Solar Install; the on-going saga

So I left the motor alone long enough to get some fridge data.



Average temp 20°C night,

25°C > 29°C Day





DSC = days since (100%) Charged.


I ran the test on a live system.
All batteries were fully charged to specific gravity 1.275 (4 years old and testing as new ...didja hear that li-ion fanciers?! )

Float current = + 0.1A
Running (parasitic) Load = 0.15A
(0.1 + 0.15) x 24hours = 6Ah system self-consumption.

Any time I've investigated a non-operational solar panel I've diagnosed a pi$$ poor install. Solar panels work for >30 years.

I'm not a proponent of inverters in installations you can touch opposite walls simultaneously. However they have their uses.

This is a proper job inverter!



This is not the story of a proper job inverter. This is a story about fit to task inverters....





Behold the world's only (that I know of) low frequency solid state true sine inverter.
220Vac 50hz 300VA (600VA Surge - 15 mins)
with auto-sleep load detection

Editable LVD & HVD thresholds (not easy)
I like low frequency inverters because you can neutalise them and then use earth leakage protective switchgear, this to my understanding is safer than the other muck switched earth or CTE high frequency flimsy el-cheapos...

Wellrr the SureSine's sure stable.






You can tell a good inverter by the resolution of the peaks.



Suresine (no load)






Studer (no load)
>10yo design I might add...newer is not better!

This is the unit it is replacing.

It's a 600VA MSW trinket I got in Maplins back in the day and have given it light duty service powering a boat anchor while underway and not much else.
Once upon a time I took a disliking to the LVD Buzzer and removed it with extreme prejudice.






That's some top of the class modification alright...:rolleyes:
Sine is not the correct term at all.








Here's my test rig. I'm using the 24-ish volt isolation transformer in my soldering iron station, that I terminated to panel mount BNC, as an isolated 10X reduced O-scope probe.




Waste not, want not...I built a Mobile 600VA Unit with it for well..you name it...it ain't fancy but tis better than burnin' dinosaurs.


600VA Inverting Square Wave
230Vac 50hz, 1.5kVA Surge

• 2 x USB Outlets (2.1A Combined)
• 3 x 13A Outlets
• 12Vdc WW LED Strip switched panel light (redundancy)
• 2 x GU10 7W LED luminaires
• Voltmeter
• 2 x Switched Inline 10A connectors (for ancilliary lighting)
• 100A ANL Fuse with see-through holder.
• Janky Croc Clips in flexible Conduit.
• 7W Solar Panel Superseal Connector (offset voltmeter)
• CTE, Double Insulated (except for inverter casing).

Need I say more?





Jigsaw,
Brushed Motor,
Phase angle fired.
No Probs...forgot to check speed control functionality.
[EDIT: Speed control works ok, torque stalls below 1.5 (of 6) setting (super bad for jigsaw motor...no cooling)]

Huzzah! Another meter!



I've been sitting on this since September.
Mostly because it's self-calibrating and I wanted to give it a chance to prove itself.

The verdict? It's pretty good. Easy install, self-calibrating.
Accuracy...hehe




Voltmeter: dead nuts on.
SOC: It's conservative, but it's only ever 0 > 10% below reality.
It dives from 100% SOC to 90% SOC, overshoots on discharge and holds low until it syncs and starts tracking again.

The TriMetric is still on the money moreso than any other but that did take me a year to calibrate. I calibrated the BMV to the TriMetric but I only use it as a volt-ammeter for all the previously stated grievances.

It has programmable relays but I don't use them and an alarm feature I've set the low to 11.8V so it disappears and won't talk to me when I abuse my battery (display turns off to save energy).

I bought it used and reset it on commissioning, after the 4 months it's generally only reading ~5% low of actual up from 10% on install.

So mostly it's reaffirmed that my TriMetric is the best you can do and my cal is pretty darn good.
Having said that for most people that don't wire their shunts properly (soldered sense terminals, twisted pair, zero disparity between sense conductors, realistic capacity assessment (-10%) and efficiency (99% @ < 95% SOC, 50 - 60% to 100% SOC).

I'd probably recommend it for the non-power users.

I now have a very informative combinational cluster.



(infering 44W going to running loads)
All of which I can read without leaving my seat :cool:

Didja hear that Schlictron, Studer, ApprenticeVolt, Xantrex, BEP?...it's a great feature I think...having a monitor you can monitor without glasses...simple inexpensive thing to do, yet most > €150 offerings on the market don't seem to think so..:rolleyes:

In other news my MorningStar Controller meter is tuned. :mad:
It's programmed to lie in certain ocassions (not uncommon...you know like battery chargers that tell you your battery is charged because they don't fancy finishing the job, I've a great example of this on the way Victron).
As soon as it drops below 12.0v it adopts a 100mV under-reading.



This is not the first time I've seen this craic from MS either...



Oh well small price to pay for the greatest charge controllers on the planet.


In any case I documented the SOC readings over a course of a week with large cycle depth, high current fast charging, extended microcycling and inclement weather (cold temp derating) and here are the results.
(Attached pdf has written values, worth a look for anyone still thinking a voltmeter is good enough)



You can take it that the Trimetric is correct, I verified it with measured replaced energy.

Another take away I find hilarious is this extract from SmartGauge FAQ:

Gibbo, SmartGauge wrote:

Why doesn't SmartGauge ever reach 100% charge status?
This is a very common problem. And the answer is very simple. It isn't SmartGauge that is at fault. It is the charging system. SmartGauge isn't reaching 100% charge status because your batteries aren't reaching 100% charge status. The solution is to sort out your charging system.

As you can see below The Smartgauge reads 100% charged when my battery is 85% Charged....:pac:......outtov 120%..:eek:



Just how bad are standard systems (rhetorical Q..)




Anna nuther thing Gibbo...

Gibbo, RE: Ah counters wrote:

However, they are useless for the purpose of monitoring the state of charge of batteries and a SmartGauge does a far better job, more reliably and is simpler to understand.

Not true..or yer talking through yer hat!
It all depends on the installer.
And of course if your AH counter reads >100% charged (only TriMetric does of all the meters in all the planet) then you can compensate imperically per cycle..

Having said that after a week, any inaccuracies in your Ah counter setup will have compounded to a rather large discrepancy where the Smartgauge will not suffer from this. ...it'll only be a consistent ~5% too low instead..


If it does sync up even better in the coming months I will update.

Running the fridge is easy peasy...falling asleep with the laptop on all night...well that's a challenge.

My lecky fridge uses 2% the energy my lemon did. I'm on my second gas bottle since I installed it. €75 saved so far and probably another €100 in unspoilt food. :cool:


I got me one of these to put to the test...



£50 Spares, repairs: Reported Fault; never finishes charging.

Rumour has it it's not a great unit compared to a simple cable upgrade.
I've asked politely for data from the Sterling and they have declined to produce any. Engineering Company you say?!?

Can it beat my Albright magnetic latching contactor over 35mm²?
Producing:







I lashed together a test rig...




Low and behold..



Now why you would want a battery to battery charger to float I don't know because it takes days to charge a FLA battery and I usually don't drive that long.

Not to worry though it does float prematurely just like the rest of the market offerings for peace of mind.

What the eff is this?



Unit Main Ground.
It's a 50A Maxi Blade fuse on the ground on 4mm² cable?
That's 40A cable!...okay 105°C rated er...ah I'll give ya 60A on a cold morning.

Whadya using it as a potential divider onna common grounded system? It's just ...not.... sigh!...



A final look at the krust. Marvel at the high current capable plumber style soldered brass studs and lugs
(and yer ProNautic ProCharge Ultra...is just as lacking...least it was back in 2015 :rolleyes:...it was the best performing off the shelf charger I've seen ever though....that one time I used it...factory refurbed unit...ie. resurrected)




Hey look Charles...this is what cable stress relief looks like...for the suggestion box like...I ferruled yer janky fans too!



The black aluminum tape is a duct for cooling. I tossed the lid so I wasn't promoting the product and also to save a few mm for shoe-horning into the lecky dept.

..let me take you back to a time in my youth, the time I was wiring a live circuit...semi-live, and blew up a 120A battery to battery charger...oh I remember it like twas thuther day...

No solace was found in the thought; I, such an expert in my field, has learned to identify the distinguishable difference in smell between plasma and fried semi-conductor...

As I was inserting my B2B lovingly on the floor,
forever more,
with a live input 12v,
a floating ground (not connected),
didn't I spot weld my shunt with the B1 output,
ina fearsome conductor wrangle...

..convince I did I "ya can't smell magic smoke from an IP68 appliance,
sher twas just probably plasma!"
...crack on I did I..
...but while lights indicated
current wasn't moving
...no charger can be this bad

Well're there's nay'er fer it but sniff autopsy...

And low..



Fet 11 stood; a pool his own entrails...

left this world so violent
his buddy beside,
poor blighter,
is plastered,
in his gubbins,
galore!









...sigh..

What is MPPT an why do we need it?

A typical solar panel is configured as a string of 36 x ~0.5V Solar Cells.




Which produce 19Volts at full load.
While our battery bulk charges in the 13.6V > 14.2V range.



What PWM controllers do is reject the power above the bulk voltage.
What MPPT controllers do is regulate and utilise it.

Why that's a lot like saying...



Well yes it is... but,
...it's more of a 36 cells operating at 83% duty thing rather than 30 are flat out and 6 are idle.

But you can use that power with MPPT...or you could get the right solar panel for your system.


Would that be the same in effect as this?



A 17% smaller panel that performs the same? Or a same sized panel that performs 17% better.
Yes, indeed, with PWM.

Unfortunately 30 cell, high quality solar panels are not a thing. 36 cell are the standard.

So were I to say a 100W standard 12v solar panel on a good PWM controller is setup to run at 83% peak performance. Mounted horizontally so derate another 30%... so it's a 60W panel on PWM, and a 70W panel on MPPT.

This is a very telling graph I pulled from a 90W solar panel.
Assuming MPPT can use everything under the curve so 100% rated.
Let's look at the MPPT is 30% better than PWM myth....



A PWM controller directly connects the battery to the solar panel. This makes the effective maximum power point of that system to be (battery voltage x solar current) fixed at around 14V in practical use.

That means only the shaded portion is available to PWM systems.
While the shaded and unshaded portion is available to MPPT sytems.

Let's interrogate that graph for more data..





Interesting findings.



How is MPPT 30% better when there's only 20% more power available from the system?

Cold weather uprating was it?



0.61V x 36Cell X 5A = 110W = 30% more better

If you ever saw 0°C and 1000Wm² solar irradiance at the same time and if the cell temperature wasn't generally 50°C hotter than atmosphere being that it lives in a glasshouse facing the sun...


What does MorningStar claim? 10% > 15%!

Here we have Normal Operating Conditions Vmp of a 60-Cell Solar panel ≥ 2x(12volt).

If this was 24V PWM system. MPPT would have a 3V advantage compared to it...

In a fortuitous turn of events I only blew up the easy to get to FETs

When I say easy to get to ...wielding a soldering iron from above a heatsink, around bodge jumper cables and capacitors in a teeny channel because the back of the board is an Epoxy slab.


Easy like not impossible..:P



I'm very glad these lived.





I've soldered the new FETs to the legs of the old. I cut them as long as I could. I've even managed not to solder any legs together (world of pain).



Seeing as the FET bus is in parallel and I didn't order enough reinforcements I ran the test at 75% complement and current limited it with a 12.5A charger boosting a fully charged battery.



It's cycling through charge stages and behaving like recovering..
I haven't load tested it or output current sensed.


She'll be right in no time...
"RS.. I need another bag of them FETs, 30 sticks of hot glue anna boxa band aids!"

I like it. The grunt and the charge tenacity.:cool:
She's sure aggressive & persistent.
Constant output, no more generation reduction from low alternator RPM in traffic, thanks to it's input load B2 regulation.


Sure expensive for a 120A 1200W Continuous DC - DC Charger.


Test Conditions:
5 year old flooded battery charged to specific gravity 1.275, watered.
C20 discharge applied for 10 hours.
100% rated capacity confirmed (temperature compensated).
B2B Charger wired on 35mm² from Alternator B+ to Service Battery Bus Bars with 35mm² feeder.
4mm² original equipment ground cable with onboard fuse installed.
Temp sensor installed.
Manually switched.


SOC at commencement of charge according to

Trimetric (trusted): 52.5%
Smartgauge (tuned low, tracking): 45%

Resting Voltage: 12.1V



B2B Inrush Surge: 92A




I can't tell who won.

The Sterling 1200W Continuous DC - DC Charger is €500, a used magnetic latching Alright SW80 contactor cost me €30.
My SW80 does E-Start, full rated load, 100% efficiency and can be fixed with either end of a hammer, also waterproof and fanless. :P


I like both.

It's extra redundancy.

High charge rate for windy motorways (alternator temperatures).
"Low" Charge rate for pootling through town (fan belt wear reducing)


This Sterling 1200W Continuous DC - DC Charger is this much better than a good cable job anna contactor.



Which is to say about 10 times better than standard and double better than my Split Charge Contactor in practical use.
It's most effective in the 70% - 90% range...it has filled under the curve very nicely in the happy battery zone.





Give me a welder, metal, a dedicated alternator and a pulley kit I can do you an even less favourable result with double the input potential, inherent redundancy and cooler alternator temps.


Or stick a 50Ω wirewound potentiometer beside yer dash ammeter on the D+ from the dash light above it.


0.5k€ for a double better charge powah source with short life capacitors in a hot sealed box?
Worth it?



I've derated mine to it's apparent nominal max operating current. It seems I mighta forgota put 40A back into the box when I reassembled it.
Dear Charles...maybe I left them on the table or is the rating 50% inflated or was it 67% efficiency? who nose? Maybe they'll come back? I'll be sure to let you know if they do!


I guess I didn't over-spec it at all. :rolleyes:

I can now replenish 90Ah of my 220Ah battery in 1.5 hours or 100 miles.
That used to take ~2.5 hours (~7x better than standard).


My Service Battery is getting a wee bit fond of it's evaporative coolant (H²O), arah about 100ml a month....the Sterling will probably raise that loss to ~125ml.

B2B Charge Profile: Flooded; 14.8V regulated...with temperature compensation 18mV per °C (fixed: conservative), what does naff all under 5°C because the output max voltage is 15.1V





Battery temperature at 60A sustained, C4 charge...nothing to get excited about.

"Soft Start" is a ~60 second activation delay.