From New SMA INVERTER SI6.OH-11, battery full 50HZ Jumps to 60Hz

[Bruce S]

SparWeb;
Sorry should've PM'd you before adding you via your previous info.
From a semi-distant point of view and someone who was around when the new forum went live and ALL the flamewars going on then,,,
I was seeing the writing on the wall and wanted to "{hopefully}" nip it in the bud as smoothly as possible.

 

[oztules]

Fairly well put Sparweb.....

The unit does not do what is supported in the literature. That should be enough to rely on.

The concept of tech support pre purchase is while technically correct, is practically useless... that s why we read the product specification sheets, we don't generally need a sales man to lie to us .. when we have got the literature......

No, it appears we have a business model that is designed to be anti competition in nature.
A 10 hz drift is almost criminal, as it will cause problems with a whole raft of equipment, and timing mechanisms that rely on stable and clean frequency environments.... it's inane.

SMA has admitted that this is deliberate, and not a bug they were unaware of.... as such, their advertising blurb should be very clear and up front that they will purposely try to destroy or at the very least disrupt your equipment connected to their SI unit if the loop control is compromised in any way.

To pretend that SMA has no culpability in this destruction is a very long bow to draw.

Reliance on the fact that non-SMA gear is attached in a normal manner, that will preclude this from being a "closed" system, is not good enough. To drive the power into the ultra dirty region as a response to the breaking of the closed loop, is deliberate... they say as much. Setting the drift default more than 1 or 2 hz is clearly an attempt to stop third party equipment from being used.

So CM is caught up in SMA's attempt to stifle competition, and are trying to tie up the market... this is not a technical issue, it is a sales issue... the same as say Microsoft  or any other OS stopping competing browsers from working properly on their OS  or hobbling them and allowing only their software to work properly or faster in their environment.

If SMA designed their software to not go off the planet when it detects that another charge controller that is doing a better job than theirs  ( and most probably do from my experience) then there is no problem. It is a problem deliberately made by the SMA business plan.... not from the technical department I suspect.

There are two courses of action.. refund is the best, as the unit is substandard to requirements. Anything that delivers such woeful frequencies is junk.. no other way of saying it.. if it were of Chinese  origin, we would hear it  far and wide and from the rooftops.....but.... as it comes from what was once a reputable firm, it is now taken as normal by "competent installers" such as Sean... which gives us a glimpse of their competency really.  If 10hz drift is acceptable programming, and if installing systems that can destroy appliances if it looses control of the loop........ hmmm.... I wonder what else is acceptable to these people.

One also wonders how on earth they got a UL listing... did they have it set to sensible limits when they submitted for testing??? how did this get through from a safety testing company with a 10hz drift for no possible reason... or was this modified later on after approval?

Did they tell UL up front about this?... oh and by the way, did we mention we have put in a destructo routine for if the poor dumb user should try to charge their batteries on a cloudy day with a non-SMA compatable appliance.... that should be fine shouldn't it? We could just shut it down, or even ignore it, but this is way more fun.


The alternative is to on sell this to someone else that does not want to use it for anything more than a dumb inverter, or wants to incorporate it into a SMA only environment, and buy something else that does not try to  blow things up deliberately.

CM will now have to tell prospective buyers to not charge their batteries from a battery charger or any other source while the SI is connected to it's grid ties... or it will try to destroy their appliances... that could crimp it a bit I suspect... but be unconscionable to do otherwise.

Secondly, just use work arounds.... your batteries will probably be better off.

Thats it in a nut shell. The thing is designed to not work properly out of the box if it does not have control of the loop... this is on purpose, and deliberately designed to cause the maximum amount of disruption so it can't be ignored..... buyer beware.... SMA is not the slightest bit interested in solving this, as they created this scenario in the first place... it could be solved in seconds... but they will make sure it won't be.

As Mary points out, they are not trying too hard to let you know about this foible either... until it is too late of course.. then they can peddle out the "should have consulted our pre sales techs first " routine.

All good reasons to buy something else really.

Thats it from me CM.

This thing is not designed to play nicely with non frequency controlled smart chargers ( read SMA stuff). They want a start to finish "plug and play"  system with no surprises. This makes it simpler for them and their "competent" installers to get it right and manage.

The only thing they have done wrong, is make their SI unusable in conjunction with other systems... and they have done that dangerously wrong. If they wanted to press the point, they should just shut down if they loose control of the loop it they are that concerned..... not cunningly kill of the householders appliances.

They should just refund the money and move along....



.....oztules

ps Sean, thats just mealy mouth drivel designed to mislead ... like fine print, ..........and primitive methods seem to work far better  and more reliably than SMA software and support... at least they don't try to destroy your appliances like the carefully crafted SMA software does while trying to charge batteries. To smart by half.

[sean_ork]

Would it not be more productive for CM if this thread didn't simply consist of a stream of opinions, as per the previous post ?

I've tried to highlight what I suspect to be a questionable feature of the wind input controls.

I've tried to point out the questionable marriage of a dominant control, and the independent wind input controllers.

Technical pre sales, and pre sales design input is always to be recommended as, if correctly contracted, design and performance responsibilities lay with the manufacturer.

Assuming the position of the competent person, and self selecting a product, without a complete and in depth understanding of a products feature set somewhat negates the right of complaint.

I can image that CMs head is spinning, dozens of folks commented on his pre purchase "is this product suited to my needs" request for information. Even more have commented during its installation and commissioning, and it appears the entire internet is awash with the post installation saga.

Let's hear what SMA said to CM pre the purchase, perhaps then CM can opt which route to take.

I suspect he is currently bidding on a few hefty dumb UPSs.

[oztules]

Well it appears he  has a dumb ups already.... its' playing up at the moment......

Whether it can be trusted to provide clean output in normal and reverse grid tie without the freq control enabled is the question.......can it hold frequency at 50hz +- .1 at most.

We also  don't know what other surprises those naughty programmers have in store for him... so lets see if you can be at least a little bit useful with your vast knowledge of these things and answer that first.... then...

1. How many transformers in the output stage.

2. Are they EI or torroids

3. How many fets on the H bridge output stage and what type. ( irfp___etc)

It is a bit light on power, but we will see what we can do with it.


...................oztules

[sean_ork]

How would the answers to those questions help CM ?

Ripping open a cheap disposable product with the aim of improving its performance might be appealing to some, I suspect CMs wife would not be best pleased if CM is encouraged along that route.

This is getting rather silly.

[oztules]

"Ripping open a cheap disposable product with the aim of improving its performance might be appealing to some, I suspect CMs wife would not be best pleased if CM is encouraged along that route."

So far this expensive dangerous product has been going through CM's wifes appliances and destroying them one by one.. I'm sure she is over the moon with her SMA already.
My wife would not tolerate it at all, and would likely have put the axe through it already, and then told me to fix it up properly.

SMA with their "kill the clients appliances programming" has made this unit a expensive pile of junk...... and we don;t even know if it will run as a normal inverter with grid tie without frequency drift... as you have evaded the question....  If this was driving engineering machinery like mine does, god only knows what would happen with 20% increased rpm, and timing sequences that rely on good power within normal specifications.

Until we know if we can get guaranteed frequency stability, we need to realise that the simple throwing out of ALL the SMA control circuitry will be the simplest and best way to get a steady performance with no hidden agenda's from your sadistic programmers.... they have proved that they have no scruples and are totally untrustworthy. They have no eye for customer safety with that sort of behavior..... so we may be best to get rid of it.

So what do we have left to work with... they may have decent transformers to work with, or they may use cheapies and use programming to get the idle currents down, and use multiple transformers. They may use good fets, and be capable of a few more kw if their transformers are any good.

Lets face it, customer service is non-existent, warranty is not worth the paper it is written on, and the unit itself is dangerous... so whats to lose?.... Thats how it may help.

It may be that the trannies and fets are below par, and the whole lot needs to be sold on to some one else that does not want to go outside SMA environments... but it looks like we are stuck with this one at the moment, and it does not do what we want and destroys Mrs CM's stuff.

I have no doubt, that used as just an inverter, it is probably a reasonable performer... but for the price it is an under performer... I could build 6 for that sort of money... more powerful, higher surge, and clean reliable frequency.... yes it was "rather silly " to buy a SMA from the looks of it... and no doubt CM's wife would have told him so by now...... poor fellow.

I know that SMA can design and build a good inverter, they are not idiots, they are very talented... but their ethics and disregard for the customer, has resulted in a very unsatisfactory unit in this case..... not worth buying from where i sit. Plenty of better stuff out there now, and will be more in future. SMA is probably well aware of the fact the world has caught up to them, and is passing them. So they are going for a niche position... via a turn key system, and hope to capitalise on that "systemic" approach.

In doing so and trying to protect this environment they have resorted to unconscionable practices, and the sooner they change their ways the better for their customers.
From the antics displayed thus far, I would not touch their stuff with a red hot poker. If I had fallen for this I would rip the heart out of it, and replace with better electronics with reliable outcomes

And it is not difficult to do now days.

I hope CM can get rid of this boat anchor without too much loss, and start from scratch with something else.... anything else would be better.

Now... answer the questions Mr "competent installer"

Can it hold frequency with a grid tie connected, and what is the guaranteed drift without trying to control the grid tie with their freq system.
Then 1,2, and 3

...........oztules

[sean_ork]

Now... answer the questions Mr "competent installer"

Perhaps your high post count excludes you from the usual requirement to be polite, maybe thats also an excuse for your aggressive tone, so you will have to forgive give me for not following your route towards a solution.

I've pointed out a number of the overall systems failings, there's been plenty of suggestions as to a workaround. Had note been taken of the many suggestions I suspect no appliance damage would be been caused.

Let's hope we hear from CM soon as this idle debate and bickering between us is pointless.

[Bruce S]

Now... answer the questions Mr "competent installer"

Perhaps your high post count excludes you from the usual requirement to be polite, maybe thats also an excuse for your aggressive tone, so you will have to forgive give me for not following your route towards a solution.

The answer to this remark is NO, no one not MODs not Superduper posters not even owners are exempt from being polite.
I have not yet checked my 15 messages on here, but I'm sure some other MOD has reached out to OZ already.

The bigger differences are that most of us who have been around since the previous incantation of this forum; OZ knows how passionate he is about Corporate  sided equipment  doing as they are supposed. He's got enough knowledge to talk to EEs at EE level and has helped tons of people with not even one request for anything beyond a simple Thank You.
IF it looks like I am coming to his defense, then possibly I am. I had my own run-ins with OZ, but they were because he's trying to make a point  I glossed over enough to start looking stupid.
I've been schooled by OZ more than once over the past decade.

 OZ , I do have to ask to go to the kinder side of posting.

[Bruce S]

With due respect to CM's UPSs, the units he posted as using.
The APCs are neither cheap to purchase nor cheaply built. Their frequency drift is no more than about ±1% and looks beautiful on an O-scope and this is on battery power. True their standby power usage isn't the best, but they do want they are normally intended for.
I personally owned one built for USA. The mains the output is trueRMS,  can fairly easily be changed from 208 to 240AC (60HZ) ( I've this to 3 separate units on different occasions at my previous job).
AND I am currently rebuilding one here at my new job, with the only part hardest to find being the cooling fan  .

   

[sean_ork]

I argree Bruce, they are grand beasts, there's plenty here still being put to good use to keep the lights on when the usually abundant winds are resting. My comment about them being dumb was in comparison to the product causing such a debate on this thread.

There's also plenty of systems here akin to that which CM is creating, which use the same product as the domanent controller, that are working faultlessly. I know they work well when correctly implemented, so I'll not be getting dragged into an anti product or anti big corp debate, others can do the ranting.

I initially posted to point out an untruth, also some concerns with the overall system.

My points have been echoed by many others, across numerous forums.

My last point will be to express a wish that CM takes the advice that is being offered by many, even if he might not agree with it - let common sense and logic prevail.

[Mary B]

Point out to me the exact location in the sales literature that details this!

Would it not be more productive for CM if this thread didn't simply consist of a stream of opinions, as per the previous post ?

I've tried to highlight what I suspect to be a questionable feature of the wind input controls.

I've tried to point out the questionable marriage of a dominant control, and the independent wind input controllers.

Technical pre sales, and pre sales design input is always to be recommended as, if correctly contracted, design and performance responsibilities lay with the manufacturer.

Assuming the position of the competent person, and self selecting a product, without a complete and in depth understanding of a products feature set somewhat negates the right of complaint.

I can image that CMs head is spinning, dozens of folks commented on his pre purchase "is this product suited to my needs" request for information. Even more have commented during its installation and commissioning, and it appears the entire internet is awash with the post installation saga.

Let's hear what SMA said to CM pre the purchase, perhaps then CM can opt which route to take.

I suspect he is currently bidding on a few hefty dumb UPSs.

[oztules]

I'm sorry Bruce and others.

It is impossible for me to see any upside or common sense in anything that has been mooted by  Sean. He seems full of self interest, rather than useful advice... particularly when he considers himself competent in all things SMA

If he is so knowledgeable, then why not provide the codes to allow CM to reprogram the defaults for the frequency shift to safer limits.... no money in that for SMA?
So the only advice is to spend more money and tie yourself totally to SMA systems to stop them behaving badly... very dangerous looking at what they have done so far.

If CM falls for the SMA dc controller as well, then it is not just more good money after bad, it  would allow the SMA to control the battery bank.... good luck with that.

I would much prefer to see the Midnight and tristar look after the bank. SMA may build inverters that can work ok, but they have not glistened with rectitude when it comes to battery management over on this side of the world... quite the opposite in fact.

Even now it would appear that the current setup shows a divergence in what the bank needs. The Midnight and the Tristar are trying to top the bank up, and the SMA is driving the frequency over the edge to try to stop them... they can't all be right can they?.... Which controller would you trust with your most expensive part of the system ( and thats not counting trying to destroy the white goods on purpose thats a separate issue).

So, if you leave your future exclusively to the SMA, SI, you have killed of all the options for future improvement, and will forever be at their whim.

If it were mine and I was certain that I was stuck with it and I could not guarantee that the frequency would not change when grid tied to the sunnyboy.... I'd rip out the SMA controller complete down to just the transformers and fets, spend 70 euro and put in another controller that provides clean power, and seamless grid tie

Reverse grid tie that, and then let the tristar/midnight control the battery bank.... and then work out a sophisticated system to control the grid tie which would not involve frequency adulteration..... I will be doing this soon anyway instead of the interim bang bang setup i have at the moment.... which is a bit adumbrate, but works better than the SI can do

CM has built a  well thought out system, so I naturally have affinity for folks that" can and do".  I have no time for people or companies or their supporters that aim to do harm for any reason. It is a matter of trust... and SMA have demonstrated that they will happily trash your trust for a few dollars more.

I cant find it in my heart to say anything nice or barely civil to folks like that. I tried and failed... my bad, but i can live with that. I still have some standards left, none of which include taking advantage of folks in trouble. The SI only needed safe set points, and they refuse to provide that...... but we have this other DC controller thing you need to fix the problem that we deliberately created for you.... there are so kind.

I'll leave this alone, and let Sean peddle his magic ....after all, he is the competent installer, and I'm the village idiot.... luck with it CM.


.....................oztules

ps Bruce, Damon  and other Mods, delete anything you feel is in poor taste with my blessing ( not that you need it..... but you know what I mean)

[sean_ork]

It is impossible for me to see any upside or common sense in anything that has been mooted by  Sean. He seems full of self interest, rather than useful advice...

Oz, clearly you are trying to provoke me, perhaps in the hope that I am put on mute - I'm far to long in the tooth to fall for such antics.

Self interest ? - quite the opposite, I've nothing to gain here, nothing what so ever.

I note your comment with regards to offering CM some codes, I'm assuming by that you mean the installer code - if you were fully aware of CMs situation you might understand why that would be a complete waste of bandwidth.

Further, having read your additional comments later in your last post could I very very politely suggest that you spend a little time reacquainting yourself with his systems configuration.

Mary, the text you have quoted does not appear anywhere with the SMA documentation.

It appears that its only me that's required to answer questions, if anyone can be bothered to scan back a little I raised a point that no one appears to think has any bearing on CMs system.

[sean_ork]

Even now it would appear that the current setup shows a divergence in what the bank needs. The Midnight and the Tristar are trying to top the bank up, and the SMA is driving the frequency over the edge to try to stop them... they can't all be right can they?...

I knew we'd agree on something - this is just one of the system configurations significant failings, another has been pointed to in one of my previous posts.

A solution for the open loop nature of this system has been detailed by many many others, elsewhere - fortunately the inverter will not need to be dissected to implement it.

[oztules]

CM... it's time to give you something to lift your spirits in case you end up getting no sense out of SMA... which looks to be the case from here.

In the last 2 days I have built a bigger inverter than that one, and it does not do anything wrong... does not try to blow the household goods up, holds frequency perfectly rock solid, and had minimal voltage drift and a nice tidy sine wave... basically everything SMA pretend to have, but without the dangerous frequency drift........ It has not got the intellectual property of SMA thank goodness.

So for the price of a packet of cigarettes and a dozen beers, we buy the controller from ebay... then we need only a transformer and fet array. I bought the fet board as well for a hundred dollars more... woopie... last of the big spenders here.

I bought a heap of old chinese grid tie inverters that had the torroid transformers in ( galvanically isolated type) They were 3kw aerosharp units.... $70 each.

Pulled them apart, unwound the transformers, and rewound them for this purpose. Epoxied the layers to make it silent in operation... and it is.... scary silent.

The aerosharp boxes are beautiful too s/steel and aluminum... the chinese are building some beautiful stuff now.....

So put together  it looks like this:



Could only find 3700 watts to stuff to plug in ... then run out of two ways...



Waveform @ no load



Waveform with a few kw.. run out of 2 ways as the scope used one of them.



Notice the figures barely change for the voltage and the wave form stays fine too.

It is good as far as I have had load. Maybe tomorrow i will find where it clips.. expect it will be in the 6-8kw range should be where we see something... depends on how good my transformer skills were.

So chin up CM..... SMA dont make clean power, they deliberately pollute it.

So chin up......If you get backed into a corner, there are options we can do, that will put you in a better place probably.

And no Sean we don't agree on anything unfortunately..... to you that quote says "system fault", to me it tells me exactly whats going on, and where it has failed. In two ways.

 I trust the tristar to get it right, and i trust the midnight to get it right.

What that shows us is that the SMA is out of step with it's battery management system, it is trying to do the opposite to the other two units..... my money is with the other units. SMA is held in low esteem over in this remote part of the world for lots of reasons, but the main one seems to be battery management.

So I don't see a system failure as much as I see the SMA out of sync (figuratively) with the algorythms used in the other two.

This then gets expressed as the SMA going silly... so the poor programming of the frequency simply highlights the poor battery  management skills of the SMA or the other two take your pick... I suspect the SMA, you will blame the other two... we don't agree.... what we do know is the SMA tells us it disagrees violently.

You also blame the system configuration. This is valid in so much as the SMA  does not agree with the other two..... if it disagreed and just plain shut up about it, the system would work flawlessly.

The SMA would get the charging wrong, but the other two would correct it... the SMA would continue to develop clean power anyway, and everyone is happy.
Now we change the system as you suggest:

ie If we were to change the system to your configuration with the SMA DC controller, then we are held captive to whatever the SMA thinks is good battery management, so we close the system, but still get moderate frequency drift... but within the 2hz range ( still ridiculous, but not dangerous). The system works perfectly... or does it

The Midnight and Tristar DC controllers being used  disagree with SMA... and so would I, but as they would no longer be needed we would not know that the SMA was mishandling the batteries by the metrics used by the other two... until it becomes too late.... seen it heard it before.....not a SMA problem our excretion has no aroma.

So fixing the system to your own design would likely result in poor battery management, and I suspect most of the world would want the midnight or the morning star to control that.

So that mixed with the pig headed and destructive programming that SMA have displayed so far..... bodes ill for the batteries.

Thats the problem with only looking at a problem as a systemic anomaly only... you  miss the real issues. You need to see the anomalies as sign posts to what the problem really is.... then look to see if it is systemic or can be contained in only a part of it..... or indeed if a part of it is responsible for the system falling down.

If the SMA had the same battery charging algorithms as the other two, they would be all happy.  If the other two were slightly retarded from the SMA rates, were still good.
I know you still cant see it, but there is a common unit that is causing problems that it has no right to do. SMA have taken it upon themselves to be disruptive.... who'd have thought..

So in this case,
1. If the SMA did what it claimed... the system is fine.

2. If the SMA didn't entirely do as claimed, then we can live with that too,  ie...charge when it thinks fit, and not charge if it disagrees... that simple, and exactly what the other two do.

3. The SMA throws a deliberate wobbly if it cant get it's way and behaves badly with dirty power and wild frequency drift.

The system is sound.... the inverter is crap... even I can cobble together a better one than that thing... in 2 days or a remote island in the middle of the roughest water in the world, using a cheap chinese controller that makes clean pure stable power... SMA has so debauched their device it no longer can claim that ... instead it is an appliance killer....mine cost maybe $500AUD that thing probably $5000.... so what do we pay for... intellectual property.... but they look very fragile there, the chinese are killing them.

You need this to fail so you can continue installing SMA... because when the rest of the world cottons on that there is nothing special anymore with SMA........and are 10 times the price............fear not, it  will take long  time. They are well embedded, but this silliness will speed it up for sure.

I might be the village idiot, but I can do as I say.....

So would I strip it down and mod it... if it had decent transformers i would in a heart beat, but your low output for the weight at low duty cycle, tend to make me think they are using EI transformers, which are half the efficiency weight wise than the torroids I use.... and in that case... probably just get rid of it.

My torroid in this one weighs in excess of 34kg... that would need 68kg of traditional transformer for the same power..... so yes beginning to think you have a few transformers in there to cover up the lousy idle currents... a small one for low power use, and switch in the biggie for power runs.   the magnetic domains required to be magnetized each cycle are much lower for a small transformer, so you get nice idle figures.

Mine has a big transformer, so much more magnetizing current... or is there?
No not that much really 30 -35 watts. The torroids use better thinner steel so the eddy currents are much lower for the same mass of steel, and there is only half required, and a much superior magnetic path, so leakage is very very low... hence the small sag.


.........oztules

[sean_ork]

So, in just a few word .....

CM, you've purchased the wrong product.

[electrondady1]

i love this other power forum.
 

[mab]

So, in just a few word .....

CM, you've purchased the wrong product.

Well, I suspect we all agree that CM could have spent him money on something better... 

that's why we're hoping he'll get his money back.

[clockmanFRA]

Very nice Inverter Oztules. Real nice to meet folk that actually do, and make stuff.

I have glimpsed in on your endeavours over the years, but have found some of your processes slightly baffling.

30 odd years ago I did build some some 200w, 12v to 230vac, using torroidal transformers, Inverters, 50HZ Quartz oscillator circuits etc,  for a particular customer. The output required was on 20w so I could fix frequency distortion and drift under load. But with the onset of SMD devices and multi multi layer bourds, my PCB board manufacturing stuff lays dormant in my chemical rooms.

3 years ago Eric, (Eric has the brain of a Planet), and I put together this 32 battery or cell, BMS device that used an Ardino and SD card to monitor each 12v battery, we also put a LVD and “once bat at Float then Divert circuit”. So I am reasonable with PCB, Voltage reg design building and the like, but Arduino Code,  hmm yea, just don't ask me to write it.



I have worked with some fantastic barmy Guys like my mate Merv, now retired but was head of R&D GEC Plessey, as I can mechanically manupliate at the 4 Micron level, Watch designer and maker you see. Shame he's retired as he would have found me some decent torroidals somewhere in Europe.

At the moment I have been working with this oldish Eltek, variable up to 100a, 56v battery charger, real nice bit of gear as a stand by charger unit for my 48v 1100ah at c10 batteries. It was given to me and the other bits are second hand etc, so no money.



Inside the SI is this lot with the upper section the Caps. The transformer sits behind this PCB ally bath unit, and is encapsulated. I think its just one Torriodal but then the whole unit is about 50kgs so maybe a charger transformer as well but I would really need to look see. But the old used SB3800 is definitely a Torridal as I really looked……whistle…



Regards the SI, I have asked for my money back, and now through the ‘European Consumer Centre' forwarded by 'UK Trading Standards'. And as I cover 3 EEC countries the appropriate Authorities in each is now dealing with the matter. So at present I am reluctant to mess with the SI, but at least it should have a 48vdc 230vac wound transformer in it. My old 3.7kW APC UPS was pulling 160watts or so.

My Future Inverter.
I will need some thing around the 6 to 8kW output, but ultimate for our community project, holiday cottages, around the 11kW range, so can we parallel up in the future ?.
Not sure how I would control the other Off Grid set GTI's and they are up to 150meters/400ft away from where the main inverter and batteries are, so not keen on laying more cable as the Mrs will kill me, I promised her this year, “I shalt not dig any more cable trenches”.

So I would like some way to AC Couple the new Inverter without the need for doubling my battery Bank.?

A Bit of Blue sky thinking……. I note that some times folk on other forums complain about their GTI PV inverters shutting down as the main utilities Grid has gone a few volts high. It seems that with this PV on roofs on housing estates, and folk on that housing estate NOT using all the generated power, then the Grid rises a few volts and shuts down some GTI's. I wonder if we could use that way as I have GRIDGUARD codes to make the Used/second hand SB GTI's sensitive to voltage control at each installation after voltage drop etc.
Just a thought and a bit safer that messing with the HZ frequency.?

In Conclusion I would be willing to have a bash at rewinding and setting up the drive PCB's etc, I have full machine shops and clean rooms.

Now where do we start I feel sure others here would want to watch this adventure.?     

[sean_ork]

Well, I suspect we all agree that CM could have spent him money on something better... 

that's why we're hoping he'll get his money back.

It doesn't need to be a better one, just the right one for his existing system.

Perhaps one that speaks to his battery charge system, in a manner understood by both.

Its a simple message.

[oztules]

CM,
You easily have the right stuff to make whatever you turn your mind too.. shame you are dealing with really strange people there at SMA. I will never understand their intransigence or their programming snafu... but they are what they are.

Inverters are essentially simple things, and there are lots of ways to get there.

Probably the most important part is the transformer, as it will dictate how things go at high power. EI transformers will work fine, but will cost in both extra weight, and magnetizing current, and sag.

Look for installers of solar systems, and see if they have dead galvanic  isolated grid tie inverters laying around. In this country the Govt made stupid moves and encouraged fly by night installers to do grid ties .. free money attracts all types, but the upshot was that one company "inspire" bought in container loads of inverters from china, and promptly went bust....this led to a glut of brand new inverters on ebay for as little as 60 dollars each for very nicely built units.... 30-40 kgs each.

So I bought a pallet of those, and found a pallet of aero sharps at the same place.... nirvana. You could not get just the enclosure for that money... and you get beautiful transformers to match... wonderful stuff... anyway moving right along.

I have started a series ( of 1 so far...) on how to build very powerful inverters for next to nothing, and the first part is here: http://www.anotherpower.com/board/index.php/topic,1044.0.html

and a story of how I got to play with these things is here lots of info in there.... somewhere.. http://www.anotherpower.com/board/index.php/topic,902.0.html

It will explain the rest.

Powerjack have the worst name in the world for sure, and most of there stuff is regrettable, and their claims are outlandish, but somewhere along the way they have stumbled on a truly amazing low frequency power card and controller. I don't know where it really came from, as there are a lot of chinese inverters made using these as the heart... but after buying one out of curiosity.. I realised they had a corker.... and they were happy to sell me just the cards for $45 for the controller, and $170 for the whole set....for less than $300 you can buy their 15kw set... just add transformer.

Their 15kw set matched with a proper transformer is a game changer. It won't do 15kw with their setup, but with your own transformers, you can certainly pull 15kw for short periods, and I have found out I was running over 8kw for long periods.... yes the 100 amp slow blow fuses I forgot I put in finally blew out.. dufus...

They take the grid ties for reverse grid tie, and I have used up to 6kw going backwards through them to the batt bank.... more than the batts wanted, but testing to find the limits is fun too..... and no... don't know where the upper boundary is... it just kept on going.

So it turns out that for a very modest outlay, you buy the complete heart and fet boards, and if you manage to blow it up, it cost peanuts to replace and go again... that gives us freedom from "the man"... no longer do we worship the inverter gods and pray it does not fail....now we blow it up for the hell of it just testing different scenarios... and I do too.

Do they last and are they reliable.

Well it appears of you don't do weird stuff or try to use the battery charging part without first turning the unit off ( like they say in big red letters) then they are basically unbreakable. Some running for 4 years non-stop in the north of the island here. They have extraordinary surge capacity ( 10hp 3 phase induction motors start no problem in rotary converter mode, industrial welders running 6.5kw seem no problem..... providing the transformer can take it  without sagging.

The transformer will be more like 230/8 or closer to 30v not 48v The big torroids I make are about 2.2v/turn.. so 14 or 15 turns of welding cable for the primary35 to 50mm sq is ok for most work, as drawing more than 160amps starts to push the battery bank more than I like. If you expect more power for longer, then thicker is better (160a is closing in on 8kw cont).

My current unit runs the house including the hot water system It draws a steady 2.5kw for 3 hrs every day between 10am and 4pm then time clock shuts it off till the next day.... during that run time, the inverter will be running minimum  3kw cont up to 8kw for short durations... (just the kettle is 3000 watts) for 3 those hours, and the transformer gets to about 48C. The transformer is key.

From what you say the SMA has torroid, and the low idle means it is much smaller than I use, which is reflected in their lower duty cycle for even 4kw.. still they are fair figures for a little  inverter. It would have been a useful little performer had they not shot themselves in the head.

Idle current can be taken down further by winding more turns on the secondary and the primary, as this lowers the magnetizing current, and consequently the magnetic domains need less coersion to magnetize... but will come at the expense of more copper loss. So we can get the idle down much lower than I have settled for... but I decided half a KWH a day is not worth worrying about, as I have over 10kw of solar... so is replaced in a few minutes when the sun get s going.

I will do another story on the solar controller... this one does over 100A @ 60v .... costs about $20-$30 in parts I suppose.

So if you get away from big ticket items and replace them with home brew items, you get to have total control as well as have a system to fiddle with.

So, I hope you get your refund, and spend the money more wisely next time...though  you weren't to know you had belligerent programmers waiting for you either. It should have been happy times... they know what they are doing... I certainly do not.

I don't care what anyone says, I think your system was adequately designed, and the thing should have worked as you intended. It is not a system fault.

The Chinese have changed the paradigm in the electronics world in so many ways. Every computer I see ( like this dell I'm typing on) now is made in china. USA cant even build their I phones at home... lack of infrastructure to do so apparently.... they were very ordinary, now they change the world ... they seem to have all the latest technology, and make stuff for nothing... amazing... so I take advantage of it and get freedom at the same time.... virtually disposable high power inverters... that last for a long time as well..

As always, have fun, and chin up.
Edit: RE:Charger transformer SMA...... No....same transformer. The synchronous switching in the H bridge that drives the transformer naturally lends itself to reverse running... ie when the fets channel turn on, it is after all just a N channel substrate... so power runs both directions... so if we put more voltage onto the secondary than was being generated in the first place, the current will effectively run backwards from the mains to the batteries. This is how the grid tie charges your batteries as well... but instead of the inverter syncing to your source, this source syncs to your inverter... so no second transformer is required.


................oztules
Gosh I rabbit on too much....

[clockmanFRA]

That's a brilliant post oztules, thanks for sharing.

And No you don't "rabbit on to much" just good to hear some one passionate about what they are doing.

I will now look at those links, and see if I can purchase some of those 'Magic cards' you mention. (if I recall you had some somewhere for sale?)

Funny that they may come from Aus, when my tracker circuits come from Dave in Cowra, NSW.
Its a small world really.

[Mary B]

WRONG I copied that text directly off SMA's UK website!

It is impossible for me to see any upside or common sense in anything that has been mooted by  Sean. He seems full of self interest, rather than useful advice...

Oz, clearly you are trying to provoke me, perhaps in the hope that I am put on mute - I'm far to long in the tooth to fall for such antics.

Self interest ? - quite the opposite, I've nothing to gain here, nothing what so ever.

I note your comment with regards to offering CM some codes, I'm assuming by that you mean the installer code - if you were fully aware of CMs situation you might understand why that would be a complete waste of bandwidth.

Further, having read your additional comments later in your last post could I very very politely suggest that you spend a little time reacquainting yourself with his systems configuration.

Mary, the text you have quoted does not appear anywhere with the SMA documentation.

It appears that its only me that's required to answer questions, if anyone can be bothered to scan back a little I raised a point that no one appears to think has any bearing on CMs system.

[sean_ork]

You might want to copy and paste from the SMA site again - the only text you've quoted was originated by me.

Post #76 relates ?

WRONG I copied that text directly off SMA's UK website!

Mary, the text you have quoted does not appear anywhere with the SMA documentation.

[oztules]

No Clockman, I never have anything for sale... just don't do that...... and secondly getting stuff off this island is not simple.

 The ship ( use the term very loosely) may not come here because of the weather, and the planes have problems with weather too.... nautically it can be a rough neighborhood.

In that post i refer to there is a english fellow under the monica of off the wal l( OTW)that does buy and sell power jack stuff.... I think.... but I know he also liases with powerjack and helps sort out warranty issues.

Unlike SMA, they seem incredibly approachable, and actually implemented a few things that were in that thread... so their bigger inverters now use only .5kwh/day idle, where before they used 5kwh hrs/day, they no longer need a zenner put across a certain resistor to stop them shutting down from o/voltage from pwm chargers, and a few other tweaks...

 I was amazed they listened, but it looks like they are genuine in their quest to improve their products... and thats in direct comparison to SMA.. who really need to do improvements as we now know for certain....

The other interesting thing they seem to have done on the last 15kw power cards they sent, is the fet banks can be replaced with a screw driver only... no solder required. Each if the four H bridges are independently screwed to the inputs and the heat sinks.....so if you do blow it up, you don't need to buy another power card, just 4 fet boards.... interesting..

So my original tale of woe I used to get them to sell me the cards ( isolation and remote location redundancy issues) seem to have rang a bell with them.

Their claimed figures are barefaced lies....well, maybe not lies as much a stretching the assumed truth.... Like SMA failing to disclose, they also fail to mention their idea of continuous is 12 seconds... so their 15kw continuous unit means 12 seconds @ 15kw, and really about 5kw and below really continuous continuous, and 7-9kw short term continuous ( 30 mins etc). With your own big transformer, then 9kw continuous is possible I suspect.

Their peak figures must be fairly close ( have not empirically tested them for that), so they claim 60kw peaks, and from what I have seen.... and how i have seen them start massive loads.. I tend to believe them. When you see a 10hp induction motor near leap off the ground at start up.... you can't help but be impressed, and as the sparks die down as the start relay  drops out  for the 1500uf 500v capacitors that give it's extra phase the extra grunt to get it going.... I actually marvel at the little fets doing it... simply amazed.

They soft start too.... they need to... torroids well made will use up to  60 times their magnetising current for the first few cycles.. so that would blow it all to hell... but they start smoothly. When I blew the 100 amp fuse, and started it again, I had forgot to release the load, and it started up normally, no sign of stress driving a straight 8.5kw load made up of hot water service,3wk kettle, fridges, freezers and all the usual house stuff.... yes I'm impressed with these things.

So they are probably not the best in the world, but gee they put a lot of others to shame. One fellow over here uses it in preference to his $10000 dollar name brand unit.... because it drives all the loads better... more particularly his wifes front loader washing machine... he is happy and sad at the same time.

Their quality control is absolutely all over the place. It is as if they have a midnight shift that cobble together their own stuff and sell it for nothing on ebay.... you never know what you will wind up with. Some will have lower transformer ratings but fan cooling to compensate... some less transformers ( missing?) and more fans... some are full of the real stuff, ie their 5kw transformers and the right number... but the cards seem to be all the same...

So moral of the story expect little and be surprised one way or the other. hey they are still easily worth it... eg a chap down on the south of the island got the bid for $420 AUD for a 15kw 48v unit shipped to here as well. When I heard the price I wondered if it would be an empty box....Have not had the chance to see whats in it, but it is huge, will have the 15kw cards and whatever transformers they decided to put in. He tells me its a meter long and weighs a ton.. so thats promising. It should have three 5kw ones ( no they are not really 5kw, maybe 2.5 cont).. but for 400dollars it is hard to believe. I will see it when I'm down that way.

Buying a cheap unit on ebay ( any 15kw unit no matter what they put inside will probably be more powerful in every way than the SMA was.....).It may be the best way for you to go to get your feet wet. You know enough to give it a flogging, and then because all the connections are in front of you, you can plan on making the big one, and use that for backup.. if you can get a 15kw unit for that price.... They are usually 1000 to 2000 buy it now..... why the huge range....one can only guess.... but the auction ones can be gotten cheaply.. insanely so.

No, I don't get any money or special deals from PJ, I buy all my needs from ebay just like anyone else... but if it can help remote living people like me to have real power without the head locks that the bigger companies try to do ...... I have done a good thing.
Bought my last three from here;http://www.ebay.com.au/itm/121372307145?_trksid=p2057872.m2749.l2649&var=420314377003&ssPageName=STRK%3AMEBIDX%3AIT

Even the locals that are going this way, I make them buy all the stuff directly, I don't want a cut. Don't need it, don't want it. The one exception is a very good friend who has an island... population 1 person and a heap of sheep.
So I have made him this... for free



It will power the island, and the extra solar will power the desalination plant we are building.. 10kw solar, home brew controller, and this inverter...and that one i finished yesterday as a backup... he can throw away the candles now.


.................oztules

[Bruce S]

Are they soy candles or coconut oil candles ?
                                   

[camillitech]

Golly gosh chaps, what a stushie http://www.thefreedictionary.com/stooshie As this thread (and ones elsewhere) has quite clearly indicated (incorrectly)  that SMA inverters can start fires and is clearly heading for some kind of litigation. Perhaps it would help Clockman's case if someone removed the photograph of his SI.6.OH inverter with seriously undersized battery cables. Two times 16mm square if I'm not mistaken instead of the minimum one times 50mm square as CLEARLY stated in the manual !!!! Would not want to compromise his solid case of it being installed correctly.

Cheers, Paul

PS, you can remove my post afterwards 

[clockmanFRA]

Hi camillitech,

Those cables were updated to 50mm/2 soon after that photo, and before the machine went active.

So all installed as per the SMA Published technical Instructions.

[clockmanFRA]

Thanks oztules,

I have ordered a set, hopefully they will get through French Customs?  and actively seeking torroids transformers.

Thanks for your encouragement.

[camillitech]

Hi camillitech,

Those cables were updated to 50mm/2 soon after that photo, and before the machine went active.

So all installed as per the SMA Published technical Instructions.

That wasn't the point CM, the point was that you've posted a picture for all the world to see of your inverter with out of spec cables fitted and you're in the middle of a rammy http://www.thefreedictionary.com/rammy with SMA. It only needs one of their employees to look at this thread, and judging by the rumpus it's not unlikely. Then the corporate lawyer will say "that man did not read the instructions, he cannot be competent, case closed" and you could be left 'holding the baby' (the big red one outside your shed) so to speak. I'm only looking after your interests.

Hope you get it sorted soon.

Cheers Paul

[DamonHD]

I am not a lawyer but that is IMHO simply nonsense.

CM's contention can simply be that the equipment is not fit for purpose since in a fairly common/normal arrangement it needlessly destroys batteries and appliances, in his opinion.  Indeed he might be of the opinion that the destructive behaviour could be construed as deliberately anti-competitive, which could bring SMA into a whole new world of pain with the EU competition authorities.

The fact that there is a photo of a different arrangement here is not relevant in the slightest.

I do wish people would stop trying to scare either CM or Fieldlines into taking down the threads.  I don't see many wild accusations or libels; we are discussing and expressing opinions on some apparently bad engineering and post-sales support from an otherwise respected international brand.  I'd like to hope that this is not typical SMA equipment or support behaviour, as reported.

Maybe SMA would like to deter people from buying SI units for domestic use at all, I don't know.

And people do use systems in ways not intended and then complain when it behaves strangely; I've had it happen to things that's I have put together, used orders of magnitude beyond the original design spec and I'm frankly amazed that it worked at all!

Rgds

Damon

[clockmanFRA]

Here we are.

[sean_ork]

Good luck CM, I genuinely wish you well with your endeavours.