Jump to content
Genetry Solar Forums

Powerjack 8k high output voltage


Rafaelmedina23
 Share

Recommended Posts

Hello, I bought this inverter, it is more backward than the previous one, on the boards it reads 2016 and the transformer 2017, it is giving 275v A/C output, here only 240 is used, I would like to know how to lower that voltage since this inverter does not It has a low voltage knob, it only has a battery knob.  thanks for your help.

20221106_073908.jpg

20221106_073901.jpg

Screenshot_20221105-230730_Facebook.jpg

20221106_073923.jpg

Link to comment
Share on other sites

Seeing as there's no easy options for adjusting the voltage (especially on these older boards--which I'm not very familiar with), the easiest way to reduce the output voltage would be to bypass part of the resistor ladder that provides the AC voltage feedback to the MCU.

image.png.625b33eda4f86bc9043479af16a8fd5a.png

Try bypassing just one block (which takes 2 resistors out of the circuit), and see how it behaves.  By increasing the sensed feedback voltage, the MCU should lower the output voltage in response.  If bypassing one block isn't enough, try bypassing 2 blocks.

Worth noting that these are 499K resistors, so there is pretty much zero current flow here.

Link to comment
Share on other sites

15 hours ago, Sid Genetry Solar said:

Seeing as there's no easy options for adjusting the voltage (especially on these older boards--which I'm not very familiar with), the easiest way to reduce the output voltage would be to bypass part of the resistor ladder that provides the AC voltage feedback to the MCU.

image.png.625b33eda4f86bc9043479af16a8fd5a.png

Try bypassing just one block (which takes 2 resistors out of the circuit), and see how it behaves.  By increasing the sensed feedback voltage, the MCU should lower the output voltage in response.  If bypassing one block isn't enough, try bypassing 2 blocks.

Worth noting that these are 499K resistors, so there is pretty much zero current flow here.

 

20221107_090546.jpg

Link to comment
Share on other sites

19 hours ago, Sid Genetry Solar said:

Seeing as there's no easy options for adjusting the voltage (especially on these older boards--which I'm not very familiar with), the easiest way to reduce the output voltage would be to bypass part of the resistor ladder that provides the AC voltage feedback to the MCU. 

Is this what the pot on the variants with a voltage control is doing? More voltage drop, lower read voltage, harder it works to hit voltage.
Solder in a few mega-ohm pot and get a simple variable voltage control. Nice, simple, workaround.

Out of curiosity, what do you think could cause a 60v increase in output? Seems to me like only the regulator interpreting the wrong voltage and overdriving to compensate...

Link to comment
Share on other sites

A v3.6 control board!  I'm jealous!  My best powerjack (still running my house) has a v3.4 control, which in my limited experience is the best control board powerjack ever made.  Your v3.6 looks just like it and can't be too different.  Hopefully you can get the voltage down to normal.  Solderiing across the points Sid circled should bring it down some, probably between 10 and 20v.  As he said, you might have to short out another pair, or even two more(just do one at a time, though).

The only other thing I can suggest is doing a thorough cleaning of the control board(top and bottom) as a bit of dirt in the wrong spot can make these things do weird stuff.

Link to comment
Share on other sites

21 minutes ago, NotMario said:

Is this what the pot on the variants with a voltage control is doing? More voltage drop, lower read voltage, harder it works to hit voltage.

You hit the nail on the head.

...the side effect being that now the CPU has no idea what the actual output voltage is.

Somewhat comically, PJ literally hand-replaces 2 SMD resistors on the control boards to switch between 110v and 220v regulation.  CPU code stays the same--and has no idea of the actual output voltage.

 

21 minutes ago, NotMario said:

Out of curiosity, what do you think could cause a 60v increase in output? Seems to me like only the regulator interpreting the wrong voltage and overdriving to compensate...

Could be anything, all the way down to dust/"dusty goo" buildup on the PCB.  The "isolation" path from the AC output to the feedback amplifier is literally >16MOhm of resistors, so mashing your fingers on those resistors will change the output voltage significantly.

 

20 minutes ago, dochubert said:

A v3.6 control board!  I'm jealous!  My best powerjack (still running my house) has a v3.4 control, which in my limited experience is the best control board powerjack ever made.  Your v3.6 looks just like it and can't be too different.

hehe, "best one they ever made"--and that's because it's a direct shameless copy of the "original design" inverter that all Chinese LF inverters are based off of!  Before any cheapening mods were made without understanding how the circuit actually worked....

Link to comment
Share on other sites

7 minutes ago, Sid Genetry Solar said:

Somewhat comically, PJ literally hand-replaces 2 SMD resistors on the control boards to switch between 110v and 220v regulation.  CPU code stays the same--and has no idea of the actual output voltage.

Hey, Little bit off topic...

My U-Power seems to be 110v regulated. (L2 does not reach control board).
What is the practical difference? Second leg can have too much voltage sag?

Link to comment
Share on other sites

3 hours ago, NotMario said:

Second leg can have too much voltage sag?

A bit.  Sag will be determined by transformer losses--as obviously, if L2 is pulling power from the core, it will cause a sag in L1 (just less the L2 winding resistance).  And the inverter will try to regulate L1.

You get no current limit on L2 though.  As though that makes much of a difference though 😉.

Link to comment
Share on other sites

6 hours ago, dochubert said:

A v3.6 control board!  I'm jealous!  My best powerjack (still running my house) has a v3.4 control, which in my limited experience is the best control board powerjack ever made.  Your v3.6 looks just like it and can't be too different.  Hopefully you can get the voltage down to normal.  Solderiing across the points Sid circled should bring it down some, probably between 10 and 20v.  As he said, you might have to short out another pair, or even two more(just do one at a time, though).

The only other thing I can suggest is doing a thorough cleaning of the control board(top and bottom) as a bit of dirt in the wrong spot can make these things do weird stuff.

Actually yes, the boards are a bit dirty, I'll try that first to see what happens. Thank you

Link to comment
Share on other sites

On 11/7/2022 at 10:35 AM, Sid Genetry Solar said:

that's because it's a direct shameless copy of the "original design" inverter

I've heard that original LF inverter was designed by Trace (before they were bought by Xantrex).  Any idea if that's true?  If so, I wonder what the brand name/model of the first LF inverter was?  And are any of them still around?  I know there are a few old Xantrex/Trace inverters still around, but are they that first "original" design, or were they copying something still older?

One more 'wonder'.  I wonder who claims to be the original designer of that original LF inverter?  They can't be blaming that on Tesla!

  • Like 1
Link to comment
Share on other sites

On 11/9/2022 at 5:06 PM, dochubert said:

I've heard that original LF inverter was designed by Trace (before they were bought by Xantrex).  Any idea if that's true?  If so, I wonder what the brand name/model of the first LF inverter was?  And are any of them still around?  I know there are a few old Xantrex/Trace inverters still around, but are they that first "original" design, or were they copying something still older?

One more 'wonder'.  I wonder who claims to be the original designer of that original LF inverter?  They can't be blaming that on Tesla!

I don't know either.  Jack might admit to it if I ask, but I don't see that as being important at the moment.

One thing is for sure, the design was well-done for what the designers had available at the time.  That little barebones 68HC11-based processor on the "original" control boards is literally completely maxed out generating an SPWM output--and a lot of support circuitry on the output was due to the internal peripherals not having it!

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...