Colombia project updated:

System
System Posts: 2,511 admin
Colombia project updated:

***
Due to the new forum and the fact that we have decided to only use wind and not solar; I have started a new thread on our Colombia project. Yes, I have included all the info necessary for newcomers who did not read about the Colombia project before.
***


Hello everyone.

We are making some interesting and exciting progress on our first project. We have chosen the following water purification system from Danfoss. If you are interested in the details of the water purification system here is the link:

www.aguayuda.org/Alibre/WaterCube.zip

Below are the Water Cube specifications:

Power Supply: 1 phase 220 – 240 V
Power Consumption: 4W per Liter
Maximum Water Production: 4,800 Liters per day

Our goal is to produce 3,000 liters of water a day which would require roughly 12kW per day. We also are either going to use UV or ozone as a final treatment system which requires roughly 400W per day.

That is a total of 12,400W per day.

As some of you know the location of our project is on the north coast of Colombia. It also happens to be an excellent candidate for wind. See the excel sheet below which also shows a comparison of other wind generators we have been looking at and their power output:

www.aguayuda.org/Alibre/WindroseEnergy.xls

After giving solar panels a hard long look, we have decided it is just too expensive and are taking a serious look at the PacWind Delta I generator:

http://www.pacwind.net/Delta-I.pdf

Though the Watercube is designed to run continuously, we realize this would be difficult to do via renewable energy. Therefore, we would like to have an energy system that can provide us with enough energy to produce 3,000 liters of water.

So my question is what would be the best way to setup a system using two Delta I’s and one bank of batteries good for 1 day without wind. Yes, we realize we should size the battery bank larger but batteries are JUST too expensive and the larger the battery bank, the more expensive the costs become.

What would the experts (Jim, Bill, Niel and others) recommend for such a setup? Some questions that come to mind are the following:

If I have two wind generators, how does that work regarding the electrical setup? Are both just connected to the same battery bank?

I am assuming I need two inverters to reach the voltage I need? Someone recommended the following inverter/charger:

http://www.magnumenergy.com/MS-AEseries.htm

Is that sufficient?

PacWind recommends the Tri-Star 60 charge controller.

Battery Bank at 24V or 48V? Besides the difference in wire size, I am still not clear of the advantage between 48V and 24V.

Dump Load is necessary for the excess energy between the two wind generators. What type is recommended here? How many do I need?

Switch/fuses?

If I forgot to provide any other important info to be able to design this system, please let me know.
Also nothing is etched in stone yet regarding this project. We are still looking at other options to run the WaterCube but the Delta seems like a strong candidate due to its simple construction, durability (marine grade) and maintenance-free aspect. As for the tower structure, the monopole with hinges seems to be the easiest setup. There is not enough room for guy-wires or a crane! : )

Look forward to your technical recommendations but don’t forget the environment is harsh so all the components need to be reliable in harsh conditions next to the ocean.

And as usual THANKS for your help and advice!


Simon

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    I am not too sure that I have much to add here--Go 48 VDC if you can get the equipment to run at that voltage--the size of the wiring becomes much less (as the current is halved wrt 24 VDC and the losses are cut to 1/4--(I^2)*R wiring losses). Obviously, the turbine and the charge controllers need to operate at these voltages too (you may have to do 24 VDC if the hardware is not available).

    Attaching two or more charging sources parallel onto the battery bus should work just fine--but check with the turbine/charge controller mfg. to make sure they are OK with it.

    The Inverter you shown also includes a transfer switch and internal battery charger (which operate off of AC line input)--would be a nice backup with fuel driven generator for low wind/turbine down applications.

    Missing the "Hours" from ratings (4W => 4W*hours, 12,400 Watt*Hours per day...). Guessing that the "cube" requires 220/240 VAC 60 (50?) Hz at 1,000 Watts (?).

    Assuming that the wind you listed in the spread sheet is accurate for your site (I thought you posted a link to wind history--but I could not find it). That is one windy site.

    Batteries--You probably want to only go to 50% discharge, so KW*H per day * 1/50% = 2x minimum daily use. A big, well maintained battery bank will probably last ~7 years (big guess). A larger bank may last 10 years (less deep cycling/fewer cycles). Need a monitor that is capable of measuring battery state (percent of capacity) to ensure that batteries are not over discharged.

    I would suggest that you ask the vendors where you plan to purchase the equipment (Arizona, Colombia, etc.) who / what they would recommend--and perhaps to engineer the RE system for you (turnkey for first unit--you could always clone/improve the system later based on your experiences).

    Backup? Generator? Water Tanks (chlorine?)? Low Wind periods?... I am sure that you know more than I about those issues.

    I am sure that Jim and others can give you some very good names in solar inverters and batteries (plus support equipment).

    You asked about dump loads--again, anything can work. But--from the Popular Mechanics Article that I "dumped" on earlier--there is this quote:
    the PacWind team recommended their brand-new, top-of-the-line Delta II turbine. It can produce 10 kw at around 28 mph and has a cut-in wind speed of 6 mph. These turbines don’t need a braking mechanism and can self-start at very low wind speeds—something similar designs in the past could never do.

    And the PacWind site says "-Safe in all wind conditions."... This would seem to imply that these Delta Turbines don't need dump loads--but there is not enough technical detail given on their website to know for sure. A dump load (heater) in a tropical/sea-coastal environment needs to be a pretty tough cookie if it going to be in an exposed location.

    In any case--I would ask/make PacWind (or their local supplier) design your generator system for you... With the supplier name and part numbers of the needed components (controller, tower, dump loads, breakers, etc.). You are going to be spending good money with them and so they should be more than happy to help. There is nothing proprietary in the basic setup.

    You could also call Tri-Star for their recommendations too... Whomever is most helpful--is probably going to be a good supplier down the road when issues arise. Those that don't return your phone call (or give obviously bad advice) are probably suppliers to avoid.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Colombia project updated:

    you had better ask pacwind what it is they will provide you to reduce up to 190vdc down to 48vdc for their turbine. you would not be able to use solar charge controllers for this. let us know what they say for their product and being you would have 2 you would need 2 of their wind charge controllers to go to the batteries. also, ask them of their available dump loads as those are big wind machines powerwise. in fact, lay the whole thing out you wish to do and get their feedback and you can run it past us too as to what they tell you.
    i'd be real curious of the costs on this.
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    You are right Niel, the Tri-Star is only rated to 125VDC max... Generally it is not a good idea to hook up something that can reach 190 volts under no load (miss programmed controller, output fuse pops, etc.).

    Sure makes it seem that a dump load would be required with this Morning Star controller. (noticed there is no voltage/current specification for the wind turbine)...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Colombia project updated:

    "(noticed there is no voltage/current specification for the wind turbine)..."



    Electrical Specifications
    Power Interface:
    D.C. Voltage output 0-190 V Peak
    D.C. Current output 0-20 A Peak
    Electrical connection 2-Wire
    Wire Type 10 AWG
    Electromagnetic Containment FCC Class B
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    I meant the I/V curves (power/load/operational curves) for the alternator... Is it a constant voltage, current, or constant power device... Or is the output voltage regulated by a modal-control field controller, etc.

    It is listed as a 2kW device on the website click to products, the PDF file shows a power curve limited at ~3,600 watts (above ~36mph)...

    2kW would be 20 amps @ 100 VDC, 3,600 watt would be 180 volts @ 20 amps--clearly outside of the range of a standard Tri-Star 60...

    Do you need a 4kW dump load/controller--or is the turbine "Safe in all wind conditions" regardless if the attached controller/loads?

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • niel
    niel Solar Expert Posts: 10,300 ✭✭✭✭
    Re: Colombia project updated:

    "Do you need a 4kW dump load/controller--or is the turbine "Safe in all wind conditions" regardless if the attached controller/loads?"

    this is why i advised him to get input from the manufacturer of the turbine as they should have the accessory products being the needs are more unique.
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: Colombia project updated:

    Simon,

    I've been on travel for several days. I'll take a look at your new plan and provide comments this weekend.

    Regards,
    Jim / crewzer
  • System2
    System2 Posts: 6,290 admin
    Re: Feed pump missing!!!

    After reading all these interesting feedbacks to my post (THANK YOU!), I realize I forgot to add in the power calculations on the feed pump.

    So the total power needed for the complete system is 15400W (I added 3000W for the feed pump per day). Sorry about that! As amazing as the WaterCube is, I keep forgetting that it does not include the feed pump. : (

    More reason to have two Delta I's.

    Simon
  • System2
    System2 Posts: 6,290 admin
    Re: Colombia project updated:

    No problem Jim. I look forward to your input.

    I tried to send you a personal message today but it said your mailbox is full.

    We maybe near your area next weekend if you have time to meet up. Send me an e-mail and let me know your schedule.

    Thanks,

    Simon
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    Simon,

    It is WH (Watt*Hours), not "W" only for your numbers here (--know what you mean)--but will confuse somebody trying to help if you mix load, like 1,000 watt load for the desalinator vs 24,000 WH (or 24 kWh) running it for 24 hours.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • System2
    System2 Posts: 6,290 admin
    Re: Colombia project updated:

    I apologize for the confusion.

    When I am using "W" I mean the load for the day.

    So the 15400 W is the total load per day not per hour.

    I will try to be clearer in the future.

    Simon
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    Actually, I do think you mean 15,400 watt*hours per day...

    15,400 Watts per Day does not really make any sense... Unless you are saying you have a 15.4 kWatt load running 24 hours per day which would be:

    15.4kW*24 hours/day = 369.6 kWhrs per day (more power than I use in a month at home)...

    Watts is roughly equivalent to Miles per Gallon. Watt*hours is roughly equivalent to Gallons.

    A more exact analogy is "watt" is equal to "gallons per hour" and "watt*hours" is "gallons"... (or the difference between miles per hour and miles driven). You can drive 800 miles in a day at 50 mph for 16 hours--but cannot drive 800 miles per hour per day...

    So, for example, you need an inverter that can support a load (rate) of 1kW (1,000 watts). If you ran it for 24 hours, then the amount of "work" it can do is 24kW*Hours per day.

    But, you could also design a system to run 1 hour per day using a 24kWatt inverter. The amount done per day would still be 1H*24kW=24kW*hours per day...

    Both are perfectly good solutions--but, obviously the system that only ran 1 hour per day would need an inverter (and desalinator) 24x as large as a smaller one that ran 24 hours per day.

    Sorry to be so pedantic--but it really can get confusing about what you are talking about when you run through the power requirements for the various pieces of equipment in the discussion...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • System2
    System2 Posts: 6,290 admin
    Re: Colombia project updated:

    Pedantic? hehe! I had to look that up. You are not being insulting Bill, that is just part of your charm.

    To produce 3,000 liters the water purification system runs roughly 15 hours a day in which it consumes 15,400 watt*hours per day

    Thanks Bill for straighting this out.

    Simon
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: Colombia project updated:

    Simon,

    The mailbox problem is fixed… new forum software, new rules… :cry:

    I’m not sure that I’m an expert in anything, and I know that I’m not an expert in wind energy. Accordingly, I’ll pass for now on assessing the generator’s suitability. I will, however, offer some comments on the battery, DC voltage, controller, and inverter issues.

    Allowing for one day of energy autonomy (15.4 kWh), 90% inverter efficiency, and a maximum 50% discharge, you’ll need a battery bank rated at (15.4 kWh / 90%) / 50% = 34.2 kWh. Assuming a 48 V nominal battery voltage, the bank’s capacity spec will be 34.2 kWh / 48 V = 713 Ah. You could build a bank of this size from 16 size L16 flooded-cell batteries, each rated at 6 V nominal x 390 Ah. Configured in an 8 x 2 bank (eight batteries in series per string, and two strings in parallel), the bank’s capacity would be 48 V nominal x 780 Ah.

    See: http://www.trojanbattery.com/Products/ProductSpec.aspx?Name=L16P
    And: http://www.trojanbattery.com/Tech-Support/BatteryMaintenance.aspx

    The main advantage of a 48 VDC system over a 24 VDC system is efficiency. For a given amount of power, the current value in the 48 V system will be ½ that in a 24 V system. This allows for smaller wires and/or less energy loss in the DC wiring. Additionally, DC-to-AC inverters are typically more efficient with higher DC input voltages.

    I find Pacwind’s specification and application documentation for the Delta product to be alarmingly thin. :grr However, borrowing from documentation for their Seahawk product, they seem to like the Morningstar TriStar controller.

    http://www.pacwind.net/SeaHawkManualNew.pdf
    http://www.morningstarcorp.com/products/TriStar/index.shtml

    Pacwind’s application of the TriStar controller is as a diversion controller and not as a series controller. Accordingly, and with all due respect to my esteemed forum colleagues and their concerns, I don’t believe that the apparent mismatch of the generator’s output voltage (0-190 V peak) and the controller’s input voltage specs (68 V operating, 125 Voc) will be an issue as long as the generator, battery bank and diversion load(s) all remain interconnected.

    But, you should read the TriStar manual to familiarize yourself with diversion controller operation. This type of charge controller application uses a diversion load – typically a resistive heating element in a water tank – to “divert” excess energy away from the batteries so as to avoid overcharge. I suspect that you’ll need two controllers and two diversion loads should you decide on two generators, but that you can use just one battery bank.

    Consuming 15.4 kWh/day of energy over 15 hour/day means that your load will be just over 1,000 W. A single sine wave inverter rated at ~2,400 W to ~3,000 W should operate at peak efficiency under that load. Assuming that the Danfoss Watercube 4800 requires 230 VAC nominal at 50 Hz, a single OutBack FGX2348ET (sealed) or VFX3048ET (vented) inverter/charger should work. I’m not familiar with the Magnum inverters, but it’s my understanding that they too are very good products.

    See: http://www.outbackpower.com/Export.htm

    Finally, I appreciate your invitation, but I'll have to take a rain check for next weekend... my brother is throwing a HUGE party for his oldest son's 21st birthday! :cool:

    HTH,
    Jim / crewzer
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    Jim, you very well may be right... The PDF just shows them connecting a standard 3 phase alternator with rectifier block directly to the battery bank (and they show fuses for the controller+diversion loads, but no fuses for the alternator connection--mistake in drawing since they call for a 60 amp fuse to the turbine).

    However, the rating of the Delta-I turbine seems to be 20 amps max and either 2kW or up to 3.6 kWatts (in a hurricane?).

    If you take the 2kW at 20 amps, that is still 100 volt output. I was thinking that they use the MPPT function to down-convert the turbine's output voltage to battery bank level. If not, then that is a lot of wasted potential power if you your battery back is less than 100 volts when charging...

    From the PDF link you posted--they list wind speeds at which these other units begin charging:
    The rotor of the SeaHawk should begin to rotate when the wind speed reaches approximately 1.8 m/s 4 mph.

    Battery charging will commence on a 12 Volt battery system at about 3.1 m/s (7 mph)[.]

    [O]n a 24 Volt battery system at about 5.4 m/s (12 mph) and on a 48 Volt battery system at about 10.8 m/s (24 mph).
    24 mph minimum wind speed to begin charging a 48vdc battery pack--that is one mighty windy location.

    All in all--the specs. seem not to be consistent--perhaps marketing is ahead of engineering for this newer product.

    I would guess that your (Simon's) spreadsheet outputs for the Delta-I either cannot be accurate, or this is based on 12 volt battery bank (a 48 vdc bank would appear to loose 1/2 the total power generated--as nothing below 24 mph even generates power), or the Delta-I has a different alternator configuration (or this style turns a bunch faster) than their earlier turbines.

    Not enough information to really make any decisions here.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: Colombia project updated:

    Bill,

    This wind generator mumbo-jumbo is a great deal of the reason why I’m definitely not a wind energy expert… Considering the range of MPPT controllers available for PV applications (Analytic Systems, Apollo, Blue Sky, Outback, Xantrex, and more to come), the corresponding dearth of controllers for wind applications suggests to me that this application remains problematic.

    Speaking of MPPT controllers, the Morningstar TriStar is straight PWM and not MPPT, so my gut feeling is that you’re right about wasted potential power. The target absorb voltage for the “48 V” battery bank will be ~58 V to 59 V.

    However, MidNite Solar's new "Classic" controller will have a "wind" mode, so perhaps it might be worth having Simon wait for this new product for his application.

    Regards,
    Jim / crewzer
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    Yes Jim, you are right (of course) about the Tri-Star being a PWM controller--I forgot to include that in my analysis of the system design in the discussion--That probably means that your conclusions that the Tri-Star is still being used as a diversion type controller is 100% right.

    "Adding" to the lack of any useful design information from PacWind for the Delta-I wind turbine.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • System2
    System2 Posts: 6,290 admin
    Wind data: It should be right!

    Wind Data for Colombia:

    http://www.windfinder.com/windstats/windstatistic_riohacha.htm

    Click on yearly button and the numbers should be very similar to the wind rose data we are using in the excel sheet.

    The Delta I power curve is difficult to read so it is not 100% accurate but close enough.

    It is VERY windy there and great sun too. I found a great website for solar information at:

    http://www.gaisma.com/en/location/riohacha.html

    Isolation is 6.55
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    The Delta-I power curve I saw is not accurate for the system you would be designing--assuming that you aim for a 48 VDC system (keeping wire gauge and currents low and efficiencies high).

    You need a 24 MPH wind just to start generating 48 volts (0-23 mph->zero watts, 24 mph 1 watt?). According to the Delta-I data sheet, it should be generating 1,500 watts at 24 MPH, not "zero" watts. Let alone what wind speed is required to reach 60+ VDC to really properly charge the batteries (just guessing).

    Again, the lack of real information (that Jim and others have found) from the PacWind site is just letting me to divine specifications and how the resulting turbine will perform... Gets to be a "self limiting" exercise in the end without actual specifications from PacWind.

    It is certainly possible that the total yearly amount of wind energy could be met by a properly sited 2kW turbine--the ability to only store a few days of electrical energy means that ~7 months of the year, its output will be substantially less than average (and 5 months well over average)--meaning that you will either need more turbines or other power sources or alternative storage (energy or product).

    Also, regarding Isolation of 6.55, you need to multiply by the "clearness factor" which gives daily average of 4.4 kW/sqm (over one year)...

    And, averages don't really help when you are looking for specific results. If you need 3,000 liters of water per day, 365 days per year, then you kind of have to look at the average worst case 5.47*0.66=3.6kW/sqm (average sun in December?)... Your worst case could be 1/2 that--so you need other sources of power--turns out December through April are your best wind months--so it can compliment solar (and keep fuel use lower).

    Again, make the Wind/RE vendor do the hard work... Having accurate wind speeds for the site is going to be very important as the power output for the Delta-I seems to go with roughly the square of the wind velocity. A linear average does not properly weigh the true energy content of the wind.

    If you can store several months of fresh water (and possibly just use chlorination to keep it safe for daily distribution)--that would help you ride out calm winds and cloudy days... And limit the requirements for fuel driven generators to make up for less than optimum seasonal RE power.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • System2
    System2 Posts: 6,290 admin
    Re: Colombia project updated:

    If you read my VERY first post, I clearly stated that we are trying to make this work with TWO Delta 1's not just ONE.

    PacWind estimated 10kWh/day when my wind dealer spoke with them about our project. But I am not sure if that was for 48V, 24V or 12V batteries. I agree with you that the PacWind data sheet is not very informative.

    As for the isolation, I was just pointing out that there is a lot of sun. Clearly, only solar would be very expensive.

    I agree with the RE vendor approach and we have already been doing that for awhile. And yes they are located in Colombia. We have another “Colombian” company designing a system with two wind generators or one wind generator with solar panels. No guarantees that the perfect solution will come out of this but we are definitely trying out all possible solutions. I don’t mean to be pedantic but certainly Bill you don’t think this forum is the only source we are using. : )

    Have a nice day!

    Simon
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    OK................
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • crewzer
    crewzer Registered Users, Solar Expert Posts: 1,832 ✭✭✭✭
    Re: Colombia project updated:
    "Also, regarding Isolation of 6.55, you need to multiply by the "clearness factor" which gives daily average of 4.4 kW/sqm (over one year)..."

    Bill,

    I'm afraid I can't agree with this statement. My take is that the 6.55 insolation spec is for solar irradiation hitting a horizontal plate on the ground. I believe this is why GAISMA's winter insolation numbers are so low for the northern hemisphere. The "clearness" factor is described here: http://www.gaisma.com/en/info/help.html

    HTH,
    Jim / crewzer
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Re: Colombia project updated:

    Jim,

    You are correct... I read the Clear Factor and thought that it was needed to calculate the local Insolation using the "top of atmosphere" Insolation:
    Insolation The monthly average amount of the total solar radiation incident on a horizontal surface at the surface of the earth for a given month, averaged for that month over the 22-year period (Jul 1983 - Jun 2005). Each monthly averaged value is evaluated as the numerical average of 3-hourly values for the given month. Source: NASA Langley Research Center Atmospheric Science Data Center.
    Clearness The monthly average amount of the total solar radiation incident on a horizontal surface at the surface of the earth divided by the monthly average incoming top-of-atmosphere insolation for a given month, averaged for that month over the 22-year period (Jul 1983 - Jun 2005); (i.e. clearness index is the fraction of insolation at the top of the atmosphere which reaches the surface of the earth). 0 = very overcast and 1 = sunny. Source: NASA Langley Research Center Atmospheric Science Data Center.

    But, it appears that Clearness is probably a factor that tries to give a number to cloudy vs sunny weather--and the "other (above) Insolation" is not a theoretical factor, but the "measured" surface Insolation.

    Sorry...

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • System2
    System2 Posts: 6,290 admin
    Re: Colombia project updated:

    I apologize for not writing sooner but I have been extremely busy as of late.

    Just wanted to share the latest on the Colombia project:

    We had a pretty interesting technical solution with the help of Bornay. Unfortunately, we were never able to realize the project.

    See below why:

    http://www.aguayuda.org/Eng/Proj/LosFlamencos.cfm

    No one said development work is easy. Thanks for everyones advice and support.

    Simon
    p.s.
    Jim we still owe you a lunch!!!