 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 16 of 23   |
|||||
| Author | Message | ||||
| Cpoc Regular Member  Joined: 28/05/2024 Location: PortugalPosts: 78 |
No worries take your time to heal. |
||||
KeepIS Guru  Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Thanks  It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
For anyone interested: I uploaded V7.3 with adjustable timed delay of restart ramp-up after a Low-Battery stop condition, see included "Latest changes.txt". Download It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
I've been thinking about Inverter low battery voltage Ramp down and restart, obviously some of us have various ways to handle this outside of the Inverter - but not all external monitoring allows for control of a large DIY Inverter. Having this setting in the inverter is common in almost every commercial Inverter I have seen, it's up to you if you want to employ it, if you think this is a bad idea, please stop reading. I'll repeat -> If you think this is a bad idea for ANY reason, please stop reading!!!! you are missing the point. Soooo for anyone who wants to employ this setting in a basic off-grid system for Inverter control, I have the following settings: Before I start and get dumped on, I guess I must state the bleeding obvious for some: If an installation has not been correctly built, you will have more trouble than just the battery Cutoff settings - knowing your setup and monitoring various state of charge and voltages under system loading is something we need to do and record if we plan to set it up correctly. Low voltage cutoff will be problematic if the battery to inverter DC path is not wired with appropriate gauge cable and correctly terminated at every point - no weak links allowed in any large inverter. Back to the Inverter Controller: Low voltage ramp down and restart voltage points are what I have been testing and pondering over, and the implementation of Low voltage Cutoff logic, it's obviously not as simple as it appears because of different Loads, battery capacity and technology employed in out individual DIY off-grid Solar and Inverter builds. Large LFePO4 bank voltage differences from charged and running under "average loads", to running at a lower SOC, is often measured in millivolts down to 40% SOC. In battery types with slightly higher internal R, the delta is often measured in volts from a running charged state down to around 40% SOC. There is voltage sag under heavy loads, and there can be high transient startup loads or short duration increased loads, such as a fridge, freezer or electric jug, and all voltages are impacted by the current state of charge and system build type and quality. The transients are easy to handle with a simple short timeout period for Low voltage cutoff setting, but you may want to extend the timeout period for that fridge or Jug, especially when batteries are getting down and operating near the cutoff voltage, this may not be to much of a problem with a good large LFP battery bank, but could pose a problem, depending on size and SOC, with older battery technology's. I've come up with a workable solution for my Inverter and I have implemented the following: These additions do not complicate the code or the operation of the controller in any way. An adjustable timeout setting "In Setup" for low battery voltage cutoff, and in extending that timeout period, I also reduce the timeout period if the voltage drop is more than 10% below the normal cutoff voltage setting. If 10% below cutoff, the inverter will stop almost immediately and wait for the restart voltage set point, an adjustable restart timer "In Setup", is then implemented to allow the batteries to attain some SOC. The restart point could be tricky if set low, and even then, with LFP and lower charge rates, or high charge rates with high current chargers, the inverter could restart with a low OR no change in SOC. Yes !! we ALL know that voltage is a very poor indicator of SOC, and especially in a dynamic situation when charging, I don't need to have that pointed out for the umpteenth time as though I'm some noob - Steps off soapbox. 1: I've implemented "low AC current" logic for Low battery voltage under 3A ac, if less than 3A and below cutoff, stop inverter. 2: If AC current is above 3A, and if the voltage drop is more than 10% below the low voltage Cutoff, stop the inverter. 3: Once the Voltage is above the restart voltage, (it's charging) and an "adjustable" Delay period "In Setup" starts a countdown before the inverter is allowed to ramp up again. The delay period can be set from 1 minute to many hours. This at least gives the batteries some time to charge, or it can simply be be used to stop a possible charge to load low voltage trip cycle condition. Again if interested, the Controller Hex code has been uploaded. Download _ Edited 2024-07-29 10:25 by KeepIS It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Just to give you an idea of why and how I use these settings in my Nano Controller. A Large LiFePO4 setup with 4 x 53v battery banks. I have them set for Low Charge rate with a Max charge setting of 25A per Bank. I usually charge to 90% SoC, evening + over night + early morning, before Solar consumption, typically results in the batteries at around 83% SoC, occasionally down to 72%. During the mornings, this will often drop to 50% SoC when the Workshop is powered up and used, Solar HWS is sometimes switched across as well, and sometimes in the evening in cold winter cloudy days. The lower max SoC, and higher SoC from overnight results in an average charge rate of around 3A to 10A per bank [ 12A to 40A ] this results in 90% SoC being reached at around 3pm, while still running OFF grid + charging + household + workshop loads, and running mostly on 4 or 5 Solar chargers and arrays - except with heavy workshop usage. The only time the Inverter would switch off and cause the ATS to transfer all loads to the Mains supply would be if SoC went below 40%, it would normally happen at night, so the "restart timer" is set at two hours. In the morning, Two hours after initial charging starts, there would normally be enough Solar to start the Inverter, and usually enough battery charge to even handle cloudy weather intermittent low solar periods by then. In either case the Inverter can never be in a state of continually being switched between Inverter and Mains supplies by the ATS, because even 10 minutes of controller restart timing will not allow that to happen. The same sort of conditions would apply when our Mains AC finally gets disconnected. In winter and poor solar conditions, charge rate and SoC is increased, having the settings in the Controller makes it so simple to adjust the inverter for these changing charge conditions. The charge rate and SoC are adjusted to get the maximum life from the LiFePO4 banks. They like charge + discharge cycling, and not sitting at a high SoC for hours, and especially when close to 100% SoC, even with a small tail current. I also have true charge-discharge tracking and SoC monitoring, it can be used to override the restart timer if the need ever arises. What must be kept in mind with full off grid systems, some of us build these ourselves because we don't want to rely on commercial equipment that we can't get parts for, or circuits or can't repair. Finally it's not an exercise in right or wrong way to do something. We make our own Solar chargers and inverters to be self sufficient, and like others, I have a lot of spare parts for any, and all repairs, and pre-built modules. I will not build a system that requires an interconnect network of ANY kind unless I designed and built it - even then I would likely not use it because of the increasing critical parts shortage, transportation, supply chain, infrastructure and economic situations that are full of read flags and increase with each passing week. If I'm not around, my Wife, family and friend must at least be able to operate the system, and someone with a bit of intelligence should be able to swap out a complete spare Power board or controller, or low voltage solar charger if needed, I make step by step videos and manuals for them. Your outlook, location, situation and views my be completely different to mine, your Inverter, solar designs, installation and needs will also be completely different to mine.  _ Edited 2024-07-30 08:16 by KeepIS It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Cabinet construction for the Dual power Inverter is underway In preparation, I decided to add an input for keeping an eye on the 2nd power board Capacitor bank charge voltage, in case I need the controller to hold of closing the Kilovac until both banks are at battery voltage. This uses analogue input A7 and a 100k - 4.7k divider, same as the Vcap input divider R23 and R25 on the Controller board. It actually works out very neat as it utilizes spare pin 6 on J5-J6 and the same calibration as the existing VCap ADC. The results are perfect, there is absolutely no ADC cross channel coupling with AC current input A6 or other ADC channel. It's completely automatic, if a voltage > 10v is on ADC 7 then the LCD displays it instead of the Code Version number: LCD displays "Ca2 53.2v", once A7 is below 10v, the LCD reverts to "Ver 7.5ks" If I find I don't need a Ca2 display (monitor the second cap bank) then it can be used for any other purpose required. _ Edited 2024-07-31 13:56 by KeepIS It's all too hard. Mike. |
||||
| Cpoc Regular Member Joined: 28/05/2024 Location: PortugalPosts: 78 |
All these improvements are great. Looking forward to see it all come together. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
I was going to start building the toroid cases first, so glad I didn't, instead I built a 5 wheel Base that will support the cases that make up the inverter. The frame is built with 90mm x 35mm pine, has three internal braces using the same timber, supporting the toroids on each side of the base, the top and bottom covers are 7mm ply. I incorporated 2 x 100mm 240vac fans in the base, one for each toriod transformer. The two fans are sandwiched between very thick dense rubber blocks that have some weight to them, but very springy texture, don't know what it is, but I removed them from some 12v LifePO4 battery boxes, they supported 4 heavy 3,2v cells on all sides. The two 240v fans are wired in series when the temperature relay not energized, and they are absolutely silent - seriously silent when mounted between rubber blocks like this. In series, these fans exaggerate the natural convection air flow through and around the toriods, when Inverter AC is on, these fans run silently all the time. I have found after a year and more of running a similar air flow setup on the existing single toroid, having this exaggerated convection air flow around and through the Toroid at all times makes an amazing difference to long term toroid running temperatures. When the Nano controller Toroid temperature control kicks in, it enables a 12v 0.8W relay in the Base housing, this relay switches the Two fans from series to parallel connected for toroid cooling, and once again, the lack of any bearing or AC motor noise through the case is amazing. Once the toroids are mounted and the toriod cover is fitted, I doubt I will even hear this very high Air flow. The light weight five wheel Base fully supports two x 3 stacked toriod transformers and 4 chokes, the fifth wheel means it will never sag, it supports two of us standing on it with ease, and I can lift it easily with one hand. My plan to build two toroid cases has changed, now just a single removable cover over the two toroidal transformers and four 6 stack toriod ring chokes, these simply sit on the base, and no, they do not and can not move, no mounting hardware needed except for a simple mount for each choke. Like most of us, I'm sure the plans will change slightly as the build continues.   _ Edit - a five wheel base, not six  Edited 2024-08-05 17:59 by KeepIS It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
I finally have the basic cabinet finished, still has to be sanded and coated. The pictures show only one 3 stack toroid, an Aerosharp toroid is there just to balance the weight as I built the cabinet. The PLY cover over the Toriod housing will be replaced with 7mm thick hardwood which is the same material appearance as the 18mm case, the same 7mm ply will be used to make a sloping front display/control panel. The idea was to make servicing or modifying as simple as possible, and make choke or toroid removal or modification easy to do. Each power board and heatsink take only minutes to remove or install.  The lower Toriod cover sits in a slot, it's held on with rare earth magnets and a lock bolt in the center top. The Big opening in the top compartment is for the Sloping panel for inverter controls, meters and various LCD displays and LEDs.  The large compartment and display panel (not fitted) simply lifts out for full access to the Toroids, that compartment will plug into the back panel (not fitted) making it simple to remove, the display panel will hinge down for access and adjustments without having to remove the compartment.  The front cross bar is held in with a big Hex screw and metal screw thread inserts in each end of the cross bar, removal allows full unhindered access for Toroid removal.   Heatsinks are retained by two metal screws on each side. Lower AC socket is for dual base Fans. Large terminal block is for the Dual toroid AC outputs. Still needs the center back cover and toroid compartment exhaust ports. _ Edited 2024-08-10 19:36 by KeepIS It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
I have decided to implement a few extra safety features into the high power Dual Toroid 32 FET build. This is still an experimental build, it may work perfectly first go, or it may require some modifications, especially at very high power. I will incorporate a Peak DC input current trip circuit for each power board, each with an over-current Latch and LED indicator, the outputs of each will feed the Nano Controller Master over-current Latch. I will do the Same for AC current trip on each Toroid, each using an AC current transformer and associated Latch and indicator, again feeding the Master over-current latch on the NANO Controller. Why? 1: I want to know which Power board tripped over-current. 2: I want to know if AC or DC current tripped the Controller, and from which Toroid AC output. 3: Switching between LOW-HI power mode (1 or 2 power boards) requires no change with respect to DC or AC over-current protection adjustments/settings. It will take all the time wasting setup, calibration and hassles of working out what tripped the over-current and why, and allow full monitoring of DC input balance between the two Power boards. _ It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Getting closer, sanding, edge round-over and stain / oil to go. I will start the wire up in a day or so, then shut down the current Inverter and transfer that 3 stack toroid into the new cabinet. I'll be using a temporary control panel that I can botch up and finalize the layout with before transferring the final design over to the good control panel, but I will have the inverter fully running with the ugly temp panel before that. I'll allow myself two days to get it up and running with both power stages paralleled, once I'm happy with that, I'll move on to adding Wiseguys Dual power control exactly as he designed it, and giving that a good test and gathering some readings under various loads for a dual unit.  _ Edited 2024-08-11 16:18 by KeepIS It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Finished a bit faster than I thought, start wiring and component layout tomorrow, dam Chinese wood oil has a strong smell (plant based). I'm thinking of taking the main Inverter off line tomorrow and just swapping the power board, Nano controller and Toriod over to the new cabinet. That should allow a build to basic test running state of a couple of days. Sacrificial mock-up front panel fitted, cabinet looks way better in real life.  It's all too hard. Mike. |
||||
Revlac Guru Joined: 31/12/2016 Location: AustraliaPosts: 1022 |
Looking good, I started building a cabinet as well but all in steal, Still changing some ideas as I go.  Cheers Aaron Off The Grid |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Thanks, it looks a little strange, but this one is all function over form, don't care if cabinet screws are showing, or timber does not match. The base area is similar to the last smaller unit, but this one is higher, so in reality it takes up no more floor space, the 5 rollers are a joy to move the weight of this cabinet with twin 3 toroid stacked transformers, 4 big heavy chokes and two big heatsinks. It may not look like it yet, but this is designed to give the cleanest and shortest heavy 2G and 0G cable flow from the battery to the kilovac, power boards, chokes, and Toroids, along with maximum natural convection airflow over two long vertical mounted heatsinks on the cabinet back, and fan enhanced convection air flow with high flow forced fan cooling through and all around the two toroids as needed. This also evolved as I built it, and now that it's built, I would not change a single thing about it, it's the cabinet I wished I had built after finishing the first build, but isn't that usually the way it happens. _ It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
The build is slowly progressing, I'm forcing myself not to take any shortcuts for the sake of "just getting it going". I've also decided to complete the Control panel and various LED indicators and display modules first before running the inverter up. I'll also build another Power board and leave the existing Inverter in a running fashion, minus its 3 stack toroid which goes across to this new Dual power stage & toroid inverter. I'm looking at another three to four days before powering it up, it's looking so good that I would hate to regret a rushed wiring or sensor mounting/location decision, just to get it finished a bit quicker. I also have to accept that I'm not as fast as I once was, even from just a few years ago  they lied about golden years, they totally suck  _ It's all too hard. Mike. |
||||
| Cpoc Regular Member Joined: 28/05/2024 Location: PortugalPosts: 78 |
Getting old sucks there is nothing golden about it. I now have to wear my old man glasses just to read anything on paper. I wish I was 20 again. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Include time wasted trying to find the right pair for closeup work as well as forgetting where I put the other pair in a the middle of a messy project build.   It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
I have been dreading cutting out the front panel, so I made a timber template first and mirrored it on the front panel at each end as a cutting guide, I had planned a day (tomorrow) for doing this because of how careful I have to be with it. But a few hours this afternoon and it's almost ready, the fancy color lettering overlay will be made and fitted after the inverter is fully tested, just in case I need another LED or switch added. A quick test fit and everything is aligned and fits perfectly. Now only two small Heat-sink temperature meters, a power switch, reset switch and LEDS, and I can wire this up for the Nano controller and other Sensors. BTW The front panel does not require any holding screws because of the way it's recessed in opening, the handles allow it to be lifted out and placed on top in an instant. This is just so easy to work on.  _ Edited 2024-08-15 18:19 by KeepIS It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
Getting close now.  _ It's all too hard. Mike. |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
The top row of meters are L to R 1: Power stage 1 DC Volts, Amps Power 2: Heat sink 1 temperature 3: Nano controller LCD 4: Heat sink 2 temperature 5: Power stage 2 DC Volts, Amps Power Two analogue meters are 600A FSD Peak DC Input current: LH - Power stage 1 RH - Power stage 2 Bottom row LED indicators (white) are Two on the Left - Power stage 1 and Two on the right Power stage 2, AC and DC current trip Indicators. Middle Green LEDs: 1: Battery 53V DC connected 2: Low Voltage Input 3: RED LED: Nano Controller Status LED 4: Main Over current RESET button plus OC Indicator 5: Dual Power Mode enabled. Center large Black knob - 3 position Inverter power: (1) OFF - (2) AUTO start - (3) ON  It's all too hard. Mike. |
||||
| Page 16 of 23 |
|||||
 Print this page |
