Vdc would be 14.3A and it would be higher at lower voltages (when the pack is discharged) so we need to be sure the bike can handle higher currents. I will also need to see if I, or they know, what connector Alta uses for the charger since it is not a standard ev plug.

 

[bluefxstc]

I received a response back from Elcon. They do provide the chargers for Alta. With a CAN controlled charger the vehicle tells the charger what to do. This is normally a voltage and current request so the BMS on the bike is communicating with the charger to tell it what it wants and the charger is communicating back to tell the bike it's status. This is from the Elcon site. " The charger expects to receive every second a message from the BMS with CAN ID 1806E5F4 and 8-byte data with the voltage and current requested. If the charger doesn't receive a valid CAN message in 5 seconds, it stops charging until it receives a valid CAN message. The charger sends out every second a CAN status message with voltage, current and status information". So another, higher current/power, Elcon charger may work with an Alta, but it depends on what the bike ask for. In my case with a EX, the standard charger is about 2700W on 240Vac which works out to a little less than 8A. If the bike is limiting the current to 8A through the CAN commands it sends to the charger, it doesn't matter how big a charger I use, the charger will only give 8A because that is what the bike asks for. If on the other had the bike tells the charger to give a higher current, say 20A, not unreasonable since the programmers know the size of the charger is the ultimate limit, the charger will give the lower of the charger limit or what is commanded. In the case of a 5Kw charger, this would be the charger which will deliver about 14.2A at 350Vdc. Does anyone know if a Alta fast charger will work, and charger at the higher rate, on a MX or EX that originally came with a standard charger. If it does, then it means that the current request from the bike are at least at the higher power level (approximately 9.4A) of the Alta fast charger.

What Elcon doesn't provide is the charger to bike connector. This is apparently put on by Alta after the charger leaves Elcon.

I hope this is somewhat clear. I am not a programmer, but if you have a question, ask away and I will try and clarify.

 

[snydes]

If I recall the 3.3kW charger does not charge an MX or EX any faster. So I guess then we can assume at this point is that the limitation is what the bike is telling the charger. Thanks for your effort looking into this.

 

[Philip]

The fast charger does charge an MX or EX a little faster.

Both offboard chargers work on all bikes (base and R models).
The R pack will benefit the most from the faster charger because the cells "run cooler."
The benefit of running the faster charger on a base bike is modest.

 

[Jon]

Just found this thread, and it looks like you all are looking for the same thing I am.

Any word on the charging protocol between the Alta Charger and Redshift? I imagine it would be something similar to the J1772 AC standard (which is very well explained here: The DIY J1772 Charging Adapter - 300MPG.org ), but with the the AC replaced with some unknown DC output. We would need to know what the bike is looking for in terms of hand shake and supply voltage/amperage.

As a starting point, do we know if the charger is outputting 350V? @Mark911 's tear down of the battery shows the 18650's arranged in a 6p16s config, which run at a more reasonable 60V that can be added up for 350V as needed. That makes me hopeful that a homebrew 6p16s module in a saddle or tail bag could be used as an 'emergency fuel supply', if only we knew what the bikes CAN was looking for from the charger.

Back of the napkin design works shows that you could fit such a module in about a 5" diameter, 10" long cylinder weighing 11-15 lbs. Add a few more inches for a BMS and you could put one in each side of a Mosko Moto Reckless 10! ... or maybe a module on one side and a fire extinguisher on the other

 

[bluefxstc]

Communication between charger and bike is CAN so more dynamic than J1772 which is basically just signal levels. The car tells the EVSE that it is plugged in, EVSE tells the car how much current it can give and charging starts. The car/car charger handles all the rest. With CAN there is a lot more back and forth communication between the charger and the bike.
Here is what Elcon says about controlling their chargers with CAN:

How can I use CAN Bus control with my Elcon PFC charger?
If your Elcon PFC charger has been programmed for CAN Bus control and you received a CAN Bus adapter with your charger, pin 1 of the adapter must be connected to CAN-Lo from your BMS and pin 2 to CAN-Hi. If your BMS does not already have a 120 ohm termination resistor across CAN-Lo and CAN-Hi, you will need to add one. The charger expects to receive every second a message from the BMS with CAN ID 1806E5F4 and 8-byte data with the voltage and current requested. If the charger doesn't receive a valid CAN message in 5 seconds, it stops charging until it receives a valid CAN message. The charger sends out every second a CAN status message with voltage, current and status information. Up to 4 chargers with different CAN IDs 1806E5F4, 1806E7F4, 1806E8F4 and 1806E9F4 can be connected to the same CAN bus and be controlled by one BMS.

As for voltage, I would guess, the final charge voltage is probably about 10% higher than the pack voltage so 375-390 Vdc, but that is just a guess. Not sure what the exact pack voltage is, but you normally charge to some higher voltage than pack voltage and then after you shut the charger off the cells settle down to the pack voltage.

 

[Jon]

bluefxstc, I totally skimmed over your first post about the CAN bus control info from Elcon, my bad! Sorry to make you repeat yourself.

Based on the tear down I mentioned above, there are at least 5 high power switching transistors which allow modules to switch from series to parallel. If those transistors are controlled during the charging process, there is hope that we are charging at a lower voltage and there may be OTS DC-DC CAN Bus controlled charge controllers that could be used to replicate the communication protocol between the bike and charger. If we are running 350V, we might be shit out of luck for DC-DC controllers, but that seem like a dicey choice for cell balancing... I hope they split the modules out!

In any event, it seems like the next step toward unlocking DC-DC charging is measuring the charge voltage and monitoring the CAN bus signal. Should be able to get charge voltage with clamp style ammeter, considering that we have a pretty good idea of power. Depending on the CAN hi and CAN lo protocol, we might be able to pull the bike's requested voltage and current straight out of the signal. But once we know what the bike BMS is asking for in terms of CAN signal, any programmable charge controller should be able to talk to the bike.

I might have some free time in the next few weeks to order a CAN bus analyzer and start looking into this. I'll update this thread when I know more.

 

[bluefxstc]

I would be surprised if they break the pack down and charge at a lower voltage. That would increase the current and the weight of the wires to carry the extra current. The bike is already slightly heavier than its gas counterparts so they would want to make it as light as possible. All of the OEM EVs that I know of charge at the pack voltage so charging at 350-400Vdc is not a big deal for the industry and Porsche is talking, or may be going, to 800Vdc. Higher voltages have big advantages in transmitting power in reduced weight and losses.

If you can sniff out the CAN traffic it should tell you exactly what the bike wants and what the charger is sending. Just need to look for the CAN ID and decode the binary number associated with it. I have a CAN sniffer, just need to get the time to figure out how to use it and get the right connector for the bike.

 

[SuburbanDad]

Afaik, any node on the bus can read the entirety of the bus, so you should be able to sniff the comms on any of the available pigtails.

Incidentally, I bought a new map selector switch and lo and behold, it has the same flat 4 canbus connector. If I get a block of free time I am going to cut the connector off the old switch and use it to sniff the bus traffic. Charging protocol would be super interesting traffic to sniff.

 

[TonyWilliams]

As for voltage, I would guess, the final charge voltage is probably about 10% higher than the pack voltage so 375-390 Vdc, but that is just a guess. Not sure what the exact pack voltage is, but you normally charge to some higher voltage than pack voltage and then after you shut the charger off the cells settle down to the pack voltage.

That’s not how modern lithium cells are handled. The BMS will not allow the charger to exceed 84S * 4.2V = 354 volts.