Thread: Smart Battery Isolator

RawPwr (thrash?) - that seems to be a UIBI - they use "S" to control the relay (switching).
Besides, if yours has worked fine for 8+ years, how can it be wrong?
If I am right, then the Cole-Hersee S-type isolate is equivalent to a relay - eg, gnd - #85; S/alt = #86; batt1 & bat2 = #30 & #87.

bratnetwork seems to have a "smart" version which is voltage sensing. They are fine until they have problems. They also cost more than the UIBI, but are the only common commercial option for stator etc systems.


The cost for ~200A-400A capacity should not exceed $20-$25 for the isolator itself.
(Smart isolators are usually more expensive with inferior performance depending on user desires and application.)

Thanks for your input OldSpark. Well, I guess I should get another 8+ years out of this set-up then, right? .........Let the games begin!

I know. With either the smart isolator or a standard RV dual battery switch or even a HD relay, if your accessories are hooked up to your auxiliary battery, then your batteries are separated when the ignition is off. Therefore you can't drain your main battery. And your starter isn't even touched. Yes--you have to turn your ignition off--those lights on the dash are a good indicator to do so.

From what you wrote you said you didn't want a "traditional one" because of diodes and efficiency loss which simply doesn't exist-you get direct power from the aux batt.

"I originally took that wire and hooked it up to my IGN switch in the car bc i wanted it to shut off as soon as i turned the key off instead of staying open till the voltage drops enough to shut off" -- it sounds like you rewired it to defeat the intention and purpose of the manufacturer.

Adrian, you may not be fully comprehending...

Originally Posted by adrianl_888

... your batteries are separated when the ignition is off. Therefore you can't drain your main battery. And your starter isn't even touched.

If controlled by the IGN, then your (two) batteries are paralleled whilst cranking, hence sharing the starter current (iac battery SOC & path resistance).
May people blow fuses because of that. (Others have auto-resetting breakers.)

Originally Posted by adrianl_888

... diodes and efficiency loss which simply doesn't exist-you get direct power from the aux batt.

The efficiency refers to the diode drop. And the need to compensate for it - eg, a single wire D+ alternator will undercharge the batteries, hence premature failure.
Diodes are an expensive and inefficient method for high currents.

Originally Posted by adrianl_888

...i wanted it to shut off as soon as i turned the key off instead of staying open till the voltage drops enough to shut off...

That's how voltage sensing "smart isolators" work. Most are set to disconnect at 12.5V or thereabouts (hence reasonable sulfation). Of course they should be user-settable.
Even if they isolate at 12.9V, why discharge the surface charge? Sure - it's only a few hundred Farads, but still. And being above 12.67V, it guards against sulfation.


The bottom line however is that many do NOT want IGN controlled relays - they do want the system to be charge controlled and automated. Hence the "charge sensing".

you know i think im finally going to go oldsparks route. not for the computer though, but i have a need to have a cell phone on 24/7, and i dont want to risk the starting battery being drained so i think ill just use one of those little 7ah agm batteries used in ups's and such... ill just switch it on with a relay thats commanded by the alternator light. it would be a $5 solution where a smart isolator would cost $80.

not sure if i need to limit the current going into the little agm though. i dont think so.... as it stand my cell phone lasts almost a full day on its internal battery in its mode that its gotta stay in (the screen stays on unfortunately.) i think it draws around 200mA while keeping the phone running and battery topped off, so its pushing the limits i would want to parasite from my starter battery. i would imagine the 7ah battery would keep it running for 3 days anyway, which should be fine. its not critical that the phone stays on, just really convenient.

I'm not sure what I like most - getting approval or complimented, or finding another smart reader. [ IMHO of course! And smart-pun intended - especially in light of the UIBI (ha-ha!) and conditional on having common charge-lamp type alternator, NOT marine/bike/RV stator controlled alternators. ]

From here you can skip straight to CONCLUSION way below, else.....

The 7AH - like any battery - should have current limiting, but rarely is that incorporated. (More detail below.)

Reserve time... using data from my 2-page Yuasa NP7-12 data/pdf:
The Yuasa AGM NP7-12 is 7AH at C20 (the common 20-hr discharge rate) to 1.75V/cell (10.5) @ 30°C. IE, fully charged, it supplies 7AH/20H = 350mA for 20 hours at 30°C with an "end voltage" of 10.5V as measured across the still loaded battery terminals (not open-circuit!).
Its capacity drops to 6.4AH@C10 (20°C); 5.9AH@C5 (also 20°C but to 1.70V/cell = 10.2V) and 4.2AH@C1 (20°C & 1.60V/cell = 9.6V).

[FYI eg: 4.2AH@C1 - 4.2AH/1H = 4.2A for one hour to FULLY rating/spec discharged; or 5.9AH/C5 = 5.9/5 = ~1.2A for 5 hours, etc.]

Its capacity drops if colder, but it will enjoy a longer life. (Don't keep batteries warm to get more capacity - a bigger battery is much much cheaper!)


Yuasa specify a max charge current of 0.25CA Amps; CA = C20-rating Amps; ie, 0.25 x 7AH = 1.75A.
This occurs for charging at 2.275V/cell (13.65V) which is its specified float aka constant voltage, but it may need current limiting above float voltage and up to its max charge voltage of ~2.45V/cell (14.7V) at 25°C (or ~2.35V/cell (14.1V) at 50°C etc).


Deep cycle versus Cranking battery considerations....
As a Deep Cycle; fully discharged to 100% DOD (Depth Of Discharge), it should last over 170 cycles, but if only 50% discharged it should last over 420 cycles, or probably at least ~2.5 times longer.
It almost trebles again (1200 cycles) if only discharged to 30% DOD (ie, to 70% of capacity remaining as opposed to 50%). 30% DOD is the usual Cranking battery spec.
I wonder how many cycles it would last if only discharged 20% as per recent recommendations by my gurus... twice as many?
And be aware that immediate recharging when not being discharged is generally assumed for battery specs (hence why car batteries last so long...)


Why did I detail the above? Isn't it about time I detailed some typical specs rather than incredulous writings elsewhere in webland?
Maybe readers might instead start to connect with batteries - especially if the Yuasa NP7-12 2-page pdf or similar is downloaded.
In retrospect, there are only a few common-sense behaviours. It's their interaction that complicates things - and their artistry - batteries are "... more of an art than a science..." (too bluddy true!).
And though I may often use traditional Rules Of Thumb and older specs, the discharge time, DOD, temperature etc inter-relationships haven't changed much for Lead-Acid batteries (whether flooded, gel, or AGM).

Specs are supposed to define the minimum typical behaviour for (maybe) 95% of production, or maybe all but one in about six-million if manufacturing to six-sigma (I wonder if that can be done yet for batteries?). But do specs provide a good representation?
For example:
I have been using a Yuasa UXH38-12 38AH AGM in my ute for maybe 12 months. It has a spec'd max recharge current of 0.24 x C20 = .24 x 38 = ~9A yet it often accepts over 40A after a 15-second crank. (FYI - my alternator is 70A.) And I have a 140A reduction starter-motor instead of the standard 240 starter.
Another UXH38-12 in a daily-driven Mitsubishi L300 van with a starter-motor of (I presume) 250A or more has been used for over 3 years and it's still fine.
The UXH series AGM batteries has a design life of 10 years and is intended for UPS applications - eg, 10-15 minute discharge. These UXH38s were obtained when 5 years old.
FYI - I only recently discovered their (local) replacement price is AUD$670 - yet the much bigger UXH100 (or UXH 120?) is only ~$500 - ie, ~4x cheaper per current/power unit!

So both cranking and initial recharge currents well exceed the UXH battery specs.
Plus if intended for UPS, they are expected to rarely if ever be used except for occasional testing (maybe every 6 months?). Yet they are still fine after 1 & 3 years respectively. And I drive fast over VERY rough roads (they ain't a vehicle battery!) and I use my starter motor to un-bog and get over steep obstacles.
(IMHO - With Optima Yellow D-75s (48AH) selling locally for $450 to $550 and Yuasa UXH100-12 (100AH) for say $500-$600, which would you choose? I'd even take the $670 UXH38 over a $400 Optima 48AH!)

So how does my Yuasa survive? Even with its lightly discharged 4-fold-excessive 40A initial charge current (at ~14.5V), it quickly drops to under 10A within 30 else 60 seconds.
Maybe that is the artistry... the combination of low DOD and quick recharge with a high burst of current, else slower & lower current discharge with slower recharge (for maximum life - excluding the odd "reconditioning" high recharge current)?


But since a 7AH is so cheap, try it and see.
7AH was the best bang-for-bucks in small batteries. I got my NP7-12s for $35 (the same price as the 4AH NP4-12, and under 1/10th the price of big yellow tops!). Ironically flooded and Gel 7AH batteries were typically $50!
These days the 15AH may be the best $B4Bs.
(Be warned, the Yuasa NP7-12 has TWO spade connector size options - 4.7mm & the standard/common 6.35mm (1/4").)

The main thing to monitor is the battery's heat. If the battery is damaged or suffering, it will get hot - ESPECIALLY AGMs!!! - thermal runaway, fire, etc... Flooded cells should boil off before getting to that stage....


CONCLUSION (???)
So if you have a 200mA load, the 7AH should last 40 hours or more, but that's FULLY discharged (say 170 cycles = ~6 months?).
You could add a low-voltage cutout (a $20 MW728?), but that adds an additional drain (10mA??; plus its relay current).
And see the previous 2 blocks/paragraphs [ "But since a 7AH is so cheap, try it and see.... ].

I'd be tempted to get 2 NP7-12s or equivalent. Not to use in parallel (despite current sharing etc) but simply to swap the 2nd when the first dies.
(The storage life of a fully charged NP is pretty good though it decreases considerably with temperature, and storage time impacts permanent capacity - it's a highly variable relationship.)


FYI - the datasheet I used was "NP7-12 DataSheet 1.cdr" by MDS Battery Ltd, Enfield, yUK.
Alas I just searched "NP7-12 DataSheet.pdf" and got the NP7-12 January 08 version from yuasabatteries.com/pdfs/NP_7_12_DataSheet.pdf.
It is MUCH better quality - I can now read the graphs far more accurately! I wonder what errors will now find in the above!


Let me know if you find the linked pdf useful. It has lots of graphs, but with careful consideration, they encompass and convey so much!! It's almost a 2-page summary of several volumes (eg, batteryfaq.com).

well its normally going to drain for about 24 hours each time... i wonder if i should just buy one of those external li-ion packs made for cell phones. they charge via usb, and they supply via usb. they might take 5 hours to charge, but i can let that run on a time delay relay... the 4Ah it would drain into the li-ion batteries wouldnt make a difference on my sears platinum battery..

this seems like it would fit the bill. its about twice the cost of the lead acid setup i would have done, but this is all inclusive and will probably last for more charge cycles. most external packs have button power switches that you need to activate every time you go to charge, but it looks like this one is just always ready-mode. thats what i need anyway, i dont want to be messing with any of this on a normal basis, i just want to keep the phone on 24/7

http://cgi.ebay.com/5000mAh-Power-Ba...item27b9e2ed5d

with this i would just put it on a 4-hour time delay relay, so when i park the car the battery would charge for 4 more hours then stop taking from the car battery and just feed the phone. this should last a few days, which is all i need since i just have to start the car once for another 4 hour charge.

That LiIon has under half the capacity (of the 12V 7AH). Not that that has a big effect if using linear 5V converters, but if using SMPS or dc-dc converters - yes, half the capacity.


Not that I recall much about LiIons... (The last I looked at was LiPo's - especially after they stopped exploding. Not bad seeing model aeroplanes with 200A to 400A motors instead of 6cc petrol engines!)

[And FYI, forget memory effects. That too is myth (based on one misunderstood NiCad event) - it is usually internal "DC-bridging" that drops NiCad (and similar?) capacities.]

If current limiting is your concern, a ~1A current limiter should be cheap - a mere voltage regulator & resistor or a couple of transistors and resistors etc (under $5). The only concern would be the additional voltage drop (~0.6V?) but it could be bypassed after a while (a timer or extra sensing circuitry).


But I know many use 7AH & smaller paralleled with their main 12V battery and AFAIK they haven't had serious life-time issues. (Only the usual paralleling issues or dangers, but the UIBI overcomes that when not charging.....)


BTW - if using the UIBI, ensure that the (charge light) circuit can source enough current for the UIBI relay - eg, 12V @ 250mA. Older mechanical voltage regulators certainly can, and most alternator-internal regulators from last millennium. But it may be that some cannot - especially ECU interactives (though they are DP etc types and not D+ or (S/I)L types but can be used in a similar fashion).
If the alternator (regulator) cannot supply the required relay-energising current, it should be easy enough to add a MOSFET which adds no load (a few uA or nA if that!) and can drive any relay. In fact 80A MOSFETs can be bought for $2-$3 and they can be used INSTEAD of a relay....

i like relays to add hysterisis, but if i were to go off my alt light i would certainly add a transistor to drive the coil. i would want to take ZERO chances of screwing up my ecu!

5ah should be fine though. the phone lasts almost as long as i need it to on its 1200mah internal battery. i figure if i can get it to last 3 days thats longer then i usually let the truck sit...