Reverse protection diode on PV input to MPPT

[ekarlson]

I am in the process of revamping the solar build on my trailer and as part of that have been looking at adding a few safety improvements. In particular, I installed an SAE solar port through the side of my trailer to support connecting a portable solar panel to augment the panels on my roof (two separate MPPTs). In researching these types of solar ports I am finding a sufficient number of reports about ambiguity regarding the polarity of these ports that I am thinking of putting in a reverse protection diode on the PV feed from the solar port, just in case someone plugs in a solar cable that is wired backwards from how the port is wired. Some MPPTs seem to have reverse protection built-in on their PV input, but the SmartSolar ones from Victron Energy do not as far as I can tell (probably to provide higher efficiency by avoiding a hard-wired voltage drop from the PV array due to a hard-wired diode in the circuit).

I will probably end up using the Roadmaster Hy-Power 790 diode as it has a built-in heatsink and spade connectors that make it easy to use as an inline diode on the PV feed. I have not been able to find any official specs from the manufacturer yet, but it would seem to be rated for 85A and I am assuming that it is a standard silicon diode with about a 0.7V forward voltage drop.

I started looking at Schottky diodes as they are supposed to have a lower forward voltage rating and the axial forms appear to be used as bypass diodes in most solar panels. Most literature indicates that Schottly diodes have a forward voltage drop of around 0.2V, which sounded nice in terms of reducing heat and power loss. But when I go looking for actual Schottky diodes that I can buy, the specs on those units indicate a forward voltage drop of 0.5V - very little difference from a standard silicon diode. It makes me wonder why bother with a Schottky diode if the gain is only 0.2V on the voltage drop as compared to a standard silicon diode. I'm looking at a system that would support an 18V 10A PV array - so a difference of 0.2V on the forward voltage drop only means a gain of 2W on the available power - hardly seems worth it.

In addition, it seems that these diodes do get rather hot: 50C to 60C from what I am reading. So I started to look at TO220 Schottky diodes as they are designed to be attached to a heatsink. However, the pin-out on the devices seems a bit awkward for an inline application. There are three pins on one end and a metal tab with a screw hole (to attach to the heatsink) on the other end. The tab and the center pin appear to be the cathode and the two outer pins appear to be the anode. In fact, the electrical diagrams that I can find seem to indicate that the TO220 is actually two parallel diodes with two separate anodes and a common cathode. So to wire this inline I think I would want to snip off the center pin and try to crimp the two outer pins to my 10AWG PV feed and then wire from the cathode tab to my MPPT. I guess it would work, but it seems a bit adhoc. Also, the diodes I've been looking at are rated for 15A to 20A which seems like a lot of current to pass through two pins - unless the pins are thicker than what the pictures convey.

Has anyone here worked with Schottky diodes and/or the TO220 form factor? Is the real-world forward voltage for Schottky diodes really 0.5V instead of 0.2V? Does that 2-pin anode design really work for 15A rated diodes?

 

[Hedges]

Some inverters have a reverse-biased diode across PV input. No current goes through it during normal operation. If PV array connected backwards it simply shorts the array. It need to be rated for and heatsinked well enough for heat buildup at Isc.

18V PV array - so system doesn't support higher voltage string, with MPPT SCC? A few MPPT are very low voltage. Usually low voltage is PWM and not as efficient. At battery voltage or 18V, the ~1V drop of silicon diode is significant percentage.

When using Schottky blocking diode in series, if PV array is connected backwards and left for a long time, diode has to keep holding off the voltage. Schottky diode will leak some current, maybe fail to protect the circuit. The leakage current multiplied by Voc of array can be significant heat dissipation in the diode, causing runaway and over-temperature failure. You have to carefully analyze the power dissipation and thermal cooling path. Start by assuming 150C or whatever maximum junction temperature. Determine if temperature will increase from there, or cool down.

You can get all sorts of diodes in various packages. Yes, the pins seem small. Some are lugs that look similar to a spark plug, some of those have a cable off one end, maybe 50A or 100's of amps. Don't just rely on listed voltage and current, read data sheet and see how that varies with temperature.

 

[ekarlson]

The MPPT is a 75/15 from Victron - its mostly the configuration of the portal solar array that governs the 18V 10A config. I've yet to find a portable array that is wired in series instead of parallel. I have considered trying to rewire the portable array, but not clear that can be done without destroying it (wiring tends to be all sealed up). I could just lug along two standard panels, but that makes for a large, bulky and heavy package to haul around.

The diode doesn't have to hold off the current for long - just long enough for someone to plug in something with reversed polarity and then realize that they are not getting any power.

The other piece that still puzzles me is the disparity between the articles that claim that Schottky diodes have a forward voltage of 0.2V and every real-world Schottky that I can find has a forward voltage of 0.5V or more.

 

[efficientPV]

The other piece that still puzzles me is the disparity between the articles that claim that Schottky diodes have a forward voltage of 0.2V and every real-world Schottky that I can find has a forward voltage of 0.5V or more.

Yes, when I hear this I know it is someone who doesn't know electronics. Typically, when you get above 50V rating there is no forward voltage advantage. There are diodes designed for bypassing panels and they reach that low voltage, but these never see more than 20V. Then look at current flow and that forward voltage climbs again.

 

[HarryN]

There are Schottky diodes with a lower Vf but you have to be willing to pay for them.

Also while the Vf might be listed as a single number - in the real world it is a curve, so for instance take two diodes:
- one rated for 20 amps at Vf = 0.5
- one rated for 200 amps at Vf = 0.5

If you look in more detail, the 200 amp one will in fact be more like 0.2 when used at 20 amps and unlike many devices, this can sometimes fall when they get hot.

The limitation is that these can't go above 100 volts.

Schottky diode - Wikipedia

en.wikipedia.org

There are companies that specialize in different types of applications and it is surprisingly specialized.

 

[andrew-d]

You could also use something like an "ideal diode" to prevent reverse polarity; either buying something from Amazon, or building your own with a known-good part like this Linear component:

https://docs.rs-online.com/6244/0900766b81382e8f.pdf

https://www.analog.com/en/technical-articles/low-iq-ideal-diode-controller-with-reverse-input-protection.html

or

https://www.amazon.com/dp/B07SGH7BP3/

https://www.amazon.com/dp/B09BK1BTFG/

 

[Selenium Dion]

Some MPPTs seem to have reverse protection built-in on their PV input, but the SmartSolar ones from Victron Energy do not as far as I can tell (probably to provide higher efficiency by avoiding a hard-wired voltage drop from the PV array due to a hard-wired diode in the circuit).

I realize this is an old thread; I happened to find it and read through it bc I am in the middle of redesigning my current system and will be adding additional panels to my current PV array. While researching different PV disconnects, CBs, fuses, etc. I have come across several instances of anti-reverse current diodes being suggested as useful, or perhaps necessary elements to a safe and efficient system.

After reading your initial post, I couldn't help but notice the mention of the Victron SmartSolar MPPT in the quote above. I realized I had not considered the possibility that the SmartSolar 150/75 I will be installing may not have a proper safeguard for reverse current back to the array. So I took a moment to look into it a bit further, and discovered the following documentation from the SmartSolar product manual on Victron's site.



According to the product manual, this model does in fact have PV reverse current protection. I am not certain that this feature is implemented into the entire SmartSolar product lineup, but as it is relevant to this discussion I wanted to share what I had found. My apologies if this has been covered in subsequent discussion, my hope is that this info may be of some use.

 

[SpongeboB Sinewave]

A typical buck converter MPPT controller will automatically and intrinsically have reverse polarity protection due to the topology of a buck converter. The top FET has an internal diode with its cathode pointing to the PV+ input. And the bottom FET or diode will have its cathode pointing towards and connected to the top FET's anode. So when you wire the PV backwards, those 2 diodes will conduct and keep the internal negative voltage at around 2 volts or so. They will get warm that way but they are well heat-sunk and there are usually multiple parts so can take it.

That is usually enough and it doesn't typically break because the PV current is limited. (not a battery) A battery would most likely break it though.

The NRTL (UL or ETL etc) will normally test this as part of UL 1741. These were done on both the OutBack and MIdNite Solar MPPT controllers.

If the unit has a series diode, then it is going to always get warm/hot during normal operation and would need heat sinking to dissipate 10 to 20 watts depending the PVs current rating. Might as well just use the internal semiconductor switches that are already there and heat-sunk and don't need to be driven in the reverse polarity situation.

boB

 

[Selenium Dion]

Thank you boB, for taking the time to go over all of that. I greatly appreciate such a thorough reply. I honestly hadn't given that much consideration to the whole reverse polarity protection issue before. I just wanted to make sure that I wasn't overlooking any critical components that will need to be included in my system redesign.

So I read through a few more threads on the subject. I now see how the pre-charge diodes could be incorporated into a system. A bypass diode could be hardwired into the circuit as a way to attenuate the inrush current; which in the worst case would be powerful enough to destroy MOSFETs and perhaps other circuitry, and in the best case scenario would subject components to unnecessary stress.

I must confess I was not too familiar with all of this information concerning different types of diodes, a week or so ago. I've always been into electronics in one form or another, but the last time I assembled a battery pack, it was a 6-cell NiCad battery pack for my Tamiya RC, so I've missed out on some major advancements in battery technology. I did read this short article here below, which seemed to cover the basics of the design and the purpose of the PCD.

https://rnoegroho.wordpress.com/2019/11/30/why-pfc-needs-pre-charge-diode/

Anyhow, back to the diodes: as with many other topics discussed here, e.g. BMS settings, wire/fuse/CB sizing, etc., there are some conflicting views on whether or not the implementation of a PCD is a strict necessity, with some forum members stating that they may not be necessary in instances where the total system voltage is less than 3kV, such as the upgrades I have planned. What say you in the matter, if you don't mind?