diy solar

diy solar

[Project] Extending EV range by offsetting auxiliary 12v loads using solar

Another update:
Finally got around to mounting the battery and solar components in a plastic ammo box.

I mounted the SCC, fuse/distribution box and breaker on a melamine coated piece of plywood.
I found that piece in the trash, which was used as a shelf for a bathroom cabinet. The fun part is that I didn't need to cut it at all. It fit perfectly in the ammo box.
20211207_160005.jpg20211207_160412.jpg

I decided to use XT60 connectors between the SCC, MC4 connectors and fusebox, and between fusebox and loads, so that it would be easier to dismantle should I need to do some changes or maintenance.

I do regret not mounting the breaker (connected to the battery positive) in a more accessible location, like maybe on the lid if the ammo box, so that power could be disconnected without opening the box.
I might do that change in the future.

Here is are pictures of the box with the battery and shunt too.
I still have to add some double sided tape to hold the battery in place, and maybe add some foam between the SCC and the battery to prevent movement.
20211207_162429.jpg

I have MC4 connectors on the back of the ammo box, connected to the SCC.
On the front, I have an XT60 for loads. This will be connected to the DC-DC boost converter box as posted previously. The problem is that I have run out of XT60 connectors, and I need one with a long cable to connect the two.

Anyways, I like how it turned out, because I can also use this as a portable solar generator, currently supplying only 12v DC loads up to 20A but in the future I would be able to add a small DC-AC inverter if I want to.

As you can see, I have a 30A mini-ANL fuse on the battery (in addition to a 30A breaker) and all connections are also fused accordingly, as they are connected to the fuse box (including the shunt's VIN line).

I powered the box just to see if all is connected well and surely enough the Juntek shunt displays voltage. I still need to test with loads/charging, to see that I can view proper current draw with the shunt.
20211207_162755.jpg20211207_162800.jpg
20211207_162731.jpg

That's it for now...
I have ordered some more XT60 connectors, but until they arrive I still can't install in my car.
But I will be doing some more testing with both "solar box" and with DC-DC converter box.
I still need to figure out how I can view both shunts from one display screen. The Juntek supports thus using addresses, but I need to learn how to set it up.
 
@meetyg, you have proven that the upgrade from PbA to LiFePO4 has improved system voltage stability! wow! thank you for proving out this concept thus far. the documentation in this thread intensely motivates me to continue with my self chosen research path as well. big up.

thank you for these updates. your methodology is commendable, and i think your sensitivity to safety while still making progress is also incredible.


as it happens, the 12V PbA battery in my car died recently, and sourced a 12V LiFePO4 equivalent capacity pack to replace it. still doing due diligence on the self install. your graph of voltage vs time has massively reinforced the decision to include continuous voltage and shunt ampere logging at some timescale like 1 per minute at least.

thank you, and i know you’re not done yet ;)

☀️??
I have done manual amperage measurements using a clamp meter, but connecting a shunt with logging would be cool :cool:

For now I just have a (hardwired and fused) simple BM2 battery monitor that logs for up to 31 days, and has a Bluetooth app.

Which LiFePO4 did you get? I'm curious...
Also, please remind me what car you have?

Are you getting along with your VIPV setup?
 
Last edited:
Yay, I got the Juntek display to work with both shunts!
For some reason, one if them wasn't connecting wirelessly at all, it took me some time to figure it out.
I initially thought that maybe because the two shunts are different models of the VAT series (one is VAT-1100 the other VAT-1050) that they can't use the same display.
But after doing a factory reset, it started to communicate! So I set up each one with a different address (A01 and A02), and now I can see both with one display (but only one at a time of course).
I also connected the DC-DC box to the solar box, and did some more testing. I simulated the car's load using the DL24 battery capacity meter. All is working well. As mentioned, in order to connect them in the car I would need a longer cable and I'm all out of XT60 connectors for now.

Here are some pics. A01 is the shunt in the solar box connectedto the battery. A02 is the one in the DC-DC box, connected to the output of the DC-DC boost converter.
(Ignore the battery capacity on A01, it wasn't calibrated at the time I took the pictures).SmartSelect_20211208-215422_Gallery.jpg20211208_134132.jpg20211208_134137.jpg

I powered the display with a separate USB cable coming from a portable power bank. As you can see the wireless icon at the top left, the display is connected wirelessly to both shunts. Pretty cool little gadget!
It also has some neat features like relay control, but I don't have usage for it currently.

That's all folks... stay tuned :)
 
Which LiFePO4 did you get? I'm curious...
Also, please remind me what car you have?
ohmmu, model 3
Are you getting along with your VIPV setup?
Still planning phase :) replacing the PbA with LiFePO4 is strongly motivating to get acting soon ? maybe after checking airbag locations the dash might be an ok spot for placing some PV like in this project.

Cheers!
 
Yay, I got the Juntek display to work with both shunts!
For some reason, one if them wasn't connecting wirelessly at all, it took me some time to figure it out.
I initially thought that maybe because the two shunts are different models of the VAT series (one is VAT-1100 the other VAT-1050) that they can't use the same display.
But after doing a factory reset, it started to communicate! So I set up each one with a different address (A01 and A02), and now I can see both with one display (but only one at a time of course).
I also connected the DC-DC box to the solar box, and did some more testing. I simulated the car's load using the DL24 battery capacity meter. All is working well. As mentioned, in order to connect them in the car I would need a longer cable and I'm all out of XT60 connectors for now.

Here are some pics. A01 is the shunt in the solar box connectedto the battery. A02 is the one in the DC-DC box, connected to the output of the DC-DC boost converter.
(Ignore the battery capacity on A01, it wasn't calibrated at the time I took the pictures).View attachment 75032View attachment 75033View attachment 75034

I powered the display with a separate USB cable coming from a portable power bank. As you can see the wireless icon at the top left, the display is connected wirelessly to both shunts. Pretty cool little gadget!
It also has some neat features like relay control, but I don't have usage for it currently.

That's all folks... stay tuned :)
nice enclosures!
 
Just an update on both my disputes:
One for the 20Ah Lifepo4 (buffer battery) and another for the 45Ah Lifepo4 starter battery.

I won them both!
I ask for a $20 refund on the 20Ah because it gave only 17.5Ah, and $100 for the 45Ah which gave only 20Ah.

For these prices after refunds, it was totally worth it!
 
Just an update on both my disputes:
One for the 20Ah Lifepo4 (buffer battery) and another for the 45Ah Lifepo4 starter battery.

I won them both!
I ask for a $20 refund on the 20Ah because it gave only 17.5Ah, and $100 for the 45Ah which gave only 20Ah.

For these prices after refunds, it was totally worth it!
good going navigating the disputes?
 
Well, I have this crazy idea that came up a while ago, and now I'm going forward to implement it.

The motivation is like this:
We all love solar, many of us like/have EVs. We would love to put some solar power to use in order to extend our EV range.
But as some of us know, connecting solar to an EVs high-voltage battery system is complicated, not so practical for the amount of solar panels we can mount in/on an average car, not to mention dangerous.
So my idea was to offload some the 12v loads of the EV (which can get as high as 20-30 amps at times) using solar. Having that most if not all EVs have a 12v system, that is supplied by an on-board DC-DC converter (taking high-voltage of the main battery pack and supplying 12v output), offsetting some if this load will in effect save some of the HV battery's capacity, and extending overall range.

I started a thread a while ago on an EV forum, so for brevity, you could just start reading about the idea there (I suggest you start from this post of mine, as my ideas have changed a bit from the original post, but feel free to read it from the beginning):


Anyways, I would like to hear your opinions and thoughts about this.
I'm adding a basic schematic of my idea here. I will keep you posted as I advance in this experimental project of mine :cool:

View attachment 69532
 
Well, I have this crazy idea that came up a while ago, and now I'm going forward to implement it.

The motivation is like this:
We all love solar, many of us like/have EVs. We would love to put some solar power to use in order to extend our EV range.
But as some of us know, connecting solar to an EVs high-voltage battery system is complicated, not so practical for the amount of solar panels we can mount in/on an average car, not to mention dangerous.
So my idea was to offload some the 12v loads of the EV (which can get as high as 20-30 amps at times) using solar. Having that most if not all EVs have a 12v system, that is supplied by an on-board DC-DC converter (taking high-voltage of the main battery pack and supplying 12v output), offsetting some if this load will in effect save some of the HV battery's capacity, and extending overall range.

I started a thread a while ago on an EV forum, so for brevity, you could just start reading about the idea there (I suggest you start from this post of mine, as my ideas have changed a bit from the original post, but feel free to read it from the beginning):


Anyways, I would like to hear your opinions and thoughts about this.
I'm adding a basic schematic of my idea here. I will keep you posted as I advance in this experimental project of mine :cool:

View attachment 69532
So just spit balling here but why not build a small trailer with fold outs and raising roof. And a 3hp hight efficiency diesel. To tow behind so no wind or steering issues way less areo drag. And could extend your range maybe 1500 to 2000 miles pluse very comfortable for all in vehicle. I could fit in a small area when stood upright. The whole thing would weight under 300lbs with leaves 200lbs for cargo.
 
So just spit balling here but why not build a small trailer with fold outs and raising roof. And a 3hp hight efficiency diesel. To tow behind so no wind or steering issues way less areo drag. And could extend your range maybe 1500 to 2000 miles pluse very comfortable for all in vehicle. I could fit in a small area when stood upright. The whole thing would weight under 300lbs with leaves 200lbs for cargo.
Thanks for your suggestion, but it's not really my goal.
First, as I drive an EV, I really don't want to use a diesel generator (efficient as it may be). I'm trying to pollute less, not more.

Second, while your suggestion might give alot more range, I was looking for a smaller way to extend my range daily by a few KM using solar. I could have built a trailer based solar system with solar panels, inverter etc...
and used that to AC charge the car while parked. But I'm not fond of towing the extra weight, and it would defeat my goal of trying to be more efficient.
With my (planned) setup, I'm hauling around only a few extra KG and that should give 1-2km more range each day.
 
Just to update here, since I haven't posted anything in a while.

Well, I've been busy with work and some other DIY projects, so this one was on halt for a while. We have very little days of good sunshine this time of year anyways, so I wasn't urgent to install my little system.

Meanwhile, I'm trying to improve the 12v LFP battery that replaced the LA battery in my EV.

The reason improvement is needed is because I have noticed that the battery can pull alot of amps (tens to 100) from the HV DC-DC converter. The BMS on this consumer (and cheap) 20Ah LFP battery is not restrictive enough, and just let's the small LFP pouch cells take a pounding.
I think it's a waste of energy to charge this battery with a hundred amps, because probably most of it will go to heat and will destroy the cells in due time.
Also, I'm worried that this will stress the HV DC-DC converter.

To confirm my speculations, I have opened up the battery, in order to put a JBD BMS with Bluetooth (had to use a Dremel tool to cut it, since it was glued shut and I wasn't able to overcome that glue).
As you can see in the pics, it's a pretty messy build with alot of insulation foam, using a no-name BMS and pouch cells in what looks like a 2p4s configuration:

20220114_114820.jpg20220114_114823.jpg20220114_115719.jpg20220114_115741.jpg

So I replaced the BMS with a small 35Ah JBD BMS with Bluetooth (that's all I had at hand at the time) so that I could monitor the battery.

I have seen it take a charge in excess of 100A for a few seconds, and then it drops to 50-80A, depending on the SOC of this battery.
This happens because I suspect the HV DC-DC converter is a constant voltage charger @14.4v, not constant current (which ramps up voltage as the battery SOC rises).

I have tried to disable charging on the battery, just to see if it works, and if I could go about using some of its capacity. It works, the car doesn't complain. But I think the capacity has gone down alot. When initially tested when I got the battery, it tested at 20Ah. Now I'm getting only around 10Ah, until voltage drops below 12v.
I suspect maybe one (or more) of the parallel cells has gone bad and is taking capacity down.

Anyways, my plan is to replace the cells too. I ordered some small 14Ah LFP prismatic cells (similar to the 25Ah Topband cells, but shorter) and will be replacing these pouch cells.
In order to control the charging rate, I will use a diode and a CC/CV DC step-down converter (which will be placed inside this battery box, which I intend to reuse).
The diode between the battery and positive terminal will allow discharging, but prevent charging directly from the battery terminal. The DC-DC step down will be connected to the battery positive terminal, and to the battery cells (in a 1p4s configuration). This step-down will allow me to take the 14.4v the HV DC-DC converter is putting out, and step it down to say 13.6-14.0v, and limit the current to say 5A.
Since the HV DC-DC is always charging when the car is on (which might be suitable for SLA batteries), I will set the DC step down to charge to a lower voltage so that the battery won't be at a high SOC for extended periods of time.

I think that this setup will not only extend the lifespan of the Lifepo4 cells, but also will not stress the HV DC-DC converter, and will be more efficient (having that Lifepo4 has a low rate of self discharge).

Another possibility with this setup, which might be even more efficient: Maybe (if capacity is good enough) I could disable charging (by the BMS control) during regular use and enable it only when the car is plugged in to AC power for charging. The car also powers the HV DC-DC converter when charging from AC, so basically I will be using AC power to charge the LFP battery and not the waste/use the power stored in the HV battery.
Again this will be possible if the 14Ah capacity will be enough to supply 12v power to the car, say for a week, since I usually charge the car once a week.

I'm kind of sorry that I didn't choose a higher capacity than 14Ah for the cells, but I wanted to keep the weight down (that was one of the goals for replacing the LA battery in the first place) and because battery price has risen, especially shipping fees.
These 14Ah were pretty cheap with a reasonable shipping fee, but I will have to wait a while for them to arrive from China.

So in short: my current goal is to improve the 12v LFP battery, make it more efficient and safe. Only then will I move on to my original plans for extending the EV range using solar and an additional buffer battery.
 
Well, it's been a while and I've been busy...
I haven't installed the solar and battery yet, as it's been cloudy, but now the sun is out, so I will be doing it soon.

I wanted to update on the 12v Lifepo4 battery swap.
As I mentioned before, the main problem with the 12v Lifepo4 was the massive charge current caused the by the onboard constant voltage DC-DC converter.

But I think I have found a solution: Integrating a CC/CV DC-DC step down converter into the battery.

So I have built a new battery, in 2p4s configuration from these 14Ah cells:
20220512_105919.jpg

The step down is a 300w XL4016 CC/CV. I have ordered and tested some other ones, but this one enabled me to set the voltage (14v) pretty close to the car's DC-DC converter (14.4v). Other step-down would only allow to set the output voltage to around one volt lower than the input voltage, which is too low (13.4v) to properly charge up the battery and enable balancing at the top.

I don't have a schematic, but in general I have a high power diode on the positive output, so that the car won't be able to directly charge the battery:
20220520_162503.jpg

But the input to the DC step down is taken from the input (the smaller cable coming out of the black fusebox).

You can see the DC step down inside the battery box:
20220520_162514.jpg

I have epoxy glued it to the bottom of the box using some plastic studs, and put extra glue from a hot glue gun just about on everything to make it more robust to vibrations.

So the DC step down gets 14.4v from. the car's onboard DC-DC converter and outputs 14v to the battery (via BMS of course).
I'm using a JBD 60A 4s BMS.

I just started out testing, and so far everything looks great!

I have attached one of the BMS's temperature sensors to the DC step-down's heat sink, just so that if it would get too hot, it would stop charging at a preset temperature via BMS.

The only minor drawback of this is that the DC step down has an LED on constantly since it's connected to the battery. It draws a few mA, so I'm not so worried about it. It's probably gonna drain the battery alot less than an SLA's battery self discharge rate.
I could have removed the LED, but really didn't want to fuss with it.

I set the constant current to about 6 amps, which is enough IMO, since the battery won't get discharged very much anyways in an EV and I wanted to be on the safe side heat wise.

After I will finish testing and charging, I will try to install the new battery in the car and will keep you guys updated.
20220520_164614.jpg
20220520_165143.jpg
20220520_165132.jpg
 
I've been busy, but here's an update:
After testing my previous build of the DIY 12v Lifepo4 for my EV, I found a major flaw:
Since the grounds were common (input and output), the DC-DC step down had no way sense the current. This meant that whatever current was at the input, was also at the output. Luckily I found this out while testing on my workbench, and not in the car (that would have fried the DC-DC step down for sure).

So after researching this issue, I think I understand it more: Taking a look at some common schematics for DC-DC step down converters I learned that there is a current sense resistor between the input ground and output ground. This means that the charging current MUST flow through the input ground (car's onboard DC-DC converter) to the output ground (battery).

I have tested this with a few different DC-DC step downs and they all behave the same.

So here's my solution: Isolate the input ground from the output. Meaning that the negative battery terminal is connected ONLY to the DC-DC step down's negative input terminal. The negative output terminal is connected to the BMS's negative (P-) connection.

Sounds great, but this also has a small flaw: This means that discharge current will actually flow through the step-down's negative terminals. Since my car is an EV, that's not so bad, because there only be around 20 amps of discharge current which is also only for a second or two until the car powers up and the onboard DC-DC converter kicks in and supplies power to the loads.

But, I wasn't really sure the DC-DC step down that I used (with fairly small terminals) will handle this. So I started searching for something with some beefier terminals. That wasn't an easy task because alot of DC-DC step-downs must have a fairly high difference between input voltage and output voltage. For example, most of them would only be able to supply 13.4v output from a 14.4v input. 13.4v is too low for cell balancing and for getting a good enough SOC.

In the end I decided to try this one:

It's a proper CC/CV charger. It looks and feels higher quality than the previous one.
It also has a large heatsink and a small temp controlled fan (although I'm not sure how much that will help to dissipate heat inside the battery box.

It worked pretty well, and can supply around 13.8-13.9v from the 14.4v input, although at this difference the amperage is throttled down a bit.
So I set it at 13.9v (to allow cells to balance) and 5 amps constant current.

Here it is inside the battery box:
20220713_172235.jpg

I also added another "ideal" diode from the positive input to the step-down, in order to minimize current flowing through the step-down positive while discharging.

The only problem is that this thing has an LED display which draws around 25 mA. I can disconnect the display board, but it seems there is some circuitry on that board, which causes the step-down not to work if the board isn't connected.

I think it's a small price to pay for shaving 12kg compared to my lead-acid battery.
The entire battery weighs only 4.5kg with 25Ah capacity. The SLA was around 16.5kg.

Well, "does it work in the car", you ask?
So yes, and a bit no...

It works great, charges up slowly to 13.9v etc... As for heat, I've noticed that the battery doesn't heat up much. It's around 3 degrees Celsius either above or lower than ambient temperature. It depends if the car has been in the sun (gives the cells time to absorb the heat) or not.

I haven't seen more than 37 deg C, on the BMS readings (has 3 temp sensors) even after being in the sun all day. Usually while driving I guess the air flow cools it down a bit, by a degree or two, as far as I've seen.

The problem is that meantime my car had gotten an ECU firmware upgrade, which doesn't like Lifepo4's. It give me a fault when powering up the car, stating "battery charging fault". This hasn't happened with any of my previous experiments with Lifepo4's, so I suspect the firmware upgrade is to blame. It seems like they've added some resistance or charging check on power up.
But the car still drives fine, no actual problems. It's just a warning but an annoying one.

To tell the truth, I pretty much fed up with this idea. I wanted to swap to Lifepo4 for the weight saving, less self-discharge and flat discharging curve.
But it turned out to be overly complicated (and somewhat costly from all of this trial and error).

What would you do in my case?
- Ditch the whole idea? This would probably mean ditching the solar part too, because if I stay with the heavy SLA, and add solar panels and a buffer battery, the added weight would probably negate any range savings.
- Switch to AGM (maybe something lighter than the current SLA battery) ?
- Leave it as it is ? (it's OK, but I think I would need to revert to the SLA every time I need to do a yearly checkup or MOT etc...)
 
Last edited:
Well, I decided to ditch the 12v Lifepo4 EV starter battery and just ordered a 12v 26Ah deep cycle AGM, that weighs half of my current SLA.

This should allow me to continue with my solar plans, reducing weight and having less self-discharge.

The 12v Lifepo4 was an adventure, but in the end too complicated. I hope car accepts the AGM without bringing up faults, as opposed to what it did with the Lifepo4.

Just to recap: My initial plans weren't to replace the 12v SLA "starter" battery, but advice given here made me think twice and go along with it.

It has taken me alot of time to fiddle with, and try to get charging right. But that was a difficult task having that the onboard DC-DC converter is a constant voltage power supply, using the battery terminals to charge and discharge. Maybe I could have found a way to isolate the DC-DC converter, but I really didn't want to make any permanent changes to the car and/or risk damage and warranty issues.

Sorry to disappoint, but that's the reality.

I will keep you guys posted as I progress with the in-car solar.
 
Just got my 12v 26ah AGM battery. So I installed it in my car and guess what? I'm getting "battery charging fault" warnings again! Pretty disappointing!

I'm not sure what causes it, but my SLA doesn't do this.
I have two speculations:

1. Ever since I got a software upgrade on the car, (that fixed some things) I think they added some battery checking algorithm.
It might measure resistance or something and it's configured to checking SLA specs. Probably anything other than SLA might give a fault.

2. Same as above, but maybe it's not checking resistance, rather surge capability.
Before installing the AGM, I connected a clamp meter to the battery positive lead. When starting up the car, at some point, there is around 50 amps load on the SLA battery. But this is very short (maybe less than one second) and then it drops down to 20 amps and decreases.

Did the same for the AGM, but only saw around 25 amps load.
I suspect that maybe because the AGM is fairly small (26Ah), it probably had a some sort of voltage sag if the car would try to pull 50 amps (which should be fine for only a second or two) and the car identified that as a fault.

I'm stumped...

My Lifepo4 was also around 25Ah, so maybe the same behavior happened? I don't know, as this all happens pretty quickly when powering the car up.

I think I will do another experiment, just to eliminate the theory that it's a capacity/surge problem, with a 100Ah Lifepo4 battery I have (which is too large to fit in the battery compartment, but just for testing).

Pretty annoying that other than the charging current, I previously was successful with swapping to another battery chemistry. But now with this ECU software update, I'm getting these faults.
 
Back
Top