Accurate Tools Inverters

[Madscientist267]

Nevermind - got it confused with another one -

When I first looked at the page, it was sitting just over $100 - http://cgi.ebay.ca/NEW-ADVANCED-POWER-INVERTER-6000-12000-WATT-DC-AC-/290529561311?pt=LH_DefaultDomain_0&hash=item43a4e94adf#ht_3914wt_892

It's closed now anyway... Sorry for the confusion/false hopes.

On the other hand... LOL maybe those suffer from the same issues? hahaha

Eh, even at 200 bucks a pop for the actual "Accurate Tools" units (+ shipping), you can probably get an idea of what you want to know without too much trouble.

And if you want to see their mode of failure, its apparently really easy to reproduce... faithfully connect line neutral to battery ground and wait for the smoke to clear, and voila!  

You're right, you'd probably be better off getting them 'pre-fried', but I don't know how easy that will be... I'm sure that almost anywhere the 'PO' motive closely reflects what we've seen here, and people are going to either want their money back or get them replaced. Neither of which renders a dead box hanging around ready for donation, unless (like you said) you can 'con the con' into giving you a few that are in ready-for-you shape for no more than shipping...

They might take that as a threat of some kind, however, and hang up on you before you get very far. Meaning that if they think you're trying to one-up them somehow, they might grow a similar response to what happened on here with that whole 'accusation' thing... ? Fixing these things so they don't cook when connected this way could mean the end of their scam...   Sad, but unfortunately a possibility.

One way or another, I wish you luck with this. I definitely can't doubt your abilities, in fact from what I'm reading in your posts, you might really be on to something here and seem to have at least some kind of plan formulated. Just seems like so much work...  

Not that I haven't done the same kind of thing... I can't talk - I stopped counting how many times I've irritated my wife (as has she) by fixing something that shouldn't have been broken to start with... hahaha

As far as the 'NEC approvable' thing, that's basically what I meant. Able to be connected in the way that the NEC would require. Not necessarily a certification.

And last but not least - "what is just as amusing to me is the face after I have done what was thought improbable" - Who you kiddin'? LOL That's my motive half the time!  

Steve

[ghurd]

What I dare ask is wrong...?

Ignoring the stackable part, most of the no-name generic Chinese units are fairly similar inside.
This may give an idea of the kind of issues likely to crop up in that type of unit...
http://ludens.cl/Electron/chinverter/chinverter.html

G-

[CompuTutor]

Yeah, thats an old one that I've seen long ago (along with a few others),
but thanks for posting that for others so they see the task at hand.

[Madscientist267]

the kind of issues likely to crop up

LMAO Is he kidding?

What, are there competitions for this kind of thing in foreign lands or something? Its almost like a sport or something. 

Buy crappy equipment from the webernets and improve upon them at home... Hmm... Do I see the beginnings of a new market up for cornering?

The banner at the electronics store would read like this, right:

Pseudo-kits. The pre-assembled electronic kits for those DIY-debuggers!

We sell it to you AFU so you also have to figure out what's wrong before you can use it!

  Hahahaha

Steve

[CompuTutor]

This isn't limited to the home consumer market, for about two decades I worked as a marine electronic technician licensed for up to two hundred foot vessels for about thirty seven equipment manufacturing companies. It was hell. This happens on radar, sonar, satellite navigation and mapping equipment. Even auto-pilots.

Most of the time, my infield repairs were a mix of replacing parts, and determining why they failed. Then the factory would spit-out a "mod" to try on the next (similar) failure with the intention of it being a required in field service update later. So no, it isn't a race to get things to market just in the consumer segment. It happens in the commercial segment too...

An inverter, is just and inverter...
But an application-specific MCU (Microprocessor Control Unit) that can be networked master/slave feeding and inverter, that is what has my interest on this item.

Even if the rest gets trash-canned, the concept of building-block inverter networks sure makes sense if properly implemented.

All your eggs in one basket, even with the best of inverters out there, is still a go/no-go gamble...

I'd rather have a nework of smaller ones that only energize upon need to keep feeding smaller loads from being wasteful, and are seperate so a failure of one is only a failure one just that one.

[Madscientist267]

If the inverters are as poorly made as it sounds, there probably is little in terms of microprocessors involved, even for synchronization.

Many of the chips I've dinked with for switch mode power supplies have a 'sync' pin, which is really all that's needed to slave one unit from another.

From there it's just a matter of paralleling the inputs to get the source of the power, and the outputs to double the current available to the loads.

Not really rocket science; fly-by-night is going to go as cheap as they can get away with so they maximize profit margin before they 'get got'.

Steve

[frackers]

I see no reason at all why there should be a connection from the 'neutral' line back to the battery -ve (or +ve) terminal.

All the modified sine wave inverters I have seen work the same way in that they have a standard DC to DC converter circuit that is replicated to achieve the power requirements of the inverter. This conversion is done though a high frequency transformer - implicitly isolating one side from the other provided the simple approach of an opto-isolator in the feedback loop is used. The (usually) MOSFET H bridge output stage then switches that DC output to create pseudo AC.

I guess the secret is using that opto in the DC-DC converter blocks to allow the output to float with respect to the input the same way as switched mode wall warts do (and we know how cheap they are!!!)

[CompuTutor]

Your absolutely right frackers,
for the pennies per thousand units produced that they saved by having direct DC feedback loops without Opto's is just unwarranted.

[Madscientist267]

Unwarranted may be an overstatement of sorts... Not to nit pick, but;

I avoid optos in SMPS feedback loops, simply because of the extra complication they bring, when I can get away with it. Not saying they are not necessary in some designs (inverters definitely fit the necessary category), but I can see the 'want' to leave them out whenever possible.

Not that using one is all THAT complex, it just undermines the KISS thought train.

I agree, avoiding them in an inverter is begging for trouble.

Really, in any application where there is a significant difference in potentials between one side and another, isolation in some form should exist. This becomes imparative when at least one of the two potentials is at a lethal level in terms of 'human consumption', and the unit in question is designed to power something that people could be directly interfacing with physically. Inverters most certainly apply to this.

When humans are not intended to be around, isolation may not be so paramount. There are UPS designs that do not isolate the batteries from line potential at all, because a human is not expected to be dinking with the battery while the unit is in operation. Worked on quite a few of them at last place of employment. Didn't affect operation one bit, but as benchies, we had to learn to respect the fact that the SLAs were not isolated from the line.

I/O wiring was as follows:

Case: Earth grounded
Ground: Case bonded, passed through from input
Neutral: No internal ground bond; connects to inverter and passed through from input (non-switched)
Hot: Switched between line and inverter (relay)

Battery negative is tied to HVDC negative potential on the secondary side of the forwarding converter for the inverter.

The relationship between battery negative and the case is unclear when the unit is completely powered down and only checked with a meter, but it will zap the crap out of you if you get caught between either battery terminal and chassis ground with it lit up.

Inspection of the circuit reveals that the only connection between the HVDC side and battery is in the feedback loop for high side voltage regulation.

So you see, it's more common than you might think.

While I'm on the subject, those units run a chip that is capable of syncing (KA3525), and therefore theoretically could be 'stackable'. Hmm.. I'll save that one for another time...

Steve

[joestue]

Unwarranted may be an overstatement of sorts... Not to nit pick, but;

I avoid optos in SMPS feedback loops, simply because of the extra complication they bring, when I can get away with it. Not saying they are not necessary in some designs (inverters definitely fit the necessary category), but I can see the 'want' to leave them out whenever possible.

Not that using one is all THAT complex, it just undermines the KISS thought train.

I agree, avoiding them in an inverter is begging for trouble.

its no trouble at all.

in fact, who gives a $#!+ if the 300 volt bus might overshoot 10 volts by the time the optical isolator turns on?
if you've got a PWM'd sine wave inverter it couldn't even care, the modulation index might drop below 90%  

on the topic of cheap msw inverters running off 12/24/48 volts, why isn't anyone selling a 7 or 9 Level Hbridge based inverter?

it seems to me the situation is either you don't care about the harmonic current and non NEC/UL approved situation, or you spend way too much money on 5-10 khz pwm inverters and isolation transformers.
but then once you get over a few tens of kilowatts then out comes the engineered solutions.

we're at the point where a hybrid car inverter hooked up to an additional EMI filter might be able pass the 4% thd requrirement for 20$ a kilowatt.. but we're still stuck with crappy MSW inverters.. or 1% THD pure sine wave inverters.. no inbetween

[TomW]

Joe;

Keep the language PG13.

Its not that hard.

Otherwise I have to start killing your posts.

Tom

[Ungrounded Lightning Rod]

on the topic of cheap msw inverters running off 12/24/48 volts, why isn't anyone selling a 7 or 9 Level Hbridge based inverter?

Because a high-frequency PWM followed by a low-pass filter does a better job and only requires ONE set of switches?