I've noticed the net amp hour reading from the wbjr is incorrect, on a idle system, and under load.
It's night time here, so I did a test, solar panel breakers off, inverter breaker off, nothing else connected to the system, yet the wbjr is showing a -3.8a draw from somewhere. Batteries are full, sg is to spec.
I've checked the battery bank with 2 different clamp meters and they both show -0.7 to -0.8 amps, certainly not what I'm seeing from the wbjr.
I noticed it when I had a known load on my system, my tv, a light, and internet, and it was showing too high an amp draw, I knew it didn't look right. So I shut the whole system down, and the controller is showing a -3.8(9)amp draw, when the clamp meters are showing less than 1 amp.
Im pretty sure I've read somewhere here that's there's a pot in the controller that I can adjust, recalibrate. Before doing that, any other thoughts as to why there may be this discrepancy. Batteries are just 2 yrs old so I really doubt it's a battery issue. Bank is 4 Deka 6v L16's 370ah in series, 24v's.
Shouldn't the wbjr and clamp meters amps read the same, or at least close?
This system was recently changed to 24 volts. The system ran for years at 12v, and I know from years of monitoring that my tv, internet, and light draw around 105w, or in around 8ish amps, and is what the wbjr would show, and was verified with a energy meter.
Well now on 24v the same tv, internet, and light show a draw of 145w, impossible. System voltage is currently 25.3v, and amp draw is showing 5.7a. If the system was drawing 105w at 12v, like it has for at least 5 years, it should still be showing 105w at 24v, yet it isn't, it's now 145w.
Interesting...
Did the WB Jr. used to read correctly when it was on the 12V system ?
There are not adjustments for the WB Jr. in the Classic or on the WB Jr. itself.
You might want to check the tightness of the screws from the WB Jr. to the shunt itself. If things are loose, that might show an unreasonably high current reading.
The WB Jr. is mounted directly onto the shunt, right ? i.e. No extension wires ?
Yes, it appears to have been more accurate at 12v, and everything is tight. That is something that isn't overlooked here, as I have considerable temperature swings from season to season in the battery room, so I'm constantly re-tightening, and or checking the tightness of all connections.
The wbjr is mounted directly to the shunt. Everything was fine until the firmware update, change to 24v, this system has ran fine for years at 12v.
Like I said, I have an energy meter so I know the watts my light, tv, and internet draw from the system, and that wattage has changed since changing to 24v (or the firmware update) , as far as the controller is showing. Those constants haven't changed for years, still the same devices, and they've always used between 100-105 watts.
What concerns me more is the close to 4amp draw the system is showing with nothing connected, that the clamp meters don't agree with. I've done a soft and hard reset using the jumpers since doing the firmware update a few weeks ago, to me the issue has appeared since the firmware update, and not the switch to 24v. The firmware was updated about a week before the change to 24v. I did notice while still on 12v (but the new firmware) in Grahams app that the net amp hours value was negatively increasing throughout the night, again with no "measurable" current through the clamp meters. I'd get up the next day after the system had been off all night, the inverter is not left on, and the net amp value in Grahams app would have decreased by 20-30 amps. I don't ever recall seeing that. Whatever my negative net amp reading was before going to bed, it was always very close to the same number the next morning, not now for some reason.
Only other thing in my mind that would affect this is my shunt, but why would it just suddenly act up. It's been in service for close to 8 years, I don't have another of the proper size here to test.
Ok, here's a couple pic's from Grahams app for the net amp hours, one from last night before I went to bed, and this morning. The system was off for the whole night, inverter off, panels off. Where did those 20 amp hours go on a disconnected system? I was not seeing this prior to the firmware update.
Hi Muskoka,
A few questions:
What DC loads are on the battery, with the inverter OFF? Just the Classic? And, perhaps, the energy monitor? Are ALL other loads, on the inverter?
Guess that your Classic is a Lite?
There may be a bit of a semantic issue. To me, Net Ah, is the total number of accumulated Ah that the battery has been discharged, or charged, since the last full-charge.
What were the Firmware versions that were used for the Updates to the Classic?
Later, Thanks, Vic
Things do not make too much sense ...
When you changed from 12v to 24v did you also change the battery capacity value for the SOC ?
Larry
Thanks for the replies guys.
Vic, loads are a DC-DC buck converter, 2000w 24v inverter, Trimetric TM-2025. At night the buck converter and inverter were off. The Trimetric was still connected, so what little power it was drawing was still present, specs say 30ma. That's it for loads.
Yes, it's a Classic Lite, with a Mngp purchased a few years later.
Ok, system is fully charged by noon, goes to float, Net ah's reset to 0, and hover around 0 for the rest of the afternoon/evening. Unless I do a equalize, then that Net ah figure climbs.
Lets say I don't do an equalize. So the sun has gone down, net ah's are at or very close to 0. I now have my tv, a 5w light, and my internet hub/router on. That for years through testing with a energy meter draws close to 100-105w or 8'ish amps per hour, obviously the amps are at 12v, prior to me switching to 24v. I've measured the draw on the new inverter, and it's basically the same as the 12v unit I was using for years, so no change there. By the time I go to bed the net ah's are usually -30 to -40 depending on how late I stay up, so the system has been discharged that many amps. Odd thing, the watt usage is now around 145-150w, and the inverter draw is basically the same. Working current of the buck converter is 25ma. The difference in watt's doesn't really bother me, and has nothing to do with the ah discrepancy overnight, as everything is turned off, disconnected, so those devices have no effect on what I'm seeing.
When I go to bed the inverter is turned off, as well the buck converter (only had this since switching to 24v). Prior to the firmware update, or change to 24v, when I'd get up the next morning the net ah's were always very close to the number when I went to bed, in other words, the system was off, no loads, very little change in net ah. Now I'm seeing the net ah decrease by 20-30ah's overnight, with all loads off.
Don't know how this factors in, but I'll mention it. The voltage throughout the night normally climbs after I shut the system down. Going by 12v numbers again, when I'd go to bed the voltage on the Trimetric would read 12.5v, get up the next morning and the voltage would read 12.6v. As soon as I'd start turning things back on the voltage would immediately drop back to 12.5v, normal for a resting system, and a healthy battery bank.
Firmware versions I just flashed are, Classic 2193, Mngp 2186.
Larry, yes the battery capacity was changed from 740ah, down to 370ah.
I'll continue to keep an eye on it, and see if the net ah's discrepancy changes over time. I've just switched to 24v, so I'd like to watch it a few more weeks and see if there's a change, or it remains the same.
Glen
Hi Glen, thanks for the added info.
One thing: What is the brand name of the Shunt on this system. Two common ones, are MidNite's 500 A one, and the other is the Deltec 500 A. These are very high quality shunts, well made and reliable.
The other thought, would be to disconnect the Trimetric (or whatever the energy monitor is), from the shunt, and its V+ connection to the battery, overnight, to see what the WbJr/Classic reports for Net Ah.
Also, did you need to extend the cable that connects the WbJr to the Classic? If so, is this cable shielded? And, are there any other wires running in any such extension cable? How is any shield grounded?
Thanks for the FW version ... cannot speak to its behavior, as we are running one or two versions behind, here.
Thanks Vic
Do you have a Fluke handheld meter around ?
IF so... While the WB Jr. is reading 4 amps or so, try reading how many millivolts are showing across the two shunt measurement screws that the WB Jr. is connected to. Also check how many mV it reads when the two probes are just connecting to each other.
Glen,
And guess that one should ask just how the DC/DC Converter is being switched Off, at night. Guess that you are using a DC circuit breaker as the disconnect (?).
Thanks, Vic
Can you draw a simple diagram of where the shunt is in your system and what is connect to each side of the shunt since you changed to 24v system.
Let me just start off by saying, everything is connected, setup properly at this end. The system has been running for years. This wasn't a complete system overhaul, I changed the battery cabling, added a buck converter for 12v lights, swapped out the inverter 12v to 24v, flashed firmware, that's it.
Vic, the shunt is a Deltec 500a, 50mv. I did remove the Trimetric Tm2025 from the system, just for a few minutes, and the reported amp load did not change, in other words it's removal made no difference. Wbjr still reported a -3.9a load at the time, with or without the Trimetric connected. No, the cables for the Wbjr were not extended. The buck converter actually has it's own on/off switch, I haven't installed a dc circuit breaker, yet. To isolate it from the system I'm removing a wire to the buss bar, it's not a high current converter, 160w max.
John, nothing has changed in the layout, nothing was moved, and it's been fine for years. One side of the shunt is connected to the last battery negative with a 18" cable, the other side of the shunt is connected to the negative buss bar with about a 6" cable. The positive and negative buss bars take all the devices cabling, nothing is connected to the battery bank, other than the buss bars.
Bob, yes I have 3 different meters here. A Fluke 87, a Greenlee DM820a, and my Extech Ma220 clamp meter. The Fluke and Greenlee were given to me by a electrician friend, who had purchased new meters, they work fine (?), have seen years of service. Another friend has a clamp meter that I used the other day, it's no longer here.
Here's the numbers, Wbjr was reading -4.1a, idle, isolated system, midnight last night.
Extech probe to probe, 0.0mv. On Wbjr screws it bounced between .1mv and .2mv.
Greenlee probe to probe .02mv. On the Wbjr screws it bounced back and forth, .38mv, .39mv, .40mv. It was always in that order, 38,39,40....38,39,40.
Fluke 87 probe to probe, 0.0mv. On the Wbjr screws it was steady at .2mv.
There's always been a "slight" phantom/ghost load on the system, 1a or less usually. Just throwing this out there, perhaps changing the battery cabling has exposed a problem, or issue with a battery. The bank was configured series/parallel at 12v, now it's all series connections for 24v, so I'm wondering if the old setup was somehow masking/hiding some issue, the way it was wired?
Logic would dictate that going to the higher voltage would decrease any amp readings across the board, they should not be increasing, but perhaps changing the configuration has exposed a "slight" issue? Doesn't make sense that at 12v the ghost load was -1a, and at 24v the load is -4a.
Or am I fishing where there's no fish.
The place to start is isolate everything you can - so only controller and batteries in circuit and have all loads disconnected or switch the breakers or fuses off.
Yes some things can draw power even if their switch is off - depends where in their circuit the switch on device is. So first thing to do is check that buck converter you said you added in . Disconnect it and see what happens - maybe nothing but you have then ruled it out.
If you have every load disconnected and no excessive current - start connecting up each load one at a time.
Larry
Hi Muskoka,
There is a strong indication, that there IS something wrong with your system.
Even though this system has worked for several years @12 V, would suggest that you focus on what was added/changed after going to 24 V.
The Buck converter seems to be one of those things added/changed.. It may have an electronic Off/On switch (as Larry noted). You mentioned that you briefly disconnected it, and the tare draw did not change.
What brand and model number is your Buck converter?
Perhaps you should leave the Trimetric connected, but disconnect the Classic from the battery (switch your circuit breaker Off) for the entire night.
This is a simple system, there just cannot be that many things to check/eliminate.
More INFO on the exactness of your system, like the diagram that John mentioned would be good info.
Without more INFORMATION about your system, many of us have probably run out of guesses.
Thanks for added INFO. Tired Vic
Are there issues flashing older firmware on top of newer versions, I'll assume not provided a factory reset is done? Would like to go back to the version I was previously on. Is there somewhere I can download older versions, can't seem to find my backups.
No issue, What version do you need?
I've looked everywhere, and can't find my backups. Brain not getting any younger, but I'm pretty sure it was 2096 (?), number sticks in my head. I was trying to think of a way to verify that, and thought I might be able to see what firmware version I was running at a given point in time in mymidnite, seems to only show the current version.
Is there a log in the controller I can read to get that information, firmware version at a point in time?
Thank you.....
Glen
Quote from: Muskoka on March 14, 2020, 10:22:05 PM
I've noticed the net amp hour reading from the wbjr is incorrect, on a idle system, and under load.
I
I noticed it when I had a known load on my system, my tv, a light, and internet, and it was showing too high an amp draw, I knew it didn't look right. So I shut the whole system down, and the controller is showing a -3.8(9)amp draw, when the clamp meters are showing less than 1 amp.
… Shouldn't the wbjr and clamp meters amps read the same, or at least close?
This system was recently changed to 24 volts. The system ran for years at 12v, and I know from years of monitoring that my tv, internet, and light draw around 105w, or in around 8ish amps, and is what the wbjr would show, and was verified with a energy meter.
Well now on 24v the same tv, internet, and light show a draw of 145w, impossible. System voltage is currently 25.3v, and amp draw is showing 5.7a. If the system was drawing 105w at 12v, like it has for at least 5 years, it should still be showing 105w at 24v, yet it isn't, it's now 145w.
Here is an additional thought:
The Buck converter is not 100% efficient. And this efficiency will probably vary, based on the amount of current that it is supplying to the load, on it. This will show up as additional current that needs to be supplied by the 24 V battery.
AND, had mentioned this before:
Just to try to be clear, Net Ah, as shown on the MNGP Wb Status page, it the total Ah accumulated, plus, minus, or zero. This is not a real-time instantaneous value -- it is accumulated, since the last reset to zero (from the last time that the Classic transitioned to Float, on its own). You know this, but just trying to make sure. The Graham App calls this something a bit different, but just make sure that you are looking at the correct display item when looking for present value current to/from the battery.
FWIW, just trying to be clear, and not insult you, Vic
Quote from: Muskoka on March 19, 2020, 08:54:25 AM
I've looked everywhere, and can't find my backups. Brain not getting any younger, but I'm pretty sure it was 2096 (?), number sticks in my head. I was trying to think of a way to verify that, and thought I might be able to see what firmware version I was running at a given point in time in mymidnite, seems to only show the current version.
Is there a log in the controller I can read to get that information, firmware version at a point in time?
Thank you.....
Glen
Here you go:
http://kb1uas.com/Hemlock57/Classic%20Programmer2096.zip
Thank you.
Well, that was an exercise in frustration. ???
The controller flashed no problem, the mngp was another story. There's enough history here about flashing issues over the years, so I won't expand on that. After multiple attempts, trying different things I gave up. Flashed the latest firmware back to the controller, and mngp, without issue.
This morning was not a solar day, so I decided to strip the system down, completely, then add back one item at a time.
It appears the Trimetric is the problem, although I had isolated it before, and really didn't see any change. Now, with just the controller connected to the battery bank, everything else disconnected, not just breakers off, wires/cables removed, the idle system current draw is down to -1.3a with just the controller on.
If I then connect the Trimetric, that shoots up to -3.1a on the Wbjr screen on the controller. The working current draw of the Trimetric is 30ma, so something has gone awry. Voltage is still showing correct. That -1.8a draw is close to exactly the increase I see in wattage since changing the system to 24v. It also accounts for some of the negative net ah increase I see overnight.
The buck converter added -.1a when I connected it, Wbjr display went from -1.3a to -1.4a.
The series connections (3) on the batteries read -0.13a, -0.10a, and -0.16a, with my clamp meter, seems fine to me? The cable from the positive busbar to the bank reads -0.13a, and the cable from the last battery negative to the shunt reads -0.33a, again, looks fine to me?
I've sent Bogart a email to see what can be done.
Thank you everyone for the help, and your input.
Glen
I may have missed it but did you try measuring the shunt sense terminal screws with a Flue meter or similar good meter ?
Hi Glen,
Thank You for reporting-in.
SO many times, when a Poster's problems get resolved, we get NO feedback.
As, we all are learning, closing the loop can benefit many.
Many DC Clamp Ammeters are not that accurate or repeatable, but they still can give a general indication.
Thank YOU for the report ! Good Luck, Vic