I'll check with Ed but from the PVOutput forum it would not appear to be unique - several users have been having hassles. It would make sense that a running total would fill in the gaps.

This is what the PVOutput.org help file (http://pvoutput.org/help.html) indicates:

The way I understand it is that the instantaneous values (V2 and V4) could be used to generated the totallized values but if you send the totallizers that is calculated by your software it compensates for loss of data transfer and it uses the totallizers as uploaded and calculate the averages from the delta in totallizers.

That is how Manie implemented it in his app and it works properly. I have a lot of internet issues at the moment but my data remains correct so I think Manie implemented it the right way.

Cobus

Jaco

I am an engineer, reliability and accuracy of the data is the primary thing in my mind. Who's doing what is of lower priority to me.

I really liked working with Manie, he was very open and eager to discuss how the software works and discussing improvements to the software. Without rigorous user input and testing I think any development runs the risk of being sub-optimal, less than it could be. That is why I am a bit concerned about "closed" development. Having open development of at least methods, even if the programming platform or language differs, leads to better overall products. A larger community can build on work already done and improve on it. Whether it is done in Python, C, etc or on Windows/Linux/etc. does not really change the fundamental method of e.g. storing, uploading, displaying, etc. of the data. Being able to look at how different people/groups implemented a function and adding improvements must lead to better solutions.

The only case where open development would not be considered is if one intends developing a commercial product. The way I've been looking at the whole solar scene I am becoming more and more sceptical whether a commercial software solution for inverters are viable, in South Africa and global. Why do I say this?

1. Most people who install these systems are not interested in what it's doing. They really don't care what the condition of the batteries are, how much yield they got, what self consumption of the equipment is, etc. There are a small percentage of solar installers who find this interesting, basically the people on Powerforum.

2. Once you add the possible cost of the software solution, whether it is outright purchase or a monthly/annual fee for storing and viewing data, adding to that possible hardware requirements (RPi, tablet, etc.) and the complexities of keeping that running, most people loose the remaining bit of interest they had in the data.

3. The really small amount of enthusiasts are often technical in background and like to tinker and build their own. As soon as a price is added to a solution they see that as the ideal motivation for a fun build your own project. A large percentage of the people here on Powerforum fall into this category or have the skills to do it.

4. Furthermore the remaining market  is too fragmented. Just think about all the combinations of inverters and other equipment.

At most I see these solutions being viable in a donationware format. I know that is not what people want to hear. The development takes a lot of time and effort and it must be worth something but in reality it is highly unlikely that it will even recover cost. Accepting this is really important as it gives an indication of the amount of effort and level of complexity we should invest in them. Building it in a developer and enthusiast environment reduces the effort as parts can be shared and re-used.

I've seen this type of process often, exercise tracking apps, weather monitoring and storing, etc. In niche markets commercial success of a "luxury" component of the system is highly unlikely. We have to remember that the end user already paid R30-50k for the inverter, R40k for panels, R20k for the installation, R50k for batteries, then he found out he had to buy a battery monitor for R3k plus a cable for another R1k. Then he finds out one year down the line that his expensive battery bank failed due him not checking the water levels and SG every week. Etc, etc.

I know we discussed this at length during the AICC development but I think being realistic about the financial potential is important. For myself, I am not interested in solutions where I do not own the data generated and can do with it as I like. If I want to publish to the cloud I want to be free to do so. If I want to store on a local database I also want to do that. People also don't want to hear this but why would I pay for a cloud solution with emoncms and pvoutput (I'm sure there are more) around?

C

TTT, sorry I've completely derailed your topic.

47 minutes ago, jdp said:

So I guess the software you are using does not keep track of what is sent over the internet. It should keep track of what it sent and if the link is down buffer the data so it can send it when it is up again. That is communications 101. But Cobus lets not tell them what to do, lets just make sure our work 100%.

It would appear so. It does not really matter to me since the only time I really use PVOutput's data is when I am away like when we were in Coffee Bay a while back.Initially I could see no merit in the data on PV Output  and joined with very low expectations. I have changed my mind. PV Output is fun and gives you a good overview of what is happening over the whole country but is certainly not my first port of call when I am analysing  data. The 5min interval of PV Output makes it a fairly coarse dataset.  

1 minute ago, Chris Hobson said:

The 5min interval of PV Output makes it a fairly coarse dataset.  

Emoncms is much higher resolution. And you can easily build a dashboard with your system data that you can view from anywhere in the world.

Hi Cobus

My world is pretty much restricted to the Republic of Roodeberg 98% of the time. If I am away I just want the piece of mind that the system is not being abused and PVOutput's data does that . If I was in New York I really would not be fussed with production and efficiency etc. One does not go on holiday to micromanage things at home.

Chris, agreed. I'm often away from home on business while the family is still at home. Being an engineer I really like looking at the data and what the system is doing. My wife calls it my girlfriend in the garage. 

Pvoutput.org may be a coarse dataset but in my case is incredibly close to what the inverter says i producted, within 0.25 to 0.5kWh on a 50 to 60kWh generation day.

Sent from my SM-G920F using Tapatalk

PVOUtput is great for seasonal trends etc but I certainly would not troubleshoot using PVOutput data. Therefore if after the 3rd day of cloudy weather interrupting my internet and PVoutput says I have produced more than I actually have I am not going to slit my wrists. For the most part my data is also within 0.5kWh of what the inverter has recorded.

You must enjoy being top of the pile on most days.

Nope, not my thread.

But same here, I want to see what is going on NOW, not in 5 minutes, on data where I can do as I please, therefor SolWEB having it's own graphs, stats, logs and historical data per user. Sending the info to PVOutput (done) and emoncms (later), that is just for those who want that.

But Cobus, you are right, this is not viable for developers to focus on long term with no financial reward, it is more for the hobbyist, yet they are but a fraction of what is out there.

And the moment you release software, you are going to need to help adhoc or more frequently, depending on a lot of factors. And to do it all for free, fun in the beginning, gets boring very quickly if you see no benefit. And testing, very time consuming and costly to keep on fixing if you missed things.

And installers ... it is a tough one.

So my point of departure is: Device independent software that gives all the information the devices can offer for the discerning connoisseur, that if devices are replaced, that they can still use the same software, keeping all the historic data.

Unfortunately that does make it into more effort, it is more complicated, and it cannot be for free indefinitely.

So I am with JDP on this one. We tried to work together here on the forum, did not go so well. The developers here are rather busy so the only way I could get it moving with my devices was with Edmund and throwing in my developers time, whom are on salaries, to get things moving faster, for I do not want to be bound with software to one or two devices. Dang, at one stage I had 6 devices sending data.

And I do not want to see data s Edmund wants to see it, my needs are different. And then what happens if I change devices, or Edmund goes for SMA? Then we would have been stuck.

So it is a very simple problem really, reminding me of this I saw at a printer cartridge supplier once:
We offer three kinds of service:
Fast – Cheap – Good
You can pick any two

Fast and Cheap service – won’t be Good
Cheap and Good service – won’t be Fast
Good and Fast service – won’t be Cheap.

Note: Nowhere does the word free enter the above. 

So really, for a few hundred a year license fees getting more value from the data that what the software costs, with the added benefit of once it reaches critical mass that more and more ideas will flow making the software even more powerful, I see no problem with that.

To get there is no easy task either.

2 minutes ago, The Terrible Triplett said:

emoncms (later),

30 seconds delay. Really what more do you need?
And it is free to use 

Just now, viper_za said:

30 seconds delay. Really what more do you need?
And it is free to use 

By all means Viper, use it! 

TTT, I am not disputing the amount of effort in the scope for your team. What I am doubting is how many people would be willing to pay for it. As I said, the market is small, the portion of that willing to pay for it even smaller. So what I'm saying, maybe the scope should be smaller and the development done via a community. I'm concerned that you will never recover the cost you put into SolWEB.

C

1 minute ago, The Terrible Triplett said:

By all means Viper, use it! 

I am and 30s gives you more than enough time to do the math's with all the values to see if they add up correctly

Ignore the low SOC, I'm playing around a bit with automation

2 hours ago, jdp said:

That is communications 101. But Cobus lets not tell them what to do, lets just make sure our work 100%.

WOW ... 

30 minutes ago, cvzyl said:

TTT, I am not disputing the amount of effort in the scope for your team. What I am doubting is how many people would be willing to pay for it. As I said, the market is small, the portion of that willing to pay for it even smaller. So what I'm saying, maybe the scope should be smaller and the development done via a community. I'm concerned that you will never recover the cost you put into SolWEB.

C

Now worries Cobus, thank you for the concern!

We do have a long play going for this is not my first rodeo with software done on a risk basis.

Nor is Power Forum our target market, even tough we do offer it for free to new testers, but are having some doubts on that front. 

AICC and Manie are the main manne here.

1 hour ago, Chris Hobson said:

PVOUtput is great for seasonal trends etc but I certainly would not troubleshoot using PVOutput data. Therefore if after the 3rd day of cloudy weather interrupting my internet and PVoutput says I have produced more than I actually have I am not going to slit my wrists. For the most part my data is also within 0.5kWh of what the inverter has recorded.

You must enjoy being top of the pile on most days.

I was initially finding a big power production discrepancy on cloudy days until I realised the efergy monitor has no idea about current direction and it was just inverter self-consumption. In the settings on pvoutput I created a rule that sets all generation below 850W to zero and my figures came right immediately. Re top of the pile, that's just size and as we all know size really does not matter, its all about efficiency.  

Viper, Lekker man!!! I like.

FWIW SolWEB had similar view months ago, mine different looking to Edmunds, as we have different devices / needs to what we each want to see. 

Ps. You are aware that if you where using SolWEB today, it would have been free, you being one of the few pilot sites. 

[quote post="11656" Power Forum our target market, even tough we do offer it for free to new testers, but are having some doubts on that front. 

AICC and Manie are the main manne here.

What do you mean by your statement?

Sent from my SM-G920F using Tapatalk

I said:

19 hours ago, The Terrible Triplett said:

Nor is Power Forum our ...

Ag, no worries Manie. Really.

Cobus made a good point, this forum is for the techies to make new things, play with new devices and have fun whilst doing it ... then there is the rest, the vast majority.

So I am just tired of this endless dissing what is done, or not, what platform is used, or not, what device is chosen, or not, free or not, why have you not used this or that, it all is really just confusing the dinges out of the one and only question: Does it work?

The last straw for me was the comment about PVoutput and something about 101. That sarcasm was uncalled for and just made me think: But why? Is this where we are heading, sarcastic comments against other members, and aimed at my SolWEB partner to boot, on a open forum?

I considered dishing my points out to all and sundry, O wait, I bought the fire pool  , and then I though, no, way too easy and no fun in thatll! 

 

The vast majority of people in this world really do not care how it works, as long as it works, and works correctly, myself included. And when, not if, when a oeps occurs, and believe you me, I can guaranteed that it will happen to the best, the only question really is: Will it be fixed?

So I sat and thought a bit about it all here, and out there, and realised that we are on the right track, just keep on quietly chipping away at what we are doing and see where that leads us for I know for a fact SolWEB and SolWEBRdr is not going anywhere but forward, one step at a time, till we get where we want to be i.e. the preferred solar monitoring system in the country, irrespective of what equipment the user wants to use, with our without Power Forum users. 

It is going to take a while, costs a small fortune, and yes, someone is going to pay us to recoup our costs and for us to make a profit, in the end.

And that brings me to my doubt. Keeping on offering free versions to testers of SolWEB, here on the forum, obviously, I have been and are wasting my time. Mistake on my part.

Everyone is waiting for the release of the new AICC, with you helping with the development and being a beta tester, your excitement is clear and infectious. We have to wait for AICC's release.

Sorry but as a new forum member I have no idea what the rant above has to do with this thread that is for the pvoutput.org SA team.

Sent from my SM-G920F using Tapatalk

Gerald, it is how it goes on forums. To keep conversations neat and tidy, is a massive challenge at the best of days.

So threads do sometimes get derailed on maybe rants (few and very far in between), more often on car comparisons  ... all in good jest.

And just to be clear, the above was not a rant, not even close. Bit of history here, read the entire forum.

If I want to rant, done it once or twice, I clearly stipulate such by stating:
Rant Starts:
Rant Ends:

My apologies if I upset you though.

TTT i would appreciate it if you keep my name out of your comments as i dont use your name in a bad way

Sent from my SM-G920F using Tapatalk

1 hour ago, Manie said:

TTT i would appreciate it if you keep my name out of your comments as i dont use your name in a bad way

What? Confused now I am.

Re-reading my post I cannot find where I could have possibly offended you. I just responded to your question, sharing my thoughts in general.

 

3 minutes ago, jdp said:

TTT we are not dissing what is done but how it is done. To take other peoples ideas and then to make as if it is your own that we as techies have huge problems with.

If a project is open source then it is open, You can not take the efforts of other and sell it as your own. And that is a pitty as that is what happened here. That is why we will no longer share any new ideas here.

I will still help people who I feel I can help. I only comment if I have the facts.

To be honest I and other members feel that sometimes you think this is your forum. You post a lot of with every comment to make it seem you are only  trying to be funny. But in fact you insult and stir the pot. I don't know why I still bother sometimes.

I came to this forum to get help with the Axperts, that I got and I pay that forward. But the rest was not fun and I have forgiven but I will never forget.

 

JDP, in all fairness (I'm not siding with anyone here!) there's nothing wrong with taking someone's idea and making your own product off it. Heck, there's 86 inverters available in South Africa right now. But each one is different, some are very different, some have minor differences. 

BUT, what you are implying is that it is wrong to take someone's project / code and say it's your own. 

Just cause some people wanted to write applications to monitor and log their equipment doesn't mean it was your sole idea. Or TTT's or "Bob"s. In fact, from that point view everyone would be in the wrong since the manufacturers first came up with the idea to monitor and log their equipment. 

 

I'm an outsider, who was looking forward to a great new app to get more out of my equipment, but people got hurt, and by the looks of it, for a silly reason. This isn't aimed at anyone specific, but if you feel like writing a new application, don't worry what other people say about it, or moan about it, just do it. Imagine is Bill Gates had to listen to all the negativity around MSDOS / PC-DOS / Windows 3.0 back then. Well... then we wouldn't have had "too many windows developers" (although I still think there are far more Linux devs than Windows devs, but that's just me). 

 

Just my 2c, do what you had in mind from the beginning, and do it for yourself. IF you feel like sharing it, great, but then be man enough to accept that people have their own ideas & requests for how it could be used in their environment as well. Don't ignore it. Don't throw tantrums about it. Don't shutdown your project about it. Decide what you want to implement, commend on it, invite people to participate, and perhaps even clearly specify what you don't want to implement. Look at any big, successful application on the market today - it takes a LOT of community input to make it successful. Some requests / suggestions get implemented, some not.