Bests & Worsts Reviews from Amazon

Here' an unorthodox but effective way to mitigate the problem of premature battery failure, and vastly extend runtime for these generally reliable UPSs. I have managed about 15-20 APC SMT-xxxx model UPSs over the past dozen years. We have a lot of the SMT-1500 models. Some of these units were purchased before APC was bought by Schneider Electric, some afterward. All in all we are getting great, long-lived performance from all our SMT-xxxx UPSs. The one problem area is with the batteries. Buying replacement batteries is akin to buying a pig in a poke. Many reviews criticize the APC branded replacement batteries for failing prematurely, and the 3rd party knock-offs are often a joke. I've tried a lot of brands and most of them disappoint. As an example, the last time I bought replacements for a few SMT-750 and SMT-1500 units the battery life was quoted as 3-5 years. I conservatively labeled the UPSs to have the battery checked two years in. Well, 1.5 years in we had an extended power outage and the batteries failed miserably on all the units. Everyone of our UPS solutions has been engineered to have a runtime of about 2 hours. That is an unusually long amount of time, but with the original batteries, the 2-hour goal was met. When the outage occurred with the replacement batteries installed, we lost power in a matter of minutes. The batteries failed way early. Here are a couple of example links to the batteries that failed prematurely here UPSBatteryCenter Compatible Battery Pack for APC Smart-UPS 1500VA 120V SMT1500 (Plug & Play!), here New Battery Pack for APC Smart-UPS 750VA 120V LCD SMT750 Compatible Replacement by UPSBatteryCenter, and here APC Smart-UPS 750 (SUA750US) 12V 7Ah UPS Battery - This is an AJC Brand Replacement. This was when I decided there had to be a better way. I had been thinking about rigging an external battery pack on the bigger 1500 units for some time. There are of course APC models designed to accommodate external packs, but they are also designed to cost a lot more money. The SMT-1500 is a sweet spot on the price curve for us. And since we already own quite a few of them I wanted an SMT-1500 based external pack solution. This wheel has been invented before, and you can find a lot of forum posts about adding an external battery pack, with the caveat that you must connect to the UPS with a serial cable and modifying the standby time. In other words, there is a body of thought that the UPS on its own won't realize that the battery capacity has been increased simply by connecting a battery with a larger amp hour rating. You have to tell the UPS that there's more standby time or the unit will shutdown before the batteries are exhausted. This of course is a consideration even before the mechanical connection to the external pack has been sorted. However, we found this conventional wisdom not to be true. The Batteries I went out and bought two 12-volt Optima Blue Top batteries (Optima Batteries 8052-161 D31M BlueTop Starting and Deep Cycle Battery). Very well reputed maintenance-free AGM batteries. This is the top of the line 75AH model, but they make a less expensive 55AH model that will likely suit most needs. The SMT-1500 UPSs use a 24v internal pack (two 12V cells connected in series), so two Optima cells were required to be connected likewise. I then tackled the job of re-programming the runtime one of the SMT-1500s and experienced failure after failure. I couldn't get a response from the UPS with a laptop connected to the unit's serial connection (an RJ-45 jack in the back of the UPS) even using the APC supplied serial cable. In retrospect I just happened to have grabbed the very oldest SMT-1500 in our stable to test with, and the serial number / firmware rev may have been to old to go along with such antics, but the project was back-burnered for a while. A Very Workable Workaround After that last power outage I decided to just go ahead and connect the Optima's to one of our UPSs and see what would happen. What happened was I found out that no re-programming of the UPS is necessary after all. All that was required was to have the Powerchute management software installed, and to perform what is termed a "Runtime Calibration" under the Smart-UPS 1500 / Diagnostics menu. If you look at the screenshots you'll note that the standby is 710 minutes. Float Voltage It should be noted that there is also a hack for modifying the float voltage of the UPS's internal charger. Differing battery chemistries require differing float voltages. As it turned out, the chargers in our SMT-1500s float the Optima cells at 13.8v, which is just at the high end of the range recommended by Optima. Got lucky there. Mechanical Connections This was a fairly straightforward process. I just removed the internal battery pack by disconnecting the gray 50-amp Anderson Connector. I then unscrewed the Anderson connector attached to the battery and clipped off the end terminals. The wires attached to the Anderson connector are 8 gauge, and since I had decided to use 6 gauge cable to connect the Optima array, I used two 8 gauge to 6 gauge butt connectors to mate the dissimilar gauged wires (Pico 2287PT 6 Gauge to 8 Gauge Blue with Red Stripe Electrical Wiring Step Down Butt Connector 10 Per Package). The 6 gauge cable was chosen based on the length of the run to the Optima cells not exceeding a 3% voltage drop. Here is a link to that fine stranded cable: TEMCo WC0201-30' (15' Blk, 15' Red) 6 Gauge AWG Welding Lead & Car Battery Cable Copper Wire BLACK + RED | MADE IN USA. For circuit protection I used two Bussman 50-amp thermal manual reset breakers positioned atop the battery (Bussmann CB185-50 CB185 Series Automotive Circuit Breaker (Plug In Mounting, 50 Amps, Blade Terminal Connection)). As a side note, I had originally purchased these 30A ANJOSHI breakers (ANJOSHI Fuse Holders 30amp 20A 40A 50A 60A 80A 100A 150A Inline Circuit Breaker for Car Audio and Amps Overload Protection Reset Fuse Inverter 12V-24V DC Replace Fuses), but they randomly tripped at current levels WELL under 30 amps, so I returned them as defective. The Bussman units seem to be of better construction, and the ring terminals required by the Bussman breakers make a far batter connection to the cable then the single set screw method of the ANJOSHI breakers. Why two circuit breakers? It's not one breaker per battery. It's one breaker for each positive feed going to a UPS. If you examine the accompanying photos you'll notice I really went off the reservation by wiring in the Optima array to two SMT-1500 UPSs! Here's why. The goal has always been for a two-hour standby window. The equipment in the computer room draws too much net current to achieve the required standby time from just one UPS. We had to use two. But the 75AH Optima batteries provide so much standby time over the stock 18AH internal pack it became clear that the Optima array could power both UPSs and easily achieve the 2-hour standby goal. If you examine the screen shot of the runtime estimate on the Powerchute interface you'll see it reads 710 minutes (the Powerchute screen for the other UPS reads likewise). That's more than enough to power the UPSs for 120 minutes each X 2. Indeed, this setup has been tested and the required standby time was achieved in real time. And nothing blew up or caught on fire. ;> Now before you ask, I'll tell you why I didn't simply take one of the two UPSs offline, given that a single UPS connected to the Optima array could now back up all the networking equipment on its own. This is partly for redundancy, but mostly to keep the load as light as possible on each UPS to avoid burning out its inverter, since it might be required to make AC for 2 hours or more. There is talk in various forums that the inverters in this class of UPS are not designed to run for that long. But we only put an 18-22% load on each UPS, so the inverters are not heavily loaded. I'll probably update this review in a year or two. If I don't, assume that I died a horrible death in a bizarre computer room fire, in which case you don't want to try this at your own shop. :> UPDATE FEB 2020: All is well. Both UPSs are still functioning. Standby time is still about the same for each UPS (!). We have had a handful of 1 hour or less outages, and perhaps two outages that depleted the batteries to their cut off voltage. I just ran a calibration before posting, and as noted still similar number of minutes of standby time.