More Surrette problems

Re: More Surrette problems

Hi Terry,

terrynew wrote: »

Hi, everyone. Here are the results of doubling my absorb time to bring my new batteries up to specific gravities near 1265. In the last 14 days, I've had 8 days with 6 hrs in absorb at 59.2 V, and four other days with over 5 hrs in absorb. So I was expecting some good news when I took out my refractometer last night and sampled four cells that were the lowest ones after my last EQ, July 11. But no, all 4 pilots had gone down (to 1235, 1235, 1240, 1240). As well, my 'good' pilot, which was 1250 after last month's EQ, was also down (1245).

Remember that equalization brings low cells back up to high cells (in theory). So, you really need to measure all cells and look for a the SG spread. I have seen 0.015 to 0.030 SG differences listed as "significant" and suggesting an equalization to restore "balance' (the range is from different battery mfg--I would say that if your batteries are outside that range, then you have "significant" state of charge differences.

Should I just keep on truckin', or do an EQ, or is this lack of progress indicative of faulty charging hardware? Over those same 14 days, the FM80 logged a max current over 40 A on only 5 of the days (but was over 35 A on all but 2 of the 14 days).

Can you put your battery bank AH and voltage in your signature--I assume 48 volt system?

On the side, I've discovered thru several calls to Magnum Tech Support and Outback Tech Support that the Total Ah Out meter on my inverter's BMK does *not* represent my total load, as I had assumed; this is because a single shunt is used for both the inverter and the cc in order for the BMK to calculate its SOC. And so my load may be over twice what I thought it was, 125 Ah/d instead of 58 Ah/d. I wish there was a reliable way to measure the actual load.

There are different points in the system where you can measure current flow/Amp*Hours/etc... They are all valid from their point of views (charge controller, inverter, DC loads, battery bank, etc.).

Personally, I push for the battery monitor (or Victron) to MONITOR THE BATTERY. That is, for most people, the most important indicator. Keep the battery happy, hopefully you will have a long battery life. If the battery is "not happy", it will commit suicide (or perhaps be murdered).

Normally the shunt to measure battery current goes from the battery bank negative common point, to the battery shunt, to the negative/ground/common bus where all return/negative wires should be attached (if you connect a return to between battery negative and shunt, that current flow "will be missed" by the battery shunt).

In any case, that doesn't explain why the extra absorbing failed to bring up the gravities.

How many parallel strings do you have (if any)? (I see 2 strings of 8 batteries below--I think your sig is not correct now with 8x batteries 400 AH @ 48 volt battery bank?)

...When the cc said 27.7 A, the ammeter read 23.9 (pos. cable) and 23.4 (neg. cable). Should I do something about this 15% drop? Is the diff between the pos and neg cables significant?

If you only have two wires (+ and -) going to the battery bank, and no others--My guesses:

• It is difficult to have accuracies better than around 5-10% per measurement. Especially if the measurements are made at different times (measure wire A, then measure wire . 15% error in measurement is getting close to being "significant"--But if you have no other wires to the battery bank, then it has to be measurement errors.
• You can place both the + and - leads in the current clamp, and then measure the "sum of the current" which should be zero.
• Also, make sure you have the ZERO Button setup correctly. As near as I can understand--A) with no wires in clamp, push zero button until "ZERO" is displayed on the LCD meter. Then connect the clamp and make measurement. If "ZERO" is not on, then you have the unzeroed bias of the magnetic sensor. To rezero meter--move wires from clamp, push zero button to turn off "ZERO", then push button to turn on "ZERO" again. Then make measurements (depending on your needed accuracy, you may need to rezero every couple minutes, or even more often).

More importantly, Bill also asked me to confirm that my two strings of 8 batteries each were getting equal current. They aren't! When the cc said 24.4 A, the cable between the first two batteries in each string measured 11.3 A and 6.7 A. Similarly, a few minutes later, the cable between the last two batteries in each string measured 10.1 A and 5.0 A. Whoa, what could be halving the current in one string?

You cannot half the current in one string (10 amps in, 5 amps out)... That has to be variation over time or measurement error. (I played with my car's electrical system because it was "easy" to setup steady state test conditions--I.e., headlights on with motor off. Start car, measure battery current with alternator running, lights on/off, etc. It is possible that the DC Clamp is picking up stray magnetic fields, but it should not be that much error (try taking a measurement, then flipping meter 180 degrees so the current flows "backwards" through the clamp--should be about the same with just a +/- sign change on the meter).

So I checked all the voltages. (This is at 10 am, full sun, just starting to absorb.) The 16 batteries ranged from 6.92 V to 7.76 V, with most at about 7.4 V on both strings. No difference between the strings, except that min and max were both on the string with the low current.

Almost a 1 volt different between batteries in a bank is significant. Could be the low voltage battery has low SG and needs more charging. It also could be the high voltage battery is fully charged (high SG) or even a high resistance cell.

To a degree, you can isolate the problem by measuring voltage at night with significant loads--And see if the low voltage or high voltage battery has a "voltage collapse".

And check the voltage drop across cables/connections (looking for bad/dirty/loose connections under heavy current).

Also, measure the difference in current between the two strings (under heavy charge and heavy discharge). 10-20% difference is probably normal, and may even change over time (as the bank charges/discharge).

Differences of a factor of 2x--Pretty close to being significant and needing to look for bad electrical connections, bad cells, etc... A factor of 10x difference in current means something is broken in the bank (during charging, a shorted cell can make that string take all the charging current; an open cell can stop current flow, etc.).

So, log the SG, voltage, and current flow (i.e., 10 amps charging, 20 amps discharging, etc.):

String/Batt/Cell / SG / Volts / charging or discharging:
A/1/1 1.265sg 7.20 volts / charging 10 amps
A/1/2 1.200sg
A/1/3 1.260sg

A/2/1 1.245 6.95 volts
....
B/1/1 1.100 6.00 volts / charging 30 amps
...

If you have a 48 volt @ 800 AH battery bank, your estimated maximum charing current is around 41 amps:

• 18 * 175 watts * 0.77 * 1/59 volts = 41.1 amps typical maximum current in good weather

Which is about 5% rate of charge--You can easily add more panels for 10-13% rate of charge (if your loads require it, and your wallet allows it ).

If you charge during the day and discharge at night--you should be OK. If you have significant loads during the day, it will reduce the average charging current to the battery bank--which can be an issue (leading to deficit charging).

-Bill

Re: More Surrette problems

terrynew wrote: »

Inetdog is right, I'm worried by one of the two strings having a much lower current than the other. But thanks to Bill's tip on using the Zero button properly on the ammeter, today I'm getting readings that aren't as different between the two strings as yesterday's were: With the ammeter on the cable between the first two batteries of String A (A1-A2), I get 7.07 A. In comparison, the current between the first two batteries of B string (B1-B2) is 9.57 A. I get the same two currents placing the ammeter between the last two batteries of each string: A7-A8 = 7.06 A, B7-B8 = 9.53 A. This is when the CC's output current is displayed as 21.4 A.

That "zero button" just does not make sense on the meter--does it. You have to turn on the "zero mode" to zero the meter then make the measurements (with zero mode on). Every other meter I have used, you just push the button (or adjust a knob) to zero the meter. Perhaps in AC mode (which I have not used) the Zero Mode Off makes sense.

Yea--that is "close enough" for government work.

I get no difference in readings when reversing the ammeter on the cables. And with the ammeter looped over both the + and - cables entering the battery box, I get 1.08 A rather than 0 A, but the ammeter loop couldn't close completely due to the cable thickness (there's a 1/4" gap in the 'claw').

Yea--the gap is a problem.

Do you have only two battery wires (+/-) from the battery bank? Or do you have other wires (such as a ground wire that leads off to a ground rod/system ground?).

Is this 25% difference in current between the two strings more than a sign of the need to equalize? Since no one battery is more than a half-volt off the average, I figure the problem must be one or more batteries with high resistance.

By itself--probably not something to have a heart attack about.

So I could equalize as is, or disconnect one string at a time and equalize each separately, or disconnect one string at a time and absorb for 6 hrs daily, or disconnect all batteries in the low-current string to see if any one has a super-high resistance causing the problem (in which case I could separately use a plug-in charger at a neighbour's on that one for 24 hrs at 15 A).

So far, I see two things that cause me concern.

First, the large voltage difference between batteries (6.92 V to 7.76 V)... that is too much and would indicate either a Specific Gravity miss-match (and require equalization) or "unhealthy" batteries (open/shorted cells, sulfation, etc.). Or possibly poor/dirty electrical connections (measure voltage drop across wires/connection points with meter set to 2 volt or 200 mV full scale).

The second--Your array size appears be undersized (2,580 watt array):

• 12*215 Watts * 0.77 panel+charger derating * 1/59 volt battery charging = 33.7 amps in full sun/middle of day typical charing current.

We recommend a 5-13% rate of charge, which for your 800 AH battery bank would be:

• 800 AH * 59 volts charging * 1/0.77 panel+charger derating * 0.05 rate of charge = 3,065 Watt array minimum
• 800 AH * 59 volts charging * 1/0.77 panel+charger derating * 0.10 rate of charge = 6,130 Watt array nominal
• 800 AH * 59 volts charging * 1/0.77 panel+charger derating * 0.13 rate of charge = 7,969 Watt array "cost effective" maximum

And, what is the maximum current you are seeing from your array (and at what bank voltage). Your array sounds like it may be under performing. You can use your new DC clamp meter to measure the current in each of string of your array and make sure that all is OK there (solar panels, unlike batteries, current share very nicely and each string should be pretty close to the other--around 10% maximum difference).

Yikes, you're asking for a lot of testing, Bill. Can I do the voltage measurements with cables connected? When you ask for the stats at both heavy charging and discharging times, do you mean during bulk charge for the former and at night with a heavy load (e.g., toaster oven) for the latter? Should I equalize before all this?

Yes, the cables are connected for load testing (this will make a "problem" measuring the resting voltage as one string could be charging/discharging the other--but you have larger problems than resting voltage at this time).

And yes that is a lot of measurements. Trying to figure out if you have a "bad cell" causing your voltage variations or simply needing equalization (battery voltage is related to both current flow in/out and SG in a good cell).

If the batteries started out with miss-matched state of charge (i.e., one battery at 95% and another in the series string was at 75% SOC, then you have a lot of initial equalization to take care of).

Another option would be to get a 6 volt battery charger and "manually equalize" the low cells (with a smaller genset to save fuel and be a bit easier on the other series connected cells.). You have to be careful because you will be doing this with a "hot" battery bank and you will have between 0 to 48 volts DC with respect to ground on those battery charger leads (the leads should be electrically isolated from the generator/AC grounds--but I always am careful with unknown, new to me, equipment).

-Bill

PS: As BC04 suggests... Do the equalization with only one string at a time--That will better match your current to the bank size and force the charging current through one string instead of worring about it being "diverted" to the other string.

Re: More Surrette problems

If you are getting 40 amps, then no reason to go on the roof.

-Bill