Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[terryo]

Has anyone tried to make a simple solar panel based charger for a Zero that uses the Anderson port? I have a 2020 Zero FX 7.2 and would like to have an off grid charging system for my bike at trailer I use in the summer. I would much prefer not to have an inverter based and/or battery buffer based solution to keep the solution as inexpensive as practically possible and not have to worry about unattended storage of buffer batteries in the winter where temperatures could dip down to -30C on occasion. The FX would only be there in the summer.

 I believe a Zero battery pack is deemed full when the pack voltage reads 116.4v and is deemed empty at about 95v. I note the simultaneous use of two external Quiq chargers and the internal charger with the FX is officially sanctioned by ZERO. I believe the three chargers collectively consume about 2.8kw, and provide about 2.2kw on the DC side. A charger powered by a maximum of 2.2kw of solar panels is what I envision.

 The simplest arrangement would to have solar panels in series where the maximum outputted voltage is at least 116.4v and stays below some maximum. In version one of this charger I would be quite happy to monitor the display on my motorcycle and safely shut down the charger when the bikes battery state of charge display hits some percentage value, like 99% or 100%, or when the green charging status light changes from flashing green to solid green. I note the charging countdown timer gives a rough estimate of when the bike will be charged so I would be by the bike when it is close to fully charged.

 Let's call the maximum voltage that should be allowed on the Anderson port during live charging VMax. What is VMax ? Can we make a reasoned guess ? Can we have a safe charging strategy even if we don't know what VMax is ? Has anyone bothered to measure VMax ? Can they tell us what it is and how they measured it ? Since this charger is basically a series of solar panels, we need to have an idea of the amount of voltage our ideal series of solar panels put out.

 The voltage a solar panel puts out varies with the amount of solar radiation collected, the ambient temperature and probably the age of the solar panel. Used solar panels a few years old and in good condition are quite inexpensive. That is what I will use. Of course I'll test them before using them. Solar panels are rated. The ratings taken under standard test conditions include Maximum wattage output(Pmax), Open circuit voltage(Voc), and voltage(Vmp). My limited experience charging nominal 12v car and motorcycle batteries suggests considerations.

 An AGM or "glass mat" 12v battery is at 100% charge when it measures 12.8v with no load. An "old school" 12v auto battery is fully charged at 12.6v. Having said that, to get the true voltage of a car battery one must let it settle with no load, maybe sit "overnight" after it is charged to measure it. A nominal 12v solar panel can measure around 20v in an open circuit. When a low "12v" battery reads 12.0v and is connected to a 5watt 12v. solar panel, the voltage across the terminals is a bit over 12.0v. Similarly if you start your I.C.E. motorcycle or car up and then measure the voltage across the battery terminals you should get a reading of around 14v. The vehicles shop manual should give a range of voltages that indicate the cars charging system is working properly. When I use a nominal 12v charger on my motorcycle or car and measure the voltage across the terminals I see a slow rise in voltage to about 14v until the solid green light of the charger deems the battery is charged. At least with lead acid batteries we can conclude that chargers apply a voltage higher than the batteries "fully charged" voltage.

 So what can we infer here? Please electrical engineers chime in , or people technically familiar with Zeroes battery packs, please chime in.

 I venture to hypothesize that the combined voltage Vmp should be somewhat higher than 116.4v. Let's call that CVmp. I expect the voltage measured across the live circuit should drop to somewhere between the current pack voltage and CVmp and then rise as the pack gets charged. I optimistically hope that the display on the motorcycle will let me know when the bike is close to be fully charged and thus when I should safely disconnect the motorcycle from the solar panels.

 Should I try this out? Should CVmp be 116.4v ? 120v ? (Both probably not) 140v ? 160v ? For instance if I put four 250W solar panels in series that are identical and made to have at least 4 in series and have a Vmp of 35v each, would that work well?

 One consideration is if the voltage across the circuit temporarily drops. Say a cloud gets in the way. When I use only my Quiq charger to charge, I connect all the wiring, then turn on the Quiq charger, then key on the bike. After the bike starts charging I key off the bike. When I then turn off the charger, after a second or so a relay on the bike breaks the connection. I don't want a cloud to break the connection. The obvious reason is the charging process will have to be started over again but more importantly, from a safety point of view, having a Kw or so of power with no load sounds like disaster. At least that is my instinct. Would a high power capacitor in the circuit prevent this from happening, or does that just mean the connection gets broken after a few seconds of low solar panel output and that this design will inherently not work. Or maybe this is not a consideration at all as light clouds might not lower the output power enough to cause the bike to trip the relay. Or maybe I should just leave the bike keyed on until I am ready to stop the charging process.

There are other considerations. What EXACTLY should I do to prevent lightning issues. Nothing will be connected to the panels when it's overcast. They will be attached to a steel roof. I don't even know if the roof is grounded.

Any helpful thoughts ?

[Specter]

SSR's might be good for switching on and off.  have fun controlling the voltage.  Granted the battery will act like a huge cap to bring the voltage down to battery volt but MPPT is going to kill you too.

I suppose though you could just hook them up in series to be above the battery voltage and let them just push it in there is enough light to create the voltage you need.

Aaron

[terryo]

I'm hoping you are right. Diodes to prevent flow out of the battery pack when there is shade solar panels with an appropriate Vmp. Someone sent me a link to a YouTube video of someone who did this about 3 years ago.



 I just sent the poster of the video an email.

[Specter]

If you get lucky and your power requirements just happen to hit the sweet spot for the panels, then no you won't need mppt, as for charge controller, what happens on a really bright sunny cold day when the panels are putting out 110 percent rated voltage, or you get a fringe effect going on?  How much of an over voltage can your batteries take if they happen to charge up and you are still pumping unregulated power to them with nothing turning them off at XX volts.

You should not have to put diodes in to block the panels from pulling power at night, as most reputable panels have them built in to them, theyll be in the conn box where the wires come out of.

Also, even though you are pushing 120 ish volts, it's still DC, try to keep your wires as short as possible or run 10 - 12 ga wire, or you'll lose a ton of power in the cables.

Aaron

[DerKrawallkeks]

I made that video, I replied to your email.

@Aaron, no, solar panels have no inbuilt diodes for this purpose. They have freewheeling diodes which serve a different purpose.
Also you're not going to lose much at all in any cables, the losses are so small they're completely negligible

[Specter]

I have bought literally 100's of solar panels with blocking diodes in them.  Otherwise, at night the panels keep nice and warm by dissipating whatever the battery is sending them back.  Not sure where you are getting yours at but most the ones I see advertised even state they have blocking diodes already built in so they don't take power at night.

why would you need a freewheeling diode on a solar panel?  Unless you have it hooked directly to a motor there should not be any sort of inductive kick or flyback into it from a battery bank.  curious why you would need that.

Aaron

[terryo]

If you get lucky and your power requirements just happen to hit the sweet spot for the panels, then no you won't need mppt, as for charge controller, what happens on a really bright sunny cold day when the panels are putting out 110 percent rated voltage, or you get a fringe effect going on?  How much of an over voltage can your batteries take if they happen to charge up and you are still pumping unregulated power to them with nothing turning them off at XX volts.

Aaron

Hi Aaron:

 I am no electrical engineer, just somebody with rudimentary electrical knowledge trying to find the relevant information.

 Anyway, according to information on the net the voltage of a solar panel does not change much with changes with luminance. The available amps do change, possibly nearly linearly. I give this link from a solar MPPT distributor ...

https://www.victronenergy.com/blog/2020/02/20/pv-panel-output-voltage-shadow-effect/

 In the article they have a sample solar panel that has a VOC of 43v at 100% luminance, that drops to 38v at 20% luminance. I would expect that "fringe effects" caused by cloud edges would only increase VOC by roughly 1v.

 As another "inmate" has let me know, as long as my battery packs SOC is below 100%, the voltage of the solar panels across the "load" of the battery pack should be barely above the battery packs voltage. If I can't figure out some automatic disconnect, I'll just have to be there before 100% SOC to manually flip a switch to break the charging circuit.

- Terry

[terryo]

I made that video, I replied to your email.

 I thank you very much for your reply.

 Quoting from your email ...

There is one problem with the clouds, that I encountered on my 2013 S. I don't know if this still applies to a 2020 FX. I was charging via the brown anderson, and I supplied 5 V to the signal pin inside the brown anderson to tell the bike it is being charged. This is needed to keep the main contactor closed. The problem is, that in addition a minimum charge current was needed on my bike. If a cloud came, and the charge current would drop below about 4 A, the green charging light would start blinking fast instead of slow, and after a minute, the main relay would open.
That's a problem, you'll have to find a solution for.

 I don't know if my bike will suffer from the same low_amperage_caused disconnect. Experiments will have to wait for the summer. One inelegant solution might be to keep the FX keyed on. Thoughts ?

[T.S. Zarathustra]

Solar charging with four 250 w cells is slow enough that checking state of charge every our or so and turn off at 90-95% is a perfectly safe method of charging.
To do a little math. If you have 14.4 kWh pack at 30% charge you need almost 10 kWh to fill up. Four 250 w solarcells would take 10 hours under ideal circumstances to fill it up. So if you arrive at your destination around noon you can plug the bike in and in and only turn off charging when the sun goes down.

[DerKrawallkeks]

Just for everyone who reads this, I am not sure wether the controller and other components will see the solar panels OCV of e.g. 140 V when the main contactor opens. Be careful not to damage your bike