Hello friends I am Halil
I have designed all in one board for open inverter project which is shown below. I follow version 3.4 controller board design and i made few changes.I am waiting your feedbacks and suggestions.
This board designed for 48v systems but it can be adapted up to 400 volt.
johu wrote: ↑Mon Apr 25, 2022 7:16 am
Oh cool, always wanted to do something similar but then OEM inverters came along.
What ratings do you expect and what file format / editor is it?
Hello johannes
Control and power section fully isolated it can work on 400 volt but need to make test about it, fets are a bit close to each other. It is same with current also 100 amper will not be issue theoretically but current pads need to fill with solder properly.
Нi. You can use such a design, but with such an arrangement of power transistors, there will be uneven heating. I have experience of scaling for large currents, the problem is that currents 10 times higher than the phase ones are walking along the minus and plus buses. Perhaps in your case, the imposition of thick copper tires on the board will help
yaroslav wrote: ↑Mon May 09, 2022 7:25 pm
Нi. You can use such a design, but with such an arrangement of power transistors, there will be uneven heating. I have experience of scaling for large currents, the problem is that currents 10 times higher than the phase ones are walking along the minus and plus buses. Perhaps in your case, the imposition of thick copper tires on the board will help
Hi thanks for your feedback. I am aware that current and heat dissipation limit of the board. Discreet MOSFET paralleling has better heat and cooling performance, with proper liquid cooling system heat problem can be fix. For PCB traces i am planning to cut metal blocks for high current paths like Curtis controllers have. But before all of that the board need to be tested
Maybe someone in forum willing to help in test process feel free to contact me
Romale wrote: ↑Sat May 21, 2022 10:37 pm
Hi! I would like to try to make it, but with an amendment to 136 volts (32s li-ion) and the most powerful mosfets possible
Hello Romale
Thanks for your interest yes we can try with 136 volt
You can share the process here so forum and me can help
Hi, so if you can fix the board to use 30 mosfets to-247 and replace the hall-based current sensors (I'll give a link to a good sensor) I will produce at least five samples at the same time, tob can be used on different equipment by conducting comparative tests!
Romale wrote: ↑Wed May 25, 2022 11:57 am
Hi, so if you can fix the board to use 30 mosfets to-247 and replace the hall-based current sensors (I'll give a link to a good sensor) I will produce at least five samples at the same time, tob can be used on different equipment by conducting comparative tests!
Romale wrote: ↑Thu May 26, 2022 7:46 pm
Thanks. Let me know when it's ready and I'll start making the board. If you want to test it too, I can make it for you as well.
Thanks
I am working on it now i will upload files when its ready
Hi! How are you doing in changing the board? I can't wait to make and test for bugs and on a motorcycle. And then summer will fly by quickly and it will be cold)))
Hi! How are you doing in changing the board? I can't wait to make and test for bugs and on a motorcycle. And then summer will fly by quickly and it will be cold)))
[/quote
Hi
I am working on it also i redesigned the board but the current sensor you shared is not suitable for pcb mount need check another one if it is ok for you
I am working on it also i redesigned the board but the current sensor you shared is not suitable for pcb mount need check another one if it is ok for you
I think that the current sensor can be put aside on the phase terminals. By the way, how are you supposed to implement the phase output? How is the arrangement of mosfets and their cooling implemented?
The Kelly KLS7230C controller uses 30 mosfets to-220 to 100 amperes and it perfectly outputs up to 300 amperes of battery current and up to 450 amperes of motor current. It seems to me that if we implement 30 mosfets of IRFP4568PBF, we will easily get 600 amperes of motor current and up to 500 amperes of battery current! the main thing is to put powerful drivers
I am working on it also i redesigned the board but the current sensor you shared is not suitable for pcb mount need check another one if it is ok for you
I think that the current sensor can be put aside on the phase terminals. By the way, how are you supposed to implement the phase output? How is the arrangement of mosfets and their cooling implemented?
Hi
I arranged the mosfets horizontal to the board a proper heatsink place it under the board there is mounting screws, for phase output i am thinking to put screw terminals
Now ı am working on BOM list
By the way ı also have some of the components i would like to sent it to you
Romale wrote: ↑Tue Jun 14, 2022 5:38 pm
The Kelly KLS7230C controller uses 30 mosfets to-220 to 100 amperes and it perfectly outputs up to 300 amperes of battery current and up to 450 amperes of motor current. It seems to me that if we implement 30 mosfets of IRFP4568PBF, we will easily get 600 amperes of motor current and up to 500 amperes of battery current! the main thing is to put powerful drivers
For high current paths ı am designing with out solder mask so we can put copper terminals for it. We can see that amount current but not to long i think:)
do you think locking drivers will be needed for heavy-gate transistors? When I did without them, the mosfet did not close for a long time and there was heating. I started using a locking driver and everything started working fine.
In this case, the pull-up resistor is made by several kilohms and RG is calculated as before. When a positive pulse is applied, it passes directly through the diode D_ON and charges the Gate of the mosfet. When the output stage of the main driver is closed, then through the pull-up resistor PULL_DOWN opens Q_OFF and instantly discharges the Gate charge through itself, which leads to the instantclosure of the mosfet!
Romale wrote: ↑Fri Jun 17, 2022 10:07 am
do you think locking drivers will be needed for heavy-gate transistors? When I did without them, the mosfet did not close for a long time and there was heating. I started using a locking driver and everything started working fine.
Hi
We can try this solution but we are also using neagative gate voltage for fast turn of.
if you have good combination of gate circuit with turn of transitor please sent it
