PLEASE NOTE: If you had an account with the previous forum, it has been ported to the new Genetry website!
You will need to reset the password to access the new forum. Click Log In → Forgot Password → enter your username or forum email address → click Email Reset Link.
Hmm, will ck the voltage as I have a fluke digital meter. It would seem that rather than risk appliances and the like, I will absorb the loss of the cost of the inverter and try another brand.
2 hours ago, Sid Genetry Solar said:This would cause a "pulsing" output ~2x sec (IIRC), and yes, it would not cause a shutdown.
The easiest solution to that would be to actually remove 2-3 turns of wire from the primary winding of the transformer (which is the outside winding, making it not too difficult), ideally reducing the primary voltage closer to 16vAC. This would give you more headroom for higher loads--and of course, also make it easier to blow the inverter up. YMMV 🤪🤯🤣
2 pulses per second sounds about right to me... Though it does cause a shutdown after about 5 seconds of continuous cycling, for "output unstable". (you don't say...)
You're saying i should open up that bad-boy and unwind a couple turns from the tranny primary windings? Hey, if it's that simple, i will jump for joy and probably put it in my camper...
Checked across the separated heatsinks and got 15.3 VAC. Just for giggles plugged in a light load (about 145W) and tested it and got 16.1 VAC.
1 hour ago, workaholic53 said:Checked across the separated heatsinks and got 15.3 VAC. Just for giggles plugged in a light load (about 145W) and tested it and got 16.1 VAC.
Interesting, that's not bad at all (actually almost GS spec, hehe, gotta love it!) They used to do 36vAC (for 48v) which would be 18vAC for 24v...
Unfortunately without a "window" into the inverter (GS WiFi board, etc.) it's pretty well impossible to figure out exactly what's going on. Quite akin to trying to fix a modern car without a OBD-2 scanner!
1 hour ago, NotMario said:You're saying i should open up that bad-boy and unwind a couple turns from the tranny primary windings? Hey, if it's that simple, i will jump for joy and probably put it in my camper...
Ha, maybe not quite. If it's down at 15vAC or so at no load, I wouldn't recommend reducing the tranny voltage spec too much. Reduce it too far, and the FETs will blow out (been there, done THAT!)
However...it would not be impossible for their voltage feedback circuit to be going wonky under load. If someone having one of these troubled inverters has a 'scope or a fast meter...I would be curious if the AC output voltage (or the transformer primary voltage for that matter!) is exceeding 18-20vAC when it's "pulsing". Because if the inverter gets false output readings, it'll "loop" the throttle regardless of the transformer primary ratio--and for that matter, can be very easily replicated by disconnecting the transformer altogether. You're guaranteed a loop-loop in that case!
(FWIW a GS inverter will give an "Xformer Volt" error if the throttle reaches 100% and there's no detected AC output voltage--it won't "loop".)
What is weird about this is that the readout on top of the inverter with 96W and 1.24A shows a .664 power factor. It would seem that the machine is sensing that it is delivering 2/3 of its power at 1.24A which would be in line with its output?
What is weird about this is that the readout on top of the inverter with 96W and 1.24A shows a .664 power factor. It would seem that the machine is sensing that it is delivering 2/3 of its power at 1.24A which would be in line with its output?
The readout LCD has absolutely no connection to the processor on the main board--so they don't talk. And the overload sensor on the inverter processor itself has no connection to the LCD--there's 2 separate AC power sensors. One set for the LCD, and another set for the main board.
But they both sense. I am now wondering if the problem here is simply a QC issue with the soldering on the control board. Perhaps an errant tail when soldering to a trace and affecting the output upstream from where the power is sensed. A low voltage error that is passed through the rest of the board. Just a thought. This is an interesting thought problem though. I wonder whether I am going to run out of patience with it and throw it in the corner with the old HF PJ 5K inverter I bought years ago and had a similar low voltage limit problem with. I swore at that time I would never buy another PJ again, but my miserly ways tripped me up again. Lol. The other anomaly here is that the fan comes on after about the first 2-3 hrs of use and stays on for about 3 min and thereafter about every hour even with minimal load?
I swore at that time I would never buy another PJ again, but my miserly ways tripped me up again. Lol.
LAYABE is a mess up PJ that no one know what is inside ? Only control board available is rev 11.3 and the connector is mess up to the mainboard .
Finally got out the ol' U-Power and opened it up to take a look...
It's actually quite an impressive build... well laid out and organized... for the most part...
Looks like it was manufactured in Dec of 2021. QC "tested" it at 5000W...
The mainboard looks almost identical to the one in a GS. Only 3 x 18000uf 50v caps... so some differences... Looks trivial to solder in a 4th cap...
The tranny is particularly interesting. Says it's a LF-ASL3-AMG4-PSW and spec'd for 16v to 115/230v. It's very similar in size and weight to the one in my GS.
I could certainly see how PJ might be "learning" from some of the GS design choices...
The mainboard looks almost identical to the one in a GS. Only 3 x 18000uf 50v caps... so some differences... Looks trivial to solder in a 4th cap...
...because the GS inverters thus far have utilized a PJ mainboard. It worked fine at 6kw in our tests, and "if it ain't broke don't fix it."
However, the next revision of GS inverters utilize a completely in-house design mainboard/MOS boards...so there will be absolutely zero "PJ" parts in GS inverters going forward.
Says it's a LF-ASL3-AMG4-PSW and spec'd for 16v to 115/230v. It's very similar in size and weight to the one in my GS.
GS ones use a larger core with more wire. But yes, I'm pretty sure they've started copying GS transformer voltage specs...
The other anomaly here is that the fan comes on after about the first 2-3 hrs of use and stays on for about 3 min and thereafter about every hour even with minimal load?
Is the fan plugged into the control board? Or do its wires lead to the transformer?
if the latter, it goes to an infamous thermal "klixon"-style temperature switch, which is known for sticking on...
worth noting that the weak gate drive on the PJ FET boards results in significantly more heat from the FETs than is necessary.
I am now wondering if the problem here is simply a QC issue with the soldering on the control board.
Not sure about this, as several other PJ customers have mentioned the same behavior.
I personally do not have any of the newer PJ inverters, and I also personally do not use PJ inverters anymore--so I really can't verify or otherwise determine what's going on.
7 hours ago, Sid Genetry Solar said:Ha, maybe not quite. If it's down at 15vAC or so at no load, I wouldn't recommend reducing the tranny voltage spec too much. Reduce it too far, and the FETs will blow out (been there, done THAT!)
Eh, i'm getting ~16.5 at no-load... with 238.1v on the high end...
DC sits at 24.02v.
Sadly, i don't have a scope. I almost grabbed one of those cheap ones but couldn't justify it at the time.
Well I pulled it apart again and turned it on after a close inspection for any odd looking components, signs of heat over soldered joints etc. and found when I powered it on the LF Driver board led and the led below it on the control board were not lit. I am guessing that might have something to do with it. As for the fan, it is plugged into the control board. 🙂
I almost grabbed one of those cheap ones but couldn't justify it at the time.
Care to specify which make/model? A 2-channel Hantek can be had pretty cheaply (comparing!) and they're decent. I have a 4-channel Siglent--again, not terribly expensive, but for what I do a 'scope is a necessity! There's of course the super cheap tiny screen "DSO138" things on eBay/Amazon, but those are the better half of a joke.
and found when I powered it on the LF Driver board led and the led below it on the control board were not lit.
LED on the LF Driver board not lighting up when powered....that might be rather significant. I've traced out the driver board schematic, and that LED had better be lighting when the inverter's turned on. It's connected to the output of the FET driver for one of the high side boards. If that's not working...yeah, that'll be some dirty, horrible problematic output power!
Can you check the LED on the LF Driver board (with the inverter off!) by putting your meter in diode check mode, and putting it across the LED? (If the LED's good, it should light up.)
Another way to check this would be to look at the LEDs on the FET boards when you turn it on. All 4 FET board LEDs should flicker/go dim/go bright when you turn the inverter on. If one of them doesn't (it'll be on the positive heatsink side), then you have a big clue...
The LED "below the LF Driver board"...is this LED4 right above the LM339 (U02)?
The LED on the driver board read out about 2.2 one way OL on the other. The LED on the control board is on the battery input end of the control board and below one of the 6 pin components at the LF driver board corner. I checked both LEDs but neither of them lit when checked. Both read about 2.2 and OL the other polarity. All the LEDs on the positive side of the heat sink on the FET heatsink were bright and didn't flicker that I saw. In fact when battery power is connected all the LEDs are lit on the FET boards. Upon further inspection and looking alongside the case the FET LEDs indeed dimmed slightly when the inverter is turned on. I didn't see a LED desiginator on the board and you lost me with the LM339.