Home
JAQForum Ver 24.01
Log In or Join  
Active Topics
Local Time 00:43 27 Nov 2024 Privacy Policy
Jump to

Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.

Forum Index : Electronics : Inverter building using Wiseguys Power board and the Nano drive board

     Page 17 of 23    
Author Message
Cpoc
Regular Member

Joined: 28/05/2024
Location: Portugal
Posts: 78
Posted: 03:38pm 17 Aug 2024
Copy link to clipboard 
Print this post

This looks great. Simple and clear setup.
Would be amazing in the future if we can get all this information remotely to home assistant via mqtt wifi.
No hurry first we get all the hardware working and tested with the firmware.Then we can get home integration working later on. When I mean WE I mean not me but the devs as I have no idea how to make this happen.
Edited 2024-08-18 01:45 by Cpoc
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 09:47am 18 Aug 2024
Copy link to clipboard 
Print this post

Well it took longer then I thought to wire up the front panel, it took a day to work out the best position for the sockets and the number of pins needed, it makes for simple removal and repair/modifications of the control panel, luckily I fitted two special very high quality 16 pin plugs and sockets, 32 wires in total and I used every one.

I'm going to fit a tiny 1A 60v to 12v module on the Control panel, same as used on the Nano Controller input, this will take the load off the Nano controllers 12v supply and help to keep it clean and isolated.

The 12v will is used for powering the extra meters, LEDs, etc, and will free up a two wires in those 16 pin plugs, just in case I need one.

Tomorrow I'll loom the wires neatly on the control panel and start wiring up the interface sensors.

This and other sh*t has put me back around two days, but it's worth it, as it takes about 10 seconds to unplug the Panel and remove or install it.

Quick pix, needs tiding up as I have just finished wiring it up and testing.



FYI: I took these photos last night and didn't post.

The control panel lifts out at the bottom and moves slightly down before standing upright and folding down. This can be done with the inverter running and allows adjustment and measurement to the Nano controller once that board is installed.

There are a four sensor boards that will be mounted in this case and all are easily accessed for testing and adjustment.  

This complete top housing also plugs into the main inverter housing, it simply lifts off, giving complete access through the back of the housing which is 620mm x 340mm, making it soooo easy to work on, build, modify or service.



That opening is 200mm x 610mm.



_
Edited 2024-08-19 08:23 by KeepIS
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 06:51am 19 Aug 2024
Copy link to clipboard 
Print this post

Eventually decided on this method for the Battery input to the inverter, Bat-Pos to the Kilovac contactor and Bat-Neg to shunts, both using very heavy copper bar.

This panel has to sanded and sealed once its finished.

The two digital meters display power, kWh, current and voltage with very high accuracy for each power stage, great for monitoring idle power and load sharing down to a few ma.

Each shunt connects to a NEG terminal of one of the Power board mounted each side of this panel. The output terminal of the Kilovac connects to each B+ terminal on the power boards. Very short current paths and connections, these connections with this setup run cold even at 8kW with one power board.




It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 04:10am 20 Aug 2024
Copy link to clipboard 
Print this post

Build is starting to pickup pace, this was the last thing I completed last night, I realized the simplest way to access the back of the two sockets for wiring the 11 sensors, fan temperature relay, Kilovac control and other incoming / outgoing external wiring to the socket's, was to use a small box hinge and simply fold the Socket mounting plate down, soldering connections or making future changes is now a breeze.

I used two long plastic standoff from an old GT inverter to lock the mounting plate into place when it's raised, this also allows the shunts and kilovac to be easily replaced.

I have two chokes still to wind, but from then on its' just point to point wiring.

I also have 3 of Wiseguys 2 stage boards to be partly populated for two Peak DC meters, with one including the FET switching, keeping these close the the Power board and Hall effect sensors, the 3rd board will have the 2 stage switching and relays located on the wall in the AC filter, circuit breakers and AC changeover cabinet.




_
Edited 2024-08-20 14:22 by KeepIS
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 08:16am 21 Aug 2024
Copy link to clipboard 
Print this post

All 11 sensors, Kilovac control and fan control are wired up into the two sockets, it's amazing how neat and simple it turned out, this is also the complete DC battery input wiring.

The only connections to the Power board not shown are the two choke leads, these come from below into the power board, and the opto drive connectors.




Edited 2024-08-21 18:25 by KeepIS
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 04:58am 24 Aug 2024
Copy link to clipboard 
Print this post

I had forgotten how long it takes to wire up a single toroid inverter, but this dual inverter power unit is ridiculous.

I finally took the main inverter off line this morning, the toriod will be coming out today and moving into the new Dual toroid cabinet.

I think the difference in build time is also due to using 4 big connectors with a total of 64 wires to be sorted. Making the full wiring diagram wiring, wiring colors and pins numbers entered on the computer as I went along, so I don't have the problems of trying to decipher my own scribble with a dozen corrections, and putting it off once it's built.

Also taking time to test every connection, running various parts of the switching circuits under voltage, testing current shunt meters, pre-charge resistors, Kilovac wiring etc.

The good news is the Nano controller is almost wired up - so quick and simple with the wiring looms sitting there ready to go, mainly due to those big plug/sockets that took me forever.

I don't want to redo anything once this inverter is up an running, basically I'm building it once with no shortcuts or temporary kludges, and the only mods needed, if any, will be around the 2 stage control - or more to the point, how I actually use it depending on typical load testing.

Although it's taking far longer than I thought, I'm still forcing myself to just take my time, think and rethink, stop if I'm tired, and quad check everything and remember to chill out.

But it's getting so close now
_
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 09:58pm 24 Aug 2024
Copy link to clipboard 
Print this post

The Nano controller, digital DC input meters, digital heat sink temp meters, Nano LCD LEDs, Toriod temperature sensors, per-charge and switching all worked first go with a brief power up of one Toriod and power Stage.

This morning I'm doing the final tiding up of wiring looms to the Nano, then moving the second toriod across.

I have to add two Peak DC input PCB and two AC trip PCB, fortunately all the wiring is loomed and in place for these boards.

I'm loathe to run this inverter without those current trips in place, so I won't!

I'll do it properly, but it was nice to see the control panel lit up and the Nano LCD indicating everything correctly
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 09:39am 25 Aug 2024
Copy link to clipboard 
Print this post

Both toriods and one choke fitted, then I got sidetracked when I found a way to use Eight EPCOS dual 3.3mH 12A @ 600V chokes to make two good pre-filters for the two AC outputs, 4 of them used in each filter in series parallel for a total of 3.3mh @ 24A per filter, a 0.015uf on the output and T-center with a 4.4uf Toroid cap on the inputs. The 2 stage control unit mounts on the back of the lower cabinet.

Current sensors on each Toriod for individual current trip and indicators, another sensor for the main Dual/single AC output lead for LCD current display only (no trip) and AC VFB also connected.

It's so easy to work on when you can open the cabinet up like this in less than 20 seconds. So tomorrow I will have the other 3 chokes fitted and the 2nd power board mounted on the heatsink. That's just about ready to hit the detonate switch.


_
It's all too hard.
Mike.
 
Revlac

Guru

Joined: 31/12/2016
Location: Australia
Posts: 1024
Posted: 10:08am 25 Aug 2024
Copy link to clipboard 
Print this post

Looks great,  watching with interest, still thinking about running 2 toroids, but on 1 power board I will have some questions later.
Cheers Aaron
Off The Grid
 
Cpoc
Regular Member

Joined: 28/05/2024
Location: Portugal
Posts: 78
Posted: 03:52pm 25 Aug 2024
Copy link to clipboard 
Print this post

That setup is great with so many protections in place and having the ability to swap out and repair any component.
 
Cpoc
Regular Member

Joined: 28/05/2024
Location: Portugal
Posts: 78
Posted: 03:52pm 25 Aug 2024
Copy link to clipboard 
Print this post

That setup is great with so many protections in place and having the ability to swap out and repair any component.
 
analog8484
Senior Member

Joined: 11/11/2021
Location: United States
Posts: 108
Posted: 04:31pm 25 Aug 2024
Copy link to clipboard 
Print this post

Nice.  I think your cabinet builds are even more impressive than your inverter builds.  It's a work of art.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 01:58am 26 Aug 2024
Copy link to clipboard 
Print this post

Thanks guys for the kind words.

I wanted to clarify my use of three AC current sensors a little more, it may look like I'm complicating the build, but in reality it's the opposite.

This Inverter can switch between a powerful single stage low "idle power" Inverter and a two stage very High power Inverter (and higher idle power).

If you used one AC current sensor on the combined AC output feed, you would have to change the current trip setting when in single or Dual stage mode, same for the DC input current trip if you are using that, and I am.

Using a separate current trip circuit for each power stage/toroid, allows you to set the AC current at say 45A on each AC sensor circuit.

It means that nothing has to be done when switching between 1 or 2 stage operation, if one stage is running, AC trips at 45A, if two stages are running, AC trips at 90A, in the real world there will be some non linear load sharing, but you get the basic idea.  

No modification are needed to the Nano Code, the center 3rd AC sensor in the photo feeds the Nano controller as normal, but D6 is removed, the 3rd sensor correctly indicates the AC current on the LCD in either 1 or 2 stage mode, it is not used for over current trip.

Instead, the Nano current trip latch is set from a high input on the EXT trip input, removal of D6 is the only change to the Nano controller.

It's then a simple matter of duplicating the Latch circuit of Q1, Q2, and a few resistors and a diode. I make four of these little circuits, One for each toroid AC sensor and one for each power stage Peak DC input current trip sensor.

The four small Over-Current boards feed a voltage via a diode over a single wire to the Master Nano OC Latch.

As a bonus, I know which stage caused the current trip, and I know if it was an AC or DC over current event. It makes setting the trip points simple, and not a single change has to made to any setting when the inverter is switched between 1 or 2 power stages.




_
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 06:49am 26 Aug 2024
Copy link to clipboard 
Print this post

One Choke to go, differences in choke height of the front two chokes is due to different toriods rings used, the ones I used for the first two were not available.

Although the choke cables appear to be different thickness, that is due to the thicker insulation on the cables, both are the same wire size. Same applies to the Toroid primary's.


_
Edited 2024-08-26 16:50 by KeepIS
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 09:50am 26 Aug 2024
Copy link to clipboard 
Print this post

FYI the new chokes use 9 toriod rings each with 4 turns. This equates to 23uH at close to 360A for each choke before the knee where the slope starts to rise a bit faster, still plenty left at 400A, and there are two chokes for each toroid.

I will only use 25mm cable for the Chokes, I want some resistance to ease the shock load on the FETS from the Toroids under huge peak startup loads and transients. All other DC cables are 35mm or 50mm.


_
Edited 2024-08-27 06:27 by KeepIS
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 05:02am 27 Aug 2024
Copy link to clipboard 
Print this post

Both Toroids, 4 chokes and all AC wiring has been finished    

The second power board is mounted on the aluminum bars ready to bolt to the second Heat sink, and the cables are ready to be connected.

So I took the time for a breather to do some minor modifications to the Code for the Dual version of the Nano.

This is mainly for the Toroid temperature LCD display  and setup options, and both power board cap banks, for dual toroid temperature and fan control.

I had already setup Analogue port 7 to read the second power board Cap Bank, and I coded the LCD display to show CB2 below the first Power board Cap bank "CB1".

I ran a lot of tests checking Fan control and air flow with the Toroid cabinet full with Toriods and Chokes - Wow! it's amazing - and soooo quite.

The Dual version code is now finished, this evening I'll get the 2nd heatsink and power board connected.

Then I can run some low current tests in TEST mode with both toroids in parallel, no I haven't quite finished the current trip boards but with low voltage running in test mode with a current limited supply, I will be able to get some useful current indications between each power board alone, and then in parallel, I'll test for any phase difference just in case as it only takes a second.



FYI Picture of the display, everything looks dark to enable me to get a decent capture of the LCD.

NOTE: Nothing has been calibrated correctly as yet.




_
Edited 2024-08-27 17:07 by KeepIS
It's all too hard.
Mike.
 
Cpoc
Regular Member

Joined: 28/05/2024
Location: Portugal
Posts: 78
Posted: 09:57am 27 Aug 2024
Copy link to clipboard 
Print this post

It’s alive it’s alive…..amazing work.
 
Cpoc
Regular Member

Joined: 28/05/2024
Location: Portugal
Posts: 78
Posted: 09:58am 27 Aug 2024
Copy link to clipboard 
Print this post

Double post again.
Edited 2024-08-27 22:01 by Cpoc
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 01:21am 28 Aug 2024
Copy link to clipboard 
Print this post

  Cpoc said  It’s alive it’s alive…..amazing work.

 

This is the view with the Nano/Control panel housing simply unplugged and lifted off.

This is completely wired up and ready to go, all AC output connections, sensors, both DC and AC, all temperature sensors and all control connections are shown.  

This is inverter heaven for me when building or repairing if any component fails.

Simply remove 4 screws from the power board  terminal connectors, unplug the 8 pin SPWM lead to a power board, the power board plus heatsink pops out of the back of the housing by removing 2 small retaining screws either side at the rear of the cabinet.

The power board cables do not have to moved at all and stay in place exactly as they are when connected.

Some final preflight checks to be carried out before ignition in test mode






_
Edited 2024-08-28 11:24 by KeepIS
It's all too hard.
Mike.
 
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 1679
Posted: 07:18am 28 Aug 2024
Copy link to clipboard 
Print this post

It powered up perfectly:

The Nano AC VFB for AC regulation comes from Toriod 1, this is always the running Toroid.
 
Fortunately both Toroids are exactly the same and both AC voltages where exactly the same with only "1 Toroid" connected to the Common AC output port.

I placed a "sacrificial" LED between the two AC active outputs and restarted the inverter:

The LED lit up (normal brightness) just as the AC output voltage stopped rising, it's obviously the variation between the two AC outputs not coupled and I'm seeing a very slight 1v or 2v difference and then only when you move the DC input voltage making the AC voltage regulate, but only enough to just light the LED.

Once the LED settles there is a very slow fade on and off effect from the LED.

The last test before I had to attend to other tasks, was bridging the two AC outputs and connecting a 200W load to the paralleled toriods:

Perfect  

Finally, idle power is 48 watts with both Toriods powered by SPWM and not paralleled, and it's the same paralleled, 48w is basically nothing to my batteries & solar  

Next task tomorrow is to get the current trips all working, then I'll run some real tests with the battery connected, lastly test Wiseguys 2 stage circuit.



Not a good photo, but I thought it should have a picture of one of it's first loads running as a dual toroid & dual power-board inverter.



BTW the difference in the round heatsink temperature meters (either side of the LCD)
is that one is seeing the cold heatsink temperature (air-con blowing in it) and the other is inside the case cable-tied to a power-board terminal connector, I plan to use this temp sensor for testing all big terminal connections and connectors at high power levels before finally reconnecting it to the heatsink.

The control panel unplugged and removed (a few seconds) - easy access and adjustment to the Nano controller, takes 20 seconds to remove the Nano controller, the Power board terminals and connections are easy access and a power board can easily be removed without removing the nano and control panel housing, still a few PCB's to be installed in there for two Peak current display meters and 4 over current control boards.


_
Edited 2024-08-28 18:46 by KeepIS
It's all too hard.
Mike.
 
     Page 17 of 23    
Print this page
© JAQ Software 2024