jakepusateri
100K Ohm Resistor
I personally do not think the car communication will be the hard part. Everest has that part solved. I can build this board https://github.com/PionixPublic/reference-hardware#yak and it should work out of the box.
DC Direct EV charger - DIY Build
- Thread starter jakepusateri
- Start date
pollenface
Solar Addict
Can you provide an example? This is unknown to me.
Warpspeed
Solar Wizard
There are reasons why certain things are never done.
For example, why are there no five litre single cylinder engines in cars ?
Surely it would be cheaper, fewer parts and only one spark plug to replace !
Nobody uses high power boost converters for one simple reason, the peak currents are just far too high to be practical.
Same reason nobody builds high power buck converters with a large voltage step down ratio.
Try to find a commercial solar controller that converts 500v down to 12v using a buck converter.
For converting 500v at 30 amps dc (15Kw) to a different voltage, only one topolgy is practical, that is a push pull transformer coupled inverter.
You can put any secondary winding you want onto that to suit the required dc output voltage.
Cal suggested this way back at post #9, and he knows of what he speaks.
For example, why are there no five litre single cylinder engines in cars ?
Surely it would be cheaper, fewer parts and only one spark plug to replace !
Nobody uses high power boost converters for one simple reason, the peak currents are just far too high to be practical.
Same reason nobody builds high power buck converters with a large voltage step down ratio.
Try to find a commercial solar controller that converts 500v down to 12v using a buck converter.
For converting 500v at 30 amps dc (15Kw) to a different voltage, only one topolgy is practical, that is a push pull transformer coupled inverter.
You can put any secondary winding you want onto that to suit the required dc output voltage.
Cal suggested this way back at post #9, and he knows of what he speaks.
fromport
Solar Addict
What about something like this ?
Specs say 20 amps, but as a concept ?
Chademo 2 went down to 50 volt: (info from 2017)
Now with the modern 800 Volt EV's up to 1000 volt could be needed!
Toyota Prius has 20kW synchronous boost converter to step up 200V battery to ~500V for motor inverter. I modeled boost (and buck) converter in PowerEsim and got around 98.5% efficiency with SiC mosfet and diode after fine tuning inductor selection. I would not do 15kW in a single power stage since that would remove the benefit of single series string MPPT. I think 10A input / output 400V converter should be reasonable. Nice thing with boost converter the efficiency gets better the closer input voltage gets to output. PV array can be sized to have MPP real close to battery voltage for max efficiency and as panels get shaded and bypassed by diodes total power going into converter drops and output current drops as well.
Warpspeed
Solar Wizard
Sure, anything is possible.
But forget about a home built boost or buck converter at those kind of power levels.
But forget about a home built boost or buck converter at those kind of power levels.
Last edited:
zanydroid
Solar Wizard
Is this AC or DC charger? Very important to be detail oriented……..
EDIT: I just clicked the link and it literally has two places where it’s advertising its AC nature. Namely the option to select single or three phase
Another example of a DC car charger connector that attaches to DC bus is Dcbel, don’t think it’s out, also $$$
Warpspeed
Solar Wizard
If you check out how those (crappy) high frequency inverters work, you usually find the incoming battery voltage is first stepped up to usually around 350v to 400v dc, and then PWM'ed directly to create the higher voltage ac output.
The 48v (or whatever) boost stage up to the higher dc voltage, always use a push pull transformer coupled switching power supply, often in multiples to generate the high voltage dc. You never see the classic "boost converter" using flyback from a dc choke where large voltage differences or very high power are required.
It does work wonderfully well at low power, or where the voltage only needs boosting by a comparatively small amount at higher power levels.
The reason being that the current and voltage requirements for the switching device become excessive for large voltage step up ratios.
Its always much more efficient to use the turns ratio of a transformer to greatly increase either voltage or current, than try to do it with just a choke, in either a boost or a buck converter.
The 48v (or whatever) boost stage up to the higher dc voltage, always use a push pull transformer coupled switching power supply, often in multiples to generate the high voltage dc. You never see the classic "boost converter" using flyback from a dc choke where large voltage differences or very high power are required.
It does work wonderfully well at low power, or where the voltage only needs boosting by a comparatively small amount at higher power levels.
The reason being that the current and voltage requirements for the switching device become excessive for large voltage step up ratios.
Its always much more efficient to use the turns ratio of a transformer to greatly increase either voltage or current, than try to do it with just a choke, in either a boost or a buck converter.
zanydroid
Solar Wizard
How would a non-crappy HF differ from crappy ones?
Wouldn't salvaging a AC-DC OBC or DC-DC converter on charging path (there are 400V<->800V ones) out of a scrapped EV count as home built? Ish.
Warpspeed
Solar Wizard
They ALL suffer from the same problem.
The switching power supply that generates the higher voltage dc creates a power bottle neck.
It has a maximum continuous power transfer rating, and that is it.
Beyond that there is NO additional surge capacity at all.
Low frequency transformer inverters can, and usually do have a massive surge capacity, often of MULTIPLES of full continuous rated power.
Many types of loads have an initial turn on inrush surge current.
A good inverter can handle that easily, a crappy one often just goes *BANG*
jakepusateri
100K Ohm Resistor
In my case I don’t need to change the voltage all that much, just +/- 20% to account for flexibility with temperature and make sure we’re getting the maximum power point. I also don’t need surge capacity since I’m controlling the load and how much I push into the battery. As for high power, each mppt will only be around 6kw max, (15a, 400v). These units could be paralleled after the mppt but that’s for later if ever.
I totally agree. Voltage rations beyond 2:1 are better served by transformer topology. We are talking about boosting ~330Vmp panel string up to 330 - 400V battery at single series string current levels. This is exactly how it's done in HV input AIO hybrid inverters.
zanydroid
Solar Wizard
If you can get CCS board, find a MPPT off AliExpress with enough programmability on the output voltage and current limit, and are competent to write the integration software, maybe you can make it work.
Warpspeed
Solar Wizard
Yes you are quite right.
Some here are talking about 15Kw and voltage boosting for potentially future 1,000 volt EV batteries.
Buck or boost converters work pretty well for slightly tweaking the voltage, more for precise voltage or current regulation than performing massive voltage changes.
Warpspeed
Solar Wizard
Maybe, but the hardware was originally built to do something reasonably specific, and there are definite limits to how far you can push that.
jakepusateri
100K Ohm Resistor
I would love to do that and not build the mppt, but I can’t find any that are for high enough voltage battery output and don’t also have an inverter in them (which I do not want, and increases the price). Happy for any links too.
zanydroid
Solar Wizard
I think the Hybrids with HV battery support are cheap-ish relative to the known alternatives. My headcanon is that they're making money off selling you the HV battery.
So going this path, you might look for something that already has one of the supported HV battery comms protocols decoded, and then mind meld that with the CCS control portion. EG, tell the HV battery what the charging voltage and limit should be (NCM is going to be a different curve from a LFP HV battery) based on what the CCS portion says.
There's going to be a lot of questionable quality github code bashed together, along with original new code gluing it together.
I'm not sure how one would incrementally test this if already the right kind of engineer/technician. Or how the wrong kind of engineer/technician incrementally gets the knowledge to work on this.
jakepusateri
100K Ohm Resistor
Most of the 800v cars have 400->800v converters onboard anyway so they can charge at 400v DC fast chargers. Minimum 50kW on the Taycan I think. I don’t plan on going near that power wise, so 400v is my target.
