Which Battery Monitor ?

[4wsilver]

Originally Posted by Mark Johnson

[FONT=verdana,geneva,lucida,'lucida grande',arial,helvetica,sans-serif]---[/FONT] put a 20a load on it until it reaches 10.5v, measured with a high quality digital multimeter. Next, do the math. A 100a/h battery should theoretically deliver 20a for 5 hours---

Mark

Mark, the amp hour is a 20 hour rating, not a 20 amp rating. To do the test correctly you should take amp hours like 360 and divide by 20 to get the amps for the test then do the math. In this case it would be 18 amps and in your case with 100 a/h it would be 5 amps. Doing it your way would result in a much lower value and could indicate a bad battery when the battery is in reality good. This also notes that for best battery life, you should size your amp hour rating not only on the average load and amp hour use in the period between charges, but on the max load X20 also. If this results in a larger amp-hour rating then you would reduce the life of the batteries.

Also earlier you stated that voltage was not good for measuring state of charge, neither is amp hour use unless it is related to rate of use. Both are useful and they do make monitors that measure the electrolyte conductivity which is better than specific gravity, just that I cannot afford them, can you? Properly used, both AH and Voltage work, but I recommend both measurements be made and compared.

[Cheechako]

All the hocus Pocus aside...The bottom line is if your battery is reading 10.5 volts at rest it IS dead!

[Extemporaneous]

Originally Posted by 4wsilver

Mark, the amp hour is a 20 hour rating, not a 20 amp rating. To do the test correctly you should take amp hours like 360 and divide by 20 to get the amps for the test then do the math. In this case it would be 18 amps and in your case with 100 a/h it would be 5 amps. Doing it your way would result in a much lower value and could indicate a bad battery when the battery is in reality good.

Also worth noting is that (as I understand it) in Europe they sometimes market their Batteries using its 10 Hour rated value!!
That would mess with the math also.

Originally Posted by Cheechako

All the hocus Pocus aside...The bottom line is if your battery is reading 10.5 volts at rest it IS dead!

Not sure what your comment means?? Is it some symbolic end to the thread.... or is it just you bowing out?


I can't see it being even remotely possible to use this information or technique manually to monitor ones batteries. That said, because it is a standard of sorts and inherent to in the design of battery monitors, I thinks it's very important to know the real capacity (based on this standard) so you can setup you battery monitor properly.

Cheers,
Extemp.

[Mark Johnson]

Originally Posted by 4wsilver

Mark, the amp hour is a 20 hour rating, not a 20 amp rating. To do the test correctly you should take amp hours like 360 and divide by 20 to get the amps for the test then do the math. In this case it would be 18 amps and in your case with 100 a/h it would be 5 amps. Doing it your way would result in a much lower value and could indicate a bad battery when the battery is in reality good. This also notes that for best battery life, you should size your amp hour rating not only on the average load and amp hour use in the period between charges, but on the max load X20 also. If this results in a larger amp-hour rating then you would reduce the life of the batteries.

Also earlier you stated that voltage was not good for measuring state of charge, neither is amp hour use unless it is related to rate of use. Both are useful and they do make monitors that measure the electrolyte conductivity which is better than specific gravity, just that I cannot afford them, can you? Properly used, both AH and Voltage work, but I recommend both measurements be made and compared.

Perhaps I gave a bad hypothetical example. In my case, when I have tested my new Trojan batteries, I put a 10a load on my 340 a/h new 6v batteries, (2 in series), which would theoretically run the load for 34 hours before reaching 10.5a. It was a bit less than this, so I contacted Trojan about my testing procedure, (which they agreed with BTW), and they said that very few batteries achieve their rating, so mine being 90 something%, was pretty good.

All of this stuff on paper is different from real world experiences. It is not etched in stone.

"A/h out" of a FULL battery, as measured by my Link 10, is a very very accurate means of knowing my batteries state of charge. It is as accurate at a 1a load, as well as my max. load ever of about 10a... MY 340 a/h house batteries are usually being discharged at a rate of no more than a few a. at the time, and seldom accumulate more than 35 or 40 a/h down overnight. Then during the day, the solar kicks in, and it is "a/h IN" until about noon, when the batteries have been brought back to 100% with a three stage smart charger. As the sun goes down again, the "Link 10" has re-zeroed itself, and once again counts the a/h out, until the next morning. It is accurate to the tenth of an amp! With my battery capacity of 340a/h and only using about 35a daily (=10%) followed by complete recharge EVERY day, I have a several day "cushion" for cloudy days, can run on 100% solar, and get over ten years out of my batteries.

I have no doubt that if you put a 35a+ load on the batteries, and discharge them 50% every day, only to then recharge to 80% full... it gets more complicated. (It never re-zeros the monitor!) It also becomes a moot point! This is damaging to batteries, causes a short lifespan, and is a setup to be left without power someday. I don't advise my clients or anyone else to use their batteries that way!

If the system is set up and used as I have suggested, it works great!

Specific gravity has been used as the standard of the industry for decades, to know the state of charge of a wet cell battery. (IF someone wanted to be that anal about it). It test each cell individually, and is quite accurate, in that once the battery is charged, you are only looking for one cell that is extremely "different" from the others to know that you have a defective battery. In the real world, no one uses them, it is too much hassle! (The Hydrometer is about $5, and going even MORE anal (= expensive) in ones testing, represents the lunatic fringe).

IF one has a large enough house bank, an energy efficient boat, relatively shallow discharges, and brings them back to 100% daily... Then a battery monitor like the "Link 10", (ONCE IT'S COMPUTER IS CALIBRATED TO YOUR BATTERIES), is accurate, reliable, and takes the guesswork out of monitoring your batteries state of charge. It will keep you out of trouble from flat batteries, and assure their long lifespan. If you can't meet the above criteria in your energy system, the "monitors" become less than perfect, but are still the best "practical" solution out there.

For folks like the one above, who disagree, and want to rely on their own calculations instead... "knock yourself out". And by all means, good luck with that.

Sincerely... MarK

[Extemporaneous]

Originally Posted by Mark Johnson

Perhaps I gave a bad hypothetical example. In my case, when I have tested my new Trojan batteries, I put a 10a load on my 340 a/h new 6v batteries, (2 in series), which would theoretically run the load for 34 hours before reaching 10.5a. It was a bit less than this, so I contacted Trojan about my testing procedure, (which they agreed with BTW), and they said that very few batteries achieve their rating, so mine being 90 something%, was pretty good.

Hi Mark,

I'm sorry, but you're “theoretically” is not right.
A 12 Volt bank of batteries (no matter how it became 12 V , i.e. 6V series/parallel ect.) which is using the 20 Amp/hr standard and stating 340 Amp/hrs capacity should be tested using [340/20 = 17 Amps.] Your batteries should have theoretically been tested with a load of 17 Amps and with the load on the batteries that time at which the battery voltage read 10.5 Volts should have been at the 20 hour mark. That's the theory. So if we talk about your batteries and consider Peukerts law, (lead acid batteries rated using the 20 Amp/hr discharge time) if you load your batteries with more then the 17 Amps, say 25 Amps, one might think you would get 340 / 25 or 13.6 hours before with that load on your batteries would reach 10.5 Volts. That is NOT the case! Because you are drawing power out of your batteries more quickly, due to Peukerts law you will get LESS run time then 13.6 hours. The good news is that the opposite is also true! If you draw power off more slowly then what the 20 Amp/hr rate suggests (17 Amps in your case) you get more hours then the straight math would suggest. So let's say one was to use a 10 Amp load, they may think that they should run for 34 hours (340 / 10), at which time their batteries would reach 10.5 Volts. NOT TRUE! One should get more (considerably, I think) hours before reaching the 10.5 Volts.
I'm thinking that Trojan agreed with your testing methods because they knew that it would either make their batteries look very good or at least not as bad.
And it's not really anything to do with one's own calculations, it's Peurkerts Law.



Cheers,
Extemp.

[Cheechako]

"It is accurate to the tenth of an amp!" hmmmm.....just curious....how would one know this?
When I was cruising my real world went like this: (batts almost never 100% charged)
* yesterday the wind was minimal but the wind gen managed to get some amps in. Rather than run the engine again we decided to just wait until morning.
* Got up a little later than expected, supposed to take the trip into town (walk then bus if we're lucky) to get the much needed dingy motor part. Managed to get 45 mins of charging in from the engine alternator before our friends arrived in the dingy.
* Hopefully the wind will blow today, but this anchorage is pretty calm.
* Ended up being a day long trip in town. No wind to speak of all day. Running the engine a good 1.5-2 hours to try to keep up with the fridge.
* The dingy part is supposed to be in today, so back into town for another day long trip. Ran the engine this am for one hour prior to leaving again.
* Had dinner in town with friends, back to the boat after dark.
* batteries have not been 100% for days now. But we are cruising!
* They will be 100% when we start moving again.
Those of you that have every possible system are lucky I guess (solar, wind, engine) But my batteries were seldom 100%

[goboatingnow]

% Discharge, yeah - though it's not hard to compute that from Ah consumed if you know the Ah capacity of your bank. The "Time" button tells you how long you can sustain current discharge - the filtering on what this means depends on proper and various setup values - like the Ah capacity of your bank.

Actually % Discharge is the hardest thing for a monitor to do Remember % discharge is in effect State of Charge , Ultimately SoC is what we want to know. Battery monitors have a problem as they cant determine SoC, so they "model " the battery.

They need to know (a) Capacity , (b) Charge Acceptance and (c) Discharge curve. Discharge curves are approximated by using Peukert Exponent, which is merely a "model" of a lead acid discharge curve. Its not a particulary good one either. Capacity is either user entered or determined by a bit of Amop counting. Equally PeuKerts can be computed by profiling discharge curves (this is what I beleive SmartGauge does). or can be entered manually. Charge Acceptance can be computed by Amp counting and full charge parameter setting or entered manually

Hence using this all a battery monitor does is "estimate" whats going on, in reality it could be different, which is why some models requires regular resets.

As to Ah or Wh it matter not really , Ah has been used traditionally to measure battery capacity so Ah is what we use. ( as in we compared used Ah to battery Ah). Energy usage in Wh really isnt a whole lot of use unless you are studying energy usage patterns or whatever. ( or want to have a comparison between different devices on different voltages).

I havent met a battery monitor yet that lives up to its name. ( and some are very misleading). The Victron comes close

Dave

[Extemporaneous]

Thought I'd put this out there as reference. I think it's from Rolls which we are not specifically talking about, but it makes the point.

Cheers,
Extemp.

[goboatingnow]

"A/h out" of a FULL battery, as measured by my Link 10, is a very very accurate means of knowing my batteries state of charge.

Mark, keep the day job, firstly its Ah, secondly blindly counting Ah's is the worst way to determine state of charge. ( as to you comment re tenths of an amp, is that becuase you have 1 place after the decimal point on your meter. really the meter could be at least out by 10% or more often way more).


Examine your statement again, Firstly the Monitor has to determine by various modeling constants , when it determines the battery is full, That can be a very innaccurate process, especially as batteries age.

Secondly Peukerts exponent has to be determined, either its manually set ( usually 1.25) or in more sophisicated monitors the exponent is recomputed based on the discharge history. Gain this is a source of error as Peukerts Law is only an approximation of lead acid discharge capacity.

Yes calulating Ah is easy, relating that elctronically to the actual available Ah and hence determining State of Charge is a very error prone process and why battery monitors get it wrong.

MY 340 a/h house batteries are usually being discharged at a rate of no more than a few a. at the time, and seldom accumulate more than 35 or 40 a/h down overnight. Then during the day, the solar kicks in, and it is "a/h IN" until about noon, when the batteries have been brought back to 100% with a three stage smart charger. As the sun goes down again, the "Link 10" has re-zeroed itself, and once again counts the a/h out, until the next morning. It is accurate to the tenth of an amp! With my battery capacity of 340a/h and only using about 35a daily (=10%) followed by complete recharge EVERY day, I have a several day "cushion" for cloudy days, can run on 100% solar, and get over ten years out of my batteries.

PS you dont need a battery monitor , all you are doing is taking 10% out and replacing it. The monitor could be lieing through its teeth and youd see everything as all right. Its only when we strees the system that we need a monitor in the first place.

LIFEPo4 is the way to go , really. Peukerts =1
Dave

[Extemporaneous]

Originally Posted by goboatingnow

As to Ah or Wh it matter not really , Ah has been used traditionally to measure battery capacity so Ah is what we use. ( as in we compared used Ah to battery Ah). Energy usage in Wh really isnt a whole lot of use unless you are studying energy usage patterns or whatever. ( or want to have a comparison between different devices on different voltages).

Can't argue with that. A couple of point though.
Wh is the "true" measure of power. Also, if your watching different equipment (and yes comparing) you should always see the same watts while the Amps could be different for the same equipment depending on other loads on your battery bank and Peukerts law and such.

I know I'm not telling you anything you don't know, but Watts does leave Peukerts law out of the equation which has value at times.

Cheers,
Extemp.

[goboatingnow]

Originally Posted by Extemporaneous

Can't argue with that. A couple of point though.
Wh is the "true" measure of power. Also, if your watching different equipment (and yes comparing) you should always see the same watts while the Amps could be different for the same equipment depending on other loads on your battery bank and Peukerts law and such.

I know I'm not telling you anything you don't know, but Watts does leave Peukerts law out of the equatio...

Cheers,
Extemp.

Hold your horses there, Wh are a measure of energy used. Ah is a measure of electric charge. Watts is a measure of power.

At a fixed voltage watt hours and amp hours are measuring the same thing

Batterie capacity can be interchangeably specified in Wh or Ah ( Sony tend to use Wh)

Peukerts law applies to the fall in capacity against increase in discharge current. It's has nothing to do with Wh or Ah. A lead acid battery specified in Wh will have it's Wh rating derated in exactly the same way

Dave

[Mark Johnson]

Originally Posted by Extemporaneous

Thought I'd put this out there as reference. I think it's from Rolls which we are not specifically talking about, but it makes the point.

Cheers,
Extemp.

I am well aware of all of these points. I was speaking in generalities. I have also been corrected for my improper spelling of terms, or failure to capitalize, as well... (petty non important points in my book). Sometimes I use what ever size load I have handy! My decades of past experience comes into play and I make the best judgement that I can. I certainly know that you get more a/h out of a battery with a long low a. test vs a short high a. test, yet all auto mechanics do a really quick test because it is practical. NOT because it is accurate. (If I wanted to win an argument, or be MOST ACCURATE, I would do the test according to the "standard"). In my case I just wanted to know if the two years they spent on the shelf in a warehouse, getting a charge only once a month, had hurt my batteries or not.

My battery actually went through several test at different a. ratings... When I said "theoretically", I meant "not necessarily @ the accepted 20 hr rate". I suppose "approximately" would have been more acceptable, but maybe not? You guys are picky.

After a week of charges, discharges, and 5 "equalizes", it's rating went up considerably, just as Trojan suggested it might. THEY didn't nit pick about the time interval, because in batteries, we are only talking about a range, and batteries producing "within tolerance". The test would also have to be done at a given air temperature, if I really wanted to be anal about trivialities. Now, if I had expected them to produce their 340 a/h as (113a. for 3 hours), THAT would have been way different results. The test I did was close enough, just like hand grenades & horseshoes!

If one says an engine produces 300 hp, that doesn't really mean that... It means under a very specific set of test bench circumstances, one of these engines did this, once. In the real world, it varies with fuel, altitude, and from engine to engine.

I have only been trying to offer help, not get into a battle over semantics. It appears that some would rather find fault with the far less relevant issues, (which, like my punctuation & spelling, I obviously give less thought to than the "point"), and they ignore the obviousness of the "point", which is totally true, and useful to those who want to use the information.

The "point" was in answer to previous post that first indicated that all you needed was to take a v. reading, and the second was that all you needed was to combine the v. reading with other calculations... so, these monitors aren't needed, and are inaccurate.

I said what I did in response to these statements: They're nonsense! The programmable "smart" monitors like the Link 10, as imperfect as they are, is far more accurate than making your own calculations based on voltage, amps, or the phase of the moon.

These devices are "spot on" if you calibrate them to "know your batteries", and bring your batteries up to 100% daily. (This starts the calculation over from "0", so there is NO running error). They will be less and less accurate in proportion to how long you go without doing this. If you bring your batteries up to 100% just once a week, the 7 days of accumulated error will still be extremely small.

Only if you "seldom if ever" top off your batteries, will the monitor have a substantial error. When I was suggesting the advantages of these devices, I was not talking to that group of people. If you fail to bring your batteries up to 100% for a long enough period of time, while making heavy deep discharges as well, the Link 10 never re-zeros itself (which starts the calculation all over again from the top). In this case the very small daily errors can add up to a sizable error over say... months of this. In this case you might as well just look at the voltage and make a guess, as the Link 10 is guessing too. I admit that.

I never pushed this negative point about them, as the folks who NEVER bring their batteries up to 100% (generally the last 20% done with wind or solar), are not the people that I was addressing, in suggesting battery monitors to.

For many folks who want to get their batteries to really last, their equipment to run better on the higher voltage, and have maximum reliability of the system, for these folks and these only... If you start with a large reliable house bank that can power you at least a couple of days, equip your boat as energy efficiently as possible, cycle your batteries fairly shallow, and top them off every day or AT LEAST once a week, (the top 20% of the charge is best done with alternate energy) THEN, one of the smart monitors like the "link 10" will reliably tell you your batteries state of charge, with extreme accuracy!

For folks in the other categories, I apologize if you thought that my advice applied to you. It certainly does not.

May your systems work well for you and reliably too.

Fair winds... M.

[Extemporaneous]

Originally Posted by goboatingnow

Hold your horses there, Wh are a measure of energy used. Ah is a measure of electric charge. Watts is a measure of power.
Dave

Okay, so you're saying they're different?

Originally Posted by goboatingnow

At a fixed voltage watt hours and amp hours are measuring the same thing
Dave

Okay, so you're saying they're the same?
Actually "at a fixed voltage" they equate but are still not the same.
Also, that is kind of the point, as your batteries loose capacity, it is not "fixed voltage".

Originally Posted by goboatingnow

Peukerts law applies to the fall in capacity against increase in discharge current. It's has nothing to do with Wh or Ah. A lead acid battery specified in Wh will have it's Wh rating derated in exactly the same way
Dave

I was thinking Ohm's Law more than Peukert's Law. Sorry.
And it makes sense you're right regarding a battery being de-rated but my point was more of monitoring equipment and which unit of measure is more meaningful.

Keep the Education Coming.
I'm continually getting sent back the the books.

Cheers,
Extemp.

[Maine Sail]

Originally Posted by Mark Johnson

Perhaps I gave a bad hypothetical example. In my case, when I have tested my new Trojan batteries, I put a 10a load on my 340 a/h new 6v batteries, (2 in series), which would theoretically run the load for 34 hours before reaching 10.5a. It was a bit less than this, so I contacted Trojan about my testing procedure, (which they agreed with BTW), and they said that very few batteries achieve their rating, so mine being 90 something%, was pretty good.

For folks like the one above, who disagree, and want to rely on their own calculations instead... "knock yourself out". And by all means, good luck with that.

Sincerely... MarK

Mark,


20 Hour Rate = The number of amps a battery can supply for 20 hours before dropping to 10.5 volts at 80F.


To test your batteries Ah rating against actual performance you divide the 20 hour amp hour rating by 20 to determine the 20 hour draw/support the battery can provide at 80F. This is also known as the "C" rate or C20 (Capacity 20 hours). A battery rated at a 10 hour rating, as many are in Europe, would have a C10 rate..

In Practice:

Battery "A" has an amp hour rating of 150, dividing by 20 = 7.5. This battery should support a 7.5 amp load for 20 hours before dropping to 10.5 volts/dead. A battery with a 60 Ah rating will carry a 3 amp load for 20 hours before dropping to 10.5 volts.

RC or reserve capacity is almost always measured using a 25 amp load until the battery reaches 10.5 volts then compared to its RC rating minutes.

I have no idea why the Trojan tech support person thought your capacity test was relevant except to say that I have talked with a couple of them over the years who are somewhat clueless. Trojan batteries are generally rated on a 5 hour rate and a 20 hour rate so for better accuracy in the test a 5 hour rated load or 20 hour load should be applied.

For a C20 test on your 340 Ah battery you should have had a 17 amp load @ 80F until the battery dropped to 10.5V or compensated for temp.

Based on accepted C20 testing procedures, & Peukerts Law, your battery tested at 10A should have far exceeded the 20 hours before falling to 10.5 volts. 90% of 34 hours might be about right but I would re-test using the 20 hour capacity test at a 17 amp draw not 10A and see what you get.

Here's a couple of quick references.
Amp Hours: Battery Amp Hour Rating ASP

Nigel Calder - Boat Owners Mechanical & Electrical Manual
"The battery is first brought to a full charge. It is then discharged at a rate of one twentieth of its rated Ah capacity (eg: 10 amps on a 200 amp-hour battery). The test continues until battery voltage has fallen to 1.75 volts per cell (10.5 volts on a 12 volt battery)."

[Maine Sail]

Originally Posted by Mark Johnson

"A/h out" of a FULL battery, as measured by my Link 10, is a very very accurate means of knowing my batteries state of charge. It is as accurate at a 1a load, as well as my max. load ever of about 10a... MY 340 a/h house batteries are usually being discharged at a rate of no more than a few a. at the time, and seldom accumulate more than 35 or 40 a/h down overnight.

Mark,

Hopefully I can explain this well. Ah consumed as referenced above is not an accurate representation of your SOC. It is only representative of Ah's used and only give a rough approximation of SOC unless your amp loads were drawn out at the 20 hour load rating of 17A steadily, and then it could be considered accurate..

If you are discharging your batts at a rate of "no more than a few amps" you will have MORE than 340 Ah's of capacity in your battery. Thus measuring Ah's consumed against a 340 Ah battery will be slightly inaccurate to SOC.

Your batts are rated at 17 amps for 20 hours to produce 340 Ah's. Peukerts Law dictates that the lower the draw, in this case any draw less than 17A, the more Ah's you will get out of the bank. With a low enough draw you may even have a 380 Ah bank or more. Anything over 17A and you will get less Ah's out of the battery. This can easily be seen in the 5 hour Ah rating where they draw 68 amps for 5 hours and only get a 303 Ah rating on an L16 Trojan. As others have suggested the most accurate screen, with most battery monitors, will be your % Charged screen, not Ah's consumed.

This stuff is not easy to explain!!