Author Topic: Wind Water Pumping in Afghanistan - video (Read 5978 times)

cslarson · « **on:** May 18, 2011, 06:59:21 AM »

Hi all,
I recently took a three week trip to Afghanistan to attempt to implement a battery-less wind powered water pumping system. We experimented with a few ways to do this. The first involved having a controller switch on the water pump once the voltage had reached a certain level (the turbine will not spin up with a load, like a pump, is connected). Start up of the pump was hit or miss depending on how much wind was available. When we watched the current being drawn by the pump during successful startup (single phase AC) it would initially be higher and then drop down. This matches my understanding of the start-up of an induction motor. During an unsuccessful start-up, though, the current would not see that drop to normal operating. Sometimes this load would cause the voltage to drop below a threshold and the controller would disconnect the pump. Other times this did not occur and the pump would draw the higher current but not start-up successfully. I concluded this was likely not a good state for the pump to be in as I guessed that power would only be dissipated as heat in the pump coils reducing the pump life.
The other attempt involved using a variable frequency drive, controller and dump load. This can be seen in the following youtube video: http://www.youtube.com/watch?v=QvPmIcb51ls. I apologize for the poor quality and lousy narration, but hopefully the general idea will come across. The controller is generally necessary to protect the VFD and deal with various states, such as low voltage error mode, restarting, etc. A dump load kicks in to prevent the voltage from getting higher than 400Vdc and blowing the capacitors. The VFD applies a soft start to the pump so start-up is reliable. An added benefit is for multiple power sources, and potentially dissimilar power sources, to work together. Drawbacks are the added complexity, cost, and potentially limited lifespan (?) of the VFD. Additionally, as it's desired to keep the voltage of the VFD DC buss relatively high but not too high, this limits the operating speed range of the wind turbine.
I'd be interested in feedback from this experiment. What drawbacks am I not considering for the VFD approach? How to achieve more reliable start-up with the direct powering approach?
Thanks,
Carl

joestue · « **Reply #1 on:** May 18, 2011, 07:52:57 AM »

i think your reliable starting problem is the variable volts/hz

i'd be willing to bet you've got enough power to start it at 200 volts if you can keep the volts/hz high.
don't worry about the motor overheating provided the current is below the nominal rated current.

here's what i would do, as it seems you're running the pump below rated power most of the time, turn up the volts/hz through the vfd.
if you can't do that, use a small transformer to boost the ac coming out of the vfd by 10, 20 or even 30%. if you're running a 230vac motor off a vfd intended for 400 volts, but starting up the motor at 250-300 volts, then its as if you're only running the motor at 250/400 volts, and this will reduce the torque significantly.

Scott · « **Reply #2 on:** May 18, 2011, 08:39:08 AM »

Very cool!

Just curious why so much effort to avoid batteries?

cslarson · « **Reply #3 on:** May 18, 2011, 10:46:20 AM »

hi joestue,
thanks for the comments. the VFD has a number of options. i stayed away from changing many because I didn't fully understand them. there is a setting for Automatic Voltage Regulation which is on by default. i imagine this is why output voltage is limited to about 230v regardless of the voltage on the DC buss. there is also a setting for Maximum Voltage (220 by default) which according to the instructions should be set to the nameplate of the pump and is also limited by the tolerances of the inverter (so 400vdc in the case of the one we used). they suggest increasing this if the motor is far away, which in the case of a well pump, it is. i had not changed this setting, but it might be worth experimenting with it.

otherwise, as far as the directly driven pump is concerned, the v/f should be higher to ensure good startup? so we could increase the turns on the stator to do this. or connect the pump between phases rather than between phase and neutral. (the pumps are all single phase as we couldn't find three phase pumps below 2kW which is too large for a single turbine).

could you explain a little more the following? the VFD has a maximum dc buss voltage of 400v.

Quote from: joestue on May 18, 2011, 07:52:57 AM

if you're running a 230vac motor off a vfd intended for 400 volts, but starting up the motor at 250-300 volts, then its as if you're only running the motor at 250/400 volts, and this will reduce the torque significantly.

hi Scott,
we've had considerable issues in the past with the lifespan of the deep cycle batteries we could supply in Kabul. recently we've had issues with the inverters I imported from China! ...of course this vfd is from China... But primarily, the cost of that type of off-grid system (with a battery bank) and the risks involved I think make it a significant challenge in a developing world context. In Afghanistan the wind is available generally when water is required for agriculture. If we can remove the batteries then I think we have a better chance at something that is both useful and economical versus the alternatives (solar and diesel generator).

electrondady1 · « **Reply #4 on:** May 18, 2011, 11:37:51 AM »

cslarson
i know you are a good man from the posts you have made in the past
but this post has made me a bit angry.

do you really think you are using an appropriate level of technology for that village?
i barley understand what you are doing and i have been studying this stuff for a long time.
is it you intention to spend the rest of your life in that village making repairs to the system your are constructing?
do you think the little boy or the old man in the video will be capable of looking after your battery less water pumping system when you are gone?

god knows the people of Afghanistan desperately need our help.
wouldn't a traditional multiblade farm windmill , pump water when required for that village for the next fifty years?
no battery required.
and everyone in that village would understand how it worked and could attempt repairs.

Flux · « **Reply #5 on:** May 18, 2011, 01:37:41 PM »

Good to see your results.

Personally I would have thought the simplest way would be to rectify the output and use a dc motor. If brushes and maintenance are an issue then a brushless dc motor. I haven't looked at the availability of pumps with dc motors.

Failing that I would have chosen a 3 phase pump motor, single phase motors are an abomination. The starting of a 3 phase motor is defined, for single phase it is guesswork.

With the VF drive it would work so much better with the 3 phase motor. In fact I can't see why it won't work with your inherent vf drive. If you choose the pole number correctly and choose a suitable voltage you should be able to run the 3 phase motor directly off the alternator. You would need one pump per turbine to do it this way but I can't see that being a major issue.

If this is a centrifugal pump the load curve should ideally match the cube law wind power. You might possibly need some start up circuit to isolate the motor from the turbine until there is enough wind to get to pumping power but otherwise the match should be excellent.

Keep up the good work, you have something that works at present i am sure you can refine it and probably simplify it as well.

Flux

cslarson · « **Reply #6 on:** May 18, 2011, 03:18:37 PM »

thanks for your encouragement, flux.
yes, i wish we could use a 3 phase motor. there is some advantage to using what is available in the bazaar, though. the smallest three phase motor we found was 1.5kW and we only ever found one. three phase seems to only be available at sizes greater than 2kW. at a much earlier point in the project I found grundfos dc pumps available at a renewable energy supplier in Kabul. they wanted $3000 for the 900watt dc pump. the chinese AC units available in the bazaar are $150 for a single phase 1kW pump. across the street is a guy working out of a shipping container that can replace the windings or fix anything else that goes wrong with them.

yes, in theory directly driven from the turbine should work well and the pump load curve matches the power available in the wind at varying speeds. i first learned about this with the following article: http://www.bergey.com/pages/wind-electric_pumping_systems_for_communities. but the only water pumping system advertised now by Bergey involves a battery bank + inverter. i think a system that worked like the article explains would be great. the problem at the moment, though, I think, is with start-up. the turbine must spin up without a load connected. no problem. a simple circuit switches some solid state relays to connect the pump at whatever point is chosen. at that point there needs to be about 250% power available for that frequency. after a few seconds it should drop down to 100%. sometimes this happened and sometimes it didn't. we installed a flywheel on one of the turbines to see if this would help. i was not there for enough time to see how this improved reliability. instead i worked on the vfd system as I thought this would solve the start-up issue as well as provide some additional benefits (combining power from multiple turbines).

hi electrondady1,
yes, maybe the vfd is too complex. the wells are 50-100m deep. that is a long shaft to drive. i do not have very good knowledge about those windpumpers, though. maybe they would be a better option. we initially developed the wind turbines for battery charging. so we had these available to work with. there should be some advantages to wind-electric over the traditional windmills, but it's true that if it doesn't work, or if the village can't fix it cheaply enough when something goes wrong, then those supposed advantages don't matter.

Flux · « **Reply #7 on:** May 18, 2011, 04:31:24 PM »

Yes in real life many considerations come into play that aren't at first obvious. The cost of dc motors tends to be high, Until Grundfos get the Chinese to make their pumps there will always be a cost issue.

Similarly the problem of 3 phase motors in small sizes is again a supply and demand situation. Perhaps you could get your tame man to convert single phase ones to 3 phase but there is a considerable difference between copying an existing winding and changing the design.

The old fashioned wind pump has some merits but it pumps very little water compared with a well set up high speed turbine. It will probably perform better in very low wind but will be very poor in comparison on decent windy days.

Flux

joestue · « **Reply #8 on:** May 18, 2011, 08:37:47 PM »

Quote from: cslarson on May 18, 2011, 10:46:20 AM

the VFD has a number of options. i stayed away from changing many because I didn't fully understand them. there is a setting for Automatic Voltage Regulation which is on by default. i imagine this is why output voltage is limited to about 230v regardless of the voltage on the DC buss. there is also a setting for Maximum Voltage (220 by default) which according to the instructions should be set to the nameplate of the pump and is also limited by the tolerances of the inverter (so 400vdc in the case of the one we used). they suggest increasing this if the motor is far away, which in the case of a well pump, it is. i had not changed this setting, but it might be worth experimenting with it.

otherwise, as far as the directly driven pump is concerned, the v/f should be higher to ensure good startup? so we could increase the turns on the stator to do this. or connect the pump between phases rather than between phase and neutral. (the pumps are all single phase as we couldn't find three phase pumps below 2kW which is too large for a single turbine).

if that's a 220-240vac 50/60 hz motor it needs to be connected line to line. line to neutral will only get you ~60% of the voltage.
(i'm rather supprised that vfd has a neutral), and this alone would account for your start up problems.

basically, the 400 volt rating is to protect the capacitors and electronics, but that's also where it is intended to run at 24/7. Generally speaking those vfd's pwm the phases at up to 80%, so you get up to 320vdc average on each phase, when modulated as a sine wave turns into an rms ac voltage of about 220. there's ways to push through this limitation but by default this is how its done.
you'd have to find the manual to read what automatic voltage regulation means.
if it keeps the v/hz constant during dc line fluctuations that's what you want, however it may not be able to do so all the way down to 250-300 volts.

oztules · « **Reply #9 on:** May 18, 2011, 10:59:47 PM »

I recently designed and built a water pumping system that runs on solar.

It provides over 34000 liters/day into a 110 psi head. (mostly sunny day ) On average, it keeps up this output in summer (cows don't need that much in winter) It puts out an awful lot on cloudy days as well...

Diode isolate the turbine DC and the panel DC, and run on both??

It uses the home brew solar panels (here ) http://fieldlines.com/board/index.php/topic,145005.0.html

So for only $1500 we had about 1.6kw or so of solar cells. (another good project for your builders)

It drives a curtis golf cart PWM controller good for 270 amps @ 36v $250 from ebay (USA)

The brush motor is from an old floor scrubber 2800rpm 2hp 36v costs nothing as a general rule............. generally thrown out by the service mobs. The ones from ADVANCE floor scrubbers are the best .... and always good for another 20 years..

The interesting part is the controller for the curtis. It involves some resistors, 2 opto isolators and a zener diode,.(maybe a dollar for the whole board)
Thats less than two grand for a village system ( you have the pump already ... just match the pulley ratio for your load/rpm) that will beat your current setup into next week I suspect

The same controller/curtis/motor set up could be as easily used with your turbine. It simply measures the input voltage, and if it is less then 33v or whatever you set for, it turns the pwm off, as it gets over 33v it pulls in the pwm. simple.... no it is not too harsh. You can control max speed by using different resistors across the pot input to the controller.

This was to replace a commercial (MONO) high tech pump system.... $30000.00. This system is easily it's equal...... and easily expandable (build more panels)

It was designed so that all the bits were cheap to get if they failed, and could be put together by anyone.

Your current high tech solution is the reason this was developed. People were sick of buying expensive goodies that failed, and then they were held hostage. There are no shops down the road here... your on your own..... not even a bazaar, or a bloke in a container to rewind your motor.

If you stick to your VFD, perhaps just find the star point and change the motor to delta instead of star. This will give you a three phase 240v motor from a 415v one.... simple.
If it was 415 in delta, you have some coils grafting to do....... but I see this route ending in tears..... just too much to go wrong.

Ditch the windmills and build solar panels.... probably cheaper and way more reliable in the hands of neophytes ..... these folks in Africa are doing it similarly to me www.mainesolar.org/EVA.pdf

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

cslarson · « **Reply #10 on:** May 19, 2011, 04:35:30 AM »

thanks for the responses. the VFD has no neutral line, just three phases on the output. But phase to phase voltage doesn't increase above 230vac. I thought normal three phase would be 380vac phase to phase and around 240vac line to neutral. there is no starting issue when using the vfd system, only when connecting the pump directly to the turbine alternator.

the wind turbine alternator was designed so that at around 50Hz the output is around 240vac line to neutral (the turbine alternator has neutral line which is what i was referring to). in reality, i think it is reasonably close to this, maybe with a slightly higher v/f.

oztules,
yeah, there's no escaping the fact that solar systems have become much more competitive. the local Kabul bazaar is selling chinese panels for around $2/watt. so that's pretty cheap. and your suggestion to build them is interesting. i hadn't really looked in to it. we were interested in wind turbines largely because they could be built in-country (though that sort of thinking doesn't necessarily lead to a successful business...)

until dc pumps become a more standard, available option, though, i have trouble seeing why are not just more expensive, and more difficult to replace. when i search for pv water pumping systems online it seems as though many of them (http://zzstc.manufacturer.globalsources.com/si/6008827234744/pdtl/Solar-powered-pump/1041641763/Solar-Water-Pumping-System.htm) are working with AC pumps so within their "controller" is some sort of inverter unit. the vfd i used here cost about $300. a 2.2kW unit would be about half that. i thought they offered that dc-ac conversion for a good price and with some additonal benefits (soft start) as well. a similar system might work to allow solar panels to power ac water pumps as well.

vawtwindy · « **Reply #11 on:** May 19, 2011, 05:32:10 AM »

what is VFD? if we need to run 3phase 410V 11.4amps 7hp motor what do we need to do?

not trying to deviate the topic, but curious, whether this is acheievable. ?

cslarson · « **Reply #12 on:** May 19, 2011, 05:55:20 AM »

A VFD, or variable frequency drive, is a device for controlling the speed of an AC induction motor by varying the frequency of the supplied current. Generally, the voltage and frequency should have a fixed ratio, so the voltage needs to vary with the frequency. The VFD converts AC to DC (with diodes, quite simple) and then that DC back to AC (more complex, with sine wave pulse width modulation and some fancy electronic circuitry) at the desired frequency.
http://en.wikipedia.org/wiki/Variable-frequency_drive

i'm not an expert with these, but maybe something like the following would work for your motor.
http://cgi.ebay.com/GE-AF-300E-7-5-10-HP-VFD-DRIVE-6KAF343007ESA1-AF300ES-/330513182034?pt=LH_DefaultDomain_0&hash=item4cf41eed52#ht_500wt_922

nconstanzo · « **Reply #13 on:** May 19, 2011, 11:15:25 PM »

cslarson
i checked out the vsd and was wondering if that drive only works of dc are can be connected to single phase 230.

cslarson · « **Reply #14 on:** May 20, 2011, 03:58:03 AM »

the unit we used had three phase, 220v input. I located the DC buss on the circuit and made the connections for DC input myself. single phase ac input could be used on any two of the three ac input connections.
-Carl

oztules · « **Reply #15 on:** May 20, 2011, 04:45:27 AM »

http://www.aliexpress.com/product-gs/392183204-DC-Solar-Pump-for-deep-well-with-all-accessories-wholesalers.html

cslarson · « **Reply #16 on:** May 20, 2011, 05:31:23 AM »

yep, i was looking at the dc pumps listed on aliexpress yesterday. the link you posted has the largest pump at 500watts for $665 (not including shipping into afghanistan and taxes @at an additional 20% of both base cost + shipping, final cost may be close double). also the following:
http://www.aliexpress.com/product-gs/289669080-CP95-1000W-DC-Deep-Well-water-pump-wholesalers.html a 1000watt dc pump for $678 that they say should work with 2200watts of panels.
so if we say cost wise that pump would cost $1000 and the panels, in the bazaar could be had for $4400. mounting system, connections, installation, profit, etc. maybe the system could be sold for $6500. that said i think it would work better than one of our wind turbines. maybe it would also still work with fewer pv panels. the guys i am working with would sell a turbine + pump + controller for just a little over half that.

how much more efficient is the dc pump over a single phase ac pump of similar size? a three phase ac pump? one limitation with the dc pumps is that larger ones don't seem to be available. one benefit of the vfd system (working with either solar, wind, or a combination) would be that it would be scalable to larger pumps and wells... but i'm hearing the skepticism

ginger48 · « **Reply #17 on:** May 20, 2011, 07:37:45 AM »

The Chinese copies of the old Aremotor water pumping windmills are very reliable, simple, inexpensive, and supply a LOT of water. See http://www.ironmanwindmill.com/

cslarson · « **Reply #18 on:** May 20, 2011, 07:50:18 AM »

any idea of cost for those?

ginger48 · « **Reply #19 on:** May 20, 2011, 08:42:25 AM »

Tom Conels sp?, owner of Iron Man and an American, has had numerous discussions on other forums. To the best of my knowledge, price has never been mentioned, however they are less expensive than the U.S. made windmills and he's also improved the design. Price naturally would depend on size of windmill and tower height. Tom makes windmills up to 20' diameter and has installed many 20 footers in China.

tecker · « **Reply #20 on:** May 20, 2011, 08:59:39 AM »

You need drip watering system . The field is not that big and a 1" heaader with 1/2 to 1/4 pipe wll work . with a deep well watering open will waste too much water. remember to test the water quality and PH .
here's a good video for rooting thornless blackberries remember to take cuttings from new growth that is not blooming .

http://www.youtube.com/watch?v=gk6CWYfFBIw

cslarson · « **Reply #21 on:** May 20, 2011, 11:17:13 AM »

thanks for the tips on drip watering. i know next to nothing about irrigation. the village currently pump using, i think, about a 12hp? diesel generator with about a 3hp single phase ac pump. when they pump they just flood the field. they also pump during the day and i imagine much of the water just evaporates. i'm not sure how often. previously when i asked i think they said it was about every other day, but on our recent visits i didn't see this. it could depend on what time of year.
becoming more knowledgeable about irrigation would be necessary if we had a successful, commercializable pumping system to work with. otherwise i need to just focus on getting a job here in the uk and quit fooling around with stuff in Afghanistan!

tecker · « **Reply #22 on:** May 20, 2011, 06:52:27 PM »

I here ya pass on this thread to one of the folks that are in control . With some info in this area regularly being posted a distribution network would most certainly follow .
This is very refreshing to follow thanks

electrondady1 · « **Reply #23 on:** May 21, 2011, 12:43:40 PM »

"I here ya, pass on this thread to one of the folks that are in control"

down in Kandahar province, there seems to be different opinions about that very thing !

cslarson ,
the only people i have met that have been to Afghanistan are Canadian Forces.
is it safe to be a westerner in and around Kabul?
what i mean is ,if i went there,would i be killed?

itsandbits · « **Reply #24 on:** May 21, 2011, 01:34:02 PM »

Quote from: electrondady1 on May 21, 2011, 12:43:40 PM

"I here ya, pass on this thread to one of the folks that are in control"

down in Kandahar province, there seems to be different opinions about that very thing !

cslarson ,
the only people i have met that have been to Afghanistan are Canadian Forces.
is it safe to be a westerner in and around Kabul?
what i mean is ,if i went there,would i be killed?
-------------------
Are you planning a trip there E.D.? I don't want to mention when our troop support is moving out but in the next 10 yrs we are supposed to be gone, as per gvmnt release.

tecker · « **Reply #25 on:** May 21, 2011, 04:05:25 PM »

No body needs to be there to pass on this stuff .
If the folks use wind power that's cheap enough these projects will for the most part be doable .

cslarson · « **Reply #26 on:** May 22, 2011, 01:34:20 PM »

generally people who travel to Afghanistan are linked to some sort of ngo (non-government organization), agency, or business. depending on who they are with they will have some sort of "security profile" which will determine how they will travel around, where they can go, and what time they have to be in bed. the organization should also provide some cultural orientation, maybe language training, etc.
otherwise, unless you're really in the wrong place at the wrong time, Kabul is safe. around Kabul, it depends where you go and maybe at what time. you wouldn't have to stray too far in the wrong direction to not be safe.
in the 4 years i was there i never had any sort of hostile incident with an Afghan. drunk, gun carrying western security contractors, on the other hand...

News:

Author Topic: Wind Water Pumping in Afghanistan - video (Read 5978 times)