Rogue MPPT, 490W and charging mysteries

Almot wrote:

Temps are not the problem - but I don't like how Tcomp works in this controller. Once again - there is one starter FLA and 3 coach AGM. I didn't care much about starter, but followed all the rules with AGMs. And have a gut feeling that default temperature is 18-20C, not 25C as the Rogue manual says. Why I think so:
1) ABS on one AGM didn't go over 14.4 - though this is weird, I did try 14.8 setpoint with same result. Not enough power in the evening to keep 14.6? The it would've switched to MPPT, like it normally does.
2) 2 other AGM went into ABS yesterday evening, and into Float this morning - unlike the 1st AGM, these 2 successfully reached ABS current 1%C. V FL was odd, though:
12C, V FL=13.7
18C, V FL=13.6 (default)
22C, V FL=13.5
25C, V FL=13.4

I kept batts in the shade so it didn't go over 25C.

And yet another trick of the Rogue. I'm still learning, my apologies.
After I've reached the FL on 2 AGM, I put on charge the other one that was tested/discharged previously. Rested overnight it had Voc=12.3, or ~53% SOC. Started at noon with 24A - 2*245W series panels are tilted less than optimal, propped on the palm suomps temporarily. Then it dropped to 18A, alright. Then suddenly MPPT 4A at 13something V - what the heck, the sun is still high, why 4A from 490W? Aha, one panel is shading another one by 10%. When panels are in series, weird stuff may happen, alright. Lifted the shaded panel up, no more shade, controller clicks and goes into ABS 14.6 that instantly drops to 14.5 (Tcomp default 18C, remember?), and the current is 12A. Lower the panel back on the stump, 10% shade again, BUT still 14.5V ABS and 12A. So I've somehow nudged the Rogue into ABS. If I didn't mess with the panel, it would probably remain "frozen" in this weird 4A MPPT in bright sun.

Temperature compensation is definitely based off of 25C. At 25C you will have no temp compensation.

As mentioned in my previous post, you should try adjusting the VTRIM setpoint to calibrate the controller. The older controllers were calibrated internally by hand, which was tedious and time consuming, so with the new ones it is the responsibility of the user (as with most other digital controllers these days). If you have the ABSORB setpoint at 14.6 and the controller is regulating at 14.4 with no temp sensor connected and you have full sun, then VTRIM should almost certainly be nudged a little. If you have the temp sensor connected, then even a couple of degrees warmer than 25C could be pushing the ABSORB setpoint close to 0.1v lower, so check setpoint accuracy with the temp sensor disconnected. If you have little sun (you mentioned evening), then the controller may simply be unable to keep the voltage at the setpoint. There is a 0.2v threshold before the controller reverts back to MPPT mode. So if the controller's calibration is off by 0.2v for starters and you're at that -0.2v ABSORB->MPPT threshold, you will be seeing 0.4v less than the ABSORB setpoint before MPPT kicks in again.

About the shading: it will certainly cause much lower output in series arrays. How much etc is hard to say, depending on how much and what part of the panel is shaded, how the cells are arranged, bypass diode configuration... Shade may also cause the controller to lose tracking...again, results will vary depending on the above factors. Most MPPT controllers deal with shade by doing resweeps of the array at a specific interval to find and locate the MPP again should shade occur. You can set the 3048 to do this as well using the SWEEP setpoint. It's disabled by default so that the controller is in dynamic tracking 100% of the time, which may not be appropriate for all installations. All of this is covered in much more detail in the manual. One other thing I should mention is that the MPP tracking is nowhere near as fast as the constant-voltage modes such as absorb and float. This is because the battery voltage needs to be kept under control in the latter modes, so the response must be fast. In MPPT mode the controller is constantly searching for the sweet spot on the array's power curve, and it takes a little time for it to move from one point to another, especially during shading when the array voltage may drop suddenly by 50 volts or more. Waiting a longer period of time after the panel is shaded might have allowed the controller to relocate the MPP on its own, so it wouldn't have required a resweep. Shading is tricky! It's basically the reason that companies like Enphase and half a dozen others now are in existence -- to make miniature MPPT controllers for each panel, so that when one is shaded, it doesn't mess with the rest of the array.

Anyway, please be sure to have a look at the VTRIM setpoint before doing more testing or you're going to end up with the same results. And if shading is going to be a real concern with your final setup, take advantage of the resweep feature.