 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 : 6Kw Ozinverter build
  Page 46 of 51   |
|||||
| Author | Message | ||||
renewableMark Guru  Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Jesus Christ, last winding of 4 layers I ended up with 80mm ID, this time it's 85mm, reckon that balloon in the centre while doing epoxy works well, 5mm doesn't sound like a big saving but it will help. Cheers Caveman Mark Off grid eastern Melb |
||||
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Every bit counts in that hole, I am aiming for 6 Secondary windings. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
How many turns? I just did the same as last time 128 Cheers Caveman Mark Off grid eastern Melb |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
114 Turns my previous one was 98. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Reckon I could fit another layer and still have more room than the last ond due to the space saving. Not too sure if it needs it though. Cheers Caveman Mark Off grid eastern Melb |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Jeez 114 turns and 6 layers, that's 684 turns!! my 128 x4 =512, yours will be 33% bigger than that. Cheers Caveman Mark Off grid eastern Melb |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
I am working on the amount of free space I had at the end of my last build, I was surprised how tight you found it with yours. Getting the inside of that hole as neat and flat as possible does make a big difference as you are finding out. I will have to set myself a schedule of 1 hour a day otherwise I might not get around to it. Will be good therapy break from coding work I need to do for another project unrelated to solar power. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Well, with my more efficient winding this current torroid ended up 5mm ID bigger gap than the last one, so I reckon it could take another secondary winding of 128 and still be bigger ID than the first torroid I wound. IF I did do that it would be 5 x 128=640 turns. Kind of tempting.............. hmmmmmmmmm. (I reckon I could take another 3 or 4mm off that with more practice and the methods below.) Anyway your idea of 6 secondaries of 114 will only be 48 turns more than that, so that should be achievable. One thing I reckon is you really need to straighten the daylights out of the wire. It needs to be really tightly wound on the spool you work with so you can really pull super tight without the wire pulling into the layer beneath on the spool which will create a kink that will mess up the neatness. And not put too much on the winding spool, so the above doesn't happen. And not be afraid of joins, hell it's piss easy to do a join on the outside where it doesn't matter a stuff about extra space. Using the expanded balloon when the epoxy dries is a good trick that may have had a big factor in my tighter winding's this time. Just wonder if there is a bouncy ball or something similar that could take more pressure to do the same job only better. Now I have learned all that I'm going to have to do another one this time just a tad better. Only have one 3kw torroid left though. Cheers Caveman Mark Off grid eastern Melb |
||||
| Mulver Senior Member  Joined: 27/02/2017 Location: AustraliaPosts: 160 |
Hey Mark! What about a small tyre tube ? Say wheel barrow? Pump up with compressor, put a oven bag over it ? FYI. I’ve been keeping up with reading everything on the forum. But all projects on hold as we decided to sell up and move house. Great work on your inverter! Been great to read! Cheers Shane |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
I have another suggestion for you Mark, unwind it all and start again   Thinking about it what I did was use a roll of chux wipes, as I added more layers I just took more off until it slid in there nice and firm. Not saying it is better just how I did it. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Oh you really are a bastard making me think about that. Cheers Caveman Mark Off grid eastern Melb |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Thanks mate, when you buy the new place make sure it has a nice big north facing roof. Cheers Caveman Mark Off grid eastern Melb |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Your going to be laying in bed tonight tossing and turning, calling me all sorts of four-letter words. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
Tinker Guru Joined: 07/11/2007 Location: AustraliaPosts: 1904 |
Mark, what *does* matter is the quality of the joint. How is your soldering coming along?  . The wire ends should be tightly twisted 2 or 3 turns and then soldered.It might be much more informative for others reading this post if, instead of bragging about number of layers, you brag about total wire area in all your layers. It is that area that counts with wire resistance and, therefore, its current capacity. My 6KW inverter used 10 mm sq (in 4 parallel wires of 1.8 mm diameter) for the secondary. This could safely carry around 40 Amps (40 x 235 = 9400Watts!) Any more is filling the hole un necessary as I doubt your battery could handle that for the time it takes for the toroid to heat up. Now, with a 9:1 turns ratio the primary should then be 90 mm sq. I did that using 6 x 15 mm sq rectangular enameled wire This would be rather tricky to do using stranded cable. So, concentrate on that 9:1 ratio *if* you want to build that perfect toroid. No point being generous with the secondary and stingy with the primary .With regard to turns, I could have done better on my 6KW unit. If doing it again I would use 12 turns for the primary and around 110 for the secondary. For the latter I would plot a graph (mA= Y axis, Volt= X axis) for voltages from 220V to 250V. Plot points for 110 turns, 111 turns, 112 turns, 113 turns, 114 turns. The plot curve is flatter for more turns, I like it just slightly inclining a the preferred voltage (235V) rather than steep to the top. If there are too many turns there would be a very flat curve telling me no more worthwhile idle current lowering with additional turns. Of course, one needs a Variac for that plotting. And, of course, the number of turns relates to the total core cross section as a starting point. Warpspeed wrote a very good post how to calculate that some time ago. Klaus |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Stinker Tinker, don't be ridiculous, no one bragging here mate, simply stating the importance of getting everything as neat as possible, and demonstrating a clear advantage of 5mm doing so. The only person I see bragging here is someone babbling about rectangular wire and his torroid that is bigger than my one. Seriously Klaus go beat your chest somewhere else. If you like to build them bigger, go for it, knock your socks off, makes no bloody difference to me. The size I made mine is more than enough to power two homes, I see little point making one bigger than that. Anyone who has followed this has seen the pics of me unwinding and using the aerosharp wire, all the calculations were done at the start of the thread. You don't expect me to state that in every post I make do you? Warp was a big help with all that and has got my unit tuned nicely at 75hz, calculated and tested the choke we made at his place. It's a well tuned setup that I am simply going to replicate. Cheers Caveman Mark Off grid eastern Melb |
||||
wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Warp I always hold you and your posts in high regard but I couldnt get my head around this post & I think there is a problem with the maths here & if you have made a mistake by halving instead of doubling it could lead others astray. Hopefully I am not making a fool of myself too! If an inverter is passing 6KW from a 48V battery that equates to 125A rms. But if the mosfets are switching and only on for 50% of the time then surely they are not passing half the rms current but double the current as for conservation of energy half the current would be half the power. The peak current of 250A but for 50% on time is 125A RMS divided across the number of FETs per leg, 4 FETs would be ~ 31Arms ea or ~ 62A peak but for 50% duty cycle? At the end of the day for the discussion about screw terminals I plan to use them for my mosfets in the power boards until they are proven to not die unexpectedly, then I may consider soldering them in permanently. The screw connectors I plan to use have 2 pins per contact into the PCB - a small consolation for exceeding their conservatively rated current. If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
BTW I'm not totally against the terminals, just a bit uneasy with the 20A rating and the worry of them coming loose. I like to just build things and not have to fiddle with them, solder seems to be more appropriate for me. Wiseguy have you got a link to the type you intend to use? Cheers Caveman Mark Off grid eastern Melb |
||||
| wiseguy Guru Joined: 21/06/2018 Location: AustraliaPosts: 1156 |
Mark, you can find them Here If at first you dont succeed, I suggest you avoid sky diving.... Cheers Mike |
||||
| renewableMark Guru Joined: 09/12/2017 Location: AustraliaPosts: 1678 |
Thanks, still reckon I'll stick with solder. Cheers Caveman Mark Off grid eastern Melb |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
That would certainly be true if all of the power was going through a single mosfet. If it was off for half of the time. To make up for that, the on current would need to be twice as high as the averaged output current. But that is not what we have. We have two mosfets in a half bridge configuration. One is on and the other is off. When one is conducting, its partner has a complete rest while it is off. So we have two mosfets sharing that 125 amps alternately. Each will be off for half of the time, so each mosfet sees 125 amps, then nothing, then 125 amps then nothing... Or a time averaged 62.5 amps. Even if there is considerable dead time, current still has to flow through one of those mosfets through its inverse diode, so it still flows through one of the mosfets. Its not quite that simple though. Although the average will only be 62.5 amps the peak current will still be 125 amps through each mosfet while it is on. And the heat generated in both the mosfet and screw terminal block will be current squared times the resistance, but for only half of the time. So everything runs about twice as hot as it would with a steady 62.5 amps, but only half as hot as it would with a steady 125 amps continuous. It will be four times as hot because twice the current creates four times the heat. (current squared) So basically the two mosfets in half bridge configuration share the load alternately, and split the heat between them evenly. The actual RMS value through each mosfet will be 0.7071 times 125 amps rms, not half 125 amps rms. (because of the current squared thing) That also applies to any terminal blocks. How many mosfets for 6Kw to share that new effective 88.4 amps rms (allowing for crest factor). Maybe six mosfets per leg perhaps ? Six 20 amp terminal blocks are going to do that 88.4 amps rms per leg very easily. Five mosfets per leg would probably be o/k too. Four mosfets per leg for a reliable 6Kw is pretty optimistic IMHO. Sorry about all the pedantic details and the math. I know that really pisses some people here right off. Cheers, Tony. |
||||
| Page 46 of 51 |
|||||
 Print this page |
