digital electricity

[StillTrying]

Someone else from 3DFS is explaining the electricky stuff now.

[madires]

The name sounds familiar: https://en.wikipedia.org/wiki/Trip_Tucker

Have they moved to fictive characters?

[Cerebus]

Reminds me of our late cat, Dillon, who went by the stage name Trip Hazard.

[james_s]

Someone else from 3DFS is explaining the electricky stuff now.

Tripp Tucker? Isn't that a character in Enterprise?

[StillTrying]

That might be why they explain everything in Klingonese.

forums.mikeholt.com/showthread.php?t=187471

[james_s]

Now that I think about it, it does sound very much like some of the technobabble in Star Trek.

[Cyberdragon]

If we can phase split the ion beams to regulate the plasma flow...it should work right?

[madires]

In this case there's some nice edutainment about discussions involving Star Trek and Mr. Spock: https://www.youtube.com/playlist?list=PLMpofmkxKHBJfta_JzekLbWGHUSLUJoLt

[StillTrying]

3DFS still providing amusement.

3DFS is the only technology that can save the world, lucky they're on the ball...

Q: What are the 26 parameters being measured/derived within the SDE system?

A: We do not maintain a list of the 26 parameters that are being measured/derived.

... or perhaps not.

techtalk.3dfs.com/t/what-are-the-26-parameters-being-measured-derived-within-the-sde-system/60

[sibeen]

WOW...just WOW..

How did I miss this thread. 30 years I've been working in power quality and now all my prayers are answered.

This tidbit did make me giggle:

*Electricity distribution is most efficient when the laws of math and physics are properly applied. This is what this technology does.*

[Brumby]

*Electricity distribution is most efficient when the laws of math and physics are properly applied. This is what this technology does.*

You know, that sounds a LOT like "Engineering" to me.

It seems curious, though, that they want us to believe they have the ability to implement what they are saying - but seem to be utterly incapable of being able to properly explain it.

[Dubbie]

*Electricity distribution is most efficient when the laws of math and physics are properly applied. This is what this technology does.*

You know, that sounds a LOT like "Engineering" to me.

It seems curious, though, that they want us to believe they have the ability to implement what they are saying - but seem to be utterly incapable of being able to properly explain it.

They are explaining it Brumby, it's just that all of us here have such tiny and primitive brains, we can't comprehend it. It's like explaining calculus to a toddler!
You need to get on their level first, and then it will all become clear how a fast ADC and a proprietary algorithm® can magically make hundreds of amps teleport between phases.

[Brumby]

You need to get on their level first...

Sorry - I'm not after a lobotomy.

[madires]

3DFS have created a self-induced filter bubble with alternative facts and I assume they would argue that non-believers are those in the filter bubble. Anyhow, the lack of reasonable technical details and proper datasheets clearly shows that they have no clue about their own technology. Presumably it's just a more sophisticated version of the Chinese energy saver plugs with a 2.2 or 3.3µF cap between live and neutral plus an indicator LED powered by a capacitive dropper. Maybe they should help Microsoft to reduce the power consumption of the AI cloud platform required to determine if a system running Win10 could be rebooted when applying updates.

[StillTrying]

WOW...just WOW..
How did I miss this thread. 30 years I've been working in power quality and now all my prayers are answered.

It was hard work, but good fun in Real Time.

Some of my favorites have always been the high speed measurements.
24-bit resolution at MHz frequencies, done through the same thin wires that supply the 3phase current correction.
It might be just 8-bit resolution per phase and someone's confused?

And of course they're claiming to control voltage and current at the same time all over the place.

[james_s]

I don't see why people are even trying to make sense out of bullshit technobabble. The only purpose is to attract investors/customers who don't know any better, it's never going to make sense to an engineer because it's just that, babble.

[krish2487]

I am extremely sorry to dig up an old and quite interesting thread...
( well, not really.. ) :-P
I spent quite a substantial time trying to go through the entire thread and I do have a couple of genuine questions.
Not just for the 3DFS folks but actually the more experienced EEs here.


1. How you can force balance a three phase load if a phase does not require as much current as the other 2 phases??
i.e How can you "allow" less current to a phase and "push" more current through the others in order to balance it??
AFAIK, the only way you can balance current equally in all the three phases is to increase the load in the less current consuming phases in order to match the higher current consumption phase...


2. Obviously, there are exceptions like SVC which are designed to run in parallel to a inductive load to correct the PF, but how does one correct current waveforms without having an active element "in series" with the supply and not parallel?


Again, I am not trolling or trying to sound like  a troublemaker..  Its a genuine question.. Either that or my education has been a massive waste of money..

[dmills]

1 You basically have three half bridge sort of arrangements one connected to each phase, feeding inductors, and hence onto a honking great DC bus with plenty of cap.

What you do is drive the switches on the least loaded phase to run as a boost converter transferring power onto the DC link bus, in the simple case with a sinusoidal current waveform, (more on that later), then you drive the other two sets of bridge switches to drive current onto the more heavily loaded phases so as to reduce the load on those phases that the panel presents to upstream. Obviously you can adjust phase angles to improve power factor to taste.

2: Remember the algebraic sum of currents at a node is zero?
If the load is drawing a mess of harmonic current and you want to avoid it heading upstream, you just draw power as a sinusoidal load and put it back as harmonic current in anti phase to the load harmonics.

Now your controls can monitor current on the upstream supply cables (Current transformers) and bus voltage waveform from which computing what waveform must be injected to correct whatever the problem is is a simple matter of an FFT and some trig, and contrary to the fluffer it does not even have to be all that quick, I mean the IGBT pucks are only usually good for a few kHz on a good day, about the only thing faster switches buys you is smaller magnetics.

Sometimes you are going to be correcting power factor, clearly trivial, sometimes moving load from one phase to another thru your energy storage caps, sometimes cancelling current harmonics, or trying to reduce voltage harmonics (Really the same thing but sensing a different input).

The maths is messy, mostly because the parasitics are not negligible, but it is at the end of the day only maths.

Clearly such an arrangement has no actually dissipative elements apart from the parasitic ones so it can be quite efficient, but of course never 100%, and the cables need to be sized for the current, at 200kVA and up that can become significant.

This is all standard, off the shelf tech made by plenty of the usual suspects making real, well engineered gear with real datasheets.

[krish2487]

Thank you. That does clear up a lot of my questions. :-)


The second reply was something interesting for me to learn.. :-)


I was under the impression that we needed active switching elements for improving power quality.TIL moment for me .. :-)
What are such equipment called?? and can you kindly link in a sample datasheet or vendor.. It looks quite interesting.. Maybe some reading when I am bored..

[StillTrying]

Surely to do it properly you'd need a load of 24 bit 200MHz error free ADCs, over 12 FPGAa, a whole new operating system, new computing methods, new formulas, internet connection, app, encryption, clouds, AI, machine learning, water, oil, and fan cooling, and a load of bonkers efficiency claims but without any actual real results or measurements of any power or money saving.

[dmills]

You do need active elements, they just don't need to be in SERIES with the load.

In post #98 I posted a bit of a datasheet for a fairly typical box from a grown up manufacturer.

FPGAs are helpful, mostly as a way to reduce the glue logic, but really any decent modern microprocessor with a brace of counter/timer peripherals and some comparators can probably do this, the maths is a bit gnarly and deciding what to prioritise when you are running out of capacity has fun engineering tradeoffs, but there is no magic.

Regards, Dan.

[krish2487]

Got it !! 


As I mentioned earlier... A Capacitor bank for large inductive loads and even boost transistor for PFC are examples of how elements (active or passive) can be in parallel to the supply and still improve power quality..

[ogden]

This is all standard, off the shelf tech made by plenty of the usual suspects making real, well engineered gear with real datasheets.

Such as? - Could you provide few pointers to plenty of the usual suspects you are talking about?

[dmills]

The search term to use seems to be "Active harmonic filter" and here are a few examples:

https://www.schneider-electric.co.uk/en/product-range/63426-accusine-pcs%2B/
https://www.eaton.com/EN/Eaton/ProductsServices/Electrical/ProductsandServices/PowerQualityandMonitoring/PowerConditioning/HarmonicCorrectionUnit/index.htm#tabs-1
https://new.abb.com/high-voltage/capacitors/lv/modular-power-quality-solutions/pqactif
https://www.tdk-electronics.tdk.com/en/pqsine_presentation?wt_mc=gaw_pqsine_active%20harmonic%20filter&gclid=EAIaIQobChMI54vMmt-s4gIVVojVCh2BHgZsEAMYASAAEgLgiPD_BwE

There are about a dozen other players, but these guys you have probably heard of.

If I was seriously in the market, any of these companies would set me up with a real applications engineer to talk to, and nothing they claim involves magic. 

Regards, Dan.