Strange issue with MPT-7210A

thanar

thanar

Member
Joined
Feb 12, 2018
Messages
155
I am currently using an MPT-7210A MPPT solar charge controller / boost converter to charge my 36V 2s10pe-bike battery from a Lead acid 12V battery, which I keep under float voltage through the grid, since my solar setup is a bit off right now. I have set the Vmp (input) on the MPT unit at 11V to protect the 12V battery (the MPT unit wouldn't work with an input below 10V anyway)and the output voltage at 41V.These units are actually boost converters, however I have not been able to find a way you can "switch" it from solar charge controller mode to boost converter mode and back.

I've been having issues with this unit, even when it was connected to my solar panels, with its output voltage reading. Hare is my setup:

image_rhyxgl.jpg

As we can see on the image, the battery voltage reading on the MPT unit is way above the one reported by the voltmeter I have connected to the brass bars. The (disconnected)battery voltage is 35.8V. That's also the reading the MPT unit gives before it starts ramping up the current. It stopped to 1.7A because the 12V battery dropped to 11V.There is no voltage drop on the cables leading from the MPT to the brass bars, since measuring the output of the MPT unit is reading the exact same 36.2V as everywhere else.

Here's where things gets a lot more interesting:

image_ivkmer.jpg

I need to run the charging port outside the shed to be able to charge the e-bike battery without removing it, soI am switching the charge cable with a longer1.5m 2.5mmpair. As soon as the MPT unit ramps up the current, the battery voltage reading on the unit raises extremely fast and reports charging as complete. Voltage readings everywhere is 35.8V -the currentbattery resting voltage- even on the MPT unit's output. What's even more strange, moving the cables around changes the reading on the MPT unit, while keeping the pair tightly together does allow for some current to go through. I have tried a different pair of chargecables of similar length with same results.

What do you people think is happening here? Could it be that I am running the unit on such a low input voltage and it fails on a reference voltage for the output? I am suspecting high frequencies boost converters use to convert DC to DC is the culprit here. Any other ideas? Do you know if Mike fromMikes DIY Tesla Powerwallhas an account on this here forums? Seems like he's done some research on these units.

Any help would be appreciated.
 
From My experience since they voltage is being tested on the same cables as the Current is going through there is volage drop due to the resistance of the wire. I have that same unit and for solar I find it likes the Vin to be set about 1-2V lower than the panel suggests. I get more overall current out of the panels that way. just food for thought
 
This doesn't seem to be a voltage drop due to current going through issue, since I am measuring the voltage at the unit's out port (before any wires) and comes out exactly the same as the voltmeter attached on the brass bars.
 
The Vmp setting of 11V will not protect the battery because it will continue to pull power until the voltage is not high enough for the unit to work. There is no "switch" because it is effectively just a boost controller which alters the input voltage level down if the power supply source is unable to deliver high enough amps.

Running this off a lead-acid battery will destroy the battery if the input float charge to your lead-acid shuts down as the unit will just run it to 10V or lower if left long enough.

While your solar is offline to charge your bike use the controller directly off the float charger and reduce the current limit.

As for the issue you are seeing you may have a "dry joint" on the controller board terminals as this seems to be one option that would cause the issue you are seeing with a high resistance. Try just moring one cable at a time near the controler to identify if it is the input or output side, then just open it up and re-heat the solder joins for that terminal and maybe apply extra solder if the contacts need it.
 
Thanx for the reply, completelycharged!

Actually, I've found that setting a Vmp of 11V on this unit doesn't allow input voltage to drop below that said value. As soon as it drops below 11V, amperage on the output drops continuously until the voltage on the input raises to above 11V. That said, I don't know if there's a minimum it can drop down to, and in case the float charger would fail then it could kill the 12V battery. Luckily, I almost never drain more than half of my 5Ah e-bike battery, so I guess the 60Ah 12V source would handle the CC phase of charging even without the float charger active.

On the "dry joint" idea, indeed, I haven't thought of it... however, would that justify the fact that the unit works more are expected (albeit not perfectly) using a shorter cable on the brass bars?
 
The volt drop on the cable is not the issue as your image of 1.7A flow indicates the drop is around 100mV to 150mV if that on the short cable to the meter.

With the 300mA reading and 41V this to me says bad joint somewhere.

If you have a multimeter, test the voltage actoss the controller and also across the bars. If you don't have a multimeter, get a cheap one as it will help a lot as it is one of the first tools to have with batteries, along with a screwdriver.
 
Of course I have a multimeter. On the second image, right at the output of the MPT unit, the voltage reading on the multimeter is 35.8V, on par with the voltmeter on the brass bars, regardless of what the MPT unit says. Even on the first picture (using the short cable to the battery), the multimeter reading on the MPT output is still the same as the voltmeter. That's my original question, what could be wrong for the MPT unit to be so off...

I'll get a couple more photos when I get the e-bike battery low enough with the multimeter plugged on the MPT output.
 
Joint issue inside the MPPT, more likely the terminals.
 
The original terminals on the MPT7210 is pretty bad. I switched mine out to XT60
 
OK, people... this thing is driving me nuts! I soldered the output cables (did the input too, just to be sure) directly on the board, and this behavior did not change:

- When using a short cable (10cm) to the battery, power goes all the way up to 100W.
- When using a long cable (1m) to the battery (it's a 11AWG which shouldn't have voltage drop), power falls to max. 20W.
- The longer the cable, the bigger the difference between the actual measured voltage at the output and the voltage reported by the unit!

Something must be broken inside, in the boost circuitry, where it's supposed to regulate the conversion to keep the voltage / amperage below the set maximums. I have messaged BigClive if he would be interested to send the unit over to make a video, so as I learn something from this whole ordeal!
 
Looks like the culprit has been found:

This issue is due to induction in the output wire. The longer the wire, the bigger the offset between displayed and actual output voltage. Seems like the two 470?F output capacitors are failing, no visual damage though. I managed to get a much better result by adding just a 4.7?F capacitor directly to the output.

Many thanx to Daniel Raynold for the help.
Here's the link to the conversation: https://www.facebook.com/groups/diypowerwalls/permalink/710030859343478/
 
You must log in or register to reply here.
Back
Top