• 16.10.2021, 03:53
  • Register
  • Login
  • You are not logged in.

 

Dear visitor, welcome to Aqua Computer Forum. If this is your first visit here, please read the Help. It explains how this page works. You must be registered before you can use all the page's features. Please use the registration form, to register here or read more information about the registration process. If you are already registered, please login here.

Aquasuite software sensor data from CSV

Thursday, December 10th 2020, 10:07am

I'm sorry if this has already been answered / is obvious. I've tried searching, but haven't found anything.

I'm wondering if it's possible to have Aquasuite software read in a CSV file as if it were a "sensor"... my reason for wanting to do this, is that I have several temperature sensors in my loop which are connected to a Corsair Commander Pro (which I use for PWM control). But I also have a Flow sensor high flow NEXT in my loop, and I'd like to be able to use one of the temperature sensors connected to my Commander Pro (the sensor at the coldest point in my loop) with the temperature from my Flow sensor high flow NEXT (which is at the hottest point in my loop), in conjunction with the flow rate to calculate the power dissipated in my loop. I was planning on using a virtual sensor as per the layout below.

However, I can't seem to figure out a way to read the live temperature from my Commander Pro. It kind of makes sense that Corsair wouldn't present the temperature sensors that are connected to it. So I thought maybe if I configured iCUE to log the temperature data to a CSV file ever 5 seconds, then maybe I could point Aquasuite at that file almost like a "sensor from file" data source. Is this possible or am I dreaming of a feature that doesn't yet exist?
peterhudson has attached the following image:
  • Capture.PNG

Kuminauha

Junior Member

Thursday, December 10th 2020, 1:05pm

Commander Pro doesn't show up in the data quick view?

Thursday, December 10th 2020, 1:57pm

Does HWiNFO not pickup the Commander Pro data?

If not, the simplest solution may be to add a physical temp sensor at the 'coolest' part of your loop, and connect directly to the High Flow NEXT. An additional benefit is the High Flow NEXT automatically calibrates the added sensor for accuracy.

Thursday, December 10th 2020, 8:38pm

I didn't know that HWInfo data was visible inside Aquasuite. I generally don't have HWInfo running all the time, so I didn't see it. But seeing your post (and the previous one) made me try it. But interestingly when I start HWInfo, then iCUE software stops working correctly -- iCUE stops reporting temperatures and fan speeds and no longer allows me to control this. This is probably my first problem to solve. But here's a screenshot showing what I mean. Normally I would see 2 temperature sensors and 3 fan controls in iCUE.

But, this does mean that if I can solve the HWInfo to iCUE problem, then it should also solve my data availability in Aquasuite problem. Is incompatibility between iCUE and HWInfo a known issue?

Regarding the idea of simply connecting a temperature sensor to the High Flow NEXT... yes that's the ideal solution, and I have every intention of doing so. However, the connection at the back of the High Flow NEXT are JST-GH sized (1.25mm pin spacing connector, or maybe it's molex picoblade which is a very similar sized connector at 0.05 inches). But my temperature sensor hardware all uses dupont connectors which are ~2.5mm pin spacing. So in order to connect my existing temperature sensor I'd need to make an adapter cable. And before going down the hardware route, I was hoping I could solve it with software. Although I agree, it would be nice to have the High Flow NEXT auto-calibrate the temperature sensors since right now they're reading a few C apart and I'd need to work out a way to calculate the offset correctly when the loop is at equilibrium.
peterhudson has attached the following image:
  • Capture2.PNG

Thursday, December 10th 2020, 8:52pm

Ok... after a little bit of quick googling it turns out that iCUE and HWInfo really don't like playing in the same sandbox. I disabled iCUE on startup and now HWInfo (and by extension Aquasuite) can see the temperature and fan speeds from my Commander Pro just fine (see screenshot). This doesn't solve the temperature calibration issue. Nor does it solve the problem of having iCUE disabled so it won't control my aRGB anymore. But it does solve my original data source problem.
peterhudson has attached the following image:
  • Capture3.PNG

Kuminauha

Junior Member

Thursday, December 10th 2020, 10:22pm

Sooo, what excatly does auto calibration do? Seems kind of impossible to calibrate any sensor without using one definitely calibrated sensor....

You can calibrate sensors by dipping them both into same jar of water and then adjust the offset so they both measure the same temperature. Also if you wan't a real world reading you can also dip a fewer thermometer into the same jar and adjust the offsets.

This post has been edited 1 times, last edit by "Kuminauha" (Dec 10th 2020, 10:26pm)

Thursday, December 10th 2020, 11:18pm

For my purposes that absolute temperature doesn't matter since the calculation for power is basically (Hot_point_temp - Cold_point_temp) * Flow... since I'm taking the difference between hot and cold points that absolute accuracy doesn't matter, so I just need to adjust for bias. I did this just now by powering my computer down for an hour or so to let the loop reach thermal equilibrium. Since my loop is quite large (it contains 2L of coolant), I figured when I powered up, the temperature of the coolant would be relatively constant for at least 20 to 30 seconds. So I powered up and quickly wrote down the temperature displayed on the High Flow NEXT (the hot point in my loop), then quickly signed into Windows and checked the temperature at the cold point in the loop from HWInfo. It turns out that there was a 1.5C difference between them. I'm not sure if one or both is reading high / low, but it doesn't matter as long as they both have the same response gain (which I certainly hope they do). So I simply biased one sensor by 1.5C and now I have my power calculation running nicely.

Friday, December 11th 2020, 3:28am

To calculate power dissipated by the radiator, fully accurate temperatures are probably not needed, but the sensors should have the same bias, if any is present.

What I believe the High Flow NEXT does, is:
  • determine when the computer is off
  • allow the loop to settle to ambient
  • then using the higher quality/calibrated sensor in the High Flow NEXT as the source
  • apply an offset to the remote temperature sensor to equalize the values
In order to do this, the High Flow NEXT requires 5V from the USB connection while the computer is off.
The procedure outlined above is pure speculation on my part, so take it for what it's worth.