15000W LF, 48volt to 240 volt, what wattage?

The PJ driver seems to solve this problem by using small gate resistors (sorta bypassing what Miller capacitance issues are left)--and violating the manufacturer datasheet requirements for the driver chips.   by NOT putting the required ceramic capacitor anywhere on the board 

The   LED  not turn on  seem to make the LF driver  work better .   The  picture of the NCEP039N10  mosfet  show  a diode  between the  D and S  .  A  15 volt  zener  diode  may be  better  to save the mosfet from  blowing up  since the  max voltage is 20 volt .   

<a href="/monthly_2021_11/1421026852_Screenshot(753764).png.6ad284ba5bc47d77fa0d520509282093.png" class="ipsAttachLink ipsAttachLink_image"><img data-fileid="882" src="//forums.genetrysolar.com/applications/core/interface/js/spacer.png" data-src="/monthly_2021_11/546784759_Screenshot(753764).thumb.png.ffb05ecbe4155dd48e710563d33a0453.png" data-ratio="56.1" width="1000" class="ipsImage ipsImage_thumbnailed" alt="Screenshot (753764).png">

This is the internal body diode (generally rated for the full FET amperage); actually for battery charge, all the output power flows through the body diodes on the FET H-bridge.  An inadvertent bridge rectifier if you ask me.

Between G-S, yes.  For any GS upgrades, we have had to install zener diodes across the G-S lines of the FET board--otherwise random FET failures are guaranteed.  (Strangely enough, this usually would occur at very small loads.)  Use of zener diodes in that fashion solved FET drive problems--at least until the magical 6kw barrier was reached.  Beyond 6kw, the crosstalk in the ribbon cable and mainboard becomes fatal to the FETs, causing extreme heat as they deadshort each other across the battery until the solder melts off the leads, FETs fail, and the whole inverter stops working.

 Beyond 6kw, the crosstalk in the ribbon cable and mainboard becomes fatal to the FETs, causing extreme heat as they deadshort each other across the battery until the solder melts off the leads, FETs fail, and the whole inverter stops working.

Since PJ inverter do not  have the  G to  S  zener diode  this mean  the  25kw  PJ  inverter  will blow up  at  continuous  8kw  .   My   15kw  at  5000 kw will  overhaet in  one hour .  PJ  do not  make the  long  mainboard  because the crosstalk  inside  will  blow the mosfet and  short the mainboard .    I use the  damaged  long mainboard  for capacitors  extra storage  .     I  order the  PJ  mini mainboard  to  try to stop the crosstalk inside the  mainboard .   It  will  be  very difficult  to make a 12kw  inverter run continuously  as  Growatt and SolArt  recommend  spare mosfets  when buying their  12kw  inverter .  In the winter  I  can  run my  15kw PJ  16 hours  at  4000 watt continuous.    I   learn a lot  about  why  the PJ  failure  on this  forum  so I  think I  can keep  the PJ   running for a long  time  even  after  this forum  wiil  shut down in a  few days  .  Thank  you .  

Can you provide links to these claims?  Definitely got my curiosity up 🤣.  Growatt, yes.  SolArk....ummm, really??

After the forum will shut down??  Who said that?

I know Sean mentioned that the forum might have to close down several months--but with sales moving, I don't think that's an issue anymore.  Especially if we get a new website with an integrated forum (eliminating that expense), that'll really help future-proof Genetry Solar.

Can you provide links to these claims?  Definitely got my curiosity up 🤣.  Growatt, yes.  SolArk....ummm, really??

Yes I  seen  on youtube  but it will take time to find  because  so many  youtube  video  on  Growatt and  SolArt review   but  no new  video on Powerjack  and a few  on GS inverter .  

Not necessarily.  Due to the reduced drive strength (assuming the voltage does not spike higher from noise induced in the wires/board), this doesn't seem to be as much of a problem.

But for whatever unknown reason, they run the high-side FETs at 18v...barely 2v under the threshold of the NCEP039N10 FETs.  (Older HY3810s have a 25v max.)

FET failure due to gate overvoltage is usually random and unexpected.  For example, Sean would load-test the early GS control board designs one day (5-6kw for 30 minutes)...and then the inverter would suddenly blow all the FETs just running 150W of lights the next day.

FET failure due to gate overvoltage is usually random and unexpected.  For example, Sean would load-test the early GS control board designs one day (5-6kw for 30 minutes)...and then the inverter would suddenly blow all the FETs just running 150W of lights the next day .   Yes my  15kw  PJ  run  6kw  and  the next day  all the  FETs blow up  when the  switch turn  with no load .    I  change  to rev 11.1  control board and  different  FETs  and  works good  except  for the alarm  with  inductive  load  but the FETs  are good .    The  closup  picture  of the DIP SWI  switch are all set  now to OFF and  no more alarm and run inductive  load  ok .  

Regarding blowing up FETs in Growatts and Solarks,  I really can't say much about the solarks but only because I've never come across one here.  There are plenty of Growatts in my area, both grid tie and standalone.  I haven't heard of any Growatts that blew up FETs without good cause, and by good cause I mean more than doing something stupid on the AC output of the inverter.  Lightning is the typical cause.

I did see a USA bsed youtuber grizzling about his Growatt 48V 5kW inverter/charger going bang, but to be honest intuition says it was a fake based on where he bought it.

Regarding blowing up FETs in Growatts and Solarks,  I really can't say much about the solarks but only because I've never come across one here.

The youtube video  show the  FETs  never blow up  because of  over ampage  limit protection like the  GS inverter .  I  have not seen anyone on  youtube  run offgrid  inverter continous  at 12kw  .   The dealer for Growatts and Solarks do not  sell repair parts but do  return on a case by case  basis .   The max  for the 2  inverters is 9kw  not thr 12kw as advertise .  The picture I  found of the  FETs in the Growatts  inverter .   MPP  inverter  do sell  repair parts  and is recommended  .  I will never buy  HF inverter . 

So nothing to back up the statement about Growatt and sol-ark recommending people buy spare FETs?

The recent all-in-one HF Growatt inverters are Voltronics.  Their LF units seem to be inhouse and I haven't been able to determine the origins of their transformerless grid tied inverters but I don't really play in that market anyway.  I have repaired damaged Growatt LF inverters, but will not repair grid tied transformerless, too much liability attaches there.

In case anyone was wondering, it was fried serial ports.

Serial ports usually cop a beating when lightning is around.  Even devices that use opto / magentic isolation can be zapped if proper attention to clearances isn't given along with something to clamp transients so they don't exceed what ever the expected breakdown voltage is.  I saw the same in the past when I was repairing PABX cards and ... networked cash registers.

Serial ports usually cop a beating when lightning is around.  Even devices that use opto / magentic isolation can be zapped if proper attention to clearances isn't given along with something to clamp transients so they don't exceed what ever the expected breakdown voltage is.

I  use  surge protector  for the  whole house and  a surge protector for the heat pump .  Do  they  use surge protector in  Australia ?   

6 hours ago, dickson said:

I  have not seen anyone on  youtube  run offgrid  inverter continous  at 12kw  .   The dealer for Growatts and Solarks do not  sell repair parts but do  return on a case by case  basis .   The max  for the 2  inverters is 9kw  not thr 12kw as advertise .  The picture I  found of the  FETs in the Growatts  inverter .   MPP  inverter  do sell  repair parts  and is recommended  .  I will never buy  HF inverter . 

Took a closer look at the video.  Appears (from what blurry detail I could get) that the main processor is a Texas Instruments TMS320F28035.  Talk about powerful processor compared to what's (quite adequate) in GS inverters...!

• 120MIPs (vs 16MIPs)
• 32-bit (vs 8-bit)

...but does cost 3x as much.

Worth noting: Growatt (at least the SPF6000) uses the exact same FET driving methodology as PJ and the rest of the Chinese-designed LF inverters.  If it works for them, great--but I just think it's so funny how little innovation there is in the Chinese inverter market:

<fileStore.core_Attachment>/monthly_2021_11/image.png.804e86e783bb9f76dc80c2c0095b70cf.png

• Yellow circle: low-side FET drivers (discrete transistors forming a totem pole driver)
• Green circle: high-side FET drivers (betcha they're the same TLP350 chips)
• Orange circle: multitap transformer-based power supply (2 floating high side taps, and at least 1 more tap for low side and logic)

While the layout is far more concise here, it is worth significant note that I do not see ANY ceramic capacitors anywhere near any of the FET driver quadrants.  Perhaps this is the trick after all, to prevent the FET drivers from providing a fast rise time to the FETs...by not giving them a strong power source in the first place.

I'd be curious to see whether this trick still works in their 12kw inverter...

Found a video of a Growatt 12kw disassembly, and guess what I noticed on the control board:

<img alt="image.png.71da04713f99762b49c2899c34603469.png" class="ipsImage ipsImage_thumbnailed" data-fileid="895" data-ratio="114.16" style="height:auto;" width="339" data-src="//content.invisioncic.com/g308908/monthly_2021_11/image.png.71da04713f99762b49c2899c34603469.png" src="/applications/core/interface/js/spacer.png" />

Yup.  Same driver circuit.

Also worth noting that they only went to 8 FETs/quadrant (same IRFB4110) as opposed to 6 FETs/quadrant in their 6kw.

Yup.  Same driver circuit  

It is part of the control board so  the design  will  always work  and not blow up the FETs .  Other  review on youtube say it is impossible to run 12kw continuous  because  the current limit will shut down  at 8kw .  This way nobody know  what happen to the FETs  at 12kw and  never get a bad review ?   It is very difficult to run  12kw continuous  that even the 30kw powerjack with the ASL11  transformer  can not run 12kw  continuous .