Since last fall I've been using the Local App to monitor my Classic 150. For a wireless bridge, I use a small Netgear Range Extender. All is well. When I add a second Classic this summer, mostly for input distribution, I'd like to monitor it wirelessly also. What is the most efficient way to add a second wireless bridge? Two small Netgear range extenders? Would they interfere with each other? Are there any bridge hubs that take more than one Cat5e input from different devices? Thank you.
Wild guess alert!!
Plug an ethernet hub into the range extender. If the extender is just essentially a wireless ethernet "cable" it could maybe work?
Set each Classic with a different port for connections?
I don't know these range extenders but that is what I would try.
Just a W.A.G.
Tom
Look at my sig. For an option.
I would use these little LAN-WiFi bridges for the wall outlet, establish the connection to the nearest router you have for best singal.
But before you buy, check the reviews of other users of a product. Some adapters are a big piece of junk.
If you have a AVM router, it is recommended to use a AVM device, thats allways works.
If you have more then 1 device on one place, you need additional a lan switch, - so it is then better to use a 2nd router (in most cases has 4 ports) and couble it with the "main router"
But nothing beats a real lan cable :) why I have a ~ 20m lan cable from my router down to the classic in the basement
added some ferrites to it and it works fine now (as the network card of the classic is a little sissi haha)
Best Router I ever had was the AVM 7270 v2 - the v3 was newer but gets warm and wifi was not as good as in the v2, now the 6330 (internet via television cable) is the worst wifi ever had, just the chinese crap is better... don't understand why.
In the beginning with the classic, I wanted to use my old 7270v2 or v3 in the basement to connect to the 6360.
The wifi connection again and again failed. Just have full internet with my phone beeneath my living room, where the 6360 is, but if you leave the room, wifi is gone...
Now with a 2nd classic, i would have to use a switch in the basement, so thinking about to buy some NOS 7270 and deactivate the 6330 wifi antennas. (also need a 3rd lan in the basement for the UPS (USV we say in germany), that will act as inverter for the 2nd system.
Thanks to everyone for the suggestions. I'll look into some options and let the forum know what I get working.
By the way. I was re-reading the manual from my second Classic 150, which arrived a couple weeks ago, about wiring (Cat5e?) a second Classic to the first and using one of the display screens remotely, to display data from both classics. Could data from both Classics be displayed, wirelessly, via the Local App in this same way? What about updating settings from the Local App on both? Also, what is the best place to get a couple of comm wires, looks like I'll need two that are not Cat5e to set up the follow me loop on the primary. Thanks guys.
Okay, I found a couple of (technically) RJ14 4-wire with RJ11 type connector that should be fine. They're a bit long at 7 feet but probably okay. Today, I powered up Classic 'B' and this will be a follower once it gets some PV, currently Solar mode is Off for B. My original Classic has the battery temp sensor and whiz bang Jr. I noticed my new Classic 150 is at FW 2079 & the original FW 2056. Should I update the original to 2079 to begin follow-me charging or can they work together at the stated FW level? The one at 2056 has been doing well for the last 4 months. Also, all my mobile devices (a netbook) run Ubuntu right now. We only have Windows 10 on a desktop tower as a dual boot. I hope I don't have to lug that thing down stairs and over to the Classics to update FW. Its monitor is a TV set & very heavy. ;D. Will the FirmWare updater work with Wine or maybe native Ubuntu? I hope I can implement follow-me while leaving the FW as it is in both Classics anyway. Thank you.
I got sour news .
Recommended to have all the same firmware in both units when in follow me modes .
As with W10 & firmware upgrades PITA as some of had . MYSELF I have a dedicated USED laptop W7 for all solar files & firmware's etc. It has all the movies & anytning worth keeping . String size to Voltage drop files also.
This keeps all the windoz junk out of any of the power info I have.
I keep a copy of the classics settings in its folder after all is setup , for each classic & season .
VT
Quote from: Powerplay on April 08, 2016, 04:31:53 PM
... Could data from both Classics be displayed, wirelessly, via the Local App in this same way? What about updating settings from the Local App on both?
I used two running iterations of the LA, each displaying its own statically defined Classic
Auto detect off, when first LA shows its Classic then go to the hidden dir for the LA running data file, remove it and start another copy of the LA, again static define its Classic
Did this on W7 and W8, no W10 back then a couple of years ago
Can't remember exact sequence of events but some faffing about should get it to work, don't use LA anymore since mid 2014
dgd
Now to monitor two or more ethernet connected Classics I use an html5 page(s) executed by local web browser on my laptop.
To extract data from each connected Classic there is a Windows background program running. It uses tcp modbus polling each Classic nearly every second for about 80 modbus register values and updates an XML file that is used by the HTML5 web page
Similar to ZB's Blackbox code its writing data collected to a log database for future analysis/reporting
Its only a data display/reporting system at present and still needs a few gremlins extracted
I'm looking at a separate web based config program for setting Classic registers etc..
The html5 gauges display is same one I used with the Arduino and Cubie web servers with rs232 connections to Classics
dgd
dgd, that looks nice, great! (http://a68k.de/xxx/icon_spitze.gif)
Can you describe more about it? Examples for us? pleeeeeeaaaasssssssseee :D
Wich program is used?
The App from Graham is the best, no question! I love it!
But sometimes in some situations it would be good, to have the ability to display more datas in one screen.
Looks like it would be possible to arrange the websit, so it displays all needed datas at ones?
So you run the program on your server and you can access to it from any web browser.
That makes things more easy :)
As known, you can only access to the classic with one app.
But if your server can serve the datas to any device connecting to it ... wonderfull!
I've got both Classics displaying on the Local App now & updated settings on B from LA. Also, ran #8 4-conductor UF to a sub panel from my recently constructed Power Center. We had just varnished some plywood panels so I was standing in the fumes while wiring. What a trip!
Back to the DC (dark) side and on to synchronizing FW. So.
QuoteI got sour news .
Recommended to have all the same firmware in both units when in follow me modes .
As with W10 & firmware upgrades PITA as some of had .
What if I put Wine on the Net book and try to run the earlier Win version of the installer? I could try (monkey around with) that and dump Wine after/if it works. I'd hate to lug my tower down to the Classic only to have Win 10 mess up the install. Actually, Win 10.1 has been less trouble than other releases for me anyway. Not that I ever let Windows do any thing important, as a rule. What sort of issues are cropping up with the Win 10 installer?
AFAIK its not absolutely necessary to have identical firmware in both Classics for follow-me to work.
So long as the versions in use support follow-me and in your case with two fairly late but different FW there should be no problems.
I didn't see any modbus register changes concerning follow-me between your FW versions.
So you should try it and see.
I also seem to remember that the loopback connection is not necessary if you really want the first Classic to be the master and the other(s) follow its charging state, also the master has the WBjr and the battery temperature proble connected.
dgd
QuoteAFAIK its not absolutely necessary to have identical firmware in both Classics for follow-me to work.
So long as the versions in use support follow-me and in your case with two fairly late but different FW there should be no problems.
Great! Yesterday I downloaded Wine to the Ubuntu (Netbook) and it appears to be able to run the earlier version of the Win (like 98) installer very well. But my motto is try the least complicated thing first and that is definitely to leave the two firmwares as they are and set up follow me to see how it works. I'll leave the installer running on the Netbook as a plan 'B' or in case I need to install FW for some other reason down the road. Now to put up the first 800W of panels for B and pull the starter cord. Thank you for the advice.
Okay, The second Classic has 800 Watts of PV now so I'm running through a charge cycle. Everything was fine through the bulk part with both using the temp sensor on the Master Classic. The system is set to roll over to float based on Wzbjr amps set at .01 * C (5.1 amps in this case). In the absorb phase, the slave follower (doesn't have the batt temp sensor) kept driving the wattage down somewhat faster than the master for about an hour and finally it went to resting with the master still in absorb throwing (Resting 38:Unknown) message. Does this sound like normal behavior? The master is still in absorb but will probably go over to float in 15-30 minutes. I'll see what it does then. Slave has produced about .5 Kw before going to resting. Master has more PV and produced nearly 2 Kw so far.
Have you verified that ALL the settings are the same on both units?
Are the input voltage readings at each unit the same?
hth
Quote from: Powerplay on April 20, 2016, 03:28:11 PM
Okay, The second Classic has 800 Watts of PV now so I'm running through a charge cycle. Everything was fine through the bulk part with both using the temp sensor on the Master Classic. The system is set to roll over to float based on Wzbjr amps set at .01 * C (5.1 amps in this case). In the absorb phase, the slave follower (doesn't have the batt temp sensor) kept driving the wattage down somewhat faster than the master for about an hour and finally it went to resting with the master still in absorb throwing (Resting 38:Unknown) message. Does this sound like normal behavior? The master is still in absorb but will probably go over to float in 15-30 minutes. I'll see what it does then. Slave has produced about .5 Kw before going to resting. Master has more PV and produced nearly 2 Kw so far.
ReasonForResting = 38 8-4-2015 External charger. zero watts @ above set point 90 sec
This was added around 8/4/2015 - I don't use follow-me, but I thought there was really no master slave and that any time any controller changed state the other was to follow. Maybe Ryan can correct me if I'm wrong.
This should really be in the Classic forum for better response.
QuoteHave you verified that ALL the settings are the same on both units?
Are the input voltage readings at each unit the same?
hth
The setting are the same on both. Unit #2 has higher input voltage readings since it is wired that way, but I did not think that this mattered for follow me since both are doing their own thing but staying in the same charging phase until one is driven near zero near switch to float. Also the temp is rising during the day to about 25 C. I least I didn't read that any where. The follower popped back to float when the light started getting lower and both units needed to produce some power to stay in float for a bit. Now both are resting (input voltage is lower than battery). Unit #2 has 4 panels in series (longer drop to get to the combiner), Unit #1 has 3 panels in series but nearly twice the total wattage. I think the system will start fine with both in bulk tomorrow. Do you think the 38: resting state can damage the second Classic? It seems to come out of it if a load is placed on the system and more power suddenly needed. Really it seems to work fine as long as it isn't hard on the CLassic.
QuoteReasonForResting = 38 8-4-2015 External charger. zero watts @ above set point 90 sec
This was added around 8/4/2015 - I don't use follow-me, but I thought there was really no master slave and that any time any controller changed state the other was to follow. Maybe Ryan can correct me if I'm wrong.
They might have both been technically still in the same state since the Absorb count down timer was still running like absorb was still the phase for both. Yet number two was resting with 38. In the latter half of the absorb phase if the loads suddenly drop then not much power is needed to maintain voltage and the set point itself is dropping with the slow temp rise. So it might be normal behavior, unless of course it isn't. ;D
Quote from: Resthome on April 20, 2016, 05:55:18 PM
ReasonForResting = 38 8-4-2015 External charger. zero watts @ above set point 90 sec
This was added around 8/4/2015
Im not sure I know what that would mean? My thought would be that (in fallow-me), Master and slave aren't communicating properly in this case. So the slave is seeing a charging source greater than itself causing it to rest, not a master telling what it should be doing?
QuoteIm not sure I know what that would mean? My thought would be that (in fallow-me), Master and slave aren't communicating properly in this case. So the slave is seeing a charging source greater than itself causing it to rest, not a master telling what it should be doing?
Only a couple kinds of events I can think of. Switch from bulk to absorb & slave sees it first. Drops to zero to allow battery to drop to absorb voltage. Master recovers first and has a lot more juice to throw at it. Slave has to stay at zero for more than 90 seconds and goes to a resting phase. Master is at a pre 8-4-2015 FW release ( it is at 2056 I'll check its date) so does not throw the resting state. Master has enough juice to cover absorb and slave can't get back in until a higher load is applied. Does that sound realistic? I'm going to try a 'sweeper' test (with a real clunker vacuum cleaner I have that draws about 1Kw) next time I have some sun to test my theory. If I turn on the sweeper and the slave snaps back into absorb supplying power to help keep up with the extra load. We'll see.
Where I have seen this new change RFR 38 that was put in recent firmware is coming out of Equalize and going to Float. The battery voltage is higher than the float set point for greater than 90 because of the surface charge on the battery from the Equalize cycle. Once the battery voltage drops below the float set point the Classic comes out of Resting and return to the Float voltage.
Are both your Classics at the same firmware version?
QuoteOnce the battery voltage drops below the float set point the Classic comes out of Resting and return to the Float voltage.
Are both your Classics at the same firmware version?
No use running my sweeper test then. I think you've nailed it. I'm not equalizing AGMs but running a normal charge cycle. The Controllers are in Absorb phase. The set point is T-comped and the Temp has been gradually rising since morning. The one I call Slave is using the Master's T-comp data and this works fairly well. But when the battery temp rises enough to cause the absorb set point to fall. Both controllers go to zero watts to allow the battery voltage to fall. If it takes more than 90 seconds the FW on the Slave (FW 2079) throws RFR 38 but the Master (FW 2056) does not have this feature so bounces back staying in Absorb. Things don't get back in sync until the master signals Float and then the FW on the Slave joins back in. So I think I'll need to update the FW on the master to (2079) and under these circumstances the system should behave better. But do you think with both controllers at RFR 38 they would return to Float or Absorb once the set point voltage falls? If I need to update Firmware (with Ubuntu/Wine), I'll take that thread to the Classic topic so folks can see how that one goes. I might be blazing new territory there.
So with a normal charge cycle should I update Master to 2079 or Slave to 2054? I likely will not be running Equalize on AGM batteries anytime soon.
The one I call Slave is using the Master's T-comp data and this works fairly well
HM, so you do not have 2 RTS's...
Not using 2 CCs I don't have any experience with the ''sharing'' of temp data..
How does it supposed to work? Or do you mean they are set the same?
I would think that each CC would have its own sensor... ???
QuoteI would think that each CC would have its own sensor...
As soon as I set the Slave to use Master's battery temp data it always agrees with what the Master thinks Battery temp is. And you're correct, I misspoke, the two controllers are set with the same T-comp settings and they always agree on the battery temp since both use Master's RTS. So this coordinates well. Sorry for the confusion.
QuoteHM, so you do not have 2 RTS's...
Only the Master has a battery temperature sensor and the Slave is passed this data. No chance to get off track here. Both set the same, share the master temp data.
I assume RTS is battery temp sensor.
QuoteI would think that each CC would have its own sensor...
If this is the case there is a small chance for them to be off a bit from each other. It is better if the Master shares its temp data with the 'Slave', IMO.
I think unrelated is when the T-comped set point falls due to rising battery temps and can seem to confuse the throwing of RFR 38 which should only be seen during Equalize if I understand it correctly. Thanks. Do you think I should revert to FW 2056 or stay with 2079?
Quote from: Powerplay on April 21, 2016, 12:21:48 PM
I think unrelated is when the T-comped set point falls due to rising battery temps and can seem to confuse the throwing of RFR 38 which should only be seen during Equalize if I understand it correctly. Thanks. Do you think I should revert to FW 2056 or stay with 2079?
It's not tied to Equalize, it's when ever the batteries bank has a high surface charge on them that have the battery voltage higher than the controller voltage set point and it stays that way for > 90 sec. It could be coming from Absorb to Float. It depends on how far apart the set points are and how long it takes the battery voltage to fall. If you have no load it may take a while if you have some load you may never see it because the battery voltage will drop quickly with the load and 0 watts.
http://kb1uas.com/mnsforum/index.php?topic=2880.msg27490#msg27490 (http://kb1uas.com/mnsforum/index.php?topic=2880.msg27490#msg27490)
Do you think I should revert to FW 2056 or stay with 2079?
I know that 2079 made some upgrades from 2056, so I would go with 2079.
Who knows just what might not align in version to version..
I DID notice a great jump in output when I went from 19xx to 2056, but not much from 2056 to 2079...
AFAIK most of the changes were around communications to MM2
8-5-2015 Classic Rev 2050
More Classic SL mode limit fix
Added Reason For Resting 38 that will force to resting if external
charging sources are present, raising Vbatt above set point V. and 0 watts
GUI.exe changed back to original version.
8-11-2015
MNGP Rev 2054 SL changes
8-17-2015
Classic Rev 2056 Factory Burn In OCP modes
8-18-2015
V4.93 replaces update_Classic-All.exe with the 32 bit version
8-25-2015
V4.94 Removing an unused file in the package that is causing false
positive anti-virus alarms. Still Rev 2056 for Classic firmware rev
11-2-2015
Classic control Version 2074 adds My Midnite 3 capability (MM3)
Networking fixes added back in.
Auto-EQ bug fixed where EQ day count would be reset if Classic went back to Float, essentially disabling Auto-EQ
11-9-2015
Classic_ALL_2074_XX0V_11-9-2015 changes WbJrAmpHourChrg,
WbJrAmpHourDisChrgadj, WbJrAmpHourChrgadj 32 bit regs to
register address 4407 from register address 4373
11-16-2015
Classic_ALL_2077C_xx0V_11-16-2015 Classic Revision 2077 with
newer network build from 11-14-2015 and ReBulk above Float Voltage
and SOC% fix
11-20-2015
Classic_ALL_2078_xxxV_11-17-2015_AZT Re-Build of 2077C
11-23-2015
Classic build 2079 Fixes MNLP (Classic Lite panel) communications glitch
QuoteI DID notice a great jump in output when I went from 19xx to 2056, but not much from 2056 to 2079...
Classic #1 was on 2056 from the start. Even with poorly angled panels I've been getting about 85.4% of rated power from it. I can't wait to see what it does in summer when the sun will be in more direct alignment. Next week I'm putting up my last 400W of panels for this round of upgrades and I'll take both classics to 2079. One always needs to know how to update firmware anyway.
@Resthome:Very helpful rundown of the recent FW upgrades. I also agree with your analysis and think I'll see RFR 38 less likely after adding more planned loads this summer. I built a rainproof 125 amp 240 volt sub-panel last week. Everyone knows where that's going. 8) I'll start a thread in the Classic topic to let people know how updating FW with Ubuntu/Wine and the Win 98 version of the updater (seems to run well in Wine) goes.
Thanks to all for the advice.
Yeah more direct sun helps but more summer heat hurts you output. The hotter the panels get the less they produce.
[quote ]
Im not sure I know what that would mean? My thought would be that (in fallow-me), Master and slave aren't communicating properly in this case. So the slave is seeing a charging source greater than itself causing it to rest, not a master telling what it should be doing?
[/quote]
I seem to recall that in "Follow Me" there is no actual Master Classic set, rather whichever unit reaches the set point(s) notifies the others?
Why it is so important they all are configured identically. Plus the voltage offsets are tweaked, etc.
Just from here.
Tom
Quote
Im not sure I know what that would mean? My thought would be that (in fallow-me), Master and slave aren't communicating properly in this case. So the slave is seeing a charging source greater than itself causing it to rest, not a master telling what it should be doing?
I seem to recall that in "Follow Me" there is no actual Master Classic set, rather whichever unit reaches the set point(s) notifies the others?
Why it is so important they all are configured identically. Plus the voltage offsets are tweaked, etc.
Just from here.
Tom
Quote
I seem to recall that in "Follow Me" there is no actual Master Classic set,
Master is just what I'm calling the one with the RTS & Wbjr.
QuoteWhy it is so important they all are configured identically. Plus the voltage offsets are tweaked, etc.
It only seems to matter in certain circumstances. Based on my limited experience they behave like this in practice:
Let's say to stay at the Absorb set point we need 1000W and both Classics can produce about 500W. Everything should be fine but what if Classic A is T-comped to a higher set point voltage than Classic B. B will see voltage higher than set point and drive the power down to zero to adjust down even though both are in Absorb phase as passed from Classic to Classic. If everything was set equally and the Master is passing temperature data along the line then they should both stay at 500W and coordinate the absorb phase. In the first case once Classic B drops power output the Classics will both fall back to bulk eventually. Then up down up down etc.
Sp they tend to coordinate very well when fine tuned without a lot of thrashing around. This is just what I'm observing and I could be way off base as to the cause.
What you are seeing with your two Classics is their normal behaviour.
No matter how you calculate the potential power into each and the power output needed there will always be a point in the Absorb state when the power needed for charging will be taken from just one Classic and the other will go into RESTING state.
Logic would make you think that they would both provide, if each is capable, 50% or thereabouts of the needed power but that never happens.
This is because there is no communications between Classics to co-operate on providing power in the form of load or charge balancing. There will always be differences in the real voltage set point values between each Classic and no amount of tweaking will ever get them exactly equal. It probably only takes a couple of millivolts difference (on high side) to make one Classic rest while other continues making power.
This situation gets much more interesting when you have 3 or 4 Classics charging the same battery bank and can get weird enough that its necessary to use an external controller (computer) to manage charge state and charge balancing
dgd
@dgd: Maybe include a quote so causality can be applied. ::) Of course there is a point where one Classic provides all the juice but we were not talking about that. The topic here is why is it important to keep the settings between them the same.
Quote from: Powerplay on April 24, 2016, 08:31:12 PM
@dgd: Maybe include a quote so causality can be applied. ::) Of course there is a point where one Classic provides all the juice but we were not talking about that.
The topic here is why is it important to keep the settings between them the same.
Maybe you didn't quite see the relevance of my reply :)
Quote
There will always be differences in the real voltage set point values between each Classic and no amount of tweaking will ever get them exactly equal. It probably only takes a couple of millivolts difference (on high side) to make one Classic rest while other continues making power.
So you can never make the settings the same, or same enough, or identical, to have two Classics follow each other exactly.
In many multi-Classic systems it is often an advantage to control the priority over which Classic contributes greater power by having the voltage set points differenent between Classics (or Kids, OBs etc)
dgd
QuoteSo you can never make the settings the same, or same enough, or identical, to have two Classics follow each other exactly.
I'm pretty sure truncated values are used for set points after the T-comp calculation. So the calculated Absorb set point & Float set points do match exactly. And My classics both report they are charging to the same truncated value. Truncated to nearest 1/10 V. But it is true they vary in their reading of the battery voltage by a few millivolt, but this just means the one with the slightly higher reading will gradually drive down to zero first over the one with the lower reading. It is a very stable progression and I'm reporting what I observe not conjecture. If the T-comp calculation were based on different settings or temperature readings then you'd get the wild swings I'm talking about since the set points would then be different numbers and, therefore, totally different targets varying by as much as half a volt or more.
Quote
Maybe you didn't quite see the relevance of my reply
I disagree.
QuoteIn many multi-Classic systems it is often an advantage to control the priority over which Classic contributes greater power by having the voltage set points differenent between Classics (or Kids, OBs etc)
I wouldn't do that. But maybe I'll change my mind when I get to that point. I'm pretty happy with the way my 2-body system is working now that I've watched it through a few cycles. After that I would probably rather start building solenoids and combining at the Battery Bank level rather than the Charge Controller level (after more than 2 Classics are needed) though.
How many Classics are you hanging off the same Battery Bank at this point?