Over the last couple of days, I noticed that my Powerwalls discharged beyond my threshold in Self-Powered mode, but I didn't think much of it. Today, in preparation for several days of potential adverse weather, I bumped up the threshold to 100% (Essentially Backup-Only mode), but once the sun went down, the Powerwalls began to discharge. I tried switching to Backup-Only mode, but nothing changed - the batteries continue to discharge. Has anyone else seen this? This seems like it must be a bug rather than a unique problem to me.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Powerwalls Discharging in Backup-only Mode
- Thread starter DHevey
- Start date
I have my four (4) powerwalls set to Backup-only 100% mode.
The 4 powerwalls currently consume 0.9 kWhs per day as maintenance. (0.225 kWhs per powerwall)
They recharge to full every other day provided there is enough PV production available (they don't recharge every day).
There are several threads that discuss whether Backup-only 100% is the same as Self-Powered 100%.
I don't know if they are the same because my walls stay as Backup-only 100% all the time so I can't observe how 100% Self-Powered would perform.
My Powerwalls recharge when they reach 96% charge every other day.
The source of my data is the Powerwall iPhone app.
The 4 powerwalls currently consume 0.9 kWhs per day as maintenance. (0.225 kWhs per powerwall)
They recharge to full every other day provided there is enough PV production available (they don't recharge every day).
There are several threads that discuss whether Backup-only 100% is the same as Self-Powered 100%.
I don't know if they are the same because my walls stay as Backup-only 100% all the time so I can't observe how 100% Self-Powered would perform.
My Powerwalls recharge when they reach 96% charge every other day.
The source of my data is the Powerwall iPhone app.
Wow....... I have never seen behaviour like that (84% discharge) since I've had my powerwalls on Backup-only 100% since November 1st, 2019. They only drop to 96% after two days and then recharge. My Powerwalls are on Gateway version 1.43.3
Your numbers say they are consuming over 8.5 kWhs PER DAY in backup-only maintenance.
That over 263 kWhs per month ... something is wrong.
No, what I'm saying is that the Powerwalls are discharging into the protected loads no matter what I have the configuration set to. It says it's in backup-only mode but continues to power the protected loads. It seems to be stuck in a previous setting, which would have been 30% reserve in self-powered mode, even though it says it's not. It's as if the server is not telling the gateway what to do but the app doesn't know it. I hope that's more clear.
Yeah, same here. Backup mode, PWs never go below 97 or so. I would either restart everything or call Powerwall support. No reason to be cycling the batteries that much for no reason.
Yes, I was just installed on Monday (jan 6th) and have changed my reserve threshhold each day as I play with it, and it takes effect in about 2 minutes.
Yes, I've been working on some automation, so been making lots of config changes, and it usually takes affect within a few minutes. I've been changing between TBC modes & Self-Powered, and changing reserve levels, and all are usually initiated with-in a few minutes, although when changing TBC/Self-Powered modes things bounce around for up to 15-20 minutes it seems until fully setting in in terms of the flows bouncing around. And in my case sometimes this takes longer to be reflected in the Tesla app vs the local gateway web server page.
My Powerwall version is v1.43.3. Other versions may behave differently.
Everyone should have 1.43.3 now or soon as they roll it out. Wish there was a way to see you have gotten an upgrade besides checking the app. I thought I knew what I had (see sig) but I then saw I had the new one and didn’t even notice. As usual the release notes show nothing useful but it may actually have functional updates. Recently I’ve noticed that one of the recent ones actually reconnected to my wifi afterwards where it didn’t used to.
I reported the problem today and it has been elevated to tier 2. Apparently, it's a know issue, so I'm not alone. I'll report back on the resolution.
charlesj
Active Member
Thanks for reporting. Will keep it in mind if this will be the case with mine if and when it gets installed.
Changing the reserve using the TEG interface involves two steps - first set the reserve number, and then tell the gateway that the changes are complete.
Since 1.43.3 any change has rarely been honoured after the first complete operation. Instead, it is usually necessary to issue the operation again sometime later. After a complete operation, there is a pause of over a minute before the Powerwall will apply a change, even if it's going to. So far, I've been applying the second complete operation after that, and that usually does the trick.
The possible relevance of that to the situation with the Ap is that changes performed by the Ap may suffer the same effect, except that it is not possible to repeat the complete operation.
Quite how Tesla have created this situation is something one can only speculate about (though the word "incompetence" comes to mind). One would have thought that they'd respond by downgrading to the previous version, but they know they have users over a barrel, and may not care that much.
Since 1.43.3 any change has rarely been honoured after the first complete operation. Instead, it is usually necessary to issue the operation again sometime later. After a complete operation, there is a pause of over a minute before the Powerwall will apply a change, even if it's going to. So far, I've been applying the second complete operation after that, and that usually does the trick.
The possible relevance of that to the situation with the Ap is that changes performed by the Ap may suffer the same effect, except that it is not possible to repeat the complete operation.
Quite how Tesla have created this situation is something one can only speculate about (though the word "incompetence" comes to mind). One would have thought that they'd respond by downgrading to the previous version, but they know they have users over a barrel, and may not care that much.
It is pretty close to immediate. I look at a simple UI I use that monitors the status - it blanks out after a few seconds as the gateway reinitializes. It then starts out with no discharging, but starts discharging after 10 seconds or so. There is a ramp-up until it is fully offsetting the house load.
The total time from changing the setting until it's fully discharging is about a minute, but that hasn't changed from previous releases.
The total time from changing the setting until it's fully discharging is about a minute, but that hasn't changed from previous releases.
I don't see any ramp on with mine.
In the data, at the point where it decides it's going to use the battery, at 23:21:14, it actually overshoorts, and sends 362W to the grid.
It seems odd that ostensibly the same hardware with the same firmware should behave differently.
The small solar current must be spurious (stray pickup by the current transformer, perhaps), given that it's night time, cloudy, and no moon. That probably explains why it changes sign.
2020/01/25 23:20:58 M Site Solar Battery Load Charge Status
2020/01/25 23:20:58 D 2540.80 2.91 0.00 2539.12 83.4211893979 187
2020/01/25 23:21:00 D 2538.78 1.98 0.00 2538.95 83.4211893979 187
2020/01/25 23:21:02 D 2542.73 1.36 0.00 2545.24 83.4211893979 187
2020/01/25 23:21:04 D 2547.60 1.93 10.00 2552.05 83.4211893979 187
2020/01/25 23:21:06 D 2546.42 1.96 10.00 2548.28 83.4211893979 187
2020/01/25 23:21:08 D 2549.51 2.23 -20.00 2547.42 83.4211893979 187
2020/01/25 23:21:10 D 2561.13 2.63 -10.00 2557.36 83.4211893979 187
2020/01/25 23:21:12 D 2558.33 2.69 10.00 2562.32 83.4211893979 187
2020/01/25 23:21:14 D -362.21 -1.87 3010.00 2944.03 83.4211893979 187
2020/01/25 23:21:16 D 28.24 -2.25 2570.00 2591.32 83.4211893979 187
2020/01/25 23:21:18 M Site Solar Battery Load Charge Status
2020/01/25 23:21:18 D 30.03 -1.98 2580.00 2602.04 83.4211893979 187
2020/01/25 23:21:20 D 17.82 -2.96 2580.00 2605.24 83.4211893979 187
2020/01/25 23:21:22 D 24.50 -2.09 2580.00 2596.77 83.4211893979 187
2020/01/25 23:21:24 D 17.73 -2.61 2590.00 2604.22 83.4211893979 187
2020/01/25 23:21:26 D 18.55 -2.61 2560.00 2591.58 83.4211893979 187
2020/01/25 23:21:28 D 10.79 -3.20 2560.00 2596.90 83.3989160294 187
2020/01/25 23:21:30 D 16.73 -2.93 2580.00 2592.88 83.3989160294 187
2020/01/25 23:21:33 D 1.26 -3.29 2610.00 2599.36 83.3989160294 187
2020/01/25 23:21:35 D 4.71 -3.23 2600.00 2610.46 83.3989160294 187
2020/01/25 23:21:37 D 21.67 -2.09 2580.00 2594.71 83.3543692924 187
In the data, at the point where it decides it's going to use the battery, at 23:21:14, it actually overshoorts, and sends 362W to the grid.
It seems odd that ostensibly the same hardware with the same firmware should behave differently.
The small solar current must be spurious (stray pickup by the current transformer, perhaps), given that it's night time, cloudy, and no moon. That probably explains why it changes sign.
2020/01/25 23:20:58 M Site Solar Battery Load Charge Status
2020/01/25 23:20:58 D 2540.80 2.91 0.00 2539.12 83.4211893979 187
2020/01/25 23:21:00 D 2538.78 1.98 0.00 2538.95 83.4211893979 187
2020/01/25 23:21:02 D 2542.73 1.36 0.00 2545.24 83.4211893979 187
2020/01/25 23:21:04 D 2547.60 1.93 10.00 2552.05 83.4211893979 187
2020/01/25 23:21:06 D 2546.42 1.96 10.00 2548.28 83.4211893979 187
2020/01/25 23:21:08 D 2549.51 2.23 -20.00 2547.42 83.4211893979 187
2020/01/25 23:21:10 D 2561.13 2.63 -10.00 2557.36 83.4211893979 187
2020/01/25 23:21:12 D 2558.33 2.69 10.00 2562.32 83.4211893979 187
2020/01/25 23:21:14 D -362.21 -1.87 3010.00 2944.03 83.4211893979 187
2020/01/25 23:21:16 D 28.24 -2.25 2570.00 2591.32 83.4211893979 187
2020/01/25 23:21:18 M Site Solar Battery Load Charge Status
2020/01/25 23:21:18 D 30.03 -1.98 2580.00 2602.04 83.4211893979 187
2020/01/25 23:21:20 D 17.82 -2.96 2580.00 2605.24 83.4211893979 187
2020/01/25 23:21:22 D 24.50 -2.09 2580.00 2596.77 83.4211893979 187
2020/01/25 23:21:24 D 17.73 -2.61 2590.00 2604.22 83.4211893979 187
2020/01/25 23:21:26 D 18.55 -2.61 2560.00 2591.58 83.4211893979 187
2020/01/25 23:21:28 D 10.79 -3.20 2560.00 2596.90 83.3989160294 187
2020/01/25 23:21:30 D 16.73 -2.93 2580.00 2592.88 83.3989160294 187
2020/01/25 23:21:33 D 1.26 -3.29 2610.00 2599.36 83.3989160294 187
2020/01/25 23:21:35 D 4.71 -3.23 2600.00 2610.46 83.3989160294 187
2020/01/25 23:21:37 D 21.67 -2.09 2580.00 2594.71 83.3543692924 187
Similar threads
