teslanv wrote: ↑
Jan 10, 2018 11:30 pm
ElectricGod wrote: ↑
Jan 10, 2018 8:40 pm
Gotta love the blatant false advertising! Who reads the fine print? West coast REALLY needs to fix their waaaaay over the top advertising!
6kw on a 12 fet controller...uh...yeah...not going to happen.
10kw on an 18 fet...same.
Those are flat out maximums with zero room for anything to go wrong.
OK...if you go with TO-247 mosfets, but in anything in TO-220...no NOT with any mosfet that exists today and certainly not with the TK72E12N1!!!
With the AOT290 which is a 2X higher wattage mosfet, I run the 12 fet controller at 4kw continuous. The 18 fet on the same mosfet at 6kw continuous. Those are realistic specs! 6 and 10Kw....hahahahahahahahahahahahaha...what nonsense! Good thing you don't carry the 24 fet, 150 volt version...that would be a 50kw controller right?
Yes, those current ratings are maximums. I note them as such. And I race them at those power levels without issue. Perhaps you would like to buy one and prove me wrong?
I already have 3 of the 12 fet, an 18 fet and a 24 fet of this exact design. I'm about to buy another 18 fet and upgrade it with the best of the best of the mosfets I can find. By chance, do you have the new 18 fet version that supports sensorless and has copper rails on top of the board? If you have that, I might just send you some money for one! My current 18 fet is the older style.
For any of these controllers, I replace the phase and battery wires with whatever is the largest I can cram in the holes. I pull the factory shunts and replace them with larger precision .005 ohm shunts. 4 in parallel gets me to .00125 ohms which is just above the 1 milli ohm limit for the MCU's current sensing circuit. I replace the mosfets with AOT290's and add a secondary heat spreader inside the shell. I also push the mosfets all the way down to the wide section on their legs into the board. I want to minimize the 75 amp leg limit as much as possible and the heat problems that result. I feel comfortable running at 4kw continuous on a 12 fet or 6kw on an 18 fet after those changes.
I too want performance out of a small package, but I'm also not delusional about reliability and longevity. Running a controller at it's maximum limits is a recipe for blown mosfets. I seriously doubt you run a 12 fet at 6kw for long without killing it. With the best mosfets I can test, I'm still skeptical of more than 4500 watts continuous on a 12 fet. I destroyed my second 12 fet doing this sort of over the top, run at the ragged edge of the limits testing. It didn't last long.
Can I run the 2 liter engine in my car at 500 bhp? Sure I can. That's just a few engine upgrades and a retune. Will it last 50,000 miles? Ummm...heck no! You're trying to do the same thing with a controller. It's just not going to last very long running it at it's maximum limits. I want to get several thousand miles of reliable use out of a controller. If I get less than 4000 miles, there's something wrong. No...running at the maximum limits is not a good recipe for reliability. THat's why I posted what I posted.
Your advertised specs are just not realistic. At least change that big bold print so it's honest. Say something like "6kw maximum, 4kw continuous"...that's realistic and not misleading. I shouldn't have to read the tiny fine print to discover that you don't recommend running a 12 fet at 6kw. Gee really? I can't run at 6kw and have it last more than a day or 2?
I'm currently in the process of testing mosfets, not just believing manufacturer spec sheets. I have a variety of TO-220 mosfets in my possession that are the best of the best in 100v. I will be testing their real Rds at Vgs(th) and 10 volts. Later I will take a mosfet of each model and load test it until failure at 50 volts and at 100 volts to see real current handling. My old temp reader died so I need to buy a new one for this test. A higher wattage mosfet should be able to get hotter before failure than a lower wattage mosfet. My test will use a square wave so that I'm simulating a fixed speed PWM signal on the gate. It won't be exactly like a real motor control signal which is PWM, but close enough that it won't matter. There will be no back feeding through the internal diode. The test will reveal what mosfet handles the most current while switching on and off at real switching speeds with the least amount of heat. In a controller or on a test bench, that mosfet will perform best. I have purchased Infineon, AOT, IXYS, Toshiba and TI mosfets to test out. What got me doing this was doing spot checks on AOT290's and seeing some of them be out of spec. That got me wondering if off spec components was common or not. The test regimen has evolved from there.
My mosfet test results will be posted here.
https://endless-sphere.com/forums/viewt ... 10&t=91589