 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 : Wiseguy New Inverter Build Nano R6
  Page 8 of 13   |
|||||
| Author | Message | ||||
wiseguy Guru  Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Couple of minor BOM tweaks - Disco4now also pointed out missing J3, changes are in RED. WG39R7A WG44R2 BOM.pdf The Variac BOM should clarify that the ACV setting POT should be a 10K Linear 3/5/10T multiturn pot for precise setting, these are ~$30 from Digikey and ~ $2-3 from Ali. If precision setting is not your thing then a single 270deg pot works fine too. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| mab1 Senior Member  Joined: 10/02/2015 Location: United KingdomPosts: 209 |
Yay! Came home from work to find father xmas had been!  All arrived safe and sound ( and very well packaged)  Thanks again Mike for all your hard work!  |
||||
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
R13, R39 - R42 33R DNL? *** 0.25-0.4W Axial 10mm (0.4" pitch) 5 *** R39-42 DNL if U6 is a 74HC86. Should this just read the following, or am I going blind? R13 | 33R 0.25-0.4W Axial | 10mm (0.4" pitch) | 1 ? FYI: Controller worked first go It's all too hard. Mike. |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
If you are using a 74HC86 its output is barely adequate to drive a power board properly. Its absolute maximum current is stated as +/- 25mA but in the fine print it says that this is not a target and if you actually try to drive 25mA the device may no longer function properly (during lower normal use). It then vaguely states that you should use it under recommended typical parameters and of course there is no parameter given for the current value. Under electrical characteristics for the HC86, with a 4.5V supply and 4mA current drain the output voltage minimum is just under 4V and when sinking 4mA the output drive to the Optos is only 3.7V from a 4.5V supply (why provide parameters @ 4.5V or 6V supply when most logic is either 3.3 or 5V??) so maybe expect max voltage drive of ~4.2V from 5V. For the FOD3182 the recommended current is 10 - 16mA. So if we aim for the centre of that range 13mA is a reasonable figure. There is no data for 5V or 13mA so I recommended the 74AC86 or 74LVC86 as their absolute maximums are 50mA and they have a similar amount of voltage droop to the HC86 but that is at 24mA not the HC86's 4 mA load and operated at ~13mA I expect the droop value to be halved. So if you use an HC86 or similar I would not use the resistors instead load U8 & U9. Then you have ~ 100 times (2A) driving ability with essentially zero droop at 25mA & 5V. If you use the AC86 or LVC86 the buffers are not really needed so just load the 4 x 33R resistors. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1679 |
So we both miss understand the other. Mine was misunderstanding the wording for resistors R39 - R42, as they are not marked on my board, or the assembly layout. I fully understand the technical issue with the 74HC86. It's all too hard. Mike. |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
OK now I get it ! I just went to look at the silkscreen and then it all came flooding back. The previous unreleased Rev7 did have the resistors on the silkscreen but only 1 x 8 way connector - there was minimal room for both. Then I made an executive decision that I wanted 2 x 8 way connectors cause one day some lunatic will drive 2 power boards for a whole bunch of power. So the resistors are to be installed in the U8 & U9 holes standing upright (more likely leaning) if using the AC86 or LVC86 devices - I just forgot to tell anyone..... I am not infallible and checking ones own work is not best practice and bending the rules by utilising PCB pads for ICs or maybe resistors you can catch yourself out. Edited 2024-06-29 13:51 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Note Edited; Here we go again, latest BOMs and schematics for Variac and Inverter controllers. disco4now has fed back some missing lines R1,2 & 3etc in the Variac BOM - dont know how or why they went missing but now they are back! I have included the extra 1K resistor and 8V2 zener in both controllers parts list BOMs that go with the 5V linear regulator with enable (KA78R05). The picture below shows what the mod might look like for the Variac PCB. To clarify a point I have never had code corruption in normal usage with either the Variac or Nano controller, ditto for Poida and KeepIS, so using the 4 pin linear regulator Mod or and supply rail monitors may be seen as overkill, so you can always leave them all out and just build it normally & see if there is a problem first before a preemptive cure for what problem ? At least if the problem appears we do know how to resolve it. The 3 holes to the right of R52 are the original mounting holes for PS2, the 3 pin Reg.  VR1 (KA78R05) is mounted by bending 2 pins forward (In, Out) to go into PS2s outer 3 holes, 1 backwards (Gnd) and Enable is a floating termination of D11 & R52 I also included the 2 x +5V supply rail monitors in the BOMs, 1 for the Nano LCD module, the second for the controller Nano module. I recommend mounting the rail monitor for the Nano controllers on the underside main PCB Nano socket pins. The LCD one does not matter if you put the rail monitor on the module pins directly or the interface PCB as they are soldered direct to the PCB. Latest BOMs and Variac Schematic with 5V Mod WG45Rev2A Schematic.pdf WG45Rev2A WG44Rev2 BOM.pdf WG39R7 WG44R2 BOM.pdf Edited 2024-07-03 12:03 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Ziki_the Newbie  Joined: 13/04/2023 Location: YugoslaviaPosts: 39 |
Great news .The package arrived yesterday. Condition not checked, because I am not at home for another 8 days. Once again, thank you wiseguy for your hard work and effort. Now to collect parts slowly. Pozdrav iz Srbije |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
I decided to continue the staged inverter discussion back on the thread where the Nano controller & Power PCB and Variac builds started. The staged startup got very complicated very quickly so I went back and had a hard think about minimising the circuitry and parts to achieve what I needed. The basic functionality I chose works like this. A threshold is set near 22A of load current (from stage 1) which sets a re-triggerable monostable with a 10 second timeout, so the first sinewave that exceeds the 22A threshold starts the 10 second timer. Subsequent pulses even if 4.5A lower will still retrigger the timer due to the hysteresis fed back from the 10 sec timer. The 10 sec timer when untriggered or timed out prevent optical pulses to the second Power PCB and also holds off the 30A relay from turning on. The instant the 10 second timer is triggered it triggers another ~ 2 second timer, during that 2 seconds the soft start relay is pulled in for the 2 second soft start interval, when the 2 second timer times out, the 30A relay is triggered to turn on and a small delay to the soft start relay turn off, ensures overlap such that the soft start relay opens after the 30A relay is engaged. The same signal that turns on the 30A relay also enables the opto drive to the second stage which should be just before the 30A relays pull in time. After there has been no more current pulses greater then 17A for 10 seconds the first monostable times out which immediately disengages the second stage opto drive and the 30A relay. A single pulse exceeding the threshold will cause the engagement of the second stage for about 8 seconds (after the 2 second soft start) before disabling the second stage again. The KW thresholds for extended on / off are ~ >5kW and <4KW The manual bypass was a bit trickier as it needed to retrigger the first monostable as long as the switch was closed so I employed the second half of the comparator as a pulsing circuit to fool the first comparator that we are continuously exceeding the trigger threshold to start the second stage. I have not built it yet - only in my head but here is the circuit I intend to build up to test as soon as I am well enough to get back in the workshop, been down with man flu   The circuit uses some MM2 logic (mickey mouse) the resistor/diode bits function as and gates, maybe not pretty but will work as required I believe. 2Stage.pdf Comments questions criticisms all welcome. Re: Cpoc and the question about a remote WIFI LCD, Poida can confirm the following, some of the Wemos type wifi chips I understand can act as a data modem and as the LCD is a uni directional receive only comms I would not be surprised if just putting the LCD data into the modem input of a TX Wemos & using a second RX Wemos to drive the data to the remote Nano LCD might just work. Re interfacing to the home assistant device though is beyond my area of expertise and interest, best left to a software specialist/hacker (talented one...). Edited 2024-07-19 01:12 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Cpoc Regular Member  Joined: 28/05/2024 Location: PortugalPosts: 78 |
Sounds like a solid plan to me. |
||||
| nickskethisniks Guru Joined: 17/10/2017 Location: BelgiumPosts: 458 |
Looking good, although I didn't dive in the schematic all the way but perhaps instead of the mosfet switches, the pwm signals could be buffered again by a 74HC86, the 74HC86 on the controller board will run close to it's limit I think when powering 2 optodrivers per chanel. |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
The source of the controller PWM opto drives are from TC4424's which are 3 AMP drivers, they are 5V signals. At first when I considered rebuffering, I would need 5V (more bits) and it was not necessary to re-buffer a 3A signal or even desirable as I wanted to introduce no extra time delays between the 2 stages drive signals. The opto drive FET interrupters look a bit unconventional but their gates are controlled with a 12V drive so if the 5V logic signal is H or L, then the 4 FETs either have 7V or 12V of gate drive with respect to their source so they will be hard on when controlled to be on for their ~13mA gate drive signals with zero delays. When the FETs are controlled to be off, yes there are 2 internal FET diodes both facing upstream so when one FETs body diode could conduct its mating FETs diode opposes the path. The words meaning did not really register to me past the comma - the HC86 if used should have the TC4424 buffer IC loaded, I would not try to drive even 1 power PCB from a lowly 74HC86 drive. By the way for anyone interested, MickeyMouseLogic Edit: The posted schematic appears to have 2 surplus parts. The same node that drives the Gate of Q6 could be connected to the Q1-4 gates, saving D3 and R13. As the logic got simplified I missed it.... Edited 2024-07-19 12:44 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| nickskethisniks Guru Joined: 17/10/2017 Location: BelgiumPosts: 458 |
Sorry, I forgot there was indeed those tc4424 in between. Little distracted by only seeing those 150R and fod3182 on the schematic.  Thanks for the MML circuits link, didn't know about that terminology. Edited 2024-07-19 17:08 by nickskethisniks |
||||
Murphy's friend Guru Joined: 04/10/2019 Location: AustraliaPosts: 648 |
Thank you for that link, MML looks simple enough even for me to understand it  I shall try it out when I get back from my caravan trip. |
||||
| Eddy C Newbie Joined: 17/07/2024 Location: AustraliaPosts: 5 |
Hi, A Brand Newbie I would appreciate the opportunity to purchase boards if another order is contemplated. 3sets x 2.5KW Variac/Auxiliary inverter 5sets x 6kW Inverter EG8010 controller? I’m retired on a rural property in Adelaide where every day is Sunday with nothing rushed. I have a good 6 months lead time before needing boards. Currently building the Oztules 5kw 16FET power Bd and 8010 controller. I found a $100 1700w C core on eBay for testing. I still need new Cu for the secondary and Toriods from AEM Cores. Hoping prices reduce with Qty. (Can only dream). I have around 20kw of surplus solar on the roof and inslab water heating wanting access. To that end I’m creating 4 local grids to sync 12 x Fronius Primo’s with zero export smart meters and power management. No Grow rooms, just keeping warm without cutting wood. I forget to check emails, 2-3 weeks can pass, if you PM, sorry in advance and will respond. Regards Mick |
||||
| rogerdw Guru Joined: 22/10/2019 Location: AustraliaPosts: 852 |
Hi Mick and welcome ... I'm in Murray Bridge so we're practically neighbours. Sorry I can't help you with boards ... that's something for Wiseguy. Sounds like you have a fairly ambitious goal with all your requirements ... 20kW of panels is a serious system. What power are the Fronius Primo's? I'm assumimng you've seen the discussion in the other thread about combining power boards to create 12kW inverters for those who want a bit more grunt. May be another alternative. We have 19kW of panels and a 15kW Warpverter ... along with a "normal" 5kW Growatt GTI AC-coupled ... and are effectively offgrid .... though it is still connected to the property and handy as a backup. We also occasionally turn on a trickle charger if there's a couple grey days in a row ... though I am working on a diesel generator to use for backup as well. Oh, and our battery is a IFLA forklift battery ... 48V 980aH. The previous owners had fitted underfloor heating but had never connected it up ... but whenever we have excess solar, it's the first thing I turn on to try and keep the place warm. There's two circuits of 3kW and they told us there's another ... but I can't find any cables ... so it's only 6kW at present. We often can run them for 5 or 6 hours a day on good solar days ... but they're a bit scarce just at present. I'm sick of chasing wood too ... so always on the lookout for ways to keep warm. Good luck with your projects.Cheers, Roger |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Hi Mick, welcome to the forum, I have sent you a private message re boards availability. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| Eddy C Newbie Joined: 17/07/2024 Location: AustraliaPosts: 5 |
Thank you for the welcomes, I’m a slow two finger typing dyslexic semi old fart compared to some on here. Sorry compiling a response is a very slow process. For background I was a 1973 PMG intake. I reside on the Sturt River, a 10 acre Do Little farm in the Sturt Gorge. The only service is a 3ph power pole. Owning a section of river I could have 100kw hydro operating in winter when the sun does not shine. Unfortunately already need to live till 150yrs to complete all the projects started. The house is 20m above the river with the hill rising a further 70m behind the house. Its on a DIY hill side cutting, hand drilled holes blasting 11m of rock. I found a retired licensed powder monkey in Mt Barker, great guy, charged by the day. I had around 8 blasting days, varied from 50kg, 100kg and 200kg blasts. Just another learning curve rattling the valley. Talk about being motivated. The hillside rock was exposed, each hole charged with 8 sticks of reogel and a lightning storm formed of the Brighton coast heading our way. It was a race to complete connecting all detonators forming a giant aerial and set it off before the lightning did. True Russian roulette that day. I must have a top ranked avatar driver. 2009 purchased my first 24 x 175w Suntecs (AU$600 each) and 11kw $6000 SMA with its 15yr warranty. Less than 12 months old, with only 12 panels installed and lightning turned the SMA off for good. 12 x Suntecs are still in their original Pallet. Had another 2kw of panels and 4.2kw Growatt installed. Learning I purchased a spare Growatt. Well lightning got that too but had the spare. The next strike left me with earth leakage when drizzly wet. The first 5kw Fronius was connected 2021. Lightning got it that first night. It still works but self consumption readings doubled and it does some funny things it should not be capable of when master in a string. Nill strikes 1992 till 2010. Big mistake extending the earth on to the roof. Everything now gets unplugged when storms approach. I had to stop at 20kw on the roof for now and fit scaffolding supports under the ridge, truss gang nails are coming out. Another 20+kw is still on pallets scattered around the block. I had a stroke and sometimes forget where bits were left. The family looked puzzled at me one time. I had bogged the bulldozer, then the backhoe and asked for suggestions re hiring a tandem tipper to toe them out. I forgot having one parked only 30m from the bogged plant. Everything is still in there, have just lost the indexing. Just have to be patient till found. I have 2 x 10kw 3Ph Frinous, 5 or 6 +1 spare x 5kw’s Primo, same for 6kw’s Primo. Each purchased new with a smart meter and a couple of spares meters. There’s lots of forgotten bits in boxes to deck them all out. Collecting (hording) since the 90’s to go off grid once batteries were cost effective. I am truly impressed with the Fronius zero export and its power management ability. I have been adding a zero export GTI with each new load. I found the Inverter :GTI 1:1 rule can be exceeded with GTI’s configured zero export with smart meters in series and some simple trickery. For another time. Theres a 6 cylinder ford turbo diesel 60 or 120kw (forgotten) big enough to drive William Creek Pub (where it came from) and everything I have. A 4 cylinder diesel trailer lightning plant with 20kw 3ph alternator. (+ two spare 20kw 3Ph alternators) 4.2kw and 5kw single petrol gensets. Two lister twin diesels, couple of single lister’s and the Rolls Royce, my retirement hobby a 7HP Blackstone oil engine. I manage to find something that starts each power failure. Slab heating, 3km of poly water pipes installed in slab 1992 and 3 years ago started connecting it. Last year I was dumping solar into 6kw of 3ph elements + 6 x 2meter water panels. The slab always wants more. I did get the slab up to 36 degrees in August 23. Saved lots of wood. I have started using VFD’s on 3PH elements to protect the Chinees 220v elements. They did not like or last long at 250v. I was working at controlling the VFD to track available solar. Now wondering if this Viarac inverter board could track and dump available solar in one step. I nearly purchased Lifepo4 cells November 23. Then realized I should have the charging in place first. I need more advice here as lots still use lead acid? I maintained Telstra batteries for nearly 4 decades. Have also collected everything to build lead acid cells in 200 and 250mm Poly pipe. There is still no change from $10k but no BMS needed. Regards Mick PS Wiseguy, response and number coming, need to consider those options first. |
||||
| Cpoc Regular Member Joined: 28/05/2024 Location: PortugalPosts: 78 |
If you are going to build any type of Lithium battery packs wether it is lithium LifePO4 or lithium ion you need a BMS I would suggest Diybms or Batrium. Diybms does have a limitation of 128 cells max however you can parallel cells to make larger packs. Just add a active balancer and a behind the scenes active balancer like the Neey balancer for larger packs. |
||||
Revlac Guru Joined: 31/12/2016 Location: AustraliaPosts: 1021 |
Hi Mick, Sounds very interesting, had lightening take out solar gear as well, Would be interested to hear a bit more about the old engines etc one day... in another thread. Cheers Aaron Off The Grid |
||||
| Page 8 of 13 |
|||||
 Print this page |
