Inverter charger, where's the charger?

[bob prohaska]

In the course of testing a second ampinvt 800 watt pure sine
inverter charger some noises came out that sounded an awful
lot like arcing. I decided to take a look inside (finding
nothing obviously amiss) but realized that there's no visible
battery charging hardware. There's one big output transformer,
but only a few very small (sub-cm^2) core transformers on the
mainboard. The rated charging power of 15 amps at 14-15 volts
would I think call for something bigger, even if it's a fairly
high frequency circuit.

One thought is that the output step up transformer might double
as a step-down transformer while the inverter isn't active, as
when line power is available. Does anybody know how charging is
really done on gadgets like this?

Of course, documentation is non-existent. The mainboard is labeled
JGY-LI1.5k V2.0 but my websearches hit nothing informative. I do
have some photos if anybody might be able to recognize the images.

Thanks for reading,

bob prohaska

[Jim Wilkins]

"bob prohaska" wrote in message news:tfooek$2fvdo$1...@dont-email.me...

-----------------------

The only gadget like that I have is the APC1400 UPS. It charges the
batteries to the programmed float voltage at up to 6A whenever plugged in,
whether On or Off. You won't know what yours is doing unless you measure the
current.

You could parallel an external charger and perhaps have more choice of
charging rate, float voltage and battery type. As long as it doesn't have a
higher voltage desulfating mode it shouldn't hurt the ampinvt, it's just
keeping a battery within its normal voltage range. If you don't trust your
wiring skills the battery leads on NOCO ($$) chargers end in ring terminals
that can be unbolted from the alligator clips and attached to the battery.

This has proven to be a good battery voltage, current and charge level
monitor, so far:
https://www.amazon.com/gp/product/B09P65M2YF/ref=ask_ql_qh_dp_hza

[Jim Wilkins]

"Jim Wilkins" wrote in message news:tfpk1d$2i7ce$1...@dont-email.me...

"bob prohaska" wrote in message news:tfooek$2fvdo$1...@dont-email.me...

Of course, documentation is non-existent. The mainboard is labeled
JGY-LI1.5k V2.0 but my websearches hit nothing informative. I do
have some photos if anybody might be able to recognize the images.

Thanks for reading,

bob prohaska

-----------------------

If it helps, the schematic for the 900W true sine APC 1400 is available
here:
https://elektrotanya.com/apc_smart-ups_1400.pdf/download.html

[bob prohaska]

Jim Wilkins <murat...@gmail.com> wrote:
>
> The only gadget like that I have is the APC1400 UPS. It charges the
> batteries to the programmed float voltage at up to 6A whenever plugged in,
> whether On or Off. You won't know what yours is doing unless you measure the
> current.
>

Mostly I was just curious how battery charging is done on gadgets like this.
It's a commodity product apparently, with lots of vendors selling very similar
devices differing only in power ratings and packaging. Many seem to use the
same display and keypad.

For now I've accepted the charging performance as "good enough". I don't
like the standby draw (11-12 W) but can live with it.

> You could parallel an external charger and perhaps have more choice of
> charging rate, float voltage and battery type. As long as it doesn't have a
> higher voltage desulfating mode it shouldn't hurt the ampinvt, it's just
> keeping a battery within its normal voltage range. If you don't trust your
> wiring skills the battery leads on NOCO ($$) chargers end in ring terminals
> that can be unbolted from the alligator clips and attached to the battery.
>

It appears that turning off the charger function (setting it to 0%) can
have some unexpected consequences. I tried it, then set the inverter to
"off", meaning no inverter output, no AC line passthrough via the
transfer switch. The input current started to periodically rise to
high (~200W) power, the alarms beeped briefly and then things went back
to normal. Clearly, 0% didn't mean off unconditionally.

The cycle repeated at intervals of a few minutes. The first time this
happened I heard crackling noises, seemingly inside the inverter, which
is what provoked my opening it up to look inside. Subsequents tests were
without any rice crispies sounds, even during power excursions

There was nothing visibly wrong and on reassembly the inverter worked
normally. There are no manual controls at all, but I do see a 4 pin header
marked 5V TX RX G so there seems to be a serial interface. Since everything
is done in software, anything is possible 8-)

Thanks for writing!

bob prohaska

[Jim Wilkins]

"bob prohaska" wrote in message news:tfqurm$2md9q$1...@dont-email.me...

For now I've accepted the charging performance as "good enough". I don't
like the standby draw (11-12 W) but can live with it.

--------------------

I'm running small AC loads like this laptop from Bestek 300W and 500W true
sine inverters that draw about 7~7.5W no-load power. The 300's fan is mostly
off, silent, at low loads, the 500's fan runs continuously. Battery
monitoring and charging are separate functions that I can change and
experiment with to maximize economy when the grid is up or performance when
it's down.

I designed and built custom industrial and military electronics for a
living, so I prefer upgradeable modular over integrated solutions to
ill-defined or evolving problems, with Wattmeters between modules to monitor
performance. I realize you may not and probably don't have the machine tools
to fabricate the metal enclosures but I can describe only what I have.

[bob prohaska]

Jim Wilkins <murat...@gmail.com> wrote:
> "bob prohaska" wrote in message news:tfqurm$2md9q$1...@dont-email.me...
>
> For now I've accepted the charging performance as "good enough". I don't
> like the standby draw (11-12 W) but can live with it.
>
> --------------------
>
> I'm running small AC loads like this laptop from Bestek 300W and 500W true
> sine inverters that draw about 7~7.5W no-load power. The 300's fan is mostly
> off, silent, at low loads, the 500's fan runs continuously. Battery
> monitoring and charging are separate functions that I can change and
> experiment with to maximize economy when the grid is up or performance when
> it's down.

As a package the Ampinvt at ~250$ is a very decent buy, nothwistanding
all my carping. If one could figure out how to talk intelligently to the
controller it might be an outstanding buy.

>
> I designed and built custom industrial and military electronics for a
> living, so I prefer upgradeable modular over integrated solutions to
> ill-defined or evolving problems, with Wattmeters between modules to monitor
> performance. I realize you may not and probably don't have the machine tools
> to fabricate the metal enclosures but I can describe only what I have.
>

Understood completely. My hope (in part) was that somebody reading
this newsgroup might recognize the Ampinvt product as derived from
something better known and documented. It'd be surprising if it's a
completely unique, clean-sheet-of-paper design.

[Jim Wilkins]

"bob prohaska" wrote in message news:tftm4n$33v8m$1...@dont-email.me...

>As a package the Ampinvt at ~250$ is a very decent buy, nothwistanding
>all my carping. If one could figure out how to talk intelligently to the
>controller it might be an outstanding buy.

https://www.pvace.com/products/ampinvt-specific-remote-control-panel-for-low-frequency-pure-sine-wave-power-inverter

I found useful help for the APC from third party sources:
https://sourceforge.net/projects/apcupsd/
but I don't see anything similar for Ampinvt.

https://diysolarforum.com/threads/correct-low-voltage-protect-and-low-voltage-restore-settings.23077/
https://diysolarforum.com/threads/ampinvt-24v-3000w-inverter-chargers.2223/page-2
https://secondlifestorage.com/index.php?threads/i-have-it-up-and-running-finally.10876/

>Understood completely. My hope (in part) was that somebody reading
>this newsgroup might recognize the Ampinvt product as derived from
>something better known and documented. It'd be surprising if it's a
>completely unique, clean-sheet-of-paper design.

I have a troublesome MPPT controller in a housing that is identical to this:
https://www.amazon.com/vdp/ad32daa4d55f486c8c5c1ca1943f6736
In very tiny blurred print under the logo at the top it seems to have
"www,gac-inverter.com". The manual is silent on who dunnit. Googling for mug
shots suggests the perp may have been Top One Power.
https://www.enfsolar.com/one-solar

[Jim Wilkins]

"Jim Wilkins" wrote in message news:tftut9$37d2f$1...@dont-email.me...

https://www.pvace.com/products/ampinvt-specific-remote-control-panel-for-low-frequency-pure-sine-wave-power-inverter

See the bottom of the page for contact info for Foshan Top One Power
Technology Co.,Ltd. and a chat link. Good luck!

[bob prohaska]

That looks like the same contact I started with. I'll start by asking
about remote control using a computer. Kinda doubt it'll go anywhere,
but at this stage there's little to lose. If it ends up looping back
to the same support rep at least the question is new.

Thanks for the good luck wish, I'll need it.

bob prohaska

[Jim Wilkins]

"bob prohaska" wrote in message news:tgdio7$1kd33$1...@dont-email.me...

That looks like the same contact I started with. I'll start by asking
about remote control using a computer. Kinda doubt it'll go anywhere,
but at this stage there's little to lose. If it ends up looping back
to the same support rep at least the question is new.

Thanks for the good luck wish, I'll need it.

bob prohaska

---------------------

There has been considerable interest and effort to understand communications
protocols for other devices, but it may be hard to find. Sourceforge, github
and sigrok are good places to look. The most useful program I found to
communicate with a multimeter was in German, which fortunately I studied in
college and practiced when I was stationed there. Instead of numerical
values the meter transmits the 7-segment LCD driver output which a Select
Case can easily decode back to digits.

For example:
https://sigrok.org/wiki/RDTech_UM_series

Those USB power meters are quite handy to show how (or if) devices charge. I
added external wires and PP45 connectors to one that measures up to 30V and
5A, making it a USB power to everything else adapter for solar charging and
cable resistance and load testing.

[bob prohaska]

As usual, got a prompt answer saying the remote port is only for
use with the optional remote control panel. Full stop.

I thanked the respondent, who was the same person who's been leading
me 'round Robin Hood's barn regarding charging irregularities and
excess no-load power consumption. He promptly wished me a good day.

Given that there's a serial port on the mainboard it seems likely
a smarter person than I am could probably figure out what's going
on. For now I can wait. The inverter's usable, just wasteful.

[Jim Wilkins]

"bob prohaska" wrote in message news:tggpko$22b6v$1...@dont-email.me...

-------------------------------

APC is like that, our control language is proprietary and you shouldn't be
using it. At least Ampinvt offers the remote control panel, APC didn't have
any solution I could use with the older model I have. Even the serial port
is non-standard, with 24V battery voltage on one pin. I suspect it was meant
for production line testing without a battery installed. You aren't alone,
though 3rd party support helped me.

If you want to pursue DIY solutions, a current sensing switch can trigger an
action when a load shuts off. They go for around $20 on Amazon. I have one
on my fridge to detect and indicate when the compressor is running. The APC
would require a hardware mod to be able to turn on when the thermostat
closes, though it will shut down on a serial port command.

I need the APC to run the 120V fridge from battery power, but for my
computer loads the 300W Bestek is a more efficient source of sine wave AC.
Auto/air power bricks will run the laptops from 12V though not the external
accessories like backup hard drives.

[Jim Wilkins]

"bob prohaska" wrote in message news:tggpko$22b6v$1...@dont-email.me...

Given that there's a serial port on the mainboard it seems likely
a smarter person than I am could probably figure out what's going
on. For now I can wait. The inverter's usable, just wasteful.

------------------------

On the APC1400 experimenting blindly with serial commands can make a mess. I
didn't try them until I had found a listing on line and knew which commands
to skip. You might make the Ampvint control language revert to Chinese.

[Jim Wilkins]

Does this relate to the model you have?
https://acmerevival.com/wp-content/uploads/2022/09/Ampinvt-Peak-Manual.pdf

I've been matching the somewhat inaccurate voltage settings of solar
controllers to the battery by changing the controller's battery type
setting. The adjustable float voltage setting for Lithium may be useful if
you don't need a temporary boost to fully charge cycled lead batteries.

[Jim Wilkins]

You previously asked about loading an inverter. Although this isn't the same
situation it may help. While testing a DC-95MHz amplifier from a hamfest I
saw the first signs of overload distortion as an RC instead of a sinusoidal
fall after the sine peaks, presumably because the +/-15V 100mA power supply
couldn't keep up and the scope showed the amplifier's tantalum slug power
input caps discharging. The amp is good for 10V RMS into 50 Ohms, or less
voltage at lower load impedance.

Doing this testing requires AC and DC variable loads, which I've collected.
If you have a Variac you can use a heating device for a sinusoidal variable
AC load, for instance my crockpot which draws 100W on High and 70W on Low.

I use hotplates and coffee pots to test DC-AC inverters and inverter
generators. Mine deliver full power into a resistive load up to their limit,
then shut off the output until reset. I haven't tried inductive or
capacitive loads yet. I used the Hantek CC-65 AC/DC current probe to explore
and measure the current output of PWM solar controllers and it seems likely
to be good for AC line current measurements. It isn't as nice as a Tek or
Fluke current probe, but it also isn't nearly as expensive.

I bought a soup heating cup at Goodwill that draws 400W, and leaks AC
slightly to the cup as did a Faberware coffee pot I bought there, so be
careful. The coffee pot had been immersed and the heating element area was
rusty, cleaning it restored the insulation to pass my 1000V HiPot tester.