Victron Question

[WorldwideDave]

I always thought my temp sensor on the battery terminal read a little hot. I put a Ruuvi tag on top of one of my batteries in the middle of the pack and feel the temps ther are more accurate. I only casually checked the accuracy with an instant read kitchen thermometer.

photo please of your victron temp sensor attached mid-pack to your batteries?

 

[WorldwideDave]

My battery warming system would like to disagree with you. I have two temperature sensors wired directly to the Cerbo GX. Where your statement may be correct is that the solar charge controllers and the Quattro are not using these two sensors. They are using the temperature sensor from the Lynx Shunt. There may be a way to tell the devices to use the Cerbo GX temperature sensors, but I haven't looked into that.

Blame ChatGPT for that response - was trying to not go through 700 pages of manuals for all devices. Happy cerbo GX has the 4 spots for the sensors - sensors that are different from the one I have attached to the shunt, by the way, as they should be.

 

[WorldwideDave]

In all the testing you did on advice from us, I don’t think you ever mentioned the heat came from victron voltage and temp sensor.

It is just the temp sensor. The voltage sensor wire is connected to the bus bar, not the battery itself. I've shown photos.

I have mentioned the heat came from the positive terminal of battery, that had a screw to mount conductors, the correct victron temp sensor attached to the screw, followed by the postiive conductor wire's lug at the battery.

We discussed and showed photos of the install order, the temp sensor location, how it was attached, the fact it was connected to a 500A shunt, etc. I didn't have cerbo GX at the time at all, and certainly didn't know a different temp sensor could be bought and attached to the cerbo GX...for that I was today years old, and just did that.

One way you could try and prove this is create a jumper/stand off of a couple ring terminals and say 10awg. Have the voltage/temp sensor stood off the actual battery terminal. There is no current flow needed through the sensor so if you removed it from the current path (no current no heat rise right?) and still notice a temp rise then you’d notice the issue.

I want to understand this better. You are saying to build some sort of mechanism to hover the temp sensor above the positive battery post, and not attach it in any way to the battery, despite the victron manual telling me to, correct?


EDIT: This would not work as the cabling that comes with the shunt temp sensor (not the cerbo/multiplus/quattro temp sensor) is required to be on the positive terminal in order to power the shunt if using a temp sensor.

The other thing to keep in mind is victron temp sensor (and most victron gear) is design carry over of FLA world where the battery/cells would dramatically heat up under charge discharge and that temp greatly factors into how you should be charging the battery.

Yes thank you

 

[WorldwideDave]

In my prior camper's LiFePO4 batteries, I had temp sensors from the BMS, the BMV-712 shunt and the Cerbo GX. All three were within 1°F. The shunt sensor was on the battery post. The other two were taped down to the top of the center cells.

Maybe I got lucky with such a small variance.

So this is where I am confused. You also say - as does the victron manual - that the shunt sensor was on the battery post. I assume positive. This is what the manual says as well (below).

Can you please post a photo of them taped down to the top of the center cells? Those were from the cerbo GX I presume? Wish there was a 'how to do temp sensors with victron' guide - think I know what I'll work on this weekend

 

[WorldwideDave]

In my prior camper's LiFePO4 batteries, I had temp sensors from the BMS, the BMV-712 shunt and the Cerbo GX. All three were within 1°F. The shunt sensor was on the battery post. The other two were taped down to the top of the center cells.

Maybe I got lucky with such a small variance.

Oops - thanks again for agreeing that the temp sensor should be on the positive post when connected to IP22 500A shunt:

 

[WorldwideDave]

It’s a voltage and temp sensor.

I provides the power for the shunt and it is able to send temp values back to the shunt as well.

If you think the sensor is the actual element creating heat you can wire it to just a “jumper” of say two ring lugs and some conductor to stand it off. See if the heat flows the sensor or stays at the battery terminal.

My thought is that due to its mass (it’s chunky unit) it’s actually acting like a heat sink, and drawing heat away from that bad connector/connection but not the cause of the heat, but to you, it seems like the source of the heat.

Yes yes yes - and also, thanks for the reminder that it HAS to connect to the positive terminal on the battery because that is the source of positive for the shunt itself! That's right!

 

[WorldwideDave]

But remember there’s a shunt temp sensor and a multiplus sensor, two different beasts.

Shunt is a voltage and temp sensor, multiplus is just a temp sensor.

Shunt sensor must be on battery positive post to send power to the shunt.

The multiplus sensor needs no power and is commonly connected to the ground

I don’t think you can plug a shunt sensor into a cerbo, actually it might let the smoke out.

Thank you and yes - I have a Quattro I am not using yet 48V so had the very long wire temp sensor that came with that does NOT need to be attached to the IP22 500A shunt. I appreciate the reminder. I'm using the appropriate temp sensor for the one that connects to shunt, and it is the positive battery source, and it is acting like a heat sink, and I'm surprised no one else has ever complained about this issue here or on the victron support portal. I am using the quattro temp sensor connected to my cerbo right now, and it is showing 1 degree difference than the shunt one, but it is not installed in the same spot.

 

[outboard1196]

I've had to re-read this 4-5 times to understand what you are saying. In your situation, it was cold out, but not freezing. You had some alarm or rule set to do something likely when it hit 0 celsius or similar. The temp sensor you had wired on your shunt indicated it was very cold out - enough to trigger something. But the other temp sensors you had - one on the BMS, and another 'air sensor' that may have been a Ruuvi or something - both were reporting that No, it was not that cold outside. You then are suggesting that the temp sensor itself, which was attached to the battery post at the time, was reporting that it was too cold because the battery conductors were keeping warmth away from the victron temp sensor. Is my summary correct?

Yes.
The sensor going to the cerbo gx is hanging in the air next to the battery.
Between the sensor in the air and the BMS sensor I get a good idea of the battery and battery box temperature.

 

[WorldwideDave]

These shunt voltage and temp sensors are multi voltage compatible. I wonder if at a lower voltage 12v it generates more heat compared to say 48v systems? Just a wild guess.

I like wild guesses. That could be, but the shunt doesn't care what voltage it is connected to, and many RVs are just 12V, so I'd think more and more would complain.

 

[WorldwideDave]

Yes.
The sensor going to the cerbo gx is hanging in the air next to the battery.
Between the sensor in the air and the BMS sensor I get a good idea of the battery and battery box temperature.

okay that's great wow.
So that I understand completely,
although the victron shunts have an option to use a temperature sensor, and that temperature sensor requires a positive connection to a battery in order for the shunt to work, because this model/option/configuration is carried over from FLA, I should never attach this to a LiFePO4 battery post, although other users have and it says nothing about not doing this with LiFePO4, and the shunt requires it and is still sold, and it is not on their site or this site listed anywhere?
Not mad just seeking to understand.

 

[outboard1196]

okay that's great wow.
So that I understand completely,
although the victron shunts have an option to use a temperature sensor, and that temperature sensor requires a positive connection to a battery in order for the shunt to work, because this model/option/configuration is carried over from FLA, I should never attach this to a LiFePO4 battery post, although other users have and it says nothing about not doing this with LiFePO4, and the shunt requires it and is still sold, and it is not on their site or this site listed anywhere?
Not mad just seeking to understand.

You can use it on lifepo4 nothing wrong with doing that. Just for me and my use case in cold climate the copper battery cables exiting the heated box skews the temperature. Even on fla it wouldn't tell you the true temperature of the cells but it was close enough for temperature compensation.

 

[WorldwideDave]

You can use it on lifepo4 nothing wrong with doing that. Just for me and my use case in cold climate the copper battery cables exiting the heated box skews the temperature. Even on fla it wouldn't tell you the true temperature of the cells but it was close enough for temperature compensation.

That's fair but as others have stated, the temp sensor attached to the battery post seems to be a heat sink, and kept on getting hotter and hotter showing false numbers IMHO. like 75 Celsius one time. My cells are pulling full capacity and charging great and are not swollen so I don't think I did anything to break/overheat them - just think it was a false alarm due to how it was installed. But not certain.

 

[HRTKD]

That's fair but as others have stated, the temp sensor attached to the battery post seems to be a heat sink, and kept on getting hotter and hotter showing false numbers IMHO. like 75 Celsius one time.

As long as the components are in proper working order, I don't think I would characterize it as a false number. In fact, if the post was that hot the internal battery may have been even hotter. That's where a second temperature probe would come in handy. When the shunt says X and the BMS says X - 20°F that's a red flag.

An exception to this would be if you had a poor connection at that post for the main cable (not the sensor). Loose connections generate heat.

 

[Q-Dog]

photo please of your victron temp sensor attached mid-pack to your batteries?

It is a Ruuvi tag.


I

 

[outboard1196]

That's fair but as others have stated, the temp sensor attached to the battery post seems to be a heat sink, and kept on getting hotter and hotter showing false numbers IMHO. like 75 Celsius one time. My cells are pulling full capacity and charging great and are not swollen so I don't think I did anything to break/overheat them - just think it was a false alarm due to how it was installed. But not certain.

My sensor on the post does warm up when charging a few degrees. What was the ambient temperature?

 

[WorldwideDave]

As long as the components are in proper working order, I don't think I would characterize it as a false number. In fact, if the post was that hot the internal battery may have been even hotter. That's where a second temperature probe would come in handy. When the shunt says X and the BMS says X - 20°F that's a red flag.

An exception to this would be if you had a poor connection at that post for the main cable (not the sensor). Loose connections generate heat.

Right aware of that, thanks. We determined in prior posts that no, it was not loose, or bad crimp, and perhaps not bad conductor, and have no idea if the internal cells got that hot and the bms has been swapped out and conductors beefed up and no longer connects to the lid of the battery box. But that's it for now.

 

[WorldwideDave]

It is a Ruuvi tag.

View attachment 275444
I

Bought one today, thanks.

 

[WorldwideDave]

My sensor on the post does warm up when charging a few degrees. What was the ambient temperature?

trying to remember but not more than 21 C or 70 F or so outside. 100 amp constant discharge (around 94 A) for a short length of time would cause this. When discharging constantly at 46 A (pool pump, by the way), it would take about 2-3 hours and reach same temp. This is with no charge on. So of course played with time the pool pump would start, etc., but with no way to tell inverter to not start unless higher SOC or sunny day or whatever, it was both draining the battery completely (twice) and overheating dramatically fast. This forum had me scared, and all the while I thought it was just this silly battery temp sensor.
Good news is I now have a diff temp sensor connected to the cerbo directly, and a Ruuvi arriving tomorrow. I'll disconnect the shunt-based one, plop the ruuvi on top, and put the victron cerbo one on top of the cells directly, turn off all SCC, and discharge at 100 A and see what happens - mostly looking for temp diff between sensors.

 

[740GLE]

Right aware of that, thanks. We determined in prior posts that no, it was not loose, or bad crimp, and perhaps not bad conductor, and have no idea if the internal cells got that hot and the bms has been swapped out and conductors beefed up and no longer connects to the lid of the battery box. But that's it for now.

You determined the negative post had lower temp than the positive. The internal wiring of the positive had 2 6awg and the negative had 3 ?8awg?

Once you used your replacement conductors the temp sensor didn’t climb nearly as high.

 

[WorldwideDave]

You determined the negative post had lower temp than the positive. The internal wiring of the positive had 2 6awg and the negative had 3 ?8awg?

Yes, negative was always lower temp. But is that because the temp sensor I was using was from a shunt and requires the positive to be connected to the battery, though.
Yes, there were 2 from positive to the post that were bigger, and 3 from negative BMS that were smaller, but I think someone said they were about identical when added up.

Once you used your replacement conductors the temp sensor didn’t climb nearly as high.

Correct in that if I did an apples-to-apples test it did not. Example: 15 minute discharge at 100 A with zero charge = fail for old bms, success for new bms. 1 hour at 100 A, I believe it climbed up still, but never up to 75 C. Given time I think it would have, but it is only a 200 Ah 12V battery.

Tomorrow I have loads off, and will be at about 7% SOC when I wake up. Forecast is cloudy. Running 3 MPPTs, though - perhaps I'll get lucky and at some point have more than say 80 Amps going to it. I have a 10 A, 70 A, and 80 A MPPT. But winter/shady/bad angles. Really bad. I want to move array/do other testing on the charge side.

Bought a 25A victron blue smart charger to supplement MPPT and do some testing with, as I won't likely get more than 60 A charge on its own with sun right now. Will test this when it arrives, probably on Friday.

 

[WorldwideDave]

An update for you all: Installed a Ruuvi sensor. Put on top of battery cells. Put the victron temp sensor from the Cerbo GX on top of the battery cells. Removed the shunt based temp sensor that was attached to the 500A shunt and the battery conductor's lug. In image below, temp battery R means Ruuvi, and temp battery V means victron. Only been there 30 minutes. It was on top of the battery box in the sun (hence the 21.3 C max) but was within .1 minimum before I started the charging or inverting.
But while charging, not invertering, I can see already a 2.3 degree C or 36.4 degrees F difference between the two sensors. They are 1/2" apart at the most. Where should I put these sensors, and how should I mount them? I felt I was getting good readings but highly inaccurate ones from having the lug installed on the positive terminal.



This is a photo of where they are now. As a history reminder, the old CHINS battery 200Ah 12V LiFePO4 we replaced the BMS and the conductors, wrapped the battery back up the way it was with Kapton tape, put the yellow plastic around it, tie strapped the plastic together (it has 6 sides and Kapton holds it together on 5 sides, just zip tied it to hold together on one side), then lowered it back into the original box with foam on 4 sides with 6 pieces, and connected the new conductors directly to the shunt (negative) and the T-Class fuse (positive).

 

[sunshine_eggo]

The Cerbo temp sensor relies on conduction from the metallic ring terminal. You are sampling the air above the battery, not the surface. It needs to be firmly clamped to a surface, i.e., a battery terminal.

 

[HRTKD]

I'm surprised that the Victron sensor is reading higher than the Ruuvi. The Victron sensor needs to be taped down to the top of the battery surface. Ideally, it would be placed directly against a cell with some insulating material on the sensor to keep it from reading ambient temperatures.

 

[WorldwideDave]

I'm surprised that the Victron sensor is reading higher than the Ruuvi. The Victron sensor needs to be taped down to the top of the battery surface. Ideally, it would be placed directly against a cell with some insulating material on the sensor to keep it from reading ambient temperatures.

that's fair. Saying the Ruuvi installed right, but the victron one should be taped to cells.
Well the cells have the black tape on top of them.
Then they have the balance leads every here and there but can find a spot.
then the yellow plastic covering.
If you are suggesting I cut it all back open again, I'm happy (ish) to do that, but would need to know where to put the Ruuvi.

 

[sunshine_eggo]

I also have news for you guys... If you expect correlation across different types/brands/positions of sensors, you're in for disappointment.



11°F variation on the same battery.

BMS is an average of 3 sensors positioned 2 ends and middle. Right now, my BMS reports a 4°C difference between min and max. Good news is e BMS uses the least advantageous sensor, i.e., for cold temp protection, it uses the lowest reading. For high temp protection, it uses the highest reading.

Shunt and Out of the Blue are temp sensors on terminals, and those are 5°F different.