Wrenches, I had promised a followup post on what I learned when I inspected this client's batteries. The short form of the original issue (original post query is at bottom): client had as recently as June been able to EQ from his array to 31.1V, now couldn't get past 30.7.
There are a few good lessons here. Upon arrival: Batteries at 78 degrees F in basement. TM2020 indicated pretty high late-morning SOC, but still charging in bulk. Cell voltages (two-cell pairs) ranged from 4.80 to 4.82, so pretty close range there. Specific gravities, though, ranged from 1.234 to 1.260, measured with a refractometer, with lowest cells getting rechecked and verified. The cells with both the lowest and highest SGs had the (slightly) highest cell voltages. About 52A input, less 10A load. Enabled EQ from MX60; temperature rose to 84F. Voltage measured at MX60: 30.7; voltage on MX60 display: 30.9 (I corrected this); voltage measured at battery terminals 30.45. The 1/4V drop is in the cables and remaining disconnects of the original and upgraded system DC disconnects and cables. My observations: 1. Small amount of liquid pooling in EPDM battery box liner. Asking lotsa questions determined that the owner was near-obsessive in maintaining his batteries, and added water every month. These large Surrette cells had replaced multiple strings of golf-carts, so old habits die hard, even with much larger electrolyte reserves. We determined that he was filling the cells too often and too full, which was causing some electrolyte spillage. As any spilled electrolyte was later replaced with water, some cells were thus being diluted. I suspect that this is the primary cause of lower EQ voltage. What made this worse was that we had replaced his charge controller with an MX60 and showed him how to engage EQ on it. All MX60s show a screen on EQ startup that says "Check Water Level". He had interpreted this to mean add water before equalizing. The bubbling and expansion of the EQ overcharge then caused more spilled electrolyte. 2. The fan on the MX60 had failed, as many of them have. While I don't know for sure, this may have either limited the unit's output due to temperature rise, or may have caused a misinterpretation of battery voltage or temperature correction; either could have caused these symptoms. The customer noted that he was familiar with the fan's sound, but hadn't noticed that it had recently stopped. So my actions: 1. Increased absorption time. 2. Disassembled and cleaned three terminals that were beginning to show corrosion (even with Quick-Cote on them). 3. Advised the customer to fill his batteries less often and less full. 4. Set MX60 Max EQ time temporarily to the max (7 hours, I think) and instructed him to do a corrective EQ from the array for at least a week. 5. Ordered a warranty replacement fan from Outback. 6. Recommended he not sell his generator but repair it, in order to get his EQ C/rate up to C/10 or so for hotter charging and winter EQ. I could also add electrolyte to the cells with the lowest SG after all this, but how much to add is a real crapshoot. Can anyone offer guidance on adding electrolyte to compensate for spillage? And one final question: In my original post I asked if deionized water was the same as distilled water. Nobody addressed this question. The customer said he investigated (Wikipedia, I think) and found no difference. Any other opinions, or (even better) definitive answers on this? Thank you, Allan at PosE -----Original Message----- Sent: Sunday, August 03, 2008 11:54 AM Wrenches, Thank you for the whole set of responses. I haven't been to the site yet, but plan to stop there this week. As the symptoms are not easily explained, I wanted to put the problem out for ideas here first, to reduce the chance of missing something. I will report back with what I learn. I will first check whether the 0.4 V discrepancy is related to battery temperature. It seems to me that the simplest way to do this is to check the MX60 settings during equalization. If the measured voltage is at the MX60's temperature-compensated EQ voltage, then a warm environment is the apparent cause of the reduced voltage. To answer a few of your many questions and fill in some details: 1. As I noted in my original post, the MX60 is temperature compensated. The VFX3524 is too, although the client tells me he hasn't run the generator since we added the 1.28 kW array. 2. To the best of my knowledge, the charge controll settings have not changed (that would be too easy!). The client hasn't changed settings, and we take care of all of their service needs. 3. The client equalizes regularly, as he notes in his description: last EQ (to 31.1 v) was in June. 4. The array is wired 48V, so hot modules and/or undersized wire isn't a voltage-limitation issue. 5. There are 6 4V batteries, 12 cells. I had conservatively figured the charge rate at about C/20, so the charge rate is adequate for off-grid. I most likely have the MX60 set to two hours absorption with a 2% done amps override to lessen gassing. After Surrette's tech bulletin about batteries in off-grid applications (recommending higher charge voltages and longer absorption times) came out, I started setting longer absorption times; I will go back and increase the absorption time on these to three hours, while keeping the 2% done amps in place. Would anyone argue with this setting change? 6. Tri-Metric voltage readings usually vary from MX60 readings, yes, but they don't tend to change over time. This situation is presented as a recent change from consistent past performance. 7. The client has not checked individual cell voltages or specific gravities. I will do that when I go there on a service call and will report back. 8. Re the quick C/5 charge idea: I would ask Jamie Surrette for his recommendation here. For sure, it hasn't happened; the client says the generator (an older Honda 5500, I think) needs service before it will start. I'll advise him not to sell it (as he plans) until we have resolved this and gotten Jamie's advice on battery care. 9. Miscalibration of the MX60 is unlikely but possible, and I will check this when on site. Thank you to the many of you who responded. Jamie Surrette responded off-list too, asking about battery temperature. I will report back after checking all of the above. Allan at Positive Energy -----Original Message----- Wrenches, In late 2003 we installed a 24V set of Rolls-Surrette 4-KS-21PS 6V 1104 A/hr batteries in the off-grid system of a long-time client. The client's system had a democracy array of around 800 watts of tracked PV. In early 2005 we added 1,280 watts of additional PV, and the client hasn't run the generator since; the system rarely drops below 80% and fills to 100% most days in sunny New Mexico. Batteries are located in a basement, so temperatures stay relatively constant, although it will vary slightly seasonally. MX60 controller is temperature compensated, with 29.6 VDC bulk and 31.4 VDC EQ set points. The problem is that the client reports being unable to achieve previous EQ voltages, and this inability appears to be a new development. The system is otherwise performing well. As the client is sharp and pays close attention to the system's operation, I'm not ready to blame user ignorance (with one possible exception; read on). This may be an issue of seasonally warm temperatures, but I'm not sure this would explain the full 0.4 VDC reduction in voltage. I told him that I would post his problem to this list, and also request advice from Surrette Battery. Here is the client's description of the symptoms: I'm concerned that I am unable to equalize batteries at this time. Up until June of this year the Tri-Metric readings during equalizing indicated holding voltage at 31.1, with maximum amps gain of around 52. Three weeks ago, when set on equalize, the maximum voltage reading was 30.7, with maximum amps of 46; this was reached around noon, and by around 1:30 pm voltage/amps began its typical afternoon drop-off. I attempted equalizing on two other occasions since then--the last time this past Saturday, with clear skies up until mid-afternoon--but having the same results. So, the question is: are my batteries beginning their decline? We've had this set for 4 1/2 years, so it would seem a bit too early for this, given that these batteries are expected to last 7-10 years. Second issue, possibly related: On another matter, I have been using deionized water for the batteries for as long as we've been using photovoltaics (since 1991), as it is my understanding that deioinized water is virtually the same as distilled water--I just checked Wikipedia on this (using "deionized and distilled water" as the search phrase), and this appears to be the case. I don't know the answer to this question, as we have always recommended distilled water only in batteries, and have never faced this question. Any help would be appreciated. Allan Sindelar _______________________________________________ List sponsored by Home Power magazine List Address: RE-wrenches@lists.re-wrenches.org Options & settings: http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org List rules & etiquette: www.re-wrenches.org/etiquette.htm Check out participant bios: www.members.re-wrenches.org
