DIY Electric Car Forums banner

1 - 12 of 12 Posts

·
Registered
Joined
·
36 Posts
Discussion Starter #1 (Edited)
As discussed in my earlier post, I'm working through the idea of constructing an electric tricycle that is powered by renewable energy (so it can't be recharged from the grid). I initially thought of using a PV panel directly on the (custom-made) tricycle. However that had several disadvantages such as extra weight, and the need to customize the trike). The main advantage of this was that once parked, the vehicle could be recharged, so you could have double the range (if you're using it for commuting purposes). I'm now concidering a different (and probably far better) approach: simply putting the PV panel on a shed in which to store the tricycle. Why haven't I thought of this sooner ? Well I have, but I thought it would be an inferior solution also because many people already have PV panels on their house (so they don't need such a shed). However what I didn't realize was that while true, many people just haven't installed PV panel installation on their house since most want to combine it with net metering too, and since the price then goes up a lot, many simply don't go trough with it.

The tricycle shed however would not be grid-connected (so no wiring needs to be layed), nor have net-metering (so cheap). Rather, it would just be used to recharge a spare battery (for the tricycle).
Here's how it works: you use your tricycle to go to work, once you get back home (and the battery is discharged), you swap it out with the spare battery that has been charging all day in your shed. If you already have a shed, you can use your existing one. If you don't have one, you can build one using a kit. It just needs to be minimally 39 x 65" (to accomodate the PV panel and tricycle).

For the work out:
Your existing li-ion tricycle battery is used, along with a spare battery you buy (same brand/model).
Any 36V li-ion battery (any amount of amps, between 13A-17A) would be usable.
A 39 x 65" panel (250watt, 36V) is placed on the shed roof (preferably facing south, tilted at winter angle).
A charge controller is placed in the shed, between PV panel and battery
A switch is placed between PV panel and battery, purpose hereof being to redirect excess power (= energy produced when battery is fully charged) to the ground by means of a stainless steel ground rod.

I'm not sure how to make the switch. Any ideas ?
Also, I assume 36V PV panels exist, but their voltage and amp output would vary depending on the sun's intensity at the given moment, so how do I best connect the PV panel to the 36V li-ion battery ? Do I use a (PWM) solar charge controller ? If I use a PWM solar charge controller, do I still need to use a li-ion charge controller as well ?
 

·
Registered
Joined
·
4,894 Posts
A switch is placed between PV panel and battery, purpose hereof being to redirect excess power (= energy produced when battery is fully charged) to the ground by means of a stainless steel ground rod.

I'm not sure ho to make the switch. Any ideas ?
We already had this discussion in the earlier thread: there is no need or reason to direct "excess power", because the panel does not produce a fixed amount of power. After the battery is charged (as controlled by a suitable charge controller), there is no more power flow... and no problem. If you still don't understand this, asking about the switch again is not going to change anything - you still need to understand that the panel doesn't need it.
 

·
Registered
Joined
·
4,894 Posts
Also, I assume 36V PV panels exist, but their voltage and amp output would vary depending on the sun's intensity at the given moment, so how do I best connect the PV panel to the 36V li-ion battery ? Do I use a (PWM) solar charge controller ? If I use a PWM solar charge controller, do I still need to use a li-ion charge controller as well ?
True, there are panels available with various operating voltages, and panels can be combined in series for higher voltage. Just as with batteries, the listed voltage is nominal, not actual. Actual voltage is lower or higher, depending on conditions.

There is nothing special about a li-ion charger - it's just a battery charger set to different limits than a charger for a different battery. One charger (or charge controller, in solar energy terms) will do.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #4 (Edited)
brian_ said:
We already had this discussion in the earlier thread: there is no need or reason to direct "excess power", because the panel does not produce a fixed amount of power. After the battery is charged (as controlled by a suitable charge controller), there is no more power flow... and no problem. If you still don't understand this, asking about the switch again is not going to change anything - you still need to understand that the panel doesn't need it.
Yes, I agreed that it probably doesn't matter, but as I said back then, I still have reservations whether or not it has an impact on the PV life expectancy, so I rather include it anyway.
Even if I were not to include it, I would still need to make the switch regardless since I need to protect the battery from overcharging, so it would need to stop charging the battery once the battery is full.

brian_ said:
True, there are panels available with various operating voltages, and panels can be combined in series for higher voltage. Just as with batteries, the listed voltage is nominal, not actual. Actual voltage is lower or higher, depending on conditions.

There is nothing special about a li-ion charger - it's just a battery charger set to different limits than a charger for a different battery. One charger (or charge controller, in solar energy terms) will do.
Yes, but the question here is about the voltage variation: with a PV panel, the voltage varies every second depending on the amount of sunshine it gets. So don't I need some sort of device to level the voltage constantly to 36V, while still allowing the the amps to change. Say it gets 30V @ 23A; the device (DC-to-DC converter) would then change it automatically to 36V @ 19,16A (wattage hereby remaining constant, ie 690W). It would work in the other way too, ie if it gets 42V @ 17A (588W), it would convert it to 36V @ 16,33A. Or well, that's how I see it. Not sure whether such DC-to-DC converters exist. If they do, the idea is to connect one of these to the li-ion charger.
 

·
Registered
Joined
·
4,894 Posts
Yes, I agreed that it probably doesn't matter, but as I said back then, I still have reservations whether or not it has an impact on the PV life expectancy, so I rather include it anyway.
If you're going to worry about things that the entire rest of the world has no problem with, I think you're going to drive yourself crazy. Solar power is not a new thing.

Even if I were not to include it, I would still need to make the switch regardless since I need to protect the battery from overcharging, so it would need to stop charging the battery once the battery is full.
That's what the charge controller does.

Even if you didn't use a charge controller and tried to stop charging manually (which would likely be a disaster), you would just disconnect the battery. This "redirecting power" idea is just pointless and potentially dangerous.

It does make sense to have manual switches to safely disconnect the panel from the controller and the controller from the battery, for maintenance. However, they're not needed, and there's no reason to switch either set manually for every charge cycle. Again, about a million solar charging installations demonstrate this.
 

·
Registered
Joined
·
4,894 Posts
... the question here is about the voltage variation: with a PV panel, the voltage varies every second depending on the amount of sunshine it gets. So don't I need some sort of device to level the voltage constantly to 36V, while still allowing the the amps to change. Say it gets 30V @ 23A; the device (DC-to-DC converter) would then change it automatically to 36V @ 19,16A (wattage hereby remaining constant, ie 690W). It would work in the other way too, ie if it gets 42V @ 17A (588W), it would convert it to 36V @ 16,33A. Or well, that's how I see it. Not sure whether such DC-to-DC converters exist. If they do, the idea is to connect one of these to the li-ion charger.
Neither the solar panel output nor the battery voltage is constant.

The panel output does depend on exposure to sunlight, but it also depends on the connected load. If you look at the specs for a solar panel, it will have
  • open-circuit voltage (VOC) - the voltage put out by the panel in full sun with nothing connected (that's what "open-circuit" means); there is no current so no power is produced. This is what happens when the controller shuts off charging because the battery is fully charged (or if no battery is connected).
  • short-circuit current (ISC) - the current put out by the panel in full sun with the output leads connected to each other (so "short-circuited" together); there is no voltage so no power is produced. This doesn't happen in normal use - it's just the extreme case of low load.
  • maximum-power voltage and current (VMP and IMP) - the voltage and current put out by the panel in full sun with an optimal load connected so that maximum power is produced. The rarely happens in reality, unless you use a controller which is a DC-to-DC converter designed to optimize panel operation - that's called a Maximum Power Point Tracking (MPPT) controller.
Any reasonable solar panel supplier will provide a graph showing the output over the continuous transition between open-circuit and short-circuit conditions. Here's one from a random online discussion:

(This curve is for a typical panel used to charge a "12 volt" lead-acid battery)

If you connect a panel directly to a battery, the panel output voltage is forced to be the battery voltage of the moment, and the difference between that battery voltage and what the battery would be when not connected (given its current state of charge) is what drives current through the internal resistance of the battery. This is not likely to result in the optimal panel output (but most solar installations don't worry about that) and it will often not result in a suitable rate of charge (which is why they use charge controllers).

Most available MPPT controllers only convert voltage down, so the system is designed so that VMP is well above the fully-charged battery voltage. Some "MPPT-Boost" controllers will step up in voltage, but are intended for large mismatches (such as nominal 12V in and nominal 48 V out).

You are looking for a charge controller, expecting one which operates in a mode that you won't find, and unnecessarily planning to use two devices in series to do the work. A simple PWM controller will limit battery charge current and voltage appropriately (if it is set for your battery configuration and type), and an MPPT controller will also do DC-to-DC voltage down-conversion as appropriate to optimize panel operation. They are available in your desired voltage range, but MPPT types are not cheap, and controllers with battery settings for lithium are not common (most solar installations are still lead-acid).
 

·
Registered
Joined
·
36 Posts
Discussion Starter #7 (Edited)
As could be read in the other post, I calculated that with a conventional (1,58m² or 39" x 65") PV panel, you get about 237 to 316 watts per second (or 237 - 316 watt/hr). Assuming we're using a 36V panel, 6,58 amps to 8,77 amps is probable.

I read that for li-ion batteries, a maximum charge C-rate of 0,5 to 1C needs to be used; so assuming a 13-17 Ah (36V) battery, the PWM li-ion charger should be a 36V 6,5-8,5 Ah charger. That seems to fit entirely with the amount of amps we can actually expect to be outputted by the Pv panel.

The conventional chargers provided with the li-ion battery don't suffice, since they use 230V input energy, and they also have a too low an amp-output (just 2A, see here and here)

I started looking online for suitable li-ion charge controllers (so 36V, 10Ah) and couldn't find any.
I assume that higher amp ratings don't matter much as the system will reduce the amp-rating itself ?
Another thing I'm thinking of is to use 2 li-ion charge controllers, one of 24V, and another for 12V, and wiring it in series. That way, I can use commonly available charge controllers like those of ZHC, SolarEpic, ALLPOWERS, ecoworthy, ...
Would that work ?
The only other option would be to use a 12/24/36/48V MPPT li-ion charge controller, which is more expensive, and even more crucially is not commonly available (just a few manufacturers make these, like ENZPOWER, ...). So I rather avoid that route.
 

·
Registered
Joined
·
4,894 Posts
As could be read in the other post, I calculated that with a conventional (1,58m² or 39" x 65") PV panel, you get about 237 to 316 watts per second (or 237 - 316 watt/hr).
If you don't understand that "watts per second" and "watts per hour" are nonsensical, you really need to go back to the basics - and I mean elementary school basics - of energy and electricity.

For a hint: "watt" is a unit of power, not energy.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #9 (Edited)
electric_driver said:
As could be read in the other post, I calculated that with a conventional (1,58m² or 39" x 65") PV panel, you get about 237 to 316 watts per second (or 237 - 316 watt/hr).
brian_ said:
If you don't understand that "watts per second" and "watts per hour" are nonsensical, you really need to go back to the basics - and I mean elementary school basics - of energy and electricity.
For a hint: "watt" is a unit of power, not energy.
I of course know that watts per second isn't watt/hr (or more precisely "watt-hr"). What I meant was that the solar energy you receive is fairly constant, so at any given moment, you get 237-316 watts out of the panel.
If I write it just in watts-hr, it means that you get 237-316 watts out over the course of an hour -this is also possible in a non-constant manner, ie getting say 118 watts/second out for 30 mins, and then getting 474 watts out for another 30 mins- ).
By writing it in watt/second it's clear that I mean you get that amount of power out at any given moment.

Let's get back to the initial questions: could I use 2 li-ion charge controllers, one of 24V, and another for 12V by wiring it in series ?

I'm also looking into the idea of using 6V solar cells rather than a 36V PV panel, to reduce costs (a 36V PV panel is uncommon and thus costs more than 12V or 24V PV panels). 6V solar cells would be cheaper as these can be made to create any type of PV panel, so they're the main building blocks for each PV panel and thus even cheaper (when comparing it to a same surface area 21 or 24V PV panel).
Numbers still don't add up: looking at conrad, I found a 2w, 6V panel for about 15 euro. So x6 that makes 90 euro for a 2w, 36V panel. x125 (in parallel) that makes 11250 euro for a 250W, 36V panel.
A 100 watt 36V PV panel costs 112 GBP at amazon, so x 3 = 336 GBP which is way cheaper.
So, I'm not entirely sure whether in practice, this theory will actually hold up (some sites claim it does, see first link, but I'm not not sure). I might just combine whatever is cheap into a 250W 36V PV panel. Perhaps the 100W 36V panels could indeed prove to be more economical, I'll look on.
 

·
Registered
Joined
·
4,894 Posts
I of course know that watts per second isn't watt/hr (or more precisely "watt-hr").
...
If I write it just in watts-hr, it means that you get 237-316 watts out over the course of an hour -this is also possible in a non-constant manner, ie getting say 118 watts/second out
...
By writing it in watt/second it's clear that I mean you get that amount of power out at any given moment.
It seems like you don't understand some very basic notation: a slash ("/") means "per" or "divided by", and a dash ("-") means a product or "multiplied by".

So no, that's not what watt/second means. If you want to refer to average power, say "average power"; power sampled at a given moment is "instantaneous" or "peak" or whatever similar word you like that has the intended meaning and not a completely different physical entity.

You still have it all very wrong. By randomly combining time units with power units, you are making nonsensical units (such as watt/second and watt/hour) and units for energy when you are not dealing with energy (watt-hour).

Next hint: "watt-hour" is a unit of energy, not power.
 

·
Registered
Joined
·
4,894 Posts
Numbers still don't add up: looking at conrad, I found a 2w, 6V panel for about 15 euro. So x6 that makes 90 euro for a 2w, 36V panel.
Why would you expect six panels, each of which produces 2 watts (peak), to produce only 2 watts in total? The combination of six panels (whether in series or parallel) will produce 6 x 2 = 12 watts (again only ideal peak conditions). That's still only 12 watts for €90.

If you combine six panels each of which produces a peak of 2 watts at 6 volts (which means 1/3 of one amp), then:
  • in series you get 12 watts (peak) at 36 volts (still 1/3 of one amp)
  • in parallel you get 12 watts (peak) at 6 volts (6 x 1/3 = 2 amps)
If this doesn't make sense, please review the meaning of series and parallel, and how power is related to current and voltage.

€ 7.43 per watt seems wildly expensive for solar panels; that's not surprising because it will never be cost-effective to build a panel out of many small pieces each with its own frame and connectors. The GBP 1.12 (€1.25) per watt for the larger panels is much more reasonable. The other obvious option is to use 12 volt panels, which are very common.
 

·
Registered
Joined
·
36 Posts
Discussion Starter #12
Having taken a look at both options again, I think I'll go with an MPPT charge controller anyway. Cost of the PV panel vs the charge controller seems well balanced out (price for PV is probably about 375 € and up, price range for charge controller is 136 to 396 €, finding manufacturers is a bit tricky but probably still doable.

brian_ said:
The other obvious option is to use 12 volt panels, which are very common.
brian_ said:
If you combine six panels each of which produces a peak of 2 watts at 6 volts (which means 1/3 of one amp), then:
in series you get 12 watts (peak) at 36 volts (still 1/3 of one amp)
in parallel you get 12 watts (peak) at 6 volts (6 x 1/3 = 2 amps)
I confused amps with watts (mainly because I'm accustomed to see only those with panel ratings).
You're indeed correct on that.

The bottom line on the panels remains the same however: using small PV panels isn't going to be cost effective.
Solar cells might be cost-effective (see the article linked in a previous post) but as you mentioned too, these don't contain any connectors or wires. Combining the solar cells DIY by soldering, ... is probably way too time-consuming and unpractical. Simply using three 12V 250W panels is without a doubt the way to go.
 
1 - 12 of 12 Posts
Top