Hi everyone,
I think the title says it all: I’d like to read the temperature of my MultiPlus. I reckon the unit must have at least one sensor in the transformer, as temperature monitoring is available too. If I can only read it in a menu, that’s already sufficient; having it available in Node-RED would be fantastic. Or is it simply not possible…?
The background is this: my three 3000s sometimes run at full load while charging, and get really hot as a result. (I know they’re designed for this. However, I believe cooler electronics last longer.) I’d like to improve the cooling and fit some fans, and to be able to assess the actual benefit, I need meaningful readings from before and after. I don’t think it makes sense to somehow cram the supplied sensor in somewhere.
hello,
unfortunately, that isn’t possible. However, you can connect an external temperature sensor to the GX and place it wherever you want to measure.
That’s how I’ve just solved it. However, there’s no point where you could measure a meaningful temperature. When the internal fan comes on, a sensor mounted on the surface quickly gets cooled down and then gives rubbish readings.
I really need meaningful readings. First and foremost to check how effective the cooling improvements are—I only want to take sensible measures and definitely don’t want to make things worse while trying to improve them. If there’s no access to the readings, I suppose I’ll have to improvise with the sensors.
I measure the temperature in a very “primitive” way at the top of the MP2’s air outlet. And even with the internal fan running, I can observe exactly how the temperature changes when I point a small fan at the MP2’s intake vents from below.
In other words, if you “stick” the sensor to the hottest components using thermal paste (and insulation if necessary), you’ll be able to measure the difference clearly. We’re not talking about tenths of a degree °C here, but a difference of several degrees °C.
Also, if you don’t want to switch off the internal fan, you’ll always see differences — unless your additional cooling keeps the MP2 operating below the internal fan’s activation threshold.
I’m not changing anything about the internal fan; I’m adding two 90 mm fans. I was planning to have both of them help extract air through the exhaust slots at the top, so that the warm air is removed more effectively.
Now, though, I want to know whether this measure makes sense or could perhaps even have a negative effect (the internal fan is running while the additional fans are stopped—for whatever reason, let’s leave that aside for now). I’d like to use the transformer temperature as a reference, because it gets really warm, has a fair bit of mass, and a sensor wrapped around it (as I believe there is one inside) would give readings that aren’t influenced by the airflow. The sensor with the cable lug, on the other hand, has a large surface area that cools down quickly—faster than the component being measured.
Hello, I know this thread started a long time ago, but I’ve only just read it, as I also want to make some changes once the warranty has expired, because the fan noise at full load is driving my wife mad. The fan on my 3000 ramps up and down; all the updates were installed, which brought a slight improvement, but the noise is still annoying. Yes, the sensor in the transformer seems to be responsible for controlling the fan. I measured the temperatures with an infrared thermometer while the Multi was open. The power heatsink was a comfortable 26°C at full charging power, but things looked a little different around the transformer: 35°C at the bottom, 68°C on the right, 61°C at the top and 67°C on the left. The built-in fan blows directly onto the bottom, of course. After about 15 minutes with the Multi open, the internal fan slowed down, making it considerably quieter. However, the transformer temperature remained almost the same, so there must be at least one more sensor somewhere on one of the circuit boards, probably monitoring it as well. But better airflow results in a lower fan speed and therefore less noise. So I fitted three 60 mm low-noise fans underneath, with speed control, to blow additional air in from below. I put the cover back on and tested it, but the improvement was minimal. As the next step, I doubled the size of the air outlet at the top so that the warm air could escape more easily. I “measured” it by ear again, and there was another minimal to slight improvement, but it still wasn’t satisfactory. In any case, it was worse than without the cover.
Now I’m doing it completely differently: I’m removing the cover, mounting it 8 cm above the unit and enclosing the whole thing with perforated aluminium sheet so that nothing can happen. This will take another one or two weeks. I hope it works just as well as with the cover off. That would be great.
As a last resort, I might build a new cover from perforated sheet metal, with a 150 or 170 mm fan fitted directly in front of the transformer. Something is bound to work. Perhaps I can also get a replacement cover from a Victron dealer somewhere.
Maybe my idea will help others with the same problem.
A transformer like that has a fair amount of mass, but relatively little cooling surface area.
As a result, it will always run comparatively hotter than other components that have good heat transfer via a heatsink.
If the fan problem is permanent, the only options are either a larger MultiPlus or reducing the load, and ideally installing it in a relatively cool location. However, this will result in higher standby consumption.
