DIY MX60

[commanda]

We all know solar panels are expen$ive, and therefore we need to extract every possible joulle (or is that jewel?) of energy we can. Whilst an MX60 is a fine piece of engineering, I have read comments on here by at least several people who have managed to destroy them. Replacement cost for those of us downunder (Australia) being in the order of AU$1000 each, I figure there's got to be an alternative for those of us proficient at things electronic.

Working forwards from this article http://www.acs.comcen.com.au/buildppt.html where he shows that it is only necessary to maintain the cell voltage at 17 volts (thereabouts) for a 12 volt panel, and marrying this idea to an ex-telecoms 48-24 volt 1KW inverter which I picked up from my friends at oatley electronics





for $44.00, I plan to put an op-amp between the cell voltage and the output voltage adjustment pot on this power supply. This will effectively adjust the ouput current to maintain a constant input voltage. This page http://www.oatleyelectronics.com/powersup.html and find SMPS1 will give more details.

Now, if the mppt point for a typical 12 volt panel is about 17 volts, then 3 panels in series will give me 51 volts, well within the input voltage specs for this inverter. Ultimately, I'd like to have 3 banks each with 3 x 80 watt panels. 720 watts, which should approach the max efficiency point of the inverter, if it follows the typical inverter.

I had originally planned to implement this with the MPPT controller I talked about previously http://www.fieldlines.com/story/2006/3/12/14840/1315 but I'm thinking this simplified appraoch is closer to the KISS ideal.

More details later after I finish the (almost finished) dump load controller.

Amanda

[dinges]

Could be an interesting project. KISS indeed, if you've got a 48-24V inverter; if you don't... Too bad we don't have Oatley's over here. Then again, I know the dumpster of our Dutch telecom company. Many nice goodies there too (notice that 19" rack for the dumpload controller? Guess where that came from). Also had some nice 48->5, +12 and -12V inverters in them (only at a few amp though).

Only problem I have with ex-telecom equipment is that they use 48V (good) but positive ground (bad), which often requires a bit of rewiring or clever thinking, to prevent magic smoke. Hope that 48V inverter of yours isn't neg. ground, or at least has completely separate input & output circuits (no common ground). The telecom companies probably have good reason to opt for positive ground, but it beats me.

Will follow the project with interest.

Peter.

[dinges]

Ehm...

I notice in the datasheet that the converter works at an efficiency of >81%. Let's just assume that it's 81%. This means that the MPPT should be able to get at least about 120% of energy out of your current panels (without MPPT) to just break even, energy-wise.

A quick calculation (panel voltage 17V; batteries 13.8V; 17/13.8=1.23 (i.e, you'd get 23% more energy out of your panels with MPPT. Or am I oversimplifying here?)

Total system efficiency: 123 x .81 = 99.63%... You lose...

Even if you'd gain maybe total 10%, I'd still don't think it would be worth the trouble; your opinion may differ.

Apparently, the REAL efficiency of that DC-converter (as opposed to the stated minimum 81%) has a major impact on the question whether the system is worthwhile to pursue.

Peter.

[Nando]

AMANDA:

Can you check http://www.acs.comcen.com.au/buildppt.html , I can not open it !!

Nando

[commanda]

720 watts, which should approach the max efficiency point of the inverter, if it follows the typical inverter.

Good point on the quoted efficiency of >81%, of which I am aware. Gotta test it. Also, looking at the circuit, it is 2 units in parallel. Initially, I may disable one half until I get a full complement of panel$.

Amanda

[commanda]

My apologies, apparently that site has disappeared. Good thing I printed the article. I'll try & scan the relevant bits & post it as a pdf.

Amanda