My First MPPT Solar Charge and the Tracer 3215BN

[jrnymn7]

Well, second actually... first one got cut short due to heavy clouds... and a long ways to go to 'full'.

First, I'd like to thank those who were kind and patient enough to help me along this solar path. As BFL once put it, there's lots to learn about solar. All the debates/discussions over pwm vs. mppt, and such, have proven very enlightening and educational, so thank you!

So, here goes:

Bank... 430Ah, four 6's, ocv @ 12.44v, 74%soc, -111Ah's

Solar... two 140w 12v poly panels, in series, manually tracked, thru 50' of 6awg to a Tracer 3215BN MPPT charge controller and MT50 Remote Monitor.

10:50am:

- clear skies

- Isc at the moment was 'about' 9a (x2). (clamp meter not that reliable)

- Voc 20.6v

10:55am:

at Panels... 7.9a @ 31.2v,

at Bank... 18.9a @ 13.0v

The highest amperage I saw all day was at 11:05am, with 19a to the bank. Mostly in the mid 17's to high 18's up until about 3pm. I was a little surprised, given the conditions. A couple days earlier, under mostly cloudy skies, I saw amps in the high 21's several times... but that may have been an edge of cloud effect thing?

At 1:15pm, I switched over to the Solar30 pwm, and saw a difference of 1a to the bank, 17.3a vs 18.3a, for a 5.5% increase by the mppt controller.

Back on mppt, throughout the afternoon amps tapered to the mid 13's range by 6:40pm, and by 7:40pm, amps to bank were in the high 7's.

(At 5:15, I laid the panels flat, east to west, and saw a difference of 5.4a to the bank, 10.4a vs. 15.8a, for a decrease of 35%, compared to tracked.)


Tracer decided the bank was full at 7:50pm, at a little over 7a @ 14.5v. According to the Trimetric; which I believe is very close to being dialed in properly (...will know for sure next time I converter charge); 6.5Ah's were still missing at 99% soc. I usually take things down to less than 4a at 14.5v. This morning, (the next day), after a night of very light loads, ocv was at 12.73v... should be more like 12.77v, so I know the bank was not truly full.

10 minutes after it shut down, the Sun went down behind some distant trees, so the panels crashed, so it wouldn't have made it to full, anyway. And it was not able to go into float mode, so I've yet to see how that goes. But this raises a couple of questions:

? How does the Tracer decide when the bank is at 100% soc?

? Is there a way to re-calibrate it?

Any help would be greatly appreciated.

bn series manual

mt50 manual

In a couple days I'll be doing another run, less deep, and an earlier start, hopefully. That way I can see how the controller handles what it believes to be a full bank.

Another thing I noticed was, when set to 14.5v "max boost voltage", the monitor said 14.6v, and it took the bank up to 14.56v, according to the dmm. When set at 14.4v, it topped out at 14.5v (monitor) and 14.44v (dmm)... which is exactly the voltage I'm thinking will sync up well with my P10 CEF setting of 85%, on the Tri.

There are some other programmable settings I'm not sure about, but I'll post them, later. I'd like to get this '100% soc' thing sorted out first and foremost, seeing as this will be my main means of top end charging.

Thanxx!

[jrnymn7]

Did a charge with the Tracer, yesterday, and everything seems well and good. No issues with controller. Looks like less than 1/2 of one percent voltage drop thru 50' of 6awg. The 30a audio circuit breaker, on the input side, shows zero voltage drop to two decimal places, at about 7a.

Controller did drop into float at one point, for about 5 minutes, and then returned to boost. It was hazy out, with the odd low lying grey cloud passing by. Apparently this caused enough power drop that it switched to float... but not sure why it would do that? It's response time to the ever changing conditions seemed quite reasonable, though.

I checked boost time, it was indeed set at 180 min., and there were no error codes recorded. I think that shut down the other day, may very well have been a result of the Sun being far down in the w/nw, at 8pm, and partially cloudy. No problem this time round. Managed to make it to near full again... weather was not very co-operative.

I switched charge mode from soc to temp comp. Interestingly, on page 4 of monitoring, 'battery', it stilled showed soc, but it went up and down with panel power. Even when it said 100%soc, it didn't go into float. I imagine that has to do with the setting, but not sure. It went as low as 50%soc at one point, when panel power was in the 50w range (out of 280w), due to it being evening, and the odd dark grey cloud passing thru.

Amps tapered earlier than I expected, but apparently that's common with mppt controllers? Nevertheless, it remained in boost for all but that 5 min switch to float and back. I saw mostly mid 15's to mid 18's throughout the late morning, early afternoon period. The most I've seen out of the panels is about 300w, tracked... well above their 280w (2x 140w) rating.

[jrnymn7]

Jim,

When the tracer is 'off' it shows 2V panel; 0A panel; 0V batt, 0A batt, as long as it's still connected.

[jrnymn7]

BFL asked,

"But the question is what happens when the controller that has the buck converter in it goes from MPPT to just PWM and there is no other load except the battery?"

Has it been established that mppt cc's actually do go into 'pwm only'? I skimmed thru a couple of threads, recently, and I got the impression not everyone is in agreement on that?

[jrnymn7]

12thgenusa wrote:
Some controllers interrupt charging while they make an MP sweep.

yeah, I wish I had caught it right away, and was therefore able to time how long it was off, (if in fact it would turn back on.) It was definitely for more than 15 sec, though.

[red31]

jrnymn7 wrote:
(round and round we go)

I suggest you leave it connected in default setting for days and stop disconnecting it till you understand what it is doing, preconcieved notions won't change what it does.

[12thgenusa]

Some controllers interrupt charging while they make an MP sweep.

[red31]

I read this review of another tracer, sleep doesn't sound good.

sounds like a software bug!

'I then put my panels in series to give me approx. 60vdc going into the charger. I found this seemed to have fixed the issue, but very randomly the charger will go into what I am calling sleep mode. It doesn't charge the batteries, current drops to 0, then suddenly it starts charging again. Happens maybe once every 2 hrs, and only for about 15 secs.'

My epsolar VS1024 hits vboost at the battery, stays there for 2 hrs then goes to float, monitor batt v and batt charge mode.

[BFL13]

jrnymn7 wrote:

12thgenusa wrote:

jrnymn7 wrote:
But shouldn't panel and output amps be pretty much equal when only bucking?

No. That is one of the advantages (some might say differences) vs a PWM controller.

soooo much to learn... thanks for your help!

Also you have 24-12 vs 12-12. But the question is what happens when the controller that has the buck converter in it goes from MPPT to just PWM and there is no other load except the battery?

The controller adjusts "to meet the demand" going off Vmp to a higher V so output amps fit the demand. The controller efficiency would need input watts to be higher than output watts. Input V is panel voltage (say 34 in 24-12 or 17 in 12-12) Output V is battery voltage, say 14.8 at Vabs now in Float at 14.8.

Ignoring controller efficiency, with input watts equal to output watts, and input V higher, then input I must be lower. But there is controller inefficiency, so output watts is lower along with the output V, so could there be a situation where the Is are now equal? (no other load except battery)

[JiminDenver]

I have never seen any of my MPPT controllers shut down completely. The Eco-w will say Sleeping and the MS will say NIGHT but they are still on.

[jrnymn7]

o.k., I reviewed that video... he was doing a 'towel test' to de-rate the panels. The unit took about 20 minutes to find the mpp, but never shut down.

[jrnymn7]

12thgenusa wrote:

jrnymn7 wrote:
But shouldn't panel and output amps be pretty much equal when only bucking?

No. That is one of the advantages (some might say differences) vs a PWM controller.

soooo much to learn... thanks for your help!

[jrnymn7]

I never thought to check fault codes 'til after tearing things down.:( I'm not sure if it keeps a record, once powered down?

I'm sitting here waiting for the clouds to disappear, hoping to do another run... seeing as the weather forecast says "mostly sunny", this morning... but haven't even seen the sun yet.

[12thgenusa]

jrnymn7 wrote:
But shouldn't panel and output amps be pretty much equal when only bucking?

No. That is one of the advantages (some might say differences) vs a PWM controller.

[12thgenusa]

jrnymn7 wrote:
Controller Rebooting;

I tried doing an 89-100% run, yesterday morning, but a thunderstorm rolled in. But before shutting things down, there were fairly light grey clouds of various thicknesses passing by. Sometime between 1 and 1 1/2 hrs in, I found the controller shut down. Not sure how long it was in that state... but less than a half hour, for sure. I disconnected both the array and bank, and then reconnected. Unit went back to normal operation. So it may be struggling with ever changing conditions, like the one in the video. I need to review that video... as I recall it took over a 1/2hr to reboot.

Sounds like you have a defective unit. Hopefully it's not just poorly designed. Can't understand why it should ever shut down. Output will cease when panels have no light but should automatically resume once there is light. Fault codes when it happened?