Surely the PV input current is regulated by the MPPT and so the max (ie possible) output of the PV is actually irrelevant.
If this is not the case the designers have done a bad job. Equally, power dissipation should be internally regulated, for example if one or more fans fail and maximum output is compromised.
Whilst unstated, this is just the hallmark of a half decent design: resistance to self-destruction.
@farmeroz you are right in principle. But the way I understand it, ISC rating is all about safety, in case something unexpected happens. In the link I quoted, Guy says, and I agree with him, that during the panelsā lifetime, it should be expected (and designed accordingly) that the panels will short.
and what difference will that make, other than reducing the PVG max line voltage?
The MPPT always has a short circuit rating published in the datasheets. This is typically lower than the PV VOC or the charging current rating. For example, itās 70a for a 250/100. The question is whether you should design your PV array for the ISC rating or the charge current rating.
Isnāt the MPPT short circuit rating short circuiting the output, where it should be automatically controlled by protection (fast current limiting and then thermal/voltage shutdown) very basic. The PV should obviously be rated somewhat below the max input rating of the MPPT, because excess is wasted, although I guess there is an argument for over-rating in bright sunlight to cover partial cloudiness, although that may be questionable.
It looks like the short circuit current rating acts on the input, but I canāt confirm that. Thatās certainly the assumption of the whole discussion. Someone wondered what if the MPPT input is protected by a properly sized fuse or breaker? Someone else responded that by the time the fuse or breaker acts, it will be too late.
Yes, the SS rating must be the output. It should be protected by active internal current&power limiting (eg as per amplifier outputs are protected against short circuit). I do not know the internal circuitry but it will almost certainly be a PWM dc-dc converter (could be buck but same applies). This will have active output protection for its own survival.
Fuses or breakers would only operate after the device had already been destroyed, too slow and too crude in power level detection.
Note that the output has a very large capacitor, circa 0.2F. Please also note thread on burning out isolators that switch MPPT and inverters connected to battery. Not clearly documented in datasheets. Actually not documented at all.
Are you sure the SC rating applies to the MPPTās output and not its input?
On the thread I quoted, many posters asked about the input and SC rating and they were never contradicted.
No. The spec is the PV ISC, which means the input.
How can shorting an input bother anything? The voltage across the contacts will be zero, and the current zero so no problem.
What am I missing if that is the case? The whole point of a MPPT is that its controls the PV voltage-current characteristic to maximise the output power.
if you go with 3 or more parallel strings, each string need to be protected versus back current by a diode.
If possible, I would go with RS450/200 with 4 long strings. This eliminates the diodes and disconnect switches as build into RS and reduce the amount of wording and surge protection devices.
With your custom panel you could attach 5-8 panels in series on each MPPT in a RS450 with Isc=15A only. If you have longer PV wires, itās less of an issue.
20-32 Panels on RS450/200
10-16 Panels an RS350/100
In the example of a 150/85, the internal components can only handle a short circuit current up to 70A on the PV side. This is if an internal component of the MPPT fails with low resistance, for example if you surpass the max PV voltage.
If your array has a combined Isc of 60A, then there should be no further issue, the failed component fails safely.
If your array has a combined Isc of 80A, then theres no guarantee that the failed component does not lead to bigger damage.
Another possibility is a short circuit in another part of the installation, for example in one of the supplying cable pairs (positive and negative of the same string) leading to the MPPT. But since bigger MPPTs all use multiple MC4 connectors, its not uncommon that a shortcircuit in a cable pair to the MPPT can also mean that part of the shortcircuit current actually flows over the MPPTs PCB. Not necessarily through the DC/DC converter, but solely through the inputs that are all in parallel.
For example, three strings of 25A Isc in parallel, with the actual parallel connection done on the three pairs of MC4 of the MPPT, would mean that a short in one of the pairs would be fed from one side by its own panels, presumably by 25A as well as through the parallel connection from both other strings with 50A combined
You therefore need to use an external limitation of short circuit energy, a fuse or an MCB for example. Since theres no panel, at least to my knowledge, that can singlehandedly output 40A, your array will most likely consist of three or more parallel strings. In this case, you need to use fuses per string anyway, to protect each string from potential reverse current from the others. So very likely you already solved the external fusing for the MPPT simply by using inline fuses before paralleling the strings.
But the above example also shows why it can be a good idea to use both inline fuses close to the panels, as well as fuses close to the point of paralleling (if the two are physically apart, for example panels on the roof while point of paralleling is in the basement)
You are right. The problem I had initially was using 5 strings of 2 series panels each, with a total ISC of 71.2A, on a 250/100 charger which is rated for only 70A ISC. Thatās no longer an issue. I now plan to use 4 strings of 3 panels in series each instead. That will give me an ISC of 56.96A which is fine for a 250/100. However, over-paneling will be 137% compared to 114% before. Thatās not perfect but itās fine with me. All strings will be appropriately fused in the combiner box.
Thank you @BjoernK for the suggestion. I seriously looked at the RS450/200. It was good for me, as it would eliminate the need for a string combiner. Then I read on this and other forums that the problem of balancing the trackers, given the RS limit, is very common. (Any of the trackers completely shuts down when not needed).
Mainly for that reason, and for reasons of redundancy, I decided to go with two 250/100 instead.
All my strings will be fused appropriately in the 2 combiner boxes. I also have a moulded case breaker in the line coming from the combiner box to the charger but I will use that simply as a disconnect as its voltage is way off (1000v!). I couldnāt find anything locally with a more appropriate voltage.
Update: I have surveyed the market again and found a 1-pole Chint MCB of 250v and 63a. Iām expecting a max current of 57a. I could use that just before the 250/100 instead of the high-voltage MCCB. I would prefer a 2-pole MCCB but I can live with that.
I really struggle to see what your problem is.
I have 16 x370W panels in two series strings, each going into one of the two inputs of a 450x200.
If the MPPT fails itās dead meat and as far as I am aware none-repairable and all that is required is to put a suitable fuse and a fire alarm, although most likely the MPPT will simplyhave died and gone OC.
My panels have a sc current of ~8A and are fused with a 10A fast fuse each, simple and easy. I never expect this fuse to blow. The high output voltage of the string (~300) means lower cable losses between panel and inverter. I do not bother with optimisers as they introduce more cost and reduce reliability. Since installation a few years ago I have come across numerous PV users (non-tech) who have been running with many optimisers down (and low output) without realising. I do have shadowing on one or another panel array mornings/evenings in dead of winter but total electricity losses are actually tiny (if annoying).
if you put three strings in parallel and one string is shadowed, the other two strings will push more current through the shadowed string than it can handle. The panels and wires can burn.
Thatās why you need diodes against back current on plus and on minus cable. So itās not just a combiner.
If you have more parallel strings, itās getting worse.
@BjoernK I didnāt think about diodes before. I understand each of my panels has 3 diodes but I donāt know if they are helpful in this case (I havenāt bought them yet). The combiners (I havenāt bought them yet either) supposedly have diodes but I donāt know how many, if they are appropriate and how they are installed.
My immediate thought is to install additional blocking/one-way diodes in the combiner box on the positive and negative wires just before they leave the combiner to connect to the MPPT. Better safe than sorry. Good thing, I donāt have shadows.
Thank you for your advice.
@farmeroz Thanks for the detailed response. Take much of what Iām about to say with a pinch of salt because itās based on unverified reports and figures I recall from memory.
Update: My potential supplier says each panel has 3 bypass diodes, but those play a different role to help protect against shadows.
The combiners will have a separate anti-reverse diode on each incoming string to prevent the backflow of current into any string. According to him, with such an arrangement, I donāt need any additional diodes as a charge controller also has such an anti-reverse diode.