To visit MidNite Solar click this link www.midnitesolar.com
Quote from: boB on March 06, 2024, 05:58:36 PMI did not make it reset the amp-hours remaining when doing a VMM. The battery should be around the same SOC% that it was before updating the firmware.I thought I saw the SOC at 100% today but now it is 0% . I wonder if I was looking at a different Classic today or if it changed ? But now I am sure it is showing 0% SOC. I did not see the battery choice either when it started up after firmware update and it kept my settings. I did the VMM just to make sure since you had mentioned it - and then I picked lithium and put in all my settings again.
Poor Wizbandit is seeing 0% SOC but he did not see a battery type choice when he started it up, he says.
At first I thought this might have had something to do with the LITHIUM battery capacity AH lock-out with battery temperature... I then did a VMM and changed to lithium (I am using flooded right now) and watched the SOC% drop down to 99% then 98% and then back up to 99% then 100% while in Absorb and Bulk MPPT so I guess that's not it.
I thought for a moment that it counted DOWN but not UP but I don't think that is the case now. This was only doing a few amps so it took a while to go up and down is all along with the efficiency setting.
boB
Quote from: boB on March 06, 2024, 06:08:50 PMSo, could the power from the Rosie have been off for 10 seconds and then come back on you think ?
I seem to remember hearing about a 3 second and 10 second off time but I don't remember what the reason is. The LOGS should show WHY this happened.
I'm not sure how to clear those event logs ?
Can you read off the Rosie FW revision number ?
boB
Quote from: Muskoka on March 06, 2024, 12:05:47 PMMines showing 64%. I'm assuming that it didn't reset after flashing and doing a VMM, cause I was in the 70% range yesterday. I flashed this morning. I've been using my computer today with no sun, so 64% seems reasonable, if I was in the 70's yesterday.My SOC is at 100% now as they are full . I will see if it tracks down okay tonight. I just updated the Hawkes Bay the other day too and it is on the same shunt so will see how they compare.
I would think after flashing and a VMM that it would reset the SOC, but it doesn't look like it did.
Quote from: Wizbandit on March 06, 2024, 06:57:14 AMI did install a MQTT Broker or "server" as it is now called on my development COMBOX, it is running but I have not tested it. It claims to be a "low bandwidth" service but that does not mean it does not need a lot of processor to operate. As it just serves to distribute "published data" to "subscribers" with no data being stored on the actual server I would need to publish to it first. With over 400 different data points I don't want to publish every one as I suspect that would load down the system to a halt.I think there is a difference between publishing the data and having a broker ?
The only way I can see to accomplish this is to have a configuration page for each device, Rosie Inverter, Barcelona, Hawkes bay, Classic, KID, DIY3024, DIY3548, DIY5048, 10K MNAIO-ONE. Each page would allow the user to turn "on" MQTT and select using "check boxes" the data they would like published to the broker, select too many and the system crawls to a halt.
I still need to work on "safe shutdown" hardware, with the increase in support calls I have had little time to work on this issue. Once it is done I could start on building the configuration pages and testing. You would need to have some way to have your external monitor systems access the broker as I won't have time to work on outside devices.
The MNS CANBUS is complicated and not real easy to understand. Where as you read a 16 bit MODBUS register and get for example "Battery Voltage" our CANBUS is 64 bits and each of these packets can hold from 1 to 8 bytes of data. For example register 0x0A0 is the "battery register", it has 2 bytes for voltage, 2 bytes for current and 4 bytes for power (watts) stored in "big-endian" whereas MODBUS uses "little-endian". I know this won't mean much to many of you but the bytes need to be combined to get the one data value.
MODBUS data is requested per 16 bit register whereas CANBUS data is "streamed" and you need to monitor for the data you want which requires multiple tasks running concurrent in the background. That being said there are configuration registers that must be requested and they return up to 64 bits of data that needs to be unpacked and combined.
It took me over a year to become really good at this...FYI
Quote from: onanparts on March 05, 2024, 02:21:18 PMI updated with the newest one released yesterday. If one wants I can send it back to factory settings and redo the update this time showing in real time the time it took.Quote from: Weldman on March 05, 2024, 11:00:48 AMSo that's why I have been fighting the last Rosie all this time, internal temperatures below 30°F and the bug wouldn't let it turn back on. Thought I was losing my mind there for a minute each time I did a update and it refused to come on.
Updated both this morning and Rosie came back online within 10 minutes at 18°F
What FW did you update to on Rosie? The latest test FW, 24.29.2 was working down to I understand -35C in Midnites environmental chamber, and Rosie boots up within a few seconds. I tried it a few days ago. At 43F Rosie is up and running within 2-4 seconds. Prior to the 24.29.2 FW Rosie would take several minutes to wake up at even 49F.