Integration with Span?

Feature Requests
81

Not an expert, but I thought that the home’s solar HAD to connect to breaker 29-31 on the panel…

I don’t believe that my own feed through is connected to anything. I’d like someone else to confirm, but my understanding is that the span panel can handle up to 90A breakers, but if you need anything bigger than that, or if you need more than 32 breakers, then you have to do it on a sub panel which connects through the feed through.

That said, since I only OWN a Span panel, and am not authorized to install one, I do not have access to the installer manuals that could give a clear answer. Although I do not believe that my feed through is connected to anything, the HA integration for the Span Panel is reporting numbers for “Feed Through Produced Energy”, “Feed Through Consumed Energy”, and “Feed Through Power”. I have no idea what those numbers would mean in this situation.

I’ve only had my Home Assistant system connected to my Span Panel for about a week now, but I am having some difficulties keeping it connected. Every few hours, it will randomly lose the connection and all sensors will report “Unavailable”. Without intervention, it will remain in that state until I restart the integration, at which point it will usually work fine. I can probably set up an automation to reload the integration any time HA finds that it has lost the connection, but that really feels janky. Anyone else have troubles staying connected?

82

Not here sir - rock solid I’m afraid

83

Ok, I’ll try this potential solution and see how it works for a few days…

image
image1121×473 17.3 KB

84

My solar system is hooked up through the downstream lugs because the breaker requirement with the 25% overage is greater than 90 amps. I would have rather had the solar system go through a breaker on Span just so I could monitor it, but this integration has the feed through lugs monitored (which their own app doesn’t even have). The setup works without issue.

As far as the integration, the stats and monitoring are pretty rock solid, but I can no longer turn a breaker on and off - they are greyed out. Not sure what is going on there.

85

This is essentially correct. The only clarification I’d add is that the feed through lugs are not the only option for getting more than 32 breakers. You could also use tandem/quad breakers, put a subpanel on a 90A breaker off of the span panel, or have the span panel itself be a subpanel off of something else. Each option has pluses and minuses. https://support.span.io/hc/en-us/articles/4411635811479

Geez, how big is your solar system?

Some calculations based on hypothetical large residential solar systems:

15.36 kW AC system = 64A at 240V. Upsizing the breaker by 25% gives an 80A breaker.
17.28 kW AC system = 72A at 240V. Upsizing the breaker by 25% gives a 90A breaker.

The span panel, like many other panels, has a 225A bus. As long as your solar feed in comes from the opposite end of the bus from the main feed, the total of both can be up to 125% of 225A, which is 281A. So if you’ve got a 200A service from the utility, you’re limited to a max of 81A of solar. I don’t know whether this is one of the situations where you can apply the “round up to next standard breaker size” rule, but that maxes you out at either 80 or 90A of solar, regardless of whether it’s coming in from the bottom-most space in the panel, or the feed-through lugs. To increase this limit, you need to reduce the size of the main breaker, perhaps to 175A or 150A.

1 Like
86

Ok, the connection has been pretty good ever since I set up the automation monitoring data coming from the panel. However, one wierd thing has happened a couple of times where the system seems to blip for a moment, inserting bad data into the logs. The result totally trashes energy usage data when HA turns it into a report.

Here’s the data for Main Meter Consumed Energy, which should be a steady upward slope, but with the one annoying data blip:

image
image565×536 19.5 KB

The result of that blip is the following chart:

image
image1660×976 76.3 KB

Is there a way of getting HA to discard outlier data here?

1 Like
87

Because the devil is in the details on how the NEC is written…

My solar is 16.2kW, so by the math you use I would squeak in there to the Amp - and I like your math! Unfortunately the is not how we saw it with my installers and master electrician when they designed it.

There are two rules that have to be followed, a 120% rule for the bus and 125% for solar overage. The 120% rule for the bus tops the Span out at a still respectable 270 Amps, but would put me under the total without derating my 200 amp service to 175 amp. In reality, even though I have three buildings and one of them a wood shop behind a 90 amp breaker - even if I had everything on I wouldn’t overheat the bus so the extra 13 amps potential aren’t worth nuisance derating of the main breaker and was signed off by the inspector as anyone would on this setup.

The 125% is what really gets me and not very happy about, but is why we attached to the downstream lugs. Solar potential is not rated by the solar panels, but by the inverters that are connected to the system. I have 2 x 10kW SolarEdge Inverters and that calculation for protection goes above the 90 amps that Span is capable of providing, so a small sub panel is placed with a breaker on each inverter before going to the lugs.

the setup has all been super safe, but with the unfortunate side effect of Span not officially monitoring the downstream lugs in their app until this integration with HA came online.

If there are any master electricians out there I would like to know if I have been led astray on any of the above. I did send on all the math to Span tech support and they confirmed that all of the information I sent was correct for what it’s worth. I am pretty knowledgeable as I worked as a generator mechanic (among other things) in the USMC - but for the nitty gritty of the NEC I only know enough to be generally safe.

Knowing what I know now, people’s power needs are going to get pretty complicated in the next 15 years as everything is electrified, battery backup, larger solar arrays, and EV’s. That is a whole lot even on a 200 Amp circuit. Next generation Span products should be developed with this in mind and have a bus capable of 300 Amp or more since it will draw from multiple directions.

88

Ahh, I didn’t realize/misremembered that there were different percentages for the bus fed from both sides vs the 125% continuous current sizing rule.

So it sounds like you’re saying that your setup technically violates the 120% of bus rating rule, but your inspector signed off on it given the particular loads connected?

Did you consider using both of the bottom positions (right and left) on the span panel, one for each inverter? That seems like that would have been equivalent to the setup you have now, but without the need for an extra subpanel.

1 Like
89

That seems like a classic case of improperly applying the value 0 to missing data.

If you want to retroactively remove the outlier data, I think your only option is using a sqlite editor to update the home assistant db. This is possible, but not for the faint of heart.

If you want to proactively keep similar issues from happening, then it’s a matter of stopping the 0 value from being recorded in the first place. The trouble is the conversion from “missing” to 0 could have happened in a few different places:

  • The span panel itself could have responded with a 0 that one particular time its local api was called.
  • The span-hacs integration could have transformed a failed api call, or one that was missing that particular value, into a 0
  • The built-in recorder could have transformed the missing data into a 0. I think this is probably the least likely, as the recorder is far better tested and less “experimental” than the other two.

Did the blip correspond to anything unusual happening in the real world? Like a power outage?

1 Like
90

I guess that could be possible, although it wouldn’t have solved much. Also, when we upgraded we downgraded from a 40 breaker panel and had to use a couple of double breakers to make everything fit. Luckily the house used to have baseboard heat so it was designed to push some power and the infrastructure is super robust.

It is also nice to have a breaker sub-panel before the mandatory external disconnect, just in case I need to stop the connection from within the garage.

1 Like
91

Haven’t been able to tell so far. Not power outages, tho…

I am really frustrated and feeling that something is wrong with the way my Span panel was installed or provisioned, but since the company doesn’t allow consumers access to the provisioning software I can’t diagnose the problems properly. Their “customer support” is so laughably pathetic that they cannot be counted on for the simplest of questions. They get away with this by expecting customers to deal with any problems through their installers.

The end result is that when one of their approved and licensed installers ends up sucking at their job there is nobody at SPAN who can represent their product. They really go out of their way to let you feel like you are wasting their time with your petty problems.

I absolutely do NOT recommend a SPAN panel for anyone who is right now trying to make a decision. If the company were to go belly up tomorrow, those of us who have wasted money on this product will be screwed.

We know nothing about the microprocessors and microcontrollers or the operating systems running the panel. We have no access to real manuals or information to diagnose problems. There is no current expectation that we will ever get either. The SPAN panel is a black box, albeit with a pretty glass cover.

1 Like
92

I have noticed on my Energy dashboard that, the timing is off. ( see Below), I have looked around and can’t figure why its showing my Energy consumption and production from solar starts at 2AM in the morning.
I’ve tried to fix it on my own, not able to, only thing i have not done as far as Knowledge goes is to rebuild HA.
Any suggestions please.

Screenshot 2023-01-27 at 8.39.31 AM
Screenshot 2023-01-27 at 8.39.31 AM3060×1832 352 KB

93

Has anyone else lot breaker control within the Span-HACS integration? I can see the breakers, but they are all greyed out and Span is no longer providing the entity for this.

Has the entity name changed, or did Span hide this in their API?

94

I’ve never had actual control over the breakers through the integration, only increasingly intermittent usage data. Are you saying that you were able to control the individual breakers through Home Assistant?

95

Yes, for a while I had everything. It was glorious, but now the entities are missing.

96

We’re back to the “black box” problem. There is some variable that is causing functionality to come and go, and it’s not just firmware. What firmware is your panel currently running? Mine is currently running spanos2/r202249/07, but I don’t know how long it’s been running that version.

97

My panel is running spanos2/r202246/04 and I am on Wez’s version of the Integration, so I am a version behind yours.

So, when I go into the Span webpage into the Circuits page, I do have direct control of the circuits and can label Non-essential/Nice to have/Essential - all things I could do before within HA integration. I was actually starting to build a custom Lovelace page with a circuit breaker panel and controls for Span, but now it doesn’t seem worth it.

Checking the logs I am getting a couple of errors:

Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/helpers/entity_platform.py", line 289, in _async_setup_platform
    await asyncio.shield(task)
  File "/config/custom_components/span_panel/switch.py", line 94, in async_setup_entry
    if span_panel.circuits.is_user_controllable(id):
  File "/config/custom_components/span_panel/span_panel.py", line 308, in is_user_controllable
    return self.json_data[id][CIRCUITS_IS_USER_CONTROLLABLE]
KeyError: 'is_user_controllable'

Source: custom_components/span_panel/span_panel.py:263 
Integration: Span Panel (documentation, issues) 
First occurred: January 28, 2023 at 12:49:35 PM (14 occurrences) 
Last logged: 3:08:23 AM

Error doing job: Task exception was never retrieved
Traceback (most recent call last):
  File "/usr/src/homeassistant/homeassistant/helpers/update_coordinator.py", line 168, in _handle_refresh_interval
    await self._async_refresh(log_failures=True, scheduled=True)
  File "/usr/src/homeassistant/homeassistant/helpers/update_coordinator.py", line 316, in _async_refresh
    self.async_update_listeners()
  File "/usr/src/homeassistant/homeassistant/helpers/update_coordinator.py", line 121, in async_update_listeners
    update_callback()
  File "/usr/src/homeassistant/homeassistant/helpers/update_coordinator.py", line 381, in _handle_coordinator_update
    self.async_write_ha_state()
  File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 556, in async_write_ha_state
    self._async_write_ha_state()
  File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 597, in _async_write_ha_state
    state = self._stringify_state(available)
  File "/usr/src/homeassistant/homeassistant/helpers/entity.py", line 562, in _stringify_state
    if (state := self.state) is None:
  File "/usr/src/homeassistant/homeassistant/components/sensor/__init__.py", line 855, in state
    value = self.native_value
  File "/config/custom_components/span_panel/sensor.py", line 146, in native_value
    value = self.entity_description.value_fn(span_panel.circuits, self.id)
  File "/config/custom_components/span_panel/sensor.py", line 50, in <lambda>
    value_fn=lambda circuits, id: abs(cast(float, circuits.power(id))),
  File "/config/custom_components/span_panel/span_panel.py", line 263, in power
    return self.json_data[id][CIRCUITS_POWER]
KeyError: '012f2a6af2f146d7bf6d4e170a7d23c7'

I am not a coder by trade, so I’m not sure if something broke in the integration or if Span stopped sending the entities.

98

The integration hasn’t changed… I’m using the same integration that you are, and I’ve always seen different results than what it does for other people. Something is now causing your Span panel to respond differently to what has likely been consistent queries from HA. The question is ‘What would cause that?’…

Did your firmware change? Probably not, since you are running an older firmware than I am so it’s doubtful that a recent update broke it.

Did anything in the provisioning change? No way of knowing since you aren’t allowed access.

Any other things that can cause such a sudden change?

99

I just noticed that my panel has updated firmware to: spanos2/r202302/03

I have not yet noted any change to functionality.

1 Like
100

I just checked on my app and I am now on spanos2/r202249/07, so I updated but not to the 2023 version you got on…

And now I am disconnected from the Span Panel again through HA, and I am now heading up North for the weekend. These door closing authentications at random times do not work, I just reverified earlier this week and now I am locked out I suspect because of this software update. I have another weekend with corrupted data and my automations break when it can’t see my solar input/net output.

We already have a cloud-based app, it would be relatively straight forward to verify local access through that at minimum as a 2FA. Still should have local only authentication and persistent access, but something more than random lockouts are necessary.

@SPAN-Shoop I’m sorry if this feels like a rant, but so far we have heard that Span wants to be in the community and work with us, but so far these are just words with what is essentially still an expensive black box on the wall. No movement on local access, no beta testing offered, not even any app updates with more granular data.

What are the updates we can look forward to?