[RE-wrenches] Fwd: Refractometer reading has me stumped

Mon Dec 21 10:36:42 PST 2015

Hi Allan,

Thanks for the complete data situation, amazing as always. 
As usual lots of basic problems that are so avoidable. 

I remember someone, on this list mentioned that refractometers are damaged by getting too cold and that they won't read correctly
I can't find that info however.  

As the situation is unclear, I would check with a fluid hydrometer to double check your battery SG results. 

Jay

> On Dec 21, 2015, at 9:16 AM, Allan Sindelar <allan at sindelarsolar.com> wrote:
> 
> Wrenches,
> For years I have carried and used a refractometer to accurately measure SG of FLA batteries. I currently have a decent one, a nearly new Bosch Robinair 75240 that has always performed as intended. 
> 
> A couple of days ago I was hired by another local company to investigate and troubleshoot an off grid system that they had recently installed that was not keeping up with the customer's needs and had crashed. I'll give more details below, but the short question is that on a very cold day, I began measuring the SG of the batteries and got readings of between 1.00 and 1.10 on all of the cells I tested, with no obvious cause. I have never seen this and am stumped as to the cause. I tested SG on about six cells in various locations and got readings nearly identical, and all off the scale on the low side. Note that I used the supplied pipette to sample electrolyte well below the surface of the full cells, and they were lightly gassing at the time, so I don't think that this is simply highly stratified electrolyte.
> 
> The system: 12 x 327 Sunpower modules (3.9 kW) on fixed rack; Schneider XW6048 with Power Distribution Panel, XW600-80 charge controller, System Control Panel, unused AGS and Schneider's new Conext Battery Monitor. Batteries are two strings of 8 Surrette S550s, for about 32 kw-hours of C/20 storage to 80% DOD. The system has been in operation for only about 3 months. The installation quality is mediocre at best, was located in an unheated TuffShed and is powering a doublewide mobile home. The inverter, controller and SCP were connected by the Xanbus system, but the Conext battery monitor was not. 
> 
> The backup generator was a very basic manual-start 6kW Briggs & Stratton that had not been able to charge through the inverter as the inverter hadn't been properly programmed and would overload it. The site is at about 8,000' elevation, so we estimated about 4kW maximum output at 240V AC. 
> 
> I hadn't seen pictures or been given an accurate component list before arriving on site, so was not fully prepared for what I found. The day was sunny and especially cold - best guess a high in the mid-20s (F). After four hours on site my fingers were too stiff to write normally. The battery SOC monitor indicated 100%, and the SCP bar graph also estimated the batteries to be nearly full. Battery voltage under charge on each 6V battery ranged from a low of 7.38V to a high of 7.50V, with a charge rate low enough to suggest absorption. The batteries had not been equalized since new, but new was claimed to be three months ago, and this appeared accurate.
> 
> The system owners are new to off grid, and while living frugally, claimed that the system worked well during sunny periods but had crashed on about the second day of cloudy weather. They had been using the forced air furnace, and when I arrived had a heat lamp (not labeled as to watts, but I assume 250W 130V) inside the leaky battery box, shining on some of the batteries. 
> 
> As an aside, the 600V Schneider controller has no built-in display/interface, so there was no easy way to determine the charge mode or anything else except through the System Control Panel. That seems pretty bogus to me. I had not seen one of these previously, nor had I seen Schneider's shunt-based SOC meter. When I have installed XWs and SW Conexts I have always used Midnite E-Panels, which have conventional 500A shunts, and TriMetrics to offer accurate SOC for the customer.
> 
> The system settings all appeared to be set to default, other than "flooded" for battery type. Here are the changes I made in the setup. A couple of things I noticed: 
>     1) with a default LBCO of 40.0V, on at least two occasions the batteries had been completely drained, and had been recharged only by the (substantial) array; but as the array is in theory (3.9kW/58V = 65A in good sun, or a c/12 charge rate, this suggests that even empty batteries will be recharged to full in 2-3 days.
>     2) battery capacity was set at default of 440AH, when it was actually about 850AH,so the charge rate would have tapered prematurely.
>     3) The bulk voltage was the default for 'flooded' - I don't know the default, as it isn't given in the XW manual and I changed it to 'custom'. I assume about 58.4V.
> 
> Setting name                                                   Previous setting           New setting
> 
> Inverter LBCO (V)                                                     40.0 (!)                  45.2
> 
> LBCO delay (seconds)                                                  10                       600
> 
> HBCO (V)                                                                  70.0                      64.4
> 
> Battery capacity (AH)                                                440 D                     800
> 
> Max charge rate (%)                                                    100 D                     72
> 
> AC2 input Vmin (V)                                                   80                           105
> 
> AC input priority                                                        AC1                       AC2
> 
> Charge control and inverter EQ (V)                           64.0 D                     62.2
> 
> CC and inverter bulk & absorption voltage (V)         57.6 D                      58.8
> 
> CC and inverter float (V)                                           54.0 D                     53.6
> 
> 
> Given all of this, I can't explain the extremely low SG readings. I tested about six or eight of the 48 cells, and all showed the same range. I admit that I trust my refractometer, but given the other readings, could I actually have completely dead cells, only three months old, showing close to 60V with little current flowing in while I measured them? WTF is going on here?
> Thank you, as always,
> Allan
> -- 
> Allan Sindelar
> allan at sindelarsolar.com
> NABCEP Certified PV Installation Professional
> NABCEP Certified Technical Sales Professional
> New Mexico EE98J Journeyman Electrician
> Founder (Retired), Positive Energy, Inc.
> 505 780-2738 cell
> _______________________________________________
> List sponsored by Redwood Alliance
> 
> List Address: RE-wrenches at lists.re-wrenches.org
> 
> Change listserver email address & settings:
> http://lists.re-wrenches.org/options.cgi/re-wrenches-re-wrenches.org
> 
> List-Archive: http://www.mail-archive.com/re-wrenches@lists.re-wrenches.org/maillist.html
> 
> List rules & etiquette:
> www.re-wrenches.org/etiquette.htm
> 
> Check out or update participant bios:
> www.members.re-wrenches.org
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20151221/1ddf8203/attachment-0002.html>