Interesting bit regarding PF and micro grids

icarusicarus Solar Expert Posts: 5,436 ✭✭✭✭

Comments

  • EstragonEstragon Registered Users Posts: 4,496 ✭✭✭✭✭
    PF stuff generally makes my head hurt, but if I'm understanding this right, the issue boils down to the utility eating 100% of PF losses, but selling <100% of the power?




    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • BB.BB. Super Moderators, Administrators Posts: 32,800 admin
    Short answer, you are correct Estragon. I could see that utilities could switch from charging kWH to KVAH (or not PF corrected power).

    Just changing that could increase electric bills by 1/0.80 to 1/0.60 times (1.25 to 1.67 times) (the typical PF for a modern home).

    The electrical equipment, more or less, responds to kVA for sizing wiring, transformers, etc... The only thing that responds to power is fuel flow (1 kWH of energy = 2 kVAH of 50% PF loads)... However, the utility has to size everything from the generator to the outlet at your house for the kVA rating (basically actual current flow).

    Between more electronics (power supplies, LED/CFL/etc. lighting, computers, and just about anything else) can be designed to consume 0.90 to 0.95 PF (near "perfect" kWatt~kVA). Pure induction motors (not electronic drives) can have poor power factor, but that can be "fixed" to ~0.9x PF at the source (home/factory/refinery/etc.) or even capacitor banks somewhere else (pole mounted Capacitor banks). Turn them on in summer (water pumping/air conditioning) for very little costs (just banks of capacitors sprinkled around some remote switching).

    Electronic loads create a non-linear wave form (induction motors are sine wave current is shifted with respect to the voltage sine wave) that is typically a non-sine current peak near the voltage peak, and no current at the shoulders of the peak to zero volts. The utilities cannot use a simple capacitor bank to "correct" non-linear current flow. The only choice is to have Power Factor Corrected Power Supplies, or larger wires/transformers/costs for "non-linear" current flow. Or, simply you put a local AC inverter at the load/home (pretty much a UPS--The UPS input is PFC, and it powers loads--Linear or nonlinear... The utility will not see any non-linear loads because you have an AC to AC inverter "in the middle"). Of course, you have to pay for an AC to AC inverter and its losses (5-10%). Or put in a battery bank, and you have a UPS system (possibly with solar/backup genset--Just depends on what you want and how much you can spend).

    My guess, more or less this became a problem because of the design of a typical GT inverter system.

    A standard generator (utility, off grid AC inverter) are "voltage sources". The "voltage source" outputs a "perfect voltage sine wave". And the load takes the current it needs (good or poor Power Factor based on its design and needs).

    GT Inverters (I am guessing a bit here--I am not in the industry) tend to be "Current Sources". The current source simply follows the Voltage (hopefully perfect) sine wave and outputs current in proportion to the sine wave value. Normally, this would work fine when the loads are near 1.0 PF (linear loads). Resistive loads, PFC loads, etc.

    When the utility has problems is because the GT Inverter does not output anything other than a perfect current wave form (1.0 PF, or just mirroring the sine wave voltage from the utility). Newer GT inverters (from what I have seen in their specifications, can be adjusted to output +/- 0.80 PF current wave form... For example, it can delay the output of the current flow by a bit, and help supply the inductive load of a motor. The Watts is Watts, but if the output current is delayed wrt Voltage (inductive load, current lagging), the GT inverter can, "for free", output a lagging current and support the Utility's energy transmission (the "lagging" current "corrects" the PF locally, just between the motor and the breaker panel). The downside is that the electronics, transformer, internal wiring for the GT Inverter has to be 1/0.80 "heavier" (1.25x) -- So the system is "more expensive" but does not generate one more Watt of power (25% heavier internals because of the "poor power factor" output to "help" the utilty kW/kVA).

    Some older hybrid inverters (older Xantrex SW series??) where voltage sources that had to "sync" with the input voltage to feedback energy the grid... But because they are a voltage source "in parallel" with the Utility voltage source, they both share poor PF current loads. But these systems are probably more expensive to build, and perhaps not as safe (if utility power "goes away", the SW would continue to supply Voltage and Current -- Figuring out when the utility power goes away is not trivial in this setup).

    I could see (I guess) a GT inverter with a current probe (transformer) at the utility meter measure the actual current flow and voltage and output current to "support" poor PF loads (basically use the GT Inverter output to "correct" the poor current waveform at the home/factory/etc.).

    But something like this would only make economic sense if charging went from kWH to KVAH charges... Then "fixing" power power factor can be done to save money.

    "Fixing" poor Power Factor is done today at major energy users (like refineries). The measure the PF of their (mostly) inductive loads and add capacitors to "correct" the PF before it hits the utilty meter.

    For some larger commercial users, more or less, there is a "correction factor" added to the energy bill. The meter measures the average PF (like worse case PF for 15 minutes out of the month), and simply multiply the bill by 1/PF. So, for the large energy consumer, they save as much money by using more efficient loads, as they do by using PFC loads.

    A lot of guessing and just enough information over the years to be dangerous... And each utility has their own billing policies--Which are also regulated by various political entities (state public utility commissions, etc.). But I think it is close enough for this discussion to understand the issues.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bill von novakbill von novak Solar Expert Posts: 891 ✭✭✭✭
    BB. said:

    But something like this would only make economic sense if charging went from kWH to KVAH charges... Then "fixing" power power factor can be done to save money.
    Or if utilities start paying customers for ancillary services, as they do now to larger generators.
  • icarusicarus Solar Expert Posts: 5,436 ✭✭✭✭
    Estragon said...

    “PF stuff generally makes my head hurt”

    I couldn’t agree more!  Tony
  • jonrjonr Solar Expert Posts: 1,386 ✭✭✭✭
    edited November 2019 #6
    GT inverters (from what I have seen in their specifications), can be adjusted to output +/- 0.80 PF current wave form

    Like capacitors, this will help fix displacement power factor,  but not non-linear.   This is speculation, but I believe it is possible to also fix non-linear, even when it occurs elsewhere.   The GT inverter can measure the voltage waveform.  Any deviation from a perfect sine wave is an indication of non-linear loads (somewhere).    The GT inverter can then inject more power in the "slightly lower than expected" portions of the voltage waveform.   This is good for the utility but puts more strain on the inverter.   

    > simply multiply the bill by 1/PF

    This overpays the utility.  Some costs (eg buying/installing more capacitors) are higher but their fuel (and numerous other) costs aren't.  So the proper adjustment is much less.

    I am available for custom hardware/firmware development

  • EstragonEstragon Registered Users Posts: 4,496 ✭✭✭✭✭
    Presumably if (regulated) utilities are allowed to bill residential for peak KVA or whatever PF adjustment, they'd be directed to make it revenue neutral (lower $/kwh rate x KVA adj.)?

    If the problem is poor PF air conditioners etc, charging for the poor PF makes sense.  A good PF air conditioner may cost a bit more to make, but why bother if the customer doesn't see any value in it?

    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • BB.BB. Super Moderators, Administrators Posts: 32,800 admin
    edited November 2019 #8
    An example of what happened with the "new and improved" lighting... Twisty fluorescent bulbs vs filament bulbs.

    100 Watt filament bulb, 1.0 PF = 100 VA (volts * amps)
    25 Watt equivalent twisty... 0.50 PF = 50 VA

    So, what "we" pay the utility is 1/4 for energy consumed (energy conservation)... Utility only "sees" a 1/2 reduction in current flow and only reduces "their network costs of providing service" by ~1/2.

    In the end, a rough guess is fuel costs for generation are ~25% of utility costs... "The rest" (plant costs, transmission costs, distribution costs, "profit", and even office overheat overhead) is probably 3/4 of cost of bill.

    Poor power factor energy saving devices are not a good thing. it distortiest the cost model. Switching to KVAH billing (similar to what some commercial customers already have paid for decades) at least better represents actual costs of the product (electricity).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • bill von novakbill von novak Solar Expert Posts: 891 ✭✭✭✭


    Poor power factor energy saving devices are not a good thing. it distortiest the cost model. Switching to KVAH billing (similar to what some commercial customers already have paid for decades) at least better represents actual costs of the product (electricity).

    -Bill
    Agreed.  And I would add that smart meters can already measure power factor, so making the change in most markets would be a matter of a software change.
  • icarusicarus Solar Expert Posts: 5,436 ✭✭✭✭
    Bill,

    I have to chuckle at your “typos”.  I often cringe reading things that I have written that have been auto corrected to near gibberish.  Most of the time, people can get the gist so I leave it, occasionally though....

    the editorial comment about office “overheat” is funny.  Pondering the politics of an “overheated”office is fun.  The “distortiest” is pretty good too.  If you had written it to read,  “it is the diesortiest cost model” it would have been perfect. “Distortiest” would be, “the most (possible) distorted model ever!

    keep up the gud wurk.  Tony
  • EstragonEstragon Registered Users Posts: 4,496 ✭✭✭✭✭
    I think some LED is poor PF as well, especially the cheapish ones with integral drivers that fit in edison base sockets.

    "Cost" can be a pretty squishy concept for an undertaking like an electrical utility.  Our grid is supplied mostly with hydro. It has huge (sunk) capital costs for dams and related transmission, but very low marginal cost.  Maybe $10bn for the first watt (Keeyask), next to nothing for the next 700mw.

    Export markets that were supposed to need the power are using cheap gas instead.  Maybe Hydro will subsidize us for getting rid of LEDs and go back to incandescent?  :(
    Off-grid.  
    Main daytime system ~4kw panels into 2xMNClassic150 370ah 48v bank 2xOutback 3548 inverter 120v + 240v autotransformer
    Night system ~1kw panels into 1xMNClassic150 700ah 12v bank morningstar 300w inverter
  • BB.BB. Super Moderators, Administrators Posts: 32,800 admin
    Oh well Tony... Back to my overheated chromebook.

    -Bill  :'(
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • BB.BB. Super Moderators, Administrators Posts: 32,800 admin
    edited November 2019 #13
    Hmmm... Maybe I am having a stroke? I don't think I can blame those on autocorrect.

    -Bill  :/
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
Sign In or Register to comment.