[RE-wrenches] DC wire sizing

[R Ray Walters]

All of us agree our goal is to increase total production and reduce cost of the system over its life. Basically : total lifetime KWH production /  total system cost. 
If I can increase the production for the dollar spent by using bigger wire, I'll do that, and if it makes more power for the money to add a module, I do that.
Basically, I think the volt drop issue comes down to the following:
Inverter to grid not more than 1.5% drop.
Not more than 5% total losses ( code suggested, but not required)
So that leaves up to 3.5% losses from the array to inverter.
anything under 1% begins to get really expensive, so I'm playing in about  +/- 1 either side of 2% for best economics.
Its all about designing the most system for the money: I do this same cost benefit math for framing and layout, tracking, module choice, MPPT, inverter choice, wire choice, etc. 
Surprised everyone's getting so concerned over a few watts on the line, when module, inverter, and layout choices will have a much larger effect on long term production.
(ie. I spend the money saved on wire to buy better monocrystalline)
 
Ray

Friends don't let friends buy amorphous......


On Apr 7, 2010, at 5:09 PM, Joel Davidson wrote:

> Hello Ray,
> I agree with Bob-O. Smaller wires between the inverter and the grid may meet ampacity requirements and your I2R design goal, but high grid-tie voltage has been a problem. Also, 1% of lost power over the life of a PV system adds up especially when grid power continues to go up in price. I agree your reasoning, but I also agree with Feed-In Tariff PV system operators who want every kWh possible and are willing to pay the up-front cost. I'm sticking with less than less than 1% wire loss between the inverter and the grid and less than 2% overall wire loss unless the customer explicitly states that he wants more loss.
> Joel Davidson
> 
> ----- Original Message ----- From: "Bob-O Schultze" <bob-o at electronconnection.com>
> To: "RE-wrenches" <re-wrenches at lists.re-wrenches.org>
> Sent: Wednesday, April 07, 2010 1:50 PM
> Subject: Re: [RE-wrenches] DC wire sizing
> 
> 
>> Ray,
>> Sorry, I'm not buying that argument over a 25+ year design lifespan. Also, -and perhaps something many folks don't consider- the NEC requires a MAX loss of ≤5% over the ENTIRE circuit. That means all the way to the mains panel. Best practices would require no more than 1.5% VD between the inverter and the mains so as to avoid any potential of overvoltage disconnection by the inverter. As it tries to push the current into the grid, it also pushes up the voltage on that line and on it's own sensors. Obviously, the larger the current flow and the higher the VD, the worse the situation could be. Bear in mind that the grid is not held to the same over and under voltage specifications as your inverter is. 3% + 1.5% is pushing it just a bit too close for comfort and I still think wasting watts in wire losses is bad design. We agree that orienting PVs to an azimuth other than 180° and a bit less than latitude elevation is less than ideal, but you "run what you brung" in terms of the orientation and roof pitch of a structure. Better to take a hit on production if we're talking about the backasswards method of incentivizing by the installed watt, than not installing PV at all. That said, we DO have control over wire sizing. IMO, throwing away watts forever just to cheapen it up a very little bit upfront is poor economy and as the price of gird supplied power increases over the years, the waste and lost revenue is even more acerbated.
>> Best, Bob-O
>> 
>> On Apr 6, 2010, at 6:04 PM, R Ray Walters wrote:
>> 
>> Just run the numbers sometime. Compare the cost difference of #6 vs. #4 wire say, and then look at how many more watts you're actually saving, then multiply that additional wattage by the installed cost per watt.
>> Very simply, once you've satisfied code requirements, there is a point at which it is cheaper to add more panels than use bigger wire.
>> Also, that point is a moving target that fluctuates with PV and wire costs. I've found recently for our projects, that that point is falling at about 3% loss.
>> I also include a cost factor for oversizing the conduit, extra labor (bigger wire is harder to handle), and any connectors needed to land the larger wire.
>> I've got very well designed systems working for decades, using this method.
>> This is how large commercial systems are designed as well.
>> You can't simply pull 1%, and then call us bad designers because we actually do an economic analysis for each wire run.
>> It used to be unheard of to install PV facing anything but due south at latitude tilt, but now we know to add a few more modules. Same concept.
>> 
>> R. Walters
>> ray at solarray.com
>> Solar Engineer
>> 
>> 
>> 
>> 
>> On Apr 6, 2010, at 1:37 PM, Bob-O Schultze wrote:
>> 
>>> Ray,
>>> A 2% wire loss is the generally accepted metric for battery based systems with relatively low PV voltage input (<150Voc). It's just plain bad design to accept more than a 1% VD on higher voltage systems. PVs ain't THAT cheap.
>>> Best, Bob-O
>>> 
>>> On Apr 6, 2010, at 11:44 AM, R Ray Walters wrote:
>>> 
>>> 
>>> Once I have fulfilled NEC min. requirements, I use a spreadsheet to analyze the cost of larger wire vs. the cost of power lost. Going under 2% is usually not worth it, if copper prices are high, and PV cost is low enough (current market). Sizing for under 2% was good economics a few years back, when PV was high, and copper was low, though.
>>> 
>>> For NEC 2011, I agree: while I readily welcome development of DC AFI, implementing code before the technology is ready, is a bad idea. But that may be the only way to get the technology in place.....
>>> 
>>> Ray
>>> 
>>> On Apr 6, 2010, at 11:15 AM, Kent Osterberg wrote:
>>> 
>>>> Ray,
>>>> 
>>>> Considering that we design PV wiring to be efficient with voltage (and power) loss typically less than 2%, the wire size is nearly irrelevant to arcing issues.   Essentially all the energy available from the PV array can be dissipated in the dc arc.   And since the current is limited by the nature of the IV curve, breakers alone usually won't clear the fault.  The best combiner breakers can do (if you have enough parallel circuits) is isolate the fault to one string in the PV array. With one string being 1 or 2 kW in many systems there is still the potential for a lot of heat.
>>>> 
>>>> With the 2011 code just around the corner and no dc arc fault protection on the horizon, it looks like our industry is again going to have a code requirement that no one can fulfill.
>>>> 
>>>> Kent Osterberg
>>>> Blue Mountain Solar, Inc.
>>> 
>>> _______________________________________________
>>> List sponsored by Home Power magazine
>>> 
>>> List Address: RE-wrenches at lists.re-wrenches.org
>>> 
>>> Options & settings:
>>> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>>> 
>>> List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>>> 
>>> List rules & etiquette:
>>> www.re-wrenches.org/etiquette.htm
>>> 
>>> Check out participant bios:
>>> www.members.re-wrenches.org
>>> 
>>> 
>> 
>> _______________________________________________
>> List sponsored by Home Power magazine
>> 
>> List Address: RE-wrenches at lists.re-wrenches.org
>> 
>> Options & settings:
>> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>> 
>> List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>> 
>> List rules & etiquette:
>> www.re-wrenches.org/etiquette.htm
>> 
>> Check out participant bios:
>> www.members.re-wrenches.org
>> 
>> 
>> 
>> _______________________________________________
>> List sponsored by Home Power magazine
>> 
>> List Address: RE-wrenches at lists.re-wrenches.org
>> 
>> Options & settings:
>> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
>> 
>> List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
>> 
>> List rules & etiquette:
>> www.re-wrenches.org/etiquette.htm
>> 
>> Check out participant bios:
>> www.members.re-wrenches.org
>> 
> 
> _______________________________________________
> List sponsored by Home Power magazine
> 
> List Address: RE-wrenches at lists.re-wrenches.org
> 
> Options & settings:
> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
> 
> List-Archive: http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org
> 
> List rules & etiquette:
> www.re-wrenches.org/etiquette.htm
> 
> Check out participant bios:
> www.members.re-wrenches.org
>