 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 : MAD/OZ AC/DC Charge Controller
  Page 6 of 9   |
|||||
| Author | Message | ||||
oztules Guru  Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Did that have the power control incorporated? If so it may be a transient in the control. Perhaps get it going again, and run it normally for a few days on the grid proper.... then the inverter. If all good, time to redesign the controller time constants or some other problem with spikes. .......oztules Village idiot...or... just another hack out of his depth |
||||
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
The OGI shut down with an under voltage fault flashing 4 times when I went out to it there was still charge in the caps and the LED was blinking even though the breaker had tripped. It is the latest version of the control board with my version of the power board. I adjusted it very slightly each time until it stopped tripping, which it had not done now for over a week. No sign of overcurrent shutdown apart from the breaker trip. I know 100A is not enough(but today it was) , I have a dual pole 125A I was thinking of fitting with poles in parallel, but 250A might be too much, I have some 2 pole 600V 80A breakers coming that might be a better option. At the time this happened the GTI may have disconnected due to low light and reconnected. When I started up the OGI I reconnected the Aerosharp and it was only putting out 2W. This is the same GTI I have had running for close to 6 months although it was not with the latest control board. Except for a very brief few seconds when the s#it hit the fan before I realised the GTI had dropped its bundle. Yes definitely will be giving it a run on the grid rather than risking another set of HY4008s. Feel like I have dodged a bullet this time. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
for normal installations, 100amps is fine. I use 2X68a in parallel for <140amps. It seems to protect things from board damage, and even driver damage when I test them to destruction. These breakers run at design amps indefinitely, and 100 amps is up in the 5kw range, so most battery banks don't appreciate that power for too long anyway. Once homegrid interactive, then the power will be way less during the day in particular, and you won;t be doing 5kw at night for long either. So 100-140amps seems to be more than enough for anything I have seen in the field so far. I only use 20 amps on the 240v output for the same reason. Thats 5kw, and I don't want to run more than that for more than a few minutes anyway, and the breaker will remind me if I over step that . It has gone off a few times when I was pulling 8kw for more than 5-10 mins... and I am happy about that too. You can forget that with the hot water, the jug, the microwave and the washing machine going, that it all adds up. So I am happy to over shoot while the jug boils and a few minutes more even, but I don;t want to see that kind of draw for long. It is unnecessary. Did the zever die, or just the o/load? Yes, the aerosharp is talkative, but pretty forgiving too. Am currently modifying a new 100 dollar inverter welder at the moment for a battery charger... impressive power outputs I must say, just dial up 0-5kw charge.. it is pretty wow for me. .........oztules Village idiot...or... just another hack out of his depth |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
I like the sound of that welder charger, will it run 5KW continuos? Just pulled the Zeversolar off the wall to check it, no shorted IGBT's so I plugged it back into the grid and it is working as it should. I will leave it running like that for a few days, hopefully, it does not trigger any issues with the smart meter to the grid. I will put the 2 pole 80 amp breaker on the Inverter DC input when I get it that is another 20 amps more, so fingers crossed it is not too much, I am running 24 HY4008s which would take quite bit more too. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
The welder sticker says 220amps@ 60% duty cycle, so 100@100% should be possible. I don't have a genny big enough to push it that far anyway. It is just foolin around really, although I can see with a small board, I could turn it into a 3 stage charger with all the bells and whistles... but it is not what I want it for. I just want brute charge for a few hours to push some ergs into the cells if needed..... but i am currently designing a board that will give it two knobs... one for current and one for voltage... not sure why I am doing this either, as it does not suit the brute charge scenario I just described. The current control is probably worth while, but it has that on the front anyway ( amp control for welding) For a 24v system, you just take it out of the box and go, for 48v, I have rewound the transformer, but I may not have needed to have done that either. The rewind is just one more turn on the secondary ( 6 turns instead of 5). It was this one http://www.ebay.com.au/itm/ARC-220Amp-Stick-Welder-DC-Inverter-MMA-Welding-Machine-IGBT-Portable-10A-Plug/272834881672?s sPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649 With the extra turn, it works straight away. I took it to 60 amps, but no more as I only put 1.8mm wire on the secondary.... at 40 amps it was fine actually... there is a strong wind inside of it ( fan). Just something else to fiddle with. ......oztules Village idiot...or... just another hack out of his depth |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
[quote] the welder sticker says 220amps@ 60% duty cycle]The welder sticker says 220amps@ 60% duty cycle[/quote] That is very likely just a heat sink issue. Something easily fixed with a much bigger fan. Cheers, Tony. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Yes Warp, the silicon can probably do it, not sure how the tranny would hold up... probably better than I expect, as it is a soda to cool It looks like this:  There are two heat sinks missing, as the smoke managed to dribble out of the igbt's, but it gives a better look at the topology anyway. Here is my tentative board to drive it, and get rid of all it's circuitry except for the power and supply circuits.  My never get around to it too. Note the skinny wire on the tranny... thats the secondary I put on, the thicker one is the primary  Edit: Thinking about it, it is rated at 220a@28v, even though it is 60voc.... so really only good for 3600 watts in it's real form. That tiny tranny is a center tapped secondary, so has to be twice as big as a single ended one for the same rating.. ie it is really two transformers in one, but at 50% duty cycles for the secondaries. Due to the duty cycles, I suspect we can scale straight back up anyway. But 3600/56= 60 odd amps at 100% duty cycle.... so not bad for a hundred bucks I guess. With my bank, I could drive it to 80 amps, and it would start to fall off fairly quickly to 60 amps anyway. Be interesting experiment anyway. I WILL end up with some sort of charger anyway, that will be too big for my genny anyway look at it. .......oztules Village idiot...or... just another hack out of his depth |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
WTF??????????? Ebays $9,199.99 for that welder There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
I knew the Federal Reserves money printing would lead to inflation.... just a few weeks ago it was only a hundred dollars and nine dollars postage. I see they have sold 98 of them... I assume they have left the country by now... off to the Bahamas....  Perhaps if I stickytape mine back up I can get a windfall price for it  ....Monies no object here  .........oztules edit I bought the cheapest one I could find... as I knew it was only going to be blown to bits anyway.... and I have succeeded  For some reason I remembered it as $100 plus 9.90 postage, but it was 109.90 free post.... I'm definitely loosing itVillage idiot...or... just another hack out of his depth |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Maybe they are trying to recover the loss of sending that one to you. There are quite a few PCBs on aliexpress all cost more than what you paid for a complete unit. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
I noticed that too. At least with that one, you get a box, a fair clamp, a cardboard mask, and a welding tool , brush cable etc.... and it works too. After seeing how nicely it worked I looked at aliexpress too, the whole unit in running condition is a better deal methinks. How it can be possible to build a nice little unit like this on the other side of the world, ship it, someone make a profit here and there, and ship it again via aust post for 109 dollars is a complete mystery to me. I cant believe the price of the ferrite rings that would make up this transformer either. .......hmm.... wonder if we could build our inverters at 50khz, and use the bipolar waves to make a small 3kw inverter using this tranny as the LF transformer.... maybe the demodulating inductors would be big instead..... but to me it seems possible.. must be a good reason it is not done.... I'm just not smart enough to know why. .......oztules Village idiot...or... just another hack out of his depth |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
As I have heard a number of times if you go to university they teach what can't be done. A Flindwegian with 2 brains might come up with DIY version that "just works" There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
[quote].......hmm.... wonder if we could build our inverters at 50khz, and use the bipolar waves to make a small 3kw inverter using this tranny as the LF transformer.... maybe the demodulating inductors would be big instead..... but to me it seems possible.. must be a good reason it is not done....[/quote] Ah, that is how my inverter works. I use large ferrite U cores in three 2Kw flyback inverters, each with own mppt coming from a virtual tracker with three sets of panels to generate high voltage +ve and -ve dc supply rails with good voltage regulation. That then gets hard switched at low frequency to generate a multi stepped sinewave. No high frequency PWM with all of its associated problems required. Its not commonly used at low power levels because its more expensive requiring a lot more individual parts. But at very high power levels at tens or hundreds of Kw its definitely the preferred way. Cheers, Tony. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Zeversolar has completed a full day on the grid and still going. Has me scratching my head about the 100A breaker popping. Is there any problem or is 100A just a little too small or was it just a fluke? Sent a question this morning to that ebay listing for the welder, still no response. Certainly way cheaper than proper power supplies. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
The 100a would only be small if you had 5kw going through it for long periods of time with extra loads somehow too???(don't know how). Interesting the 20a did not blow... if it was from o/load. One can only assume there is something there it does not like in the pwm of the controller.... can't see how though. It should go through the fet, then into the boost controller, and that should even out the pulses from the controller... thats what boost converters do best.... make it into 100% PF, from a grubby waveform. To that end...... I think I would change the 25 in the code to zero so that in the twilight, it turns off and just suffer the three minute bursts if it is running in poor light... system safety first. Anything under 200 watts is fairly pointless, and I expect it will go down to 50w or 60w before it drops out anyway.... at 200w it takes 5 hours to make a kwh... not worth the effort on your system size. Warp, every time I think I have something... it has all been done before.....  ..........oztules Village idiot...or... just another hack out of his depth |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
[quote]Warp, every time I think I have something... it has all been done before..... [/quote] Its sometimes really interesting reading very old engineering text books, how they solved certain problems with very primitive technology. Some of the ideas were quite ingenious and could be used again today, but with a modern twist. Cheers, Tony. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Oz The charge controller is on 100% at that time of day, so it is not making any difference. The GTI reports as low as 2W and there are very cloudy days where it might in that 200W region for quite some time, it is not much but it can keep it in a state where there is little or no draw from the batteries. At the time it stopped and the 100A breaker tripped there was less than 1KW being used. Even if there was a sudden spike wouldn't there be enough power in the caps to cause the overload on the control board to trigger? Warp There are some videos on youtube of old very high power AM radio transmitters in the US. They are incredible pieces of engineering work. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Back before the electronic age (and amplification) there was a need for playing phonograph records at reasonable room volume. A stylus followed the record groove and the minute vibration operated an air modulating valve fed from a silent air blower. The resultant fluctuating air pressure was fed into a horn loudspeaker. The sound pressures produced could be made deafeningly loud. Later, this same idea was used to produce extremely high sound pressure levels for public address. It could be made so loud the whole thing was outlawed as being a public nuisance. It was used in aircraft during the Vietnam war to broadcast propaganda and could literally be heard miles away. "The voice from the sky" It was also used briefly in Australia from small aircraft flying over beaches in Sydney to warn swimmers of sharks that had been spotted from the same aircraft. Around the mid 1960's I think it was. Today rock music at open air festivals they have a massive walls of loudspeakers fed from tens of kilowatts of audio power amplifiers to create deafening sound pressure levels. The whole thing could be done a lot more simply using an old method everyone has forgotten about requiring very little power and something quite small. One day someone will rediscover this and probably be hailed a genius. Cheers, Tony. |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Good news, the Zeversolar GTI is back in action after having just 2 IGBT's replaced. Yesterday I discovered that there was a grounding fault on the PCB with the MOSFETs mounted on it for my charge controller. This is a part of an old inverter PCB that I cut up as a quick way to mount the MOSFETs and associated resistors. The screws through the PCB which has large areas of copper connected to the source and drain pins on either side of the PCB had arced as per the picture. This was only on one side, the other array did not have any damage.  I don't know if this was the cause of the big failure I had, however, the GTI did not want to operate due to low resistance to ground. It was measuring 3.5m ohm on the multimeter. This fault was not being detected before the big bang, so I think there is a very good chance this happened at that time. Either it caused the failure or was a result of it. This was not the PCB's I had made I will get back onto getting one of them up and running, they can't fail this way as the MOSFET screw holes are no covered by PCB as shown here.  There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| oztules Guru Joined: 26/07/2007 Location: AustraliaPosts: 1686 |
Arcing plus mosfets means death as a general rule.... so I think you have the reason it failed right there. New boards look the part too. .........oztules Village idiot...or... just another hack out of his depth |
||||
| Page 6 of 9 |
|||||
 Print this page |
