20,000watt AMG version PowerJack LF PSW SP

ups delivered it today November 27th, 2021. the box says 84 pounds so it can not be anything like my other powerjack 20k 48-volt inverter that weighs about 160 pounds.

must be a tiny transformer compared to my other one, have to open it up but no time today.

must be a tiny transformer compared to my other one, have to open it up but no time today.

Like to know if  the transformer is 32vac  ASL6.5  or ASL9.0    which would be better for  lithium  IRON phosphate battery  56 volts .  The control board should be rev 11.1  .  Like to know if the  LED on the LF driver always stay OFF .   Thank you .  

16 hours ago, pilgrimvalley said:

the box says 84 pounds so it can not be anything like my other powerjack 20k 48-volt inverter that weighs about 160 pounds.

must be a tiny transformer compared to my other one

It's like a box of chocolates. You never know what you'll get.

Powerjack these days will literally put anything into an inverter and call it whatever size/rating they feel like.  It's why I won't buy any more new powerjacks from now on.  (Still like my older pjs!)

... but it'll probably be brown and sticky.

LOL indeed. 

I wonder why nowadays they always put one less capacitor on the mainboards than there are positions for? Just another penny-pinching excercise?

I've had this on both of the 8kW (claimed) mainboards that I have had in the past year - one 10.3 and one 11.1,  and I notice the same on your mainboard. I have added the 4th cap myself on both boards as they don't cost much and my assumption is that 4 caps are better than 3! Nothing untoward has happened yet despite several months of use.

However I do agree the old ones are better - my old 2013 PJ 8k is still the best performer and doesn't have any heat problems since I replaced the aged unbranded fan with a Delta 24V 1.5A one last year. I just don't use the old one over winter as the high idle current of the twin transformers wastes too much of the precious little solar power that we get Oct-Feb.

Can't argue with that!  The older large mainboards had 6 caps, then they started only putting in 5 of 6.  Then they changed to the small boards with 4 spots and almost immediately only filled 3 of 4.  Maybe 3 is "enough" but I feel better with 6.

Comes down to acceptable levels of AC load induced ripple and droop under abrupt load increases.  Lots of ripple is not good for the longevity of the battery (LA, li, what ever type you like).  Yes, they might be using capacitors with high capacitance and better ripple current ratings to counter the lower number... but I seriously doubt that 🙂

Given that PJ is mostly a work of fiction anyway I doubt that any of this this is given even 1ms of consideration.

17 minutes ago, TheButcher said:

Given that PJ is mostly a work of fiction anyway I doubt that any of this this is given even 1ms of consideration.

And you would be right.  Measuring capacitor ripple current is not the easiest, but I'm pretty sure the difference would be easily spied with an oscilloscope on an inverter under load.

They had a spate of bad caps recently (the black ones, not the brown/purple ones they're using now)...must have been cheaper, but they ended up with a rash of capacitor failures causing FETs to blow out--yeah, it was colorful.  Caps would leak electrolyte / corrode off their own leads, etc.  It was bad enough that they went back to the purple ones.

On 11/30/2021 at 11:09 PM, Sid Genetry Solar said:

And you would be right.  Measuring capacitor ripple current is not the easiest, but I'm pretty sure the difference would be easily spied with an oscilloscope on an inverter under load.

They had a spate of bad caps recently (the black ones, not the brown/purple ones they're using now)...must have been cheaper, but they ended up with a rash of capacitor failures causing FETs to blow out--yeah, it was colorful.  Caps would leak electrolyte / corrode off their own leads, etc.  It was bad enough that they went back to the purple ones.

Thankfully all three mainboards (the v1.4, 10.3 and 11.1) that I have are all fitted with the same brand of 'brown' capacitors. Not seen any purple ones. The ancient v1.4 had 4, but the newer two were only fitted with 3 out of 4. I wasn't able to find a source of the brown so the ones I bought for infill of the missing caps were these:  https://www.ebay.co.uk/itm/384324739356  

They look different to the photo's of the ones I have seen on blown PJ boards so hopefully will last, especially as there there are three of the 'good' ones already on each mainboard. But I bought 4 so I have a couple of spare ones. Been running the house and garage during daytime and evening on the v10.3 for nearly 3 months now with some fairly heavy loads on a daily basis. checked this morning and the new cap still looks ok (no obvious leakage).

Been running the house and garage during daytime and evening on the v10.3 for nearly 3 months now with some fairly heavy loads on a daily basis. checked this morning and the new cap still looks ok (no obvious leakage)

I  added  6 more capacitor outside of the  inverter for a total of 12  capacitors  and no problem  running for a month now with the rev 11.1 control board .   12 capacitors help  start  inductive load  like microwave oven and  5hp  air compresser .    

still have not opened it (the 20K AMG version inverter) but got more solar panels mounted on the array structures (unistrut and 2x6 wood triangles at 45 degrees fixed ground mounts with some concrete blocks to level them), and 28 of 32 LiFePO4 Eve cells LF280K have arrived along with some other battery necessities. going toadd more ballast but not much problems although the wind has taken one and added a dent to it before i got it mounted. Thirty  250-watt PV panels mounted now>>>> a feat for this old one man operation.

thanks for all your posts and comments!

cheers all from South Dakota. looks like the sun is going to shine so time fore more solar panel and battery and inverter work out in the off-grid solar build. 😎

I guess they are a lot more "brown" than "purple."  Here's a side-by-side between the dodgy caps and the ones they're currently using:

16 hours ago, Paul said:

so the ones I bought for infill of the missing caps were these:

Those are 22000 uf at 50v to replace 10000uf at 80v?  Seems a bit off spec, especially if you are mixing them with original 80v caps.  Just saying....

So long as the voltage rating is higher than expected peak voltage input by a safe margin it shouldn't be a problem.  The large increase in capacitance may present a problem for the PCB if the inverter is ever dead dropped onto a battery etc without precharging, depending on the track width and copper thickness.  Fuses / breakers may need to be uprated likewise.  Of course it should never be done but its the things that are never done that always get you.

Presumably Paul has a 24 or 36 system?