Solar Install; the on-going saga

The 12v alternator/solar changeover relay is in. The make and break relay is out...soon as it dawned on me it wasn't doing anything the c/o couldn't. Swapped in a 40 Amp breaker on the TS output.

I resistance checked my DC breakers and now understand why but not how they are "never really off" as Solar Bob describes. They're just big resistors. My 16A breaker was reading 3KΩ open. ...Meh! I still prefer them to fuses for their switchability for select applications.
Still haven't had a decent day to test the tiltage bonus. On the whole harvesting is very low, been too busy to keep track but I reckon it's under 5Ah daily with all these grey skies usually I see more volts than amps. Still though it's keeping the batteries above 12.4v which is enough to halt sulphation even if it is just surface charge (so goes my theory anyways).

I've tried massive capacitors an assortment of ferrite cores on the DC and AC sides of the traffo - no improvement and it's still head wrecking. I reckon it's the Tristar PWM picking another fight.
The new c/o relay buzzes too when the TS is operating with no forward current across the relay. I am running them backwards to tradition ie. power in the 87 terminals and power out the 30 terminal I doubt this has anything to do with it though since the coil is as prescribed. Whatever about the traffo the relay won't last long at that. I suspect it's inductive interference. So now I have to isolate the relays from the traffo without diodes or relays ...short-term solution seems to be a screwdriver :rolleyes: ....ooh vacuum tubes maybe :D:cool::eek:

If I wasn't so stubborn I'd just buy a good mains charger....

...think I'll email Morningstar instead

Yeah I'm having lots of issues with the c/o relay, I need an opportunity to tinker some more with it before I write it off. It's supposed to have a resistor in parallel with the coil to possibly prevent what I am experiencing but it's doesn't so I'm going to add one rather than ship it back to the UK as it's not what I ordered even though it has the same serial number.
Basically my electrons are fighting all the time. Tristar PWM plays havoc with everything digital upstream. The c/o relay is buzzing when moving from energised alternator charge to normal open solar charge, I think it's latent current screwing with the electromagnet coil. So I have to open circuit the solar module and reset the solar controller to make it stop. Same happens when I turn on the traffo charger...which I'm still looking for a way to soft start. Any ideas?
It's a 70A continuous or 100A instantaneous relay but judging by the heat of operating at 15A continuous I'd happily derate it to 35A/50A.

The company I ordered it from messed up half the order and I messed up the rest so it's all somewhat frustrating.
Post #104 has the most recent wiring diagram "whole kit and kaboodle" except the M&B relay is superfluous and has been removed & the 24v c/o is currently being returned to sender as the wrong gubbins.

holyhutzpa wrote: »

Sounds like it's very difficult to get the solar to charge both the leisure and starter batteries (you had a manual switch at one point?)

Nah that's easy! (see last link in post #89).
I'm trying to charge two different size, type and variation of batteries from three different generators with minimal manual interaction with the same charger that is a solar controller not a charger with just one (or two) switch(es) to rule them all!

There's a manual direct 20A switch that links the positives of all batteries this has many handy applications.

holyhutzpa wrote: »

I'm sure you'll turn your nose at the inferior charge controller. What am I missing out on? The ability to charge with solar while driving isn't a big issue for me - and not too bothered if the solar only tops up the starter when manually switched over (when the van is idle for a few weeks, for example)
It probably won't get as much out of the leisure as the tristar controller would...? I think the tristar is slightly smarter in this way.

I use it as a digital regulator PWM 3-stage charger while driving it's much better for the batteries than the alternator constant voltage charge....mmm look at that heat sink it's the largest component!

Yup, yer dead right, I think that solar controller is junk. Have a look at morningstar's lite controllers they have 10 amp SunSavers I'd recommend. That controller won't top up the batteries it'll just tease them. At the very least have a look at the Steca controllers that Hollybrooks have and ask that to be bundled instead. I'm not recommending them either but they're better than that dreadful Chinese one.

Beware if you Silkaflex your module to the roof you'll have a very hard time tilting it at a later date unless you have a pop-top or a crowbar.

holyhutzpa wrote: »

What am I missing out on?

2 years battery service life due to under/over-charge (without knowing the product..it's just a pay peanuts get monkeys philosophy) & enough cable 6m = 2 x 3m 2-core. I'd consider going thicker and longer.
I think one ought to be considering the importance of a good solar controller more than photovoltaics if not equal. The charge controller is the keystone.

holyhutzpa wrote: »

I've also been eyeing up the trimetric battery monitor too. However, it sounds like a good application for an arduino - an arduino should be able to measure current leaving and voltage - and add up the current leaving over time, and you should be able to calculate a virtual %age discharged (however this will be based on an assumed capacity)... Finding the time to make one might be difficult though.

Apart from the SmartGauge every other decent amp hour counter needs to be shunted which is problematic as I have no perfect place for a shunt. I can't isolate the grounds as the alternator doesn't always charge the leisure bank as is my preference. Even if it did it would effect the bank capacity value. So alternator charge always throws the meter out. The TM works best in the Summer when I don't need the alternator...or a monitor for that matter because it usually says full. SmartGuage cleverly uses an algorithm based on Peukert's equation and is a unique approach that I think may be best. It's also cheaper. The TM works for Solar Bob because he has only one generator (solar) and one battery bank, I've got three generators so far and split batteries. Once I get my 48v domestic off the ground I'll probably swap for a SmartGauge into the camper and put the TriMetric on the forklift batteries.

Capacity is never what it says on the tin and changes with age and temperature, after a while with an Ah counter you'll deduce your capacity and you can verify it with a hydrometer and googleations.

I've been reading up more on the SmartGauge (love that website that dude truly understands lead acids).
The downfall of the SmartGauge is it has no ammeter just a processor. This means you can't read instantaneous charge/discharge so it'd be best coupled with a digital or 30 - 0 - 30 (depending on system requirements) analogue ammeter, externally shunted to save on wire distance.

The diode didn't work as per previous image (that's for a much lower power application I reckon).

Now that I've had a chance to look at the system I found solar's been down 5 days. Last year's set up and batteries woulda been pancakes at that rate...I was only 20% discharged and 12.5v, didn't even notice other than the "is it really that cloudy?" suspicion.

I can confirm the buzzing was the relay coil alright. I reckon I've knocked about 1000 operations off my 700 000 :pac:. The the buzz was the electromagnet firing at high frequency.
I'm not sure why solar was down but probably I wired it ar$eways because it shouldn't have been affected. I know I definitely wired it ar$eways repeatedly today as I was having difficulty with inverting the pin out to look at the back of it :rolleyes:.

Anyways, the way to get the coil to behave is to build an Resistor-Capacitor Snubber, it's a soft short that forces the latent current into heat dissipation through the coil and resister and the capacitor holds the gate.

I don't think both the diodes are necessary but i used them anyway because I had them and the one on the coil ground was the nearest battery ground so I thought no harm!

[Edit: Confirmed! The diode on the +ive side of the coil is only preventing back flow to an open switch....the reason I put it there is I was copying a diagram from a dimmer circuit and I had one handy. Neither diode is necessary for this circuit.]




I got most of the details of this from google, I wasn't bothered mathiplying anything. The resistor I guessed.
I bought the 1N4001s diode, pulled the capacitor off an expired DC laptop charger (and noticed a few more in the Halfords charger), the last Diode is the 60A power diode.
[EDIT: the power diode is definitely not needed. It's actually in parallel not series as the diagram indicates with one side going to ground the far side going to battery +ive...so rather than me changing the wiring diagram you could imagine them not being there ]


C/O relay works perfectly now. I had 1.4A coming down this morning in direct sun 11am with the module tilted.



The Capacitor is soldered in-line as well as the resistor, you can see the cap, held by a cable clip above the relay holder. The reason my relay holder looks in rag order is because I've had to modify it to accept uber-relays.

I've been lurking in the MorningsStar web-site.

The TriStar does soft start.


The problem is it's over spec-ed.
It doesn't read my 180Ah bank as ever in need of bulk charge because it's designed to charge a 345Ah - 1000Ah bank.
Which is really only a problem for mains charging when source current is relatively abundant.
I could solve the problem with more batteries .
Or soft starting from solar/alternator.
Automatically regulating a traffo voltage ramp still eludes me. Custom profile perhaps. Hmmmm....

Noise control;

MorningStar Corp. wrote:

Pulse Width Modulation (PWM) charging may cause a buzzing noise in the system during charging. PWM switching noise can sometimes be heard in the controller itself, the wiring or wiring connections, or the system loads. AM radios are especially prone to PWM interference. Noise in the controller or wiring is caused by mechanical resonance when current is switched through the circuits. There is little that can be done to remedy this issue.

However, PWM noise in the system loads can usually be reduced or eliminated by the following:
1) Minimize cable runs between components
2) Twist power pairs(+/-) to reduce radiated noise
3) Good system grounding
4) Add capacitance across load power(at load input, 22000uf or more)
5) Add a line filter. some have had success with car audio filters that eliminate alternator whine

If none of the above work there's always the option of cracking the TriStar cover and setting a dip-switch to a 1hz on-off charge waveform instead of PWM temporarily. (Items; 1 and 3 already done, 4 tried with a lesser cap. I'm dubious twisted pair will do any better than ferrite cores...that leaves power line filters and a bigger cap.)
I understand the traffo is not a system load but for all intents and purposes its a closer approximation than a solar module.

I'd just use an online calculator myself.
Yeah pretty big alrighty but so's my charger. Pity I can't test it in this weather because it's too cold to over-heat the fridge...might have to block the vents (because I never added the over-ride on switch....nor the potentiometer for that matter).
I calculated a 180Ω resistor for 720mA of fans (but they run lower than spec-ed) tried this first and got negligible drop. The 1kΩ was giving me a drop from 14.7v to 13.4v on charge and 12.8v to 11.6v when resting without running the fans. This keeps me inside the 13.8v < 6.0v thresholds even with temp. compensated open lead acids so pending further testing I think it works. Any idea why this is? It's in series.

Please PM ryme

I suspect this should come with a "Don't try this at home kids" waiver...the more I think about it though the more I'm inclined to think no-one would be bothered anyway.

I've been looking at these last two solutions;

MorningStar Corp. wrote:

4) Add capacitance across load power(at load input, 22000uf or more)
5) Add a line filter. some have had success with car audio filters that eliminate alternator whine

A line filter is just an inductor - capacitor. So that just leaves point 5.

So...erm yeah a 25amp inductor...I'm going to have to make this because it's not been invented yet mostly because I suspect it's the wrong approach.

I found this DIY filter;




Except mine had to be 25A coiled at full load :eek:.

Ferrite cores don't often get this big either so erm... steel = iron = ferrous = galvanised steel 2ton shackle :rolleyes:



Gubbins



Inductor; 10mmSq cable 105°C.

The filter doesn't mute the noise it just moves it to the new inductor :rolleyes:
....sound proofing!!!



Sound proof box.



20mm pine - cardboard - Thermawrap - glue - Thermawarp - wet sand - noisy thing!


It's reduced the traffo noise to maybe half with less rasp!
Damn magnetrostriction!
That's it now. I'm going to have to draw the line here and say that a further investment on this is pointless better spent on a real charger.

Dip switching PWM to ON - OFF charging is quiet.

I may or may not install my frankenductor depending on how well it fits into a dark hole I have nearby.

Oh yeah and don't forget to discharge those monster capacitors before you wire them kids..:pac:

spaceHopper wrote: »

Should you not have a reversed biased diode across the relay coil to stop any back emf when you switch it off.

OOps yes! There is a diode actually but it's between the TriStar and the coil because the TriStar was powering the 24v side of the traffo.

Hrm hold this - put that there - Bosch

I'm starting to regret not planning my "electrical cupboard" better now I'm getting pressed for space.
Then again there was no way I could have seen the end of this project from the beginning.
I'm spending a lot of time fishing for screws with a magnetic grabber.

24v C/O relay is in, I need the power diode in series with the traffo as well as the RC in parallel with the coil to get the relay coil to behave. Induction-tastic lately.

This relay was originally to prevent the traffo powering the solar module. It's not harmful just wasteful.
It also protects the starter battery from inadvertent 24v constant voltage charge :eek: , as well, it makes mains charging plug and play.

Sir Liamalot wrote: »

I put this on the fridge fan live in...

spaceHopper wrote: »

It's a 100 K Ohm resistor, bit big?

The aulde sacred penny dropped; I was checking open circuit voltage.
You're right Space, thanks for pointing that out. ...pull it apart and try the circuit voltage with a 180Ω resistor so...
...that can wait til Summer :pac:

I'm starting to have issue with the charging system being too good.
I have to turn the solar module off on occasion to prevent short cycling.

Ever notice how battery manufacturers never complete the first 10% of their DOD vs cycle life graph?