 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 : Toroid winding questions for a WG Inverter
| Author | Message | ||||
| rogerdw Guru  Joined: 22/10/2019 Location: AustraliaPosts: 951 |
Merry Christmas and Happy New Year to everyone. I'm in a facebook group of people focused around using forklift batteries for solar and a handful are keen to build a WG Inverter and are making inroads. Here are some questions that came up this morning. I numbered them to make reply reference a bit easier. Thanks. "Can we talk voltage here for a bit, as there are a few slight unknowns here which are worth clarifying before commencing permanent windings. 1) Does the different control chip in the WG build require any adjustments to the transformer design? I assume not, but I mention it because P18 par 2 of the book (see below) mentions different target voltages for PJ processors vs the 8010 chip. 2) How important to the WG build is the 15% voltage safety margin mentioned at the bottom of page 16? (see below) 3) There is a comment on page 27 that also influences the design: (words to the effect of) leaving out one primary turn will help out as the battery discharges and produces a lower voltage, by allowing for a higher step-up voltage vs the design. 4) Should we also target a slightly higher step up ratio to allow for battery voltage decline? 5) What about battery overcharge to say, 52V? 6) Finally, referring to Wiseguys' recent comments, is it better to target 240V or 230V output? For reference, my understanding of the step up caculations is: we are aiming for 240V RMS AC output which is 339V peak AC, plus 15% safety allowance for 390V peak AC which is 8.1:1 step up from a 48V DC input." These quotes are from Leslie Bryan's OzInverter book P18 par 2 Page 16 Cheers, Roger |
||||
KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1996 |
Merry Christmas Rodger, just a few quick thoughts. Not really, the WG design will operate over a wide range of voltage inputs, only the Nano Setup parameters and the Toriod design need to be changed to match the desired system DC input/output voltage range, exactly as you would do for any LF Controller & Inverter power stage design. This is a function of Toroid design and is the same for any Inverter controller, primarily as indicated in the info you posted, to keep the Toriod further away from saturation and to lower magnetising currents by winding the output to a voltage higher than required, IE: winding for a lower flux density of around "1 T" instead of cost cutting commercial Toroid values of around 1.4 T. IE: The Toroid must be able to produce (Example 240Vac) at the "lowest" DC input voltage that the batteries are expected to run at without the Inverter reaching 99% PWM drive duty cycle, and if wound correctly, it will also have a lower Idle current at the nominal operating DC voltage. 3 and 4 are similar, if you find you cannot get 240Vac regulated output at the design DC input you chose, you can remove a Primary turn to help fix the problem. All of this comes back to making sure you don't hit 99% PWM running at the lowest voltage input you selected for the chosen AC output voltage, and allowing a little headroom, IE a PWM duty cycle of less then 99%. Both Battery over-voltage cutoff and AC over-voltage cutoff are adjustable in the Nano Settings, as are under-voltage cutoff on both DC input and AC output. If running off grid, I personal like around 230 to 235Vac since most new equipment seems to state 230Vac +/- x%. If you are switching to the Grid power in an emergency then a voltage closer to the "average" grid voltage at your location might be advantageous. My Toroids regulate at 240Vac down to 40VDC, which is totally unnecessary with LiFePO4 as these batteries doe not sag much over their operating SOC. The older design controllers did not have a neat PWM indicator like the WG-Nano design, instead you used a DSO to confirm drive PWM, Poida designed the WG inverter Nano controller to only drive to 99%, at which point regulation stops and AC output sags if load increases. Which translates to, you should not blow the Inverter if it runs out of drive due to a miscalculation. TEST mode in the WG inverter can be used to safely test the Toroid and the main Inverter limits and settings on the bench with a small Current regulated supply and a small CAP bank, this was truly a brilliant design decision as it almost eliminates the Initial power-up "Big Bang" if Test Mode is used correctly. . Edited 2025-12-30 11:34 by KeepIS NANO Inverter: Full download - Only Hex Ver 8.2Ks |
||||
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1250 |
I would like to add some comments to the information from KeepIS - who does know what I am about to write. A) If this is taken too literally just a few hundred watts load could cause the 240V to droop if the test was done unloaded. B) If the Nano reaches 99% and cannot drive the toroidal tranny any harder a small droop down from the set value ie 240V may be quite acceptable. C) Changing the setting for 235V or 230V output instead of 240 may hold the set output for the same load that previously caused it to droop (and be less than 99% drive) without removing a primary turn. I did some tests yesterday to confirm the SPWM drive for the EG8010 & Nano to the actual AC output (initially regulated to 230VAC) when output drive was at 99% and the output was at 226V, set by varying the input DCV to the H bridge for both tests. Then I began to reduce the input voltage from 46V where I knew it was well into regulation until the output voltage drooped down to 226V from 230VAC both with a 100W load. One drooped to 226V with 43.56V in, the other was 226V @ 43.48V in. From this I deduce that each controller must have almost identical SPWM max. modulation. For extra checking I decided to check an arbitrary conversion from input DCV to ACV equivalent RMS for the Toroid primary AC voltage at max 99% duty cycle. So working backwards, the numbers were 42.0VDC input and 29.3VAC output at the Poida 99% modulation maximum. So for the conversion of AC back to DC, we multiply the 29.3VAC by 1.414 and then divide the result by 0.99 (for 99%) and we get 41.84V close enough to our 42V in! I suspect that the dead-time of the power stage in my inverse optocoupled H-Bridge drive is the cause of any minor discrepancy. The summary of the above is that the exact same transformer Primary and Secondary will give the same results interchangeably for EG8010 & Nano controllers. I will add that the EG8010 controller is my modified topology (inverse opto drive) to imitate the Nano drive and allowed me to simply swap controllers and re-run the test. It should also hold true to the Oztules configuration using the EG8010/EGS002 drive configuration but I did not test this. Edited 2025-12-31 11:34 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| KeepIS Guru Joined: 13/10/2014 Location: AustraliaPosts: 1996 |
Thanks for clarifying/correcting that point, reading back I can see that 99% PWM in the context of a running Inverter might be misinterpreted  NANO Inverter: Full download - Only Hex Ver 8.2Ks |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1250 |
I was going to PM Roger with the following but decided to make it public. He has some colleagues that were considering building an OZ inverter, he has convinced them that going down the proven Nano path is a good plan, but maybe some of the group not being overly technical may be interested in the following. I have done more A/B testing between the nano controller and my EG8010 controller and I am actually suitably impressed with the EG8010 performance and stability. The Power stage current for a 230VAC output, mirrors the DC current when using the Nano controller. One slight difference, the nano controller draws ~ 158mA at 12V whilst the EG8010 controller draws ~144mA, nothing to get excited about. But it occurred to me that there are other newbies and watchers who may feel a little overwhelmed at the concept of programming the Controller and LCD Nano's. Of course, then there are the required code value tweaks etc for calibrating which some may also see as rather daunting. The EG8010 controller option allows for simple pot tweaks to set battery low voltage cut-out and fan temperature starts. There are plenty of relatively low-cost AC and DC voltmeters, some also have Watts and Watt-hour accumulators built in. So the loss of the Nano display information can be restored using available add on meter modules. Perhaps just soldering in an EG8010 and not needing a learning curve for programming and calibrating a Nano might help entice a software challenged observer to give it a go and not be discriminated against. I believe KeepIS initially used one of my earlier EG8010 controllers before turning to the Nano. The other observation is that I have not heard a grumble/grunt from the toroid during multiple power up events with the EG8010, which does not have the AC ramp down that the nano has, but it appears that the built in ramp-up seems to work adequately. For the thrill seekers there is also two output/opto drive connectors for driving more than 1 Power stage. If anyone is actually interested in playing with or using the latest version of the EG8010 controller, please make it known and I will order some. The price delivered to you in Australia for 5 blank PCBs is under $10. Edited 2025-12-31 18:08 by wiseguy If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| rogerdw Guru Joined: 22/10/2019 Location: AustraliaPosts: 951 |
Thanks very much Mike and same for you too ... and Happy New Year to all as well. Sorry for my slow response but my home PC has blown up and trying to respond on my phone is very difficult. Gotta keep sneaking out to the workshop to use the PC out here. And thanks for your very detailed answers to all the questions. I was hoping some of the guys I sent the link to would have responded ... but they are probably as shell shocked as me in trying to understand all this.  I hope by reading and rereading this I will finally get a better understanding of all the finer points of operation ... and I have made some inroads. I've been pretty lazy with a lot of this stuff ... if I can get instructions to build something, that's not so hard. But if I have to delve deeply into the theory, I get lost really quickly. Thanks for reminding me of this. After having a few blowups of my own with the Warpverter, having a documented safe test procedure should take away a lot of the fear in building this one. Thanks for going to all that trouble Mike ... and I guess it's encouraging to see the result. I wonder how much money and how many man hours were poured into developing the EG8010 ... and then a handful of enthusiasts had a go and did just as well or better. That has to be encouraging for anyone winding a toroid to know that it probably isn't quite as critical as we worry that it is. No excuses now. Cheers, Roger |
||||
 Print this page |
 |
The Back Shed's forum code is written, and hosted, in Australia. | © JAQ Software 2026 |
