What to trust?

Ralph Day · March 2010

Hi all

I've been using a "good"hydrometer for years for my SG readings. I've always found it hard to charge that last bit up to 1.265. Maybe I know why now. I purchased a refractometer today. After doing the zero-ing adjustments it reads the pilot cell acid at 1.270 while the hydrometer reads 1.260. Another cell also reads out at .01 higher than the hydrometer.

My question to any and all is: What instrument do I put my trust in?

If the refractometer is correct I've been overcharging for a considerable period of time (a year or so?). Shouldn't be any sulfation anyway:D

Any thoughts or suggestions?

Ralph

Vic · March 2010

Re: What to trust?

Hi Ralph,

Interesting. Does the Refractometer have a tolerance spec? And since you have been at this for years, am sure that you have compensated the hydrometer for any temp differences from 77 or 80 degrees F.

Which refractometer did you snag, and if you do not mind my crassness, roughly what is the going rate for such an instrument ? Thanks for the post, Vic

Ralph Day · March 2010

Re: What to trust?

Hi Vic
The unit and instructions don't mention a tolerance spec. Mind you it appears to be a direct Chinese to English translation:''If the consumers use the instrument in accordance with the mentioned of usage, it guarantee that instrument can't break down." Familiar sounding to reading person?

http://www.robinair.com/distributors/index.php

The link takes you to the US site, for Canadian distributors it showed CarQuest...(like NAPA only in Canada). The price in Canada was $109 plus 13% sales taxes.

The temp compensation appears to be inherent when you test your null (distilled water sample) and acid sample at the same temp. When using the hydrometer temp comp was seldom necessary as the batteries are in the basement of the house with a pretty constant temperature of about 23C. Only when doing a heavy eq (once in 6 years) did the acid temperature get up above 100 deg F.

Ralph

edited to provide link for California distributors but only goes to initial distributor's page...select location and search

BB. · March 2010

Re: What to trust?

Probably the answer is --- "it is all relative"... If you are charging and the SG has not risen in the last 30-60 minutes... You are done.

Cells should be balanced within 0.030 of each other (per Trojan). If more than 0.030 difference, attempt to equalize. After one hour and the SG not increasing, you are done.

If you have measured the SG when the batteries were first installed and charged--and use the same hydrometer over the years--then you are as accurate as you will ever be.

-Bill

Ralph Day · March 2010

Re: What to trust?

Oh Bill,
If I'd known then what I know now (or think I know).

Started off with just the T500 meter from Xantrex. After some months I Bought a cheap $5 hydrometer, then a few months later a ''good'' one $45 or so. Broke that and bought another.

From startup I didn't have any SG readings and didn't know what a forum was apart from the big one in Rome

. The battery set is Surrette Series 5000 8CS 25 PS 820ah 20 hr rate, 1156 100hr rate. Overall a leap of faith for a first battery set...but what did I know. Maybe I haven't treated them too badly (I've seen others treated a lot worse), and they'll last another 15 years or so?

The microFIT feed in tarrif plan in Ontario has altered the long term plans for the whole domestic system here. We have to have some grid service for the 10 kw install on-going, so decided to have a load service installed at the same time. And since a load service will be present (cheaper to charge batteries than with diesel) I figured why not net meter the domestic solar and wind. Well you have to have an existing load account before you can apply for a net meter service...blah blah blah. Bureaucracies can be fun, if you're on the outside watching, but when you're on the outside and want in again...they're a colossal pain. But as Marty Feldman said in Young Frankenstein: ''It could be worse, it could be raining.''

Ralph

Vic · March 2010

Re: What to trust?

Hi Ralph,

Thanks for the info. I would want to believe that the refractometer would have superior accuracy, especially when calibrated with DI water.

Bill is of course correct, altho, on my C bank of batts (12 V S460s), one tardy cell was still below 1.265 comped, and had not risen in in the previous hour, letting that bank float for a few days, and then EQing that battery by itself again, the SGs came up a bit more in an additional hour of EQ. Good thing that FLAs are fairly tolerant.

To me, given that the replacement cost of the large bank(s) here have doubled in the past few years, (now about 15.5 K$) it seems like it is difficult to become TOO OBSESSIVE about one's bank (literally a BANK !).

Thanks again for the test. If you had a scale, and graduate, you could probably take a sample of electrolyte and ratio it against the same volume of distilled water as an additional check.

Thanks again, Vic

Ralph Day · March 2010

Re: What to trust?

FLA owner's spouses have to be tolerant as well. Mine isn't as fixated on daily sg pilot readings as me...can I make toast? gets asked on a semi regular basis.

I have a bathroom scale and graduated from high school chemistry but I guess that doesn't count. A buddy is a high school science teacher so I should be able to get hardware from him, distilled water, known sg acid sample etc.

Ralph

Now must enjoy the sun while it lasts!

BB. · March 2010

Re: What to trust?

Oh--I have a few batteries in the early death grave yard too--And I have a grid tied system. :roll:

Probably the #1 cause of early traumatic battery death is inattention (not enough charging)... Followed up by #2, obsessive attentiveness (over charging/equalization)...

-Bill "

: B.

PS: Above is an attempt at humor... May not translate well. And my humor may not be funny anyway...

Ralph Day · March 2010

Re: What to trust?

Perhaps early chronic battery death. Traumatic battery death sounds like the wrench dropped across the terminals scenario...followed by death or dismemberment of clumsy oaf who dropped the wrench.

Ralph

BB. · March 2010

Re: What to trust?

Ralph Day wrote: »

Perhaps early chronic battery death. Traumatic battery death sounds like the wrench dropped across the terminals scenario...followed by death or dismemberment of clumsy oaf who dropped the wrench.

Ralph

Here we go--I have to explain "traumatic" to the owner of the battery (as in expensively traumatic)... The trauma to family finances for a replacement battery bank, etc...

Humor explained, is humor that is not funny.

-Bill "

" B.

Ralph Day · March 2010

Re: What to trust?

Shorthand for you Bill:

$Traumatic battery death

the $ shifts the mental template from chronological to financial. Either way one is in deep scatological territory.

Ralph

bryanl · March 2010

Re: What to trust?

The basic question here is that of precision versus accuracy. It is very common in battery related discussions to see the assumption that the precision of a measure (such as specific gravity) is also the accuracy of a measure. That is very seldom the case.

When it comes to lead acid batteries, one should never expect an accuracy better than 20% of the measure taken. It is indeed possible to narrow this down a bit by controlling conditions and adjusting for various parameters such as temperature but the fact remains that there are always measurement errors, instrument errors, and battery variance.

These issues are particularly of concern with smaller battery banks as a statistical matter. It is nice that we have instruments with four or five digit precision but, without a proper interpretation of the measure and consideration of the inherent accuracies involved, it is easy to be misled about the results we achieve.

Vic · March 2010

Re: What to trust?

Hi bryanl,

I must not understand what you are saying by the 20 % reference.

On my batteries, 1.265 is the SG for full charge (yes this is nominal). 0.8 X 1.265 is 1.01 nominal -- essentially PLAIN WATER.

Surely you are NOT saying that expecting better than 20 % tolerance/repeatability for SG readings is a fallacy !

There can many problems with the mechanics of taking SG readings using a float type hydrometer, but the the tolerance is probably better than 1 % of the reading, if the readings are carefully made and the person taking the readings is aware of the pitfalls that can lead to errors. Anyting else would render this acitvity useless.

JMHO, I must not understand your statement. Thanks Vic

Cariboocoot · March 2010

Re: What to trust?

As with any measuring system, it's all arbitrary and relative.

Your panels don't put out their rated Watts, except when they do or go over.
Your charge controller's readings are optimistic or pessimistic depending on which one you've got.
Your batteries don't really have their rated Amp/hrs and may last years longer than the warranty or fail tomorrow.
Your hydrometer reads Specific Gravity +/- some random percentage depending on its manufacture, age, condition, temperature, atmospheric pressure, your eyesight, and whether or not there's a full moon.
Your loads are either above or below the ratings on the spec plate attached.

For something that's supposed to be an absolute precise science (physics), once you get it out into the real world solar electric seems more like magic, doesn't it?

The trick is to learn which direction to err on what numbers in order to be safe and have a reliable system.

BB. · March 2010

Re: What to trust?

I think he is saying that absolute accuracy in/with every measurement will still only give, at best, an estimate in of battery AH state within 20% of the true battery AH state...

Batteries are just not that precise of device. That is the reason way it is recommended to not take a bank below 20% state of charge--The variation in cell capacity is such that some cells may be at 20% state of charge and others are near 0% state of charge--Any further discharge will "reverse charge" the low cell(s) and ruin the cell/battery unit.

-Bill

Vic · March 2010

Re: What to trust?

OK, well I do not have to be correct on this.

Surrette says one should not expect hydrometers to be better than .005 points of actual SG, Expecting that a cell with 1.265 SG to really read within the range of 1.260 and 1.270. This might be ignoring the accuracy (tolerance) of the hydrometer, or be ignoring reading errors. But to me implies that one can read actual SG within .005 points if one tries.

Was waiting for bryanl to clearify exactly what he was saying ...

"When it comes to lead acid batteries, one should never expect an accuracy better than 20% of the measure taken. It is indeed possible to narrow this down a bit by controlling conditions and adjusting for various parameters such as temperature ..."

I read this as regarding a "reading" ie Voltage or SG etc, not inferring SOC from readings.

Off to the remote (internetless) location. Not to nit-pick ... Thanks Vic

bryanl · March 2010

Re: What to trust?

Cariboot and BB did a pretty good job explaining my caveat I think.

Another confusion that is showing here is that of mixing absolute and relative. Consider specific gravity. The measure range is from about 1.1 to 1.3 or over a range of 0.2. Twenty percent of that is 0.04. A change in specific gravity of that magnitude covers about a quarter of a battery's capacity.

Then there is the problem of what you are trying to measure. SG is usually used as a measure of state of charge. From a practical point 4 significant digits in an SG measure does not mean 4 significant digits in measuring state of charge - and you don't have that much precision in an SG measure, anyway.

If you are taking comparative measures, such as suggested for equalization, you can get a better precision because factors such as temperature, age, and so on can usually be considered common to all cells being measured. But even then, you need to keep in mind measurement errors.

The point is to not get hung up in numbers or with numbers, even with numbers about numbers.

Vic · March 2010

Re: What to trust?

OK bryanl,

Well, I have no need to determine SOC at all accurately. My quest recently, has been to determine if my upward creeping SG measurments are real, and am overcarging my a bank a bit, or if the hydrometer's error has increased. Can usually fully charge the bank from PV, altho this Winter has been unusually cloudy, so have needed to run the genset more often, and the bank has seen deeper discharges.

Some acknowledge that parameters such as Asorb time, and even Vasorb need to be tweeked depending on changes in DOD.

Recently have returned to terminating Asorb based based on charge current, as this is a fairly well-known value in systems with light loads during Asorb, and so on.

Vic

Ralph Day · March 2010

Re: What to trust?

Vic
I tried the end amps charge cutoff (to float) once with my MX60. The problem comes when the wind turbine is also inputing amps the MX doesn't ''know about". It will cause the MX to cease the absorb and drop to float because it doesn't have to put solar amps into the battery bank, then the wind gust dies down and you're left with float charge when your really needing more absorb time.

Ralph

Vic · March 2010

Re: What to trust?

Hi Ralph,

Well I do NOT have the luxury of input in addition to PV, altho, this end amps terminating Asorb is also not so good with larger loads. I may change back to Asorb Timer. I still need to do weight check on a sample of electrolyte, just for kicks.

The refarctometer seems cool. Thanks Vic

Ralph Day · March 2010

Re: What to trust?

I think I'll be comfortable trusting the refractometer. Monitoring the charging amps I found that at 1% of the 20hr rate (820 20hr rate=8.2amps) gave an SG reading (by refractometer) of between 1.265 and 1.270. That's just where it should be according to more experienced minds lurking here.

Results in reduced absorb time...less water usage (checked monthly, so not apparent yet), time will tell. Headed for 4-5 days of rain, so it will be interesting to see how quickly the SOC returns to full next week.

Ralph

bryanl · March 2010

Re: What to trust?

re: "I have no need to determine SOC at all accurately"

that's good, because you can't

the point I am trying to make is that there are limits of what can actually be measured and that needs to be considered in engineering systems effectively.

mike95490 · March 2010

Re: What to trust?

I'm not sure I live in an area where i could use the refractometer:
Range 32°F to -60°F
I was able to find the robinair version on Amazon for ~ $45, which is good compareing to a fragile, hard to read, rubber rotted off, hydrometer. i'll have to invest in a lot of rags, can't use water to wash it off, just a clean rag.

Dave Angelini · March 2010

Re: What to trust?

mike90045 wrote: »

I'm not sure I live in an area where i could use the refractometer:
Range 32°F to -60°F
I was able to find the robinair version on Amazon for ~ $45, which is good compareing to a fragile, hard to read, rubber rotted off, hydrometer. i'll have to invest in a lot of rags, can't use water to wash it off, just a clean rag.

I think Mike makes a good point here! If your using old auto type hydrometers and getting non repeatable data or expecting too much from the data, you are setting yourself up. And back to Amazon, the EZ read hydrometer is 10 bucks and is very repeatable (3 measurements and average) and will last a decent amount of time before it is gone.

Perform a cal at the solstices and tweak your absorb times after the end of the float day, if you really want to minimize water use and be full! I have tricks for new customers that keep them from going astray or wasting their valuable time with this but that is another story.

10 to 15 years of daily cycling is what I expect from a battery.

BB. · March 2010

Re: What to trust?

Dave,

What brand/type of battery bank (deep cycle, forklift, flooded or agm, etc.) and depth of cycling to you like to design for to get long bank life?

-Bill

mike95490 · March 2010

Re: What to trust?

Dave Sparks wrote: »

......
Perform a cal at the solstices and tweak your absorb times after the end of the float day, if you really want to minimize water use and be full! I have tricks for new customers that keep them from going astray or wasting their valuable time with this but that is another story.......

Tell me a story - it sounds worthwhile

Ralph Day · March 2010

Re: What to trust?

Here, here...we'd like to hear here.

Ralph

Dave Angelini · March 2010

Re: What to trust?

BB. wrote: »

Dave,

What brand/type of battery bank (deep cycle, forklift, flooded or agm, etc.) and depth of cycling to you like to design for to get long bank life?

-Bill

I told you once and it is somewhere in this forum! Basically you design for cloudy days in December. For solar you have to be in the southwest or similar and probably not coastal or fog prone. You only discharge 10 to 20 percent from full so that you charge 350 days a year. You have to have a big enough tracked array to do this or, you have multiple arrays that can charge you in the pm when it is bad in the am and of course the reverse. The only battery I have used is the Surrette L16's for over 25 years now for myself and the 35 offgrid homes that I have done.

I really love being remote which certainly is not for everyone!

What to trust?

Comments

Categories