Grid tie panel for off-grid without MPPT

stephendv
stephendv Solar Expert Posts: 1,571 ✭✭
Right now the prices for off-grid panels in Europe are quite steep, in the order of 0.8 tokens/Watt. But I can find grid-tie panels through wholesalers/surplus/left overs/etc for around 0.3 - 0.5 tokens/Watt.

I'm designing a small system of about 1.5kW and was wondering if there are problems with using a mixture of panels in order to keep costs down, but also keep the battery absorbing and EQing in summer. Looking at the IV curve of a typical grid tie panel (bottom of: http://www.solarsystems-usa.net/solarpanels/jetionsolar/jt230pce/#.Uw3Ee_RdVF4), 230W with Vmp of 30V and Voc of 37C. In summer at a panel temp of 60C, there's a 10% V drop and adding another 2% losses for cables and charge controller, that would bring the Vmp down to 26.4Vmp and Voc to 32.5V.

Firstly absorb at 28.8V:
Since 28.8 is between Vmp and Voc, the question is, how many Watts reach the batteries at that voltage?
28.8V is 10% away from the temp corrected Vmp, so looking at that IV curve, the point at 33V STC would be equivalent of the temp corrected 28.8V (if you're still following).
So 33V at STC seems to produce upwards of 200W, which really isn't so bad at all. Which seems to say that using on-grid panels in an off-grid 24V setup will be fine for absorb.

Next up EQ at 31V:
31V is 17% higher than the temp corrected Vmp (26.4V). Which puts it at 35V on the STC chart, which seems to produce 150W. So a 1.5kW array using grid-tie panels only has about 900W available to maintain EQ at 31V in summer.

This seems fine, so where have I gone wrong?

As an alternative, if the tolerances are a bit too tight, then why not split array into 10% off-grid panels and 90% grid tie? That way the off-grid's just provide the extra V boost needed in summer to complete an EQ.

Comments

  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie panel for off-grid without MPPT

    Well let's see. With Voltage above Vmp a panel's current output drops like a rock. So what you get is the "30 Vmp" panel pulling the batteries up to whatever Voltage it can, and as it does so the current falls lower and lower because even that 30 Volts isn't actually making it to the batteries even with the best of wiring on a cool day.

    So first of all the hot panel doesn't put out its Vmp. Second you lose more Voltage in the wiring. Third the charging Voltages desired are above the eventual Voltage that reaches the panel.

    How much power will the batteries actually get then at Absorb Voltage level? Try none. You're asking for 2 Volts above the effect Vmp at that point, by which time the I*V curve of the panel has fallen to the point where it will be totally ineffective. Never mind EQ because if you don't get the panel charged you can't do an EQ.

    Try this simple experiment: attach a 30 Vmp panel to two 12 Volt batteries (fully charged at 25.5 Volts) and measure the Voltage and current. Even when the panel is producing current and charging the Voltage is at battery level, not Voc. It never gets there.

    I know this because of the large number of people who have tried using 30 Volt panels in this manner and have had to replace their batteries in the first year because they were never, ever fully recharged.

    As for using some 35 Vmp panels and some 30 Vmp panels ... did you ever read any of the discussions about the effect of Voltage differences in panels in parallel?

    My advice is don't be a cheapskate and do it right. A 1.5 kW system is not all that small and would fully justify the cost of an MPPT controller (it would produce about 48 Amps).
  • vtmaps
    vtmaps Solar Expert Posts: 3,741 ✭✭✭✭
    Re: Grid tie panel for off-grid without MPPT
    stephendv wrote: »
    I'm designing a small system of about 1.5kW and was wondering if there are problems with using a mixture of panels in order to keep costs down
    <snip>
    As an alternative, if the tolerances are a bit too tight, then why not split array into 10% off-grid panels and 90% grid tie? That way the off-grid's just provide the extra V boost needed in summer to complete an EQ.

    I'm not exactly sure what you're trying to do... Are you planning to use a single PWM controller for a mixture of panels? If you're trying to save money, why not buy all grid-tie panels and use an MPPT controller?

    --vtMaps
    4 X 235watt Samsung, Midnite ePanel, Outback VFX3524 FM60 & mate, 4 Interstate L16, trimetric, Honda eu2000i
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie panel for off-grid without MPPT

    And in case anyone is wondering, 1500 Watts @ 30 Volts is 50 Amps and that current at that Voltage would have to travel from the array to the charge controller which may be a fair distance making the Voltage drop a significant issue requiring huge and expensive wire because you need 8 AWG just to handle the current. 8AWG would only do the job for about 10 feet. I'm thinking this is a real 'false economy' here.
  • stephendv
    stephendv Solar Expert Posts: 1,571 ✭✭
    Re: Grid tie panel for off-grid without MPPT

    Thanks all. vtmap, yes I'm trying to use grid tie panels with a PWM controller, no MPPT. The thing is that the client is considering one of them fancy new inverter/charger/charger-controller all in one units: http://www.hmsistemas.es/shop/catalog/inversorcargador-hibrido-tab-vx1-2000-p-274.html which has a PWM controller built in.

    Perhaps grid tie panels, some on the built in PWM charger, and the rest on a small MPPT charger? Will have to contact suppliers and check the difference in price between small and large MPPT. Just saw that morningstar released a 30A version.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie panel for off-grid without MPPT
    stephendv wrote: »
    Thanks all. vtmap, yes I'm trying to use grid tie panels with a PWM controller, no MPPT. The thing is that the client is considering one of them fancy new inverter/charger/charger-controller all in one units: http://www.hmsistemas.es/shop/catalog/inversorcargador-hibrido-tab-vx1-2000-p-274.html which has a PWM controller built in.

    Perhaps grid tie panels, some on the built in PWM charger, and the rest on a small MPPT charger? Will have to contact suppliers and check the difference in price between small and large MPPT. Just saw that morningstar released a 30A version.

    Those all-in-one units sound like a good idea. Until you see how poorly they work. Avoid them.
  • stephendv
    stephendv Solar Expert Posts: 1,571 ✭✭
    Re: Grid tie panel for off-grid without MPPT
    Until you see how poorly they work. Avoid them.

    Interesting, could you expand ?
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie panel for off-grid without MPPT
    stephendv wrote: »
    Interesting, could you expand ?

    Let's start with your first encounter: a PWM type controller in it. Some of them have MPPT types built in, which you'd expect. Really though, in that sort of system size PWM doesn't make sense. Look at the questions you've already had trying to work out a way to get cheap panels to work with it. Saving money? I don't think so.

    There have been a couple of forum members (both from Australia as I recall) who got saddled with these things. Limited to the current output of the built-in controller, which was not as much as a stand-alone unit can handle. This restricts you battery bank size to what the company thinks is right. What's more they weren't fully programmable, which restricts your battery selection.

    That word "restricts" comes up often in examining these things. They simply do not offer the flexibility found in individual units. As you know each and every solar install is unique, thus you want the flexibility of being able to pick the components best suited to the particular application. Here you have an all-in-one which clamps down on that flexibility.

    You can always add an external charge controller as well, but then why buy one built-in? Might as well skip it and get the stand-alone units to start with.

    What happens if the built-in controller fails? Might never happen, but how easy will it be to fix? Controller on the wall fails you just swap another on in place. Inverter on the wall fails, same thing. You don't have to down the whole system while waiting for replacement parts or repair.

    Now if one of the 'reputable' companies made one of these with full functionality, total integration (including shunt-type battery monitor controlling both solar and gen charge sources), and in-field serviceability you'd have something.

    But to date my experience with them has been helping people try to get the things to work within the built-in limitations, and that has been frustrating. I don't remember just who it was that had them either, but there's a chance those threads can be found on here somewhere or maybe they'll even see this and chime in.

    Nice concept, but not well-executed.
  • stephendv
    stephendv Solar Expert Posts: 1,571 ✭✭
    Re: Grid tie panel for off-grid without MPPT

    Thanks Marc. There isn't much technical detail available online about this unit so have asked the distributor to provide it. The advantage of this unit is that even without the integrated charge controller, it's still significantly cheaper than other inverter/chargers of the same capacity. So after I get some firm prices in, it looks like I may go for external MPPT, and then just use this as an inverter/charger.
  • Cariboocoot
    Cariboocoot Banned Posts: 17,615 ✭✭✭
    Re: Grid tie panel for off-grid without MPPT

    I'd still be concerned about the fact you'd need all the array power (1.5 kW) to go from the array to the built-in charger as low Voltage at high current: with 30 Vmp the current will be 50 Amps and that means some very large wire. Normally you'd use an MPPT controller on something that large, put the Voltage up to 60 and drop the current to 25 which means less V-drop over the distance and smaller wiring. That little factor could be a killer right there.