A Forum run by Enthusiasts of MidNite Solar

Hello,

I was informed by a member over at the NAWS forum that the Charge Controller companies have conversion data available that would allow me to figure out conversion losses with a particular charge controller. I am looking at the classic 200. If my array was wired at about 108 volts (6 145 watt panels in series, 18.7 vmp)  and fed in to a 12 volt battery bank (12 volt system voltage) via the classic 200, what kind of losses will I see from the down conversion process in the charge controller?  Is there a chart or calculator I could use to find this answer?

Hi ea..

Believe that the PV modules are DM Solar 145s:
Vmp 18.7,  Imp 7.75,  Voc 22.3

Recall that you are at 5200 ft elevation,  but do not know the lowest recorded temps for that location.

With 6 of these PVs in series,  you certainly in the MN Cl 200 range.  Know that you have a fairly long run from the PV to the rest of the electronics.  It is possible to locate the CC,  batts and inverter near the PVs and run the AC output of the inverter to the point of use.  Believe that you are concerned about cost of wire vs all else.

Believe that the Efficiency Chart is not yet available.  It is correct that converting from 108 ish string V to 12 v batt,  is not as efficient as converting this V to a 48 V bank.   It is known that this  chart data is difficult to collect,  and often some other mfgs are very optimistic with their "data".  Furthermore,  there appears to be NO industry standard for the collection of this data.

But,   if the string V is 108ish V,  given your location,  you really have very few alternatives to the Classic 200.

Little time now,   more later,   Vic

hi Vic,

Yes they are the dmsolar 145s at 5200 ft ele. Cariboocoot suggested that between the conversion loss and v-drop that I might lose 6% of max power (not sure I know what that means yet). I wanted to run 10 awg wire to save money and thought that the higher the voltage from the array the smaller the wire I could get away with. According to a v-drop calculator the 100ft run with 10 awg at 108 volts would be less than 2%. So the remaining question would be how much power would be lost in the controller to convert the 108 volts to 12 volts for the batteries? Searched high and low but can't find that info anywhere.

Also when I use a v-drop calculator should I use the one way distance of the wire 100ft or the loop distance 200ft?

hi again,

I forgot to mention that yes I had the idea of locating the controller, batteries and inverter below the array, possibly in a tiny shed of sorts. I didn't get any feedback from the naws group on this subject so I thought it might not be a good idea. One problem I could foresee is the cold temps and keeping the batteries at a reasonable temp in the winter. Much of the winter is below freezing day and night. Security is also a big concern.

Hi eajones,

Sorry,   just scanned the NAZW&S post,  forgot that this remoteing of the electronics had been mentioned.  If you can keep the batteries fairly fully charged in the Winter,  they should do well in cold.  They really hate the heat,   tho.

Regarding the Volt Drop calcculator,  normally,  it is speced right up front what length to use,  RT or One Way.

Believe that niel,  a member here,  and there,  has a link to his Drop Calc program.

It is quite possible that the conversion losses added by using higher string V,  12 V batt,  and the Classic 200 will add one or perhaps even 2% additional conversion loss.  I am no expert on this.

There is a large Solar Trade Show in FL,  and believe that a number of key MN employees were there,
EDIT:  The show ends on 9/13 ...     Perhaps,  in a day or so,  some of the MN wizards will respond.

Because the Classic is such a new product,  it benefits from a lot of experience that the founders of MN have gained (they also founded OB Power,  and were at Trace/Xantrex),  So conversion loss is one factor,   but  believe that the Classic is very good at maximizing the power extracted from the PV (or Wind,  Hydro,  etc).  So it is not only conversion efficiency.

EDIT: Efficiency Curves have been a topic here in the past,  here is a link to one:
http://midnitesolar.com/smf_forum/index.php?topic=91.msg589#msg589

If you search this site  for " efficienty curves " (no quotes)  you should find more.

More later,  Vic

Here are my latest thoughts on the set up. According to the voltage drop calculator if I use the 10 awg wire it will only cause a 1.6% voltage drop with 6 panels in series. So even if I lost 2% more in the controller conversion, I am looking at less than 4% power loss. Does that sound about right?

So the original Cariboocoot approved setup was:

4 dm solar 145 watt panels 18.7 vmp 7.75 amps  $640 (already bought these)
Tristar 45 mppt controller $400 plus another $125 for display, battery temp sensor and shipping
8 awg wire  250 foot roll $450 plus shipping?
Fed in to 12 volt system (battery bank)

This would give me about 31 nameplate amps out of the contoller for $1615 or $52 bucks an amp

Scenario two:

6 dmsolar 145 watt panels $960
Midnite Classic 200 $610 with free shipping
10awg wire 250ft roll $240
Fed in to 12 volt system

This would give me about 44 (46 minus 4%) nameplate amps out of the controller for $1810 or about $41 bucks an amp.

I think the Classic controller would also give me more flexibility for future expansion although I don't think I would ever need more than six panels. I could always upgrade the battery system voltage to 48 volt with some new batteries and a inverter purchase. But the panels are so cheap right now that it seems to be the cheapest upgrade to do.

Also thanks Vic for that link to the post regarding efficiency curves. As bob stated in the post some of the efficiency is lost regardless of the array voltage due to "idle losses", etc. But that was interesting about the losses due to higher voltages causing the unit to run warmer. Good catch finding that. Since I would be running a fair amount of wattage, my guess is I would not notice the losses from running the higher voltage into the 12 volt system.

Quote from: eajonesyk2 on September 12, 2012, 05:49:43 PM
According to the voltage drop calculator if I use the 10 awg wire it will only cause a 1.6% voltage drop with 6 panels in series. So even if I lost 2% more in the controller conversion, I am looking at less than 4% power loss. Does that sound about right?

You are confusing voltage drop with power loss.  With solar panels the voltage drop is proportional to the power being produced by the panels.   The power loss is proportional to the square of the power being produced.

That is why in many/most situations it is better to use the lower voltage array configuration... only at the highest power are the line losses greater than the controller losses. 

The idea of a power shed is good.  If you do that, be sure to run two conduits... one for AC power and a separate conduit for control and monitoring.   Another advantage of a separate power shed (even if it is right next to your house) is that you won't have to hear the noise (hum, fans, clicking relays) of your electronics.

--vtMaps

vtmaps,
although it is somewhat small in a well designed system, a v drop does pose a power loss. when one is putting out the high power via higher current (which you would want) the loss increases just as you state and it is less when the current is smaller. make no mistake about it that it is a power loss as whatever the v drop is multiplied by the current passing at that time is the power lost at that time. it is fluid just as you say in that it changes with the output of the pv(s). if the wire was not present and the pv output went straight to the load more power would reach the load to be available for it as the wire is a resistor even though it is a smaller one.

Hi vtmaps and niel,

Thanks for adding to the discussion. I am still having a hard time understanding the differrent terms and concepts with all this solar stuff. I think I understand the voltage drop that occurs in wiring and have done the calculations on running 10 awg from the array to the controller 100 feet away (200ft of wire). With the six dm solar panels wired in series the voltage drop was under 2%. So i am trying to get a number (%) of the loss that is going to occur in the controller? I have exchanged a few emails with Tom C at Midnite and I think he may be getting back to me with a number I can work with.

A gent over at the NAWS forum suggested 6% but I think this was including the voltage loss from the wire. I have read other post on this forum suggesting the Classic controller losses are greater than 6% by itself.

If I can get a somewhat firm number on the controller losses from running a high voltage array into a 12 volt system then I will be able to make decisions on how to proceed.

Ok looking back at the posting Vic linked to. boB provided this answer for somebody who had an array running 140vmp into a 48 volt system. He suggested 96 to 97% efficiency which is super in my book.

"Hi Stephen dv...    No efficiency "curves" quite yet, but depending on how much power you are putting out at those voltages,
the efficiency will be around 96% to 97% or so.  I am stating sort of on the low side here (as far as I remember), as I have seen higher but do not want to OVER state.
At low output powers, say, less than a couple hundred Watts, you might get below 90%, but at higher powers, you will be in the high 90's percentage wise.

I am also assuming a Classic 200 for this usage where the MPPV can be around 160 Volts with a 200V Voc

At low output powers, the idle losses become more important and eat into the efficiency numbers.
Also, the higher voltages eat into efficiency somewhat because the controller  will run warmer.

I hope to get efficiency curves for many different powers and voltages real soon.

boB"[/i]


So could I expect similar efficiency for my planned system of six 145 watt panels wired in series (18.7 vmp) going in to a 12 volt system?

wrong voltage drop calculation.

I think my v-drop calculation was off because I didn't use the total length of circuit (200ft instead of 100ft). so using 10 awg will yield a voltage drop of just over 3%. I would be happy with that if the controller efficiency is still decent converting the 108 or so volts down to the 12 system voltage.

Quote from: eajonesyk2 on September 14, 2012, 01:04:22 PMI am still having a hard time understanding the differrent terms and concepts with all this solar stuff.

With the six dm solar panels wired in series the voltage drop was under 2%. So i am trying to get a number (%) of the loss that is going to occur in the controller?

Solar panels essentially provide full voltage from dim light to full sunlight.  The current is what varies with insolation.   

Therefore to calculate the voltage drop in the cable use ohm's law.   Your cable has a resistance of 0.2038 ohms.  With 6 panels in series (Vmp= 112.2) the current at full power is 7.75 amps.  Thus the voltage drop is 7.75 times 0.2038 = 1.579 volts.  That voltage drop is 1.41% of your Vmp. 

The power that lost in the cable is the current times the voltage drop = 7.75 times 1.579 = 12.241 watts.  That is 1.41% of your 870 watts.

If you drop to half power use the above formulas, but that 7.75 amps is now 3.875 amps:

thus, the voltage drop is now 0.790 volts which is a 0.70% voltage drop.  The power lost is 3.06 watts which is 0.70% of 435 watts.

When you are producing only a few hundred watts of power, the efficiency curves for the classic may be in the low 90's.  Lets say 94% efficiency.  That means 6% power loss.  Even when it is running at 97% efficiency, that's a 3% power loss.   How do these losses compare to you cable losses?

If you calculate the power losses for two strings of three panels (Vmp=56.1  Imp=15.5) you can see what happens to the cable losses.  Unfortunately, I don't have the efficiency curves for the classic, so I don't know what will happen in the controller. 

I am familiar with the Outback controller efficiency curves.  When I compare configurations on that controller, I am surprised at how often the lower voltage array configurations win out over the higher voltage configurations.

Bottom line: I think that 200 ft of 10 gauge cable is a bad situation, and I suspect that putting 6 panels in series is making the best of that bad situation.  A better situation would be heavier cable and 2 strings of three panels in series.

--vtMaps

Thanks vtMaps,

Very good info thank you. Now i can skip the calculators and do the wire loss math myself. As a matter of fact I did the math on the six panel system using the 8 awg that was recommended by cariboocoot on the naws forum. If I did the math right, my wire loss would be about 7 watts instead of of the 12.241 you came up with using the 10 awg.

Now you mentioned the 10 awg as being a "bad situation" but if our math is correct I would only be saving about 5 watts or so of power by going with 8 awg over 10 awg. Granted percentage wise this a 45% less loss in power but it is only 5 or so watts. The additional cost of 8 awg wire over 10 awg wire is over $200. So that works out to about $40 a watt. Does this sound about right?

I originally wanted a system that could generate 1 kwh of useable power a day during the warmer months and less in the colder months so I was ok with the 4 dmsolar 145 watt panels I already have. A morningstar tristar 45 mppt was recommended but would not have allowed any more panels to be added in series.

I understand a higher system voltage will allow the controller to work more efficiently but it would require new batteries and a new inverter and I am not sure the cost would justify the benefit. That is why I was trying to find out what the losses would be in the controller. I think you are probably close with your 94% estimate.

So lets use the numbers we have so far and look at the potential power from the system. I will start off with 6 dmsolar 145 watt panels wired in series for 870 nameplate watts.  Let's say actual watts may be closer to 670. Running 10 awg wire from array to Classic 200 controller will rob about 13 watts so I am down to 657 watts. Using a 94% efficiency rate for the controller will bring my power down to 617 watts. Let's say the xantrex prosine inverter acheives a 80% efficiency rate. That will leave me with very close to 500 watts of useable power per hour of full sun. In closing two hours of full sun a day would give me my 1kwh of power needed.

Dose this sound reasonable? 

Quote from: eajonesyk2 on September 14, 2012, 10:16:33 PM
Now you mentioned the 10 awg as being a "bad situation" but if our math is correct I would only be saving about 5 watts or so of power by going with 8 awg over 10 awg. Granted percentage wise this a 45% less loss in power but it is only 5 or so watts. The additional cost of 8 awg wire over 10 awg wire is over $200. So that works out to about $40 a watt. Does this sound about right?

yes, but you are comparing 10 gauge to 8 gauge, but only looking at 6 panels in series.  What happens if you make two parallel strings of three panels in series with 8 gauge wire?  Either of us can calculate the cable loss, but neither of us knows what the controller will do.  The only thing for sure is that the controller will run cooler and more efficiently.   btw, two strings through 8 gauge wire will lose 30.8 watts at full power.  That's about 18 watts more than one string through 10 gauge.  Will the increased efficiency of the controller make up for that?  (I don't know)  Remember, these numbers are at full power.  The cable losses will be much lower at more typical power levels but the efficiency gains in the controller will persist at the more typical power levels.

When I said "bad situation" I meant that trying to push 800 watts through 200 ft of 10 gauge wire is a bad situation.  Wiring your 6 panels in series is probably the best way to deal with 10 gauge wire.  Better would be to use heavier wire and then you would not have to wire 6 panels in series.

--vtMaps

eaj, there are other factors to consider, wrt the daily amount of input you will get, and that is the time of year and your location.

High latitude = shorter days and colder temps.

Will you get a full 2 hours of input on Dec 21?
(I note you will have the array a godly distance from the house just to get full sun)

Do you get -40* temps?

hth

Hi hth,

The property is located in a high desert area but heavily forested (mostly ponderosa pines) at about 5200 ft ele. While we get sunny days a vast majority of the year winters are quite cold and snowy at times. Snow levels average about 4 feet a year and some of that snow stays on the ground from about December through March. I have owned the property since 2008 and the coldest temp I have seen was -12 f. Usual low winter temps are in the teens with the coldest temps early am (5:00 to 7:00 am). Daytime winter temps average in the 20s and 30s

Reason for array being located a 100 ft distance from house is because of trees. Some are mine and some on neighboring property. Cutting them down is not a likely option:). The answer to your question about 2 hours of sun on Dec. 21 is I don't know. I don't live on the property right now. Most of my trips down there occur between May and October. I will be making a trip down there this Dec to check out the potential locations of the array.

I am not to worried about the sun that late in the winter because my power requirements will be much lower in the winter. Power requirements will be highest in the summer which will be when I have more sun then I know what to do with.

Spring and Fall should be fine too. I was just down there on Labor Day and the 100ft location was still getting sun from around 10:00 to 5:00. The current panel on my roof was only getting sun from about noon to 3:00 and the sun was not to far from going behind the southern tree line. By the end of October I expect the panel on the roof to be shaded by tree line except for a small clearing where driveway is, maybe an hour of clear sun at best. By Dec 21 panel on roof may have no direct sun at all.

I took pics at 15 minute intervals at each solstice and equinox, should have used a tripod so they would have melded together to 'watch ' the shadows move. .  Very interesting.  Picked up a couple of faint shadows I had not 'seen' previously.  Those faint ones can make significant drops in output , especially if in series.
Will do it again this year as some trees are now gone

Quote from: vtmaps on September 15, 2012, 05:54:10 AM

  Remember, these numbers are at full power.  The cable losses will be much lower at more typical power levels but the efficiency gains in the controller will persist at the more typical power levels.

--vtMaps

However,  when the CC is in a voltage-regulating mode,  the load on the PV is lighter,  and  the string V rises,  with some CCs,  this Vin rise  can cause more heat loss in the CC,  than when it is in Bulk or when in EQ prior to achieving the target V.  It is conceivalbe that under this situation,  that added losses in the CC due to increased Vin could be higher than the wire loss increase twix Absorb and Bulk.  This appears to be some exponential relationship of Vin -- it is not linear with the change in Vin.

While I have never used ALunimum wire for power distribution,  it might be worth considering a somewhat larger guage AL wire for this run -- taking proper precautions with terminal compatibility,  and use of NoALox or similar compounds.

Will look at the pricing of 8 GA THHN Building Wire,  BUT,  last 10 AWG THHN Stranded that I bought was US$ 105/500 ft....

EDIT;  8 GA  at HD is $188.  for a full 500 ft spool.
Thanks eaj for the added info about your location .. end edit
 
Note eaj,  if you could add a signature line in your profile,  it would help all of us to make better guesses, Your location,  etc are  important items. Thanks,  Vic

...  AND, eaj,    Do realize that you could well  be planning on NOT running these PV conductors in conduit,  such that THHN wire prices may not really apply.   Just going on how I'd probably do the job.
YMMV,  Vic

Vic, I will work on a signature line.

I was planning on using wire that is suitable for direct burial (not planning on running conduit). I was basing wire cost on prices found at solar-electric.com. Actually I completely goofed my estimated wire cost. I was thinking I would need 200ft of wire but disregarded the fact that the wire is two conductor wire so I will only need 100ft. The 10 awg wire I am looking at is 1.20 a foot so $120 for the run. The 8 awg is 2.26 a foot or $226 for the run. So I would only be saving about $100 bucks going with the 10 awg.

So I guess I can stick with the four panels after all and run the 8 awg wire. This would give me about a 3% voltage drop but more efficiency in the controller? 6 awg is definitely not in my budget so 8 awg is as good as it will get. I could squeak by with a 3%v-drop for now, especially if the controller is running a little more efficiently. Running two strings of panels would require 2 awg wire to keep the v-drop under 2% and I am pretty sure that will never be in my budget

I will purchase the Classic 200 either way in case I want to add two more panels at a later date.  Hopefully I can increase my battery system voltage to a 48 volt bank down the road and then maybe add two more panels in series at that time, if needed.

Hi ea ..

Thanks for doing the signature,  it should save you quite a bit of time answering questions.

Do you already own the Deka Gel batts?
Gells are fragile batteries.  They must be charged very carefully.  The Absorb voltage setting has a tight tolerance,  and MUST be temperature compensated on every charge source.   Usually,  the maximum charge current must be limited to about 5% of 20 Hr capacity.  Please carefully read and follow the recommendations of the mfg.

You may know these things and more about sealed batteries.  Usually a good Battery Monitor is recommended,  as there is no way to directly measure the SG  or SOC in an active Off-Grid system.

If you do not yet have these batteries,  you may want to think some more about the batt purchase.  YES,  I do have an attitude.
Good Luck,   Vic

Quote from: Vic on September 15, 2012, 02:47:49 PM
However,  when the CC is in a voltage-regulating mode,  the load on the PV is lighter,  and  the string V rises,  with some CCs,  this Vin rise  can cause more heat loss in the CC,  than when it is in Bulk or when in EQ prior to achieving the target V. 

I think that is one more argument in favor of using the lower voltage array configuration.  If the Vin is going to rise on a 12 volt system, I would think that it is better to start rising from 56 volts than from 112 volts.

Quote from: Vic on September 15, 2012, 02:47:49 PM
It is conceivalbe that under this situation,  that added losses in the CC due to increased Vin could be higher than the wire loss increase twix Absorb and Bulk.  This appears to be some exponential relationship of Vin -- it is not linear with the change in Vin.

When the CC is in voltage regulating mode, the current is decreasing while Vin is rising.  Both of those will reduce the cable loss.  What that will do to the controller is unknown for the classic.  On the outback the answer to the question depends where on the power curve you are.  The outback hits its peak efficiency at about 300 watts.  If you are producing 600 watts when voltage regulation begins, the rise in Vin will make the outback less efficient, but the reduction of power being drawn from the panels will make the outback more efficient.

--vtMaps

vt,

YES,  this is what I am saying,  indirectly.

This string V is high,  especially on a 12 V system.

Yes,  was saying that the lower input current does cause less cable loss,  BUT probably even more loss in the CC (due to Vin rise),  than cable loss.  This may seem somewhat moot,  (as by definition,  at this point,  the unused power is unneeded) --  if the power is not needed,   then system losses must not matter -- but the tehrmal impact is negative.

AND,  my exprience with the MX-60  with almost identical string V as the proposed system by eaj,  is that the only time that the internal fan in the MX runs,  is under quite light CC loading -- usually in Float,  delivering a few hundred watts,  but  since Vin has risen to 120-ish Vin,  the CC needs to rid itself of more heat than when delivering 2KW,  causing Vin to be dragged considerably lower.

Know that this seems like Angels and Pins.

YES,  believe that if there is a reasonable way to run a more moderate Vin,  especially on a 12 V system,  it should be done.

IMHO,  I would strongly consider running the PV conductors in PVC conduit,  allowing the use of,  perhaps,  AL cable,  which could be quite a lot less expensive.  AL  is harder to pull,  and would probably require some extra splicing giblets to make transitions from PV/CC leads to AL,  etc ...  This could still reduce the cost of increased wire size to support lower Vin with increased currents.

However,   the higher voltage Classics may well be more well-behaved at high Vin than the venerable MX and other CCs.  One might guess that this is the case,  but dunno.

Final thing is the ambient temps in the place where the CC is located.  Cooler temps are better,  especially with high string V.  The power rooms,  here have priority for A/C,  but would guess that with the proposed system of eaj,  there is NO power budget for A/C.

Final thought,  all CCs,  and other power electronics make some acoustic  noises.  It is a very good idea to try to locate these items away from living areas if at all possible.   Vic

Vic, yes I already have the gel batts. I bought a small used pv system from a guy who had used the system for a prototype portable charging station he was trying to build. He ended up giving up on it and I bought the components. I like the idea of the gel batts for various reasons but now understand the charging limitations. When it is time to replace the batteries I will definitely purchase wet cells golf cart type. Since my budget is not expected to loosen up for a while, I have to work with what I got for now.

Regarding cc operating temps. During a majority of the year the local ambient temps are somewhat cool so that should help the cc maintain reasonable operating temps. During July, August and early September when the ambient temps will be warmer the panels will also be warmer and that should cause the panels to operating at a lower voltage, which should cause a little bit less heat stress on the cc. Does that sound right?

I will price out some other wire options including al and running conduit but if my v-drop calcs were right I would need to 2/0 copper or 4/0 al wire to run two strings of two panels each at 200ft of total wire length. I just can't see how these increased wire costs could possibly be justified with increased cc efficiency and reduced wire losses. Maybe on a much larger array this could be the case but not on 600 to 900 watt system. I will be happy to hear I am wrong?

This post is getting a little long and I thought that I would post a new post with some of the recommendations I have received so far and see if I can get some feedback on costs per watt. Thanks all for the help.

I will post my request for estimated costs per watt feedback over at the naws forum since the subject isn't mainly related to the midnite controller. thanks again for all the help here.

Quote from: eajonesyk2 on September 16, 2012, 01:27:54 PM
Maybe on a much larger array this could be the case but not on 600 to 900 watt system. I will be happy to hear I am wrong?

eaj, I am seeing a disconnect in you statement above and your sig line.
If you are planning this install to be a permanent home you are investing in the future by putting in wiring that will NOT have to be replaced or dug up or ??  You may not need it right now but you save yourself a lot of work and double expenditures...
for example what if you just want to add 50% more panels, what are the implications to losses by having smaller wire?
One sure thing is that loads grow over time... thus so may your PV
your choice..

Hi ea.

Great on the batteries.  I do like that they appear to GC2 sized 6V batts.

It is easy for folks like to snipe at others' systems.  We all do run what we got FBOW.

You will probably do fine with the batts.  IIRC  this makes two strings @24 V.  Just try to keep strings balanced by  using good,  well made batt cables of equal lengths.  A Clamp-On DC  Ammeter can help by allowing easy measurement of individual string currents.

Much of what I mentioned about reduced efficeincy of CCs relates to the prior generation(s) of CCs.  The HV Classics have been tailored for more efficient operation at higher Vin,  so they would have to be better at the high Vin operation.

There are losses in every system.  The only thing,  is that the losses accumulate every day,  and accumulate over time.  But,  PV is getting less expensive,  and if one has enough PV,  and if losses do not over-stress components,  then,  ignorance in my case IS bliss!  Although, from an investment standpoint,  additional Cu  pays off every day of operation,  to the point of diminishing returns etc.

If you find a use for almost every watt available from PV,  then this rise in Vin will not be a problem,  as the PVs will almost always be heavily loaded.  Mileage WILL Vary.   Vic

Hi westbranch,

Sorry for any conflicting ideas i have thrown out there regarding the system usage. This will be my permanent home upon retirement or possibly layoff. I have been lucky to have a caretaker staying at the property since the winter of 2009/2010. The cabin was broken in to twice before that. Besides security he has provided me with the awareness of what is necessary to survive down there. Up until Labor Day weekend he had zero watts of ac power available to him, except when I visited solo and he was able to stay during my visit. He has made do with alkaline batteries for the lantern, radio and nature provided refrigeration during the winter. He doesn't even use propane. He seems perfectly content and is happy to have a warm place to stay (wood stove for cooking and heating rooms and water). I desire a little bit more of the creature comforts the rest of us are used to. But I have no expectations of living like I do know with the only power concerns being the price of the electric bill each month.

Anyhow I installed a single 64 watt panel on the roof of the cabin Labor Day weekend and hooked it up through a simple shunt type controller and in to a 12 volt marine/rv deep cycle battery. I supplied him with a 150 watt Samlex pure sine inverter so he can run the CFL bulb in the cabin now and charge a nicad battery pack for the lantern. I also suggested he buy some rechargeable AA batteries for the radio which he can charge through the inverter as well. I won't be back down there until October to see how it is working for him but I haven't heard of any problems yet.

Moral of the story. I expect to live just fine with 1kwh or less a day so any future upgrades to the planned pv system will be very limited. I will take in to consideration the maintenance and life expectancy of wire I put in the ground so maybe conduit will work it's way in to my system after all, we'll have to see.

Thanks for your input. Also check out my planned systems over at the naws forum to be posted later today, I would really like to get the cost per watt nailed down.

Thought I might as well throw these potential pv systems up here too since you guys helped give me the basis for arriving at these numbers. They all include the Classic 200 CC as well. Let me know if these useable watt expectations and costs sound reasonable. Also posted at naws, thanks.

Proposed System 1. Estimated watts coming out of controller 407 watts/component cost $836=$2.05 per useable watt.
Calculated useable watts as follows: Name plate panel watts 580x.77=446watts Wire loss 446x.97=433 watts CC efficiency 433x.94=407watts
4 dmsolar 145 watt panels wired in series.
100ft of two-conductor 8awg wire ($226 plus shipping)
Midnite Classic 200 charge controller ($610)



Proposed System 2. Estimated watts coming out of controller 510 watts/component cost $1050= $2.06 per usable watt
Useable watts calculated as follows: Name plate panel watts 870x.77=584 Wire loss 584x.97=567watts CC efficiency 567x.90=510 watts
6 dmsolar 145 watt panels wired in series. ($320)
100ft of two-conductor 10 awg wire ($120 plus shipping)
Midnite Classic 200 charge controller ($610)



Proposed System 3 Estimated watts coming out of controller 420 watts/component cost $1260= $3.00 per useable watt
Useable watts calculated as follows: Name plate watts 580x.77=446 watts Wire loss 446x.97=433 watts CC efficiency 433x.97=420 watts
4 dmsolar 145 watt panels wired in two strings of two panels
200ft of 2 awg welding cable ($600 plus shipping)
pvc 1 1/4 inch conduit 100ft ($50)
Midnite classic 200 charge controller ($610)