I'm looking at my laptop now (remote app) and the battery's are on float holding a rock solid 55V as the battery temperature is spot on 25 deg. The wattage going into the battery's from the charge controller (that's showing in the remote app main screen) to hold float is fluctuating between 20 - 38w for the past 1/2 hr :?:
Why so low, how can this be just 38W!
Lets look at the loads. Modem 15w, router 10w , wifi repeater 5w, septic air pump 5w, standby loads 3w, inverter always on self consumption 65w. That adds up to 108w of loads then it always takes a extra 20 or so watts extra over the base loads to hold float. Around 130- 150w. When the fridge kicks in I usually see 230 - 250w.
For a moment I started thinking to myself hmm the inverter must have shut down and I have lost all my loads on the battery. Then I gave it some thought its imposable. The inverter must be on as I am remotely monitoring the system live so the modem, router and wifi repeater all have AC power. So that covers that one idea.
Then about 45 min of this strange low 20 - 38w float wattage it jumped up to 130w (I that's the fridge kicking in as that adds about 100w ) Only to drop back down to 20 - 30w 1/2 hr or so later.
Its got me buggered how its managing to hold 55v float voltage on a 48v big 1330ah flooded led acid battery + cover 108w of loads with just 20 - 30w going in ? Its even dropping as low as 5w for a few seconds but spending most of its time around 30w and no more than 38w. it doesn't add up as the voltage should just be dropping with this low float wattage as its not enough to cover the loads.
Strange behavior could I have a problem with my classic 150
i took a short 1min video of it below holding 54.8v (slight temp comp at the time) with between 0.2 - 0.6A going into the battery its to low to be possible.
http://www.youtube.com/watch?v=wFiRUEiQQrA&feature=player_detailpage (http://www.youtube.com/watch?v=wFiRUEiQQrA&feature=player_detailpage)
Ideas?
several hrs later the same thing floating between 5 - 38w holding float voltage of 54.9v. yet covering 108w of loads
Jumping up by 100w every so often to cover the compressor cycle of the fridge then dropping back down.
My thinking is ..One its reading the battery voltage wrong but unlikely. Two its showing the wrong charge wattage - Amps going into the battery just on the local app. but its never done this before.
Time will tell when I get to the house this afternoon and find a discharged battery or not :o
What kind of internal shunt dose the classic use to monitor the amps its feeding the battery and what would make it do crazy things?
Kurt
I don't think there are current shunts in the Classic, I suspect it's using Hall effect current measuring devices for in and out current. Not exact but apparently the amps to battery is within and amp or two of the real value.
That 100+ watt load you calculated as the normal load, how accurate is this?
I was surprised to see my loads were about 45w instead of 150w as calculated. Device such as router and switches and laptops seem to use next to nothing once they get to working temp and are idle. Each time I used the laptop and sleeping parts woke up the 5w used went up to 9w.
Anyway just a thought...
Dgd
Well the 6000w selectronics inverter charger consumes 65w when on and doing nothing. I remember throwing all the small loads on a watt meter some time ago and the numbers I am giving is from memory but even if you 1/2 them all and it took no power to keep the 63kw battery at float its still 1/3 the power going in to whats going out.
So tell me then. Because the classic uses a Hall effect to do current measuring perhaps the reason for it being very inaccurate below say 100w or so.
I have seen very strange behavior from hall effect watt meters. Especially when measuring a low wattage load and you put a additional load on like a old style 240 - 12v transformer to drive a 50w down light and the wattage actually drops say 20w rather than going up 50w when you flick the switch that's funky but its true. They have some down sides.
Anyhow another hr has passed and now its floating at 11w Im sure its not accurate.
Kurt
offgrid;
I wonder if that 65 watt draw for the inverter being on goes away as the load gets near its value?
I do not know but it just seems that the idle draw would be reduced as the inverter gets active? As in you have a 50 watt load does the inverter draw 115 watts (65+50) or some value between 65 and 115 or the original 65?
I think you may be chasing ghosts or just seeing differences in measuring or something.
I am no expert.
I use the shunt in my epanel to watch the "net" current to my batteries and that helps me see what is actually happening. My Outback FX inverter only shows load in 100 watt steps so tracking what it is using with the internal system is not very accurate. And comparing it to what the classic says it is putting into the batteries just will not work. With the shunt and a zero center ammeter I can tell at a glance if I am balanced, drawing or charging.
Just how I do stuff here.
I think It would be very difficult to be certain what is going on without live simultaneous measurements of the loads and the source values.
Just from here.
Tom
The Classic uses Shunts to measure current and is generally very accurate. Not saying it can not drift but it should recalibrate every time the relay opens.
Ryan
65w is just the dc consumption to power up the inverter. Its what it takes to run the beast . Its interactive inverter and charge + a data logger. I think its just what it takes.
Ryan,
Thanks for that well I'm not sure whats going on then. I'm at the house now so I will see how it plays out tomorrow as i can watch the loads real time but im 99% confident i know the base loads when I'm not at the house as it hasn't changed in the past year.
but 30w isn't enough to maintain (float) voltage and cover the loads that's for sure especially as it did this from 10am - 5pm.
Kurt
Well remember 1 amp is 48-58 watts depending on battery voltage. So say the Classic was off by 0.5 amps this would be close to 30 watts. I also wonder like Tom did if the inverter becomes more efficient as loaded so the loads are not really seen at the 60-70 watt range?
Ryan
QuoteI also wonder like Tom did if the inverter becomes more efficient as loaded so the loads are not really seen at the 60-70 watt range?
Like I said above it just takes some power to run the inverter even under no load. The manual puts this down as 65w and the shunts seem to confirm this. Though I do kind of see what your getting at that is can provide say up to 50w load without consuming any more dc power than it dose to sit idle.
I'm not to sure on that perhaps (getting my head around it)
The big difference today is I monitor the system most days and for the past 115 days I have had the classic installed. The lowest float wattage I see is around 120w ( that kind of added up ok to me) This sudden drop today was out of the ordinary.
Kurt
I agree when you do something that many days in a row and it changes it makes you back up and say HUH?
I guess till you get there and can monitor it we will have to wait patiently :o
Yes I am on location now but got to the house just as the sun was going down.It's 9:45pm Friday night where I am and its dark so no action from the controller. Tomorrow once the battery's hit float I will go to the power room and see if i can nut it out (that's if it dose it tomorrow)
Kurt.
Quote from: Halfcrazy on February 22, 2013, 06:11:09 AM
The Classic uses Shunts to measure current and is generally very accurate. Not saying it can not drift but it should recalibrate every time the relay opens.
Ryan
Shunts generally do not drift and are physically calibrated when manufactured.
If there were two 150amp shunts in the classic box then I think they would be physically large enough to be noticed. Drifting occurs with Hall effect current sensors and recalibration becomes necessary on a regular basis.
The input voltage and current displayed by the classic are quite suspect as often when compared to output volts and amps the output wattage exceeds input wattage :o
Dgd
Trust me the Classic uses Shunts. I do not have a picture now but will try to find one
Ryan
Quote from: Halfcrazy on February 22, 2013, 09:19:24 AM
Trust me the Classic uses Shunts. I do not have a picture now but will try to find one
Ryan
Ok, they must be well buried under those circuit boards... ;D
Dgd
It would be interesting to see the inner workings and the two shunts. Just to give me a better understanding of the product and the basic layout.
I tend to agree that unless your chasing milliamp accuracy most shunts I have used even on cheep imported devices tend to be surprisingly accurate over the years without any recalibration. though I would imagine if a shunt was overloaded and over heated that could mess with its accuracy.
I do have a few very small devices $30 (the size of two match boxes ) with a 130A shunt inside of them. Along with lcd screen and other electronic circuitry. So I believe the classic has lots of space for two shunts. Even the cheep devices are accurate to within a few watts (comparing 4 of the same devices on the same fixed load)
My view is perhaps the classic knows whats going on but there is things getting skewed between what the classic is reading and what the local app is displaying. though i don't have anything to back that up :o
Kurt
QuoteThe input voltage and current displayed by the classic are quite suspect as often when compared to output volts and amps the output wattage exceeds input wattage
Agree with David, I've seen several times (last time today) 8A output on classic, +9A on my batt monitor and around +9.5A on clampmeter, just one array, no other source.
Erik
I will come back later and update this thread with photos and more detail. But basically I just did some live testing and my accurate(tested and agrees with another device) shunt is showing 40A going into the battery's while the classic is showing 55A going out. Its all over the place.
I use end amps to trigger float each day. Today just befor I hit float. My pack was showing 13A going into the external shunt. At the same time the classic is showing 21A. That's not good when the classic is using this data to trigger something like end amps that I have set for 15A and it cant be trusted.
let me get a 3rd and perhaps even a 4th shunt on the line and see what device is telling fibs .
will update later.
Kurt
Quote from: offgridQLD on February 22, 2013, 08:09:29 PM
I will come back later and update this thread with photos and more detail. But basically I just did some live testing and my accurate(tested and agrees with another device) shunt is showing 40A going into the battery's while the classic is showing 55A going out. Its all over the place.
I use end amps to trigger float each day. Today just befor I hit float. My pack was showing 13A going into the external shunt. At the same time the classic is showing 21A. That's not good when the classic is using this data to trigger something like end amps that I have set for 15A and it cant be trusted.
let me get a 3rd and perhaps even a 4th shunt on the line and see what device is telling fibs .
will update later.
Kurt
If the Classic is showing 21 amps and another meter is showing 13 amps, that may be because of external loads.
Unless it got out of here without being tested and calibrated, that much difference just doesn't happen.
You also might want to check your reference shunt/meter. You also can not measure the common negative
wire coming out of the Classic. You must measure the positive output wire itself if the PV negative an battery negative
are shared.. If there are no loads or inverters on the battery side, then a negative leg battery shunt should be OK.
Current measurement is not easy and we have gone to great expense to try and get the Classic's output
current measurements very close. It ~should~ be at least within an amp or so.
That being said, if for some weird reason, the Classic got out of here without being calibrated, then it
could be off by a bunch. I would hope that would be extremely rare and of course want to know
about it.
Yes, the Classic uses internal shunts and they are our own design and custom parts.
Shunts can be made to work well but you still have to take great care in PCB layout
and circuit design just reading those shunts.
boB
Hi Bob,
Im not trying to be smart or have a go. I'm sure all effort is taken to get the classic reading as close as it can for its intended purpose. Though I have been seeing some strange output behavior in my numbers so I investigated. perhaps i just got a uncalibrated unit.
Perhaps this video will prove my point a little more. both shunts are calibrated and read identical , you will notice the inverters shunt is average over 2 or 3 seconds and my 2nd calibrated external shunt and reader samples a little faster but they both agree with each other. All readings are independent of loads and measuring output amps before the inverter. I also tested my calibrated external shunt reader along side two other shunt readers that all agree with each other. So that's 4 against one at the moment.
You can see the classic is always showing more amps going into the battery .
link to video
http://www.youtube.com/watch?v=aHZtXVUGL6k&feature=player_detailpage (http://www.youtube.com/watch?v=aHZtXVUGL6k&feature=player_detailpage)
Kurt.
My question is are all the other shunts between the classic and everything else or are they between the battery and everything? I ask this just to be sure you are accounting for inverter draw and loads?
Never-mind I was able to follow the video after a couple goes. So the Classic is reading high by 80-100 watts? Seems a little high but? I also don't see how it could have been a calibration issue at Midnite if it ran good for 3-4 months? I would say check it against a couple meters at a heavier load and see what it shows.
Ryan
The inverter was consuming about 5 or 6amps ( DC consumption) and perhaps 1A or less to keep the batterys at float you can see the -5 A in the inverter LCD that's the inverters DC load at the time so no I am not reading Net or my external shunt would only be showing 1 or 2A going to the battery if the classic was correct.
If you like a can tune the inverter off to prove this.
edit: that was a test at under 10A and at 50A this morning it was off by more than 10A though it was off by more at times!
Photo taken at same time. Shunt 1 on the inverter is befor the load (it's not NET) Net is shunt 1 number minis the (INV AMP) - load number
(http://farm9.staticflickr.com/8513/8498532437_5b8d628456.jpg)
(http://farm9.staticflickr.com/8111/8499637902_fa2bcdf510.jpg)
kurt
Ok so at the time in the video the batteries where essentially taking 0 amps. The inverter was using all the classic was making? I would be interested to see if it still shows off by the same 1.5 or so amps when it is making say 30-50 amps?
Ryan
Yes it was actually 5.2Amps not 6 A (you can see this on the inverter LCD in the video) so the battery's were taking around 1A or a little less to hold them at float.
Kurt, in that video I am seeing the Classic saying about 2 amps higher than your meters.
About 8.5 amps vs about 6.5 amps...
Try that again, but go to the Classic's mode menu and turn it off and back on again or
maybe, in main status, press the ENTER button to force a sweep. I think it may do
a zero re-cal at that time also but for sure if you turn it off and on again.
The offset is what the zero calibration takes care of.
Thinking out loud here....
If this was a gain error, then at an actual 13 amps I would expect to see
about 17 amps output and not 21 amps from the Classic.
Thanks !
boB
Ok bob,
I have removed the external shunt reader (it was very temporary setup - dangerous with cover off wires every where ;D) I did it just to prove my inverters external shunt 1 readings were ok.
So now I will do your test but just compare classic to inverters shunt 1 as we know its good now.
It doesn't look to be a offset shift its more like a 10% shift . though its not that consistent. I will make some more comparisons at different AMP levels 10 - 20 - 30 by applying loads while on float.
Back soon....
Kurt
Well, if that Classic really does have a gain problem, I can tell you what pot to turn.
If gain problem, then it's going to be off more amps at higher output and off by
lower amps at low output.
boB
Thanks bob i just did your test.
batterys on float consuming 1A to hold float
I tested .. as close as I can get jumping up and down between the two meters but you get the picture. The classic is roughly 20% over the shunt 1 at different loads.
Base loads 4.8A shunt 1 vs 6.4A classic
water pump + base loads 14.1A shunt 1 vs 17.8 classic
water pump + toaster + base loads 45A shunt 1 vs 55.7A classic
Gain problem? What pot do i need to turn?
Kurt
Quote from: offgridQLD on February 22, 2013, 10:42:14 PM
Thanks bob i just did your test.
batterys on float consuming 1A to hold float
I tested .. as close as I can get jumping up and down between the two meters but you get the picture. The classic is roughly 20% over the shunt 1 at different loads.
Base loads 4.8A shunt 1 vs 6.4A classic
water pump + base loads 14.1A shunt 1 vs 17.8 classic
water pump + toaster + base loads 45A shunt 1 vs 55.7A classic
Gain problem? What pot do i need to turn?
Kurt
You will need a fairly small, adjustment screwdriver.
The left side towards the front is a hole gaining access to the trimpot labelled GBATT.
(see picture)
Turn very slightly counter-clockwise (I think) and watch the current come down.
Easy now... be careful !
Make sure the offset hasn't been affected (it shouldn't) by turning mode to off
and back on again to recheck the reading. If it reads closer, great.
boB
Quote from: boB link=topic=1055.msg7749#msg7749
The left side towards the front is a hole gaining access to the trimpot labelled GBATT.
(see picture)
boB
So does the trim pot labeled G-PV adjust the current input reading?
Dgd
Quote from: dgd on February 23, 2013, 04:41:53 AM
Quote from: boB link=topic=1055.msg7749#msg7749
The left side towards the front is a hole gaining access to the trimpot labelled GBATT.
(see picture)
boB
So does the trim pot labeled G-PV adjust the current input reading?
Dgd
Yes, but it is gain, not offset. For the input, the error is more of an offset problem which would be
a software fix.
These adjustments are usually cal'd as good as can be at the factory and shouldn't normally be
tweaked. It appeared as though offgridQLD has a gain anomaly that should be fixable.
I would not advocate adjusting these unless you really know what you're doing and have
a good reference. I'm not sure how good those meters are that offgridQLD has, but if
he absolutely trusts them, then, so be it.
boB
Thanks for that picture Bob.
At least I know a little more about what might be happening and how to adjust it if need be. The likelihood of my two shunts both being wrong and by the same amount is slim. So I am reasonably confident . That said I will do some more tests tomorrow Using a 3rd shunt reader just to be 110% sure before tweaking the pot.
Thanks
Kurt
The part that confuses me though was the first issue was the Classic was reading to low? I would probably get an eyeball on it over a period of time before tweaking to much.
Ryan
Yes that was just on Friday It was a oddball day when it pumped out only 30w for most of its time on float and its a mystery that one. I wasn't at the house to see things live. Just using the remote app.
What I have always noticed is the classic showing more amps going into the battery than what the inverters shunt 1 reads. It wasn't until today that I took the time to add a 2nd shunt to see which was out.
Kurt
If you guys don't get praised for hands-on customer service, you should.
Quote from: Vern Faulkner on February 23, 2013, 09:39:32 AM
If you guys don't get praised for hands-on customer service, you should.
Thank you Vern !
This particular problem isn't all that big of a deal but it could be some family freezing in
the cold on a weekend and we try not to let something like that happen if at all possible.
I am still a little boggled here. If I am doing the math correctly I am showing the 1300 AH battery is using 1.2 Amps to maintain Float voltage? This seems to low? Where you measuring current flow on the battery positive leaving the classic with that shunt?
Ryan
QuoteI am still a little boggled here. If I am doing the math correctly I am showing the 1300 AH battery is using 1.2 Amps to maintain Float voltage? This seems to low? Where you measuring current flow on the battery positive leaving the classic with that shunt?
Ryan
Yes that's what it boils down to when you subtract the loads.
To be honest I'm not sure what to make of my complete system when it comes to making any logical sense from the data. I haven't made any changes to the trim pot to adjust the live amp reading of the classic to line up with what my inverter and test shunt are showing.
The reason for that being. I checked the total DC KWh logged by the inverters shunt 1 for Sunday .The number lined up perfectly with the classics output for the day 11.4kw and wouldn't you know 11.4kw on the inverter :o So it got me thinking they where both recording the same total DC kwh numbers for the day but just off in how they displayed the amps live on the screen. (that wouldn't be a issue to me as its the 24hr totals that are important to me)
The only issue is when I checked back on the past week the totals for each day don't line up between the inverter and classic (but Sunday did spot on) So its all a bit of a mess. I wish I could make some sense of it all as it makes establishing reliable patterns difficult.
To recap on why i am trying to establish what is doing what each day. All stems from me trying to work out AC KWH consumed each day VS DC KWH needed to cover that load each day .This is when I picked up on a few odd behaviors like one week day where the battery floated at just 30w and the discrepancy between the live Amp reading on the classic and my inverters shunt monitor.
What I can confirm 110% is that weekdays when I'm not there the house consumes 2kwh AC (confirmed on my inverters logs and the homes analog kwh meter) This is a very stable number that hasn't changes in the 60 weeks we have owned the house and doesn't change more than 2 or 3% from one day to the next.
The strange thing is the DC KWH the classic feeds the battery's each day can range from just 4kwh to as high as 7.5kwh DC to cover this load. (we always have more sun then we know what to do with so the fluctuation is not due to weather variation). So Why the variation and why so much more DC kwh in that AC kwh consumed where my initial questions?
After some research I came to the conclusion that because I have 63kwh 1330ah at 48v of battery and my loads are so small weekdays that the battery spends all its time in the 95% - 100 SOC range and this is a very inefficient range to charge in - most likely around 50% efficient or less. Along with 4000w of PV and a ton of sun the battery's spend most of the day on float . Now this would explain the need to pump so much more dc KWh in than I take out in AC.
What it doesn't explain is why I get so much variation each day in what the classic puts in to replenish the battery's even though the loads for 5 days of the week are identical. I should see reasonably consistent recharge pattern Monday to Friday as I have repetitive and consistent loads.
So even if I ignore all my components and just trust what the classic is telling me the classic its showing different KWH charge totals each day(and different by a big margin) yet it has a consistent load each day.
It's this condition that's got me digging around for answers.
Kurt
Kurt
I suspect the variations are just the amount of Float time etc. Maybe it was sunnier the larger days. I would expect a 1330 AH battery to suck down several Amps just to maintian float voltage.
So back to the readings and the shunt. In your video the Classic is it wired to the dc box with 3 wires or 4? Basically is there a common negative running back to the DC box and the PV- is landed in the box?
Also Where was the shunt connected in that video Classics Battery positive?
I am still suspecting the issue is a differential current from Common DC negatives as the numbers start to add up for me when I do that. The PV was at like 77 volts so the differential voltage would be 25 volts and the controller was showing 6 amps on your shunt so half that is 3 for a total of close to 9 amps so maybe it is still just the placement of the 2 external shunts you have?
Ryan
I will go over the wiring this weekend and draw up a little sketch of how all the shunts were connected. To see if there is something I am overlooking in how they are wired.
Regarding last Fridays (strange behavior) I forgot that I actually recorded a small video last Friday when my battery's were on float for several hrs.
As I mentioned earlier in this thread it was very strange as they were only taking around 30w (as displayed on the remote app live view) to hold float voltage (55v) and cover base loads as mentioned before around 60w AC + the power necessary to just run the inverter. It was strange as usually find it takes a good 100 - 130w to cover this and maintain float voltage and 200- 230w when the refrigerator cycles.
Have a look at the video and you can see why on Friday I was thinking something strange is going on as its only taking 30w and it continued to do this for several hrs until the sun went down. It would jump up to 130w for 15 min or so when the refrigerator cycled and then drop straight back to 30w for 1/2hr to 1hr or so until the fridge cycled again . Its just not enough wattage to cover 1330ah - 48v of battery at float + my AC base loads.
video link
http://www.youtube.com/watch?feature=player_detailpage&v=wFiRUEiQQrA (http://www.youtube.com/watch?feature=player_detailpage&v=wFiRUEiQQrA)
Kurt
Kurt, I can't read the display in that video.
What kind of inverter is it that you have again ?
Does that kW-Hour reading from the inverter only cover the
DC amps and volts from its battery shunt ? It doesn't
include its AC kW-Hour readings ?
boB
Sorry it was a video of my laptop screen yes its hard to see but you can just make out its feeding around 30w to my battery.
My inverter is a 6000w/48v selectronics PS1 inverter /charger. The Kwh AC reading (its a separate reading in the inverters menu) This gives me the KWh AC what the inverter has produced in to the house sub board (confirmed by also reading the homes analog AC kwh meter) Both showing 2kwh AC each day + - a few % very consistent. So I trust this is my actual AC kwh consumption as sounds corect for the loads I have weekdays and the inverter and analog kwh meter agree,
I have taken this clear video to show how the inverter DC shunt 1 reads battery amps, consumed amps and total dc amps into the inverters shunt.
http://www.youtube.com/watch?v=z1MIa05aWD4&feature=player_detailpage (http://www.youtube.com/watch?v=z1MIa05aWD4&feature=player_detailpage)
Kurt
Quote from: offgridQLD on February 26, 2013, 06:24:36 PM
I will go over the wiring this weekend and draw up a little sketch of how all the shunts were connected. To see if there is something I am overlooking in how they are wired.
Regarding last Fridays (strange behavior) I forgot that I actually recorded a small video last Friday when my battery's were on float for several hrs.
As I mentioned earlier in this thread it was very strange as they were only taking around 30w (as displayed on the remote app live view) to hold float voltage (55v) and cover base loads as mentioned before around 60w AC + the power necessary to just run the inverter. It was strange as usually find it takes a good 100 - 130w to cover this and maintain float voltage and 200- 230w when the refrigerator cycles.
Have a look at the video and you can see why on Friday I was thinking something strange is going on as its only taking 30w and it continued to do this for several hrs until the sun went down. It would jump up to 130w for 15 min or so when the refrigerator cycled and then drop straight back to 30w for 1/2hr to 1hr or so until the fridge cycled again . Its just not enough wattage to cover 1330ah - 48v of battery at float + my AC base loads.
video link
http://www.youtube.com/watch?feature=player_detailpage&v=wFiRUEiQQrA (http://www.youtube.com/watch?feature=player_detailpage&v=wFiRUEiQQrA)
Kurt
Kurt
A wiring diagram would be super. It will help us assure the Shunts are reading what we think they are reading.
Ryan