Hey buddy! (Izzymonster i assume? Me=Vommbat;)
Thanks for taking the time to reply. Yes, i was planning to do this but what you say makes sense!! So i will hook purple up to the 3rd pin . Is it ok if i post your reply on the discord. I was literally advised it was fine like i planned it;)
Heya. Yep izzymonster. Sorry I don’t use discord. Since you are just getting to it, you should know there is a small mistake in the first installation diagram above. You’ll see the purple wire is going to the pin closest to the balance connector, which is incorrect. It is correct in the next diagram, below, where the purple wire goes to the pin closest to the xt connectors. That threw me off but I just looked at the stock harness.
Wait…what? Which diagram? I installed it exactly like the diagram by Haverr(?). Purple from pin closest to charge XT60 to port3. No luck. Charger blinks red/green couple times then goes steady green but pack is not charged.
I’m also having trouble. The Haverr diagram should be right as far as I understand it, pin closest to XT connector. Here is my implementation with a +XR 4206 BMS.
OWIE is working and reporting correctly. What does not seem to be working correctly is the charger. My understanding of Li ion chargers is that as the the battery approaches the charge voltage, the output current is reduced. This allows the battery to get closer and closer to the charge voltage without going over 4.2V per cell. In my case the charger is not reducing output so it cannot get to max charge voltage without the BMS shutting off due to a cell reaching 4.2.
Here’s how it goes. Charging normally
Almost at max charge voltage but not reducing current:
Max charge voltage. No reduction in current. BMS is about to cut out at 4.2V
BMS cuts off. No charge current.
After this the battery voltage will settle for a bit and then BMS will reengage. It will charge for a short time before the BMS disengages again for the same reason. It will repeat this cycle a certain number of times before the BMS faults and needs to be power cycled.
I feel like there is something I am not understanding about how the charge interacts with the BMS. I will try a DC-DC charger I have that I know profiles the current correctly. It also allows me to adjust the output current and voltage. I will let ya’ll know what I find. I appreciate any other insights.
1 Like
Disclaimer: I am not an expert. Consider that what I say might be wrong. Make your own educated decisions. Be safe.
After some head scratching I figured this out. The short of it is that there is another electrical element in the +XR controller that is in series with the BMS. This element causes a small voltage drop before the BMS so that the BMS receives a voltage slightly less than the charger is outputting. If you remove this element, the voltage delivered by the charger is now higher than the BMS expects. The BMS will cut out before the charger has a chance to reach maximum charge voltage where it begins to reduce output current and enter the balancing phase.
Now the long version. You’ll see from my previous post that the BMS cut out just above 62.7V. This is what the BMS wants to charge to. The +XR charger on the other hand puts out about 63.3V. So where does the 0.6V go? To figure that out we can look at this trace diagram of the +XR controller. We are particularly interested in this section:
Here we can see from batt+ we head up into a black rectangular element then to a solder joint which is the + charger connector on the other side. Looking up the text on the element, 10H10, we find this. It is a vishay 100V rectifier/diode. This diode is rated for 100V and 10A with a forward voltage (or expected voltage drop across the element) of .88V. Taking a closer look at the documentation we can see in Figure 2 a forward current of 3A (our charge current) will result is a power loss of about 2W. 2W divided by 3A is 0.67V which is just about the difference between what our BMS wants and what our charger puts out.
To solve this I am going to add a diode to my charge adapter. I am trying to match the rating of the 10H10 and the forward voltage so that I have a similar drop. I have found this and this. Although the forward voltage is slightly less than what I want. If it is giving me trouble because the voltage is too high I plan to trim with ~30mOhm resistor to reduce .1V at 3A.
I feel like this may have been discovered before. I could have sworn I’ve seen a video of surfdado installing a diode but the reason didn’t click with me at the time. Also you should know that this could be different with different models of the BMS. From pictures I can see on the pint BMS there is most likely a diode on that instead of the controller. The model of board I pulled from is 4206 +XR.
1 Like
I’ve looked over the wiring and did the flash for the esp. Got the esp to establish a wifi and show the balance firmware but not getting any data from the BMS. I’ve tried a few things but with no luck.
Screen shows a connection time. with no amps no cell data.
I have a FM 4209 BMS and following this guide.
You can double check your work with the github if you haven’t already. I’ve seen some wiring mistakes in different guides but the git and jwffm chip stuff is all correct. You might also check conductivity on your purple wire and white wire from the esp to the BMS board to see that your connectors are making good contact.
1 Like
To make a long story short. I have an Error 16 that created the first task for buying a new BMS. After that one too fried, the decision to go to a Vesc-operated build became the deepest concern, which has been so epic to ride.
The second-bought BMS was not sending any cell data, so the wiring was correct. I checked this using the original FM BMS and got some data, but it had the same lousy cell data that caused me to purchase the second BMS.
Have a good read from the esp to the owie onto phone or web browser. Luckily I have an old XR lying around going to use that one till a good BMS is available that is easy to use and reliable.
So I went ahead and tried adding the diode. I was not able to get it to work but I learned a lot. I am going to move on to a smarter BMS so I have better control and more functionality. But I will unload what I learned in case someone wants to take it further.
I believe there are differences between generations of FM BMS. Earlier BMS like the 4206, and maybe the 4208, and 4209 +XR have more integration with the controller than the later 4210, 4211, 4212 +XR and pint BMS. This is because they moved the charge port from the controller to BMS for the pint model. To reduce the number of inventory parts FM probably chose to make the same change to the XR BMS that they executed for the pint.
Here is a 4206
Here is a newer model (unknown).
You will notice that there are some missing elements on the 4206 around the center of the board where the pint charger would connect to the board.
Without these elements to reduce the charge voltage, the voltage delivered to the BMS is higher than the BMS expects. This makes it overshoot. Once a cell hits 4.2V the BMS will shut off current flow between the XT connectors. What is supposed to happen is as it approaches the charge voltage, the current is reduced from 3.1A to .2A, but since the charge voltage is too high, current never reduces. This reduction in current is what allows the BMS to balance so it will not balance.
I was on the right track but I don’t think my part selection was exactly right. What you need is a simple circuit that reduces the charge voltage from 63.3 to about 62.0 at 3.1A. From there, as the battery approaches the charge voltage the circuit should reduce the voltage from 63.3 to about 62.7-62.9 at .2A. This will allow for a slower approach to max voltage and enable balancing.
3 Likes
Oh shit I’m on a 4206. I’ve got a charge/discharge setup right now that’s been causing me a few small issues, but I’m about to switch over to charge only. Did you ever discover anything else about this?
I just did a 4206 last week following the original guide without any issues.
1 Like
If you want to added a diode to reduce the incoming voltage slightly then you can do that as some have reported issues, but I haven’t seen this problem. Could also be related to some chargers being calibrated slightly differently I suppose.
Are your cells balancing that way? It sounds like the way I need to wire everything up, but if I’m in there I want to only have to open the battery box a single time if I have to.
Did you use the diode to lower the incoming power or did you wire things up without it?
I’ve done it with and without the diode and it works both ways for the ones I’ve done. I just always make sure to put a 10a fuse on the positive coming from pin 3 of the XLR.
1 Like
I did not do any more work on this after trying my 4206 and 4209 with the same issue. I charges but not to 100% because of the high voltage cut off. I just rock an external, charge only LLT BMS now.
I think I might see the same behaviour you’re talking about where the charger cuts between green/red/off, but mine all eventually charge all cells all the way to full and balanced between 4.20-4.23v
I think it still is applying full 3amps possibly, but hard to tell as it is flashing. I may consider adding diodes going forward, but I have not experienced any issues other than the odd blinking of the charger and possibly supplying more than desired current near the top end.
Another solution could be to tune the charger, but I’m not sure if that is easy to do on a FM charger… I might look into that instead of adding diodes.
That makes sense. Now that you mention it, it is really a charger issue not a BMS issue because the BMS is doing what it is supposed to. The FM charger output is just a bit higher than we need for these older models.
I have switched over to using the following combo for charging. I have been using the DC-DC module for years for mobile charging. It is very reliable and simple to use. You can just set whatever voltage and amperage you want. It will automatically lower the amps as it approaches the voltage. Power supply, DC-DC
What pin should be used to jump the purple wire on the 16pin connector? Also pins for the charge wire (blue, black, white, green)? Or does this setup need custom harness with cable gland?
You talking about the wire that goes from the 3rd xlr to the 16 pin? I believe it’s the farthest to the left on top. Probably should double check, I know for the momentary setup that’s the one I got hooked up, that tk talked about using so that the momentary setup isn’t (highly recommended) tbh that’s is a great setup other than you gotta take the max to 25/26 &-30 in order to get the thing not to shit off when sitting in curbs. But once that was figured out it rode like a dream. We’re going ti hope it hooks up tonight and powers on otherwise … I can’t keep riding this damn fm xr controller it’s so dirty and feels like a barbarian board that Neanderthal used to run from dinosaurs during the ice age.
1 Like
FYI on the OWIE github page there is a link that starts a tool to flash the esp board with the click of a button. makes it super simple, no need for the long tutorial.