Replies: 2 comments 12 replies
-
|
have you connected a inverter or something else? |
Beta Was this translation helpful? Give feedback.
-
|
Alright, the tests are in, and you are correct... a pure DC load, in this case 3 x 12 V light globes and one 10 Ohm 100W resistor in series produced this result: The 0 A is when the battery, or rather the BMS negative is removed from the inverter-charger. I didn't run this longer, as the result is clear; silly me blew the 4th bulb, acting as a fuse for the others, as they did not like to be connected to 54 V with 4 x 12 V in series; hence, the 100 W resistor I could find and add; however, it got hot darn quick. :D As a result some math needs to be employed to fix this wobble business. Thank you for your feedback. BTW: what is considered a quality BMS, ideally provinding a smooth current value? |
Beta Was this translation helpful? Give feedback.
-
|
ah great, thank you.
|
Beta Was this translation helpful? Give feedback.
-
|
Hi, I see an issue with doing some kind of averaging in the control/smart system. In my case I have the Wemos set to a reporting interval of 30 seconds, of course I could average over those data points but having more point gives a "better" average. I believe you mentioned before (in one of the bug tickets) that your code get's data a lot more often than that from the BMS. |
Beta Was this translation helpful? Give feedback.
-
|
The 1,800 data points I have in my O.P. have a 1 second interval. I have to median 15 values to arrive at a reasonable result. Looking at this graph again, I see more clearly a variation in 'wobble' frequency, which must be related to the 'dirty' consumers I have. I can say for sure that I have a lot of them; many little switching power supplies for controllers,, a fridge/freezer with a scroller, computers, displays, and the list goes on. :) |
Beta Was this translation helpful? Give feedback.
-
its a polling function, the code loops the workers, every time the worker gets firered (about 400-900ms related to the rest what todo) it cn get new data |
Beta Was this translation helpful? Give feedback.
-
|
I have a 60A Daly BMS with a small 15S 48V battery and a HM300 inverter connected. Here an example how the power values reported as "Pack Power" from the BMS look like versus the DC power reported by the HM300 via OpenDTU.
|
Beta Was this translation helpful? Give feedback.
-
|
How can anyone use this Daly output to control something?! It's useless, even if the Daly data is absolutely accurate. |
Beta Was this translation helpful? Give feedback.
-
A few days ago I started writing some code to connect via CANbus to a Daly Smart BMS 250 A for my 400 Ah 48 V battery.
Not only does it have a weird CAN protocol, but the data it publishes/provides is all over the shop.
The following listing shows current undulating almost like a sine curve, yet the battery is being discharged at a constant load of 700 W. It event reports charging when this is definitely not the case.
What do other people experience? Seeing this project, I am inclined to ditch my project and adopt this one. However, if this incorrect current reporting is the issue for others, I may well replace this BMS with another brand. BTW: the Bluetooth app shows the same variations, however, its update cycle is slower, hence, the variation is somewhat less strong, compared to the recording below.
Any hints appreciated, thanks.
The following graph shows current over 20 minutes and 1850 data points; weird indeed; actually it is unusable.
Beta Was this translation helpful? Give feedback.
All reactions