• 27.04.2024, 09:14
  • Registrieren
  • Anmelden

  • Sie sind nicht angemeldet.

 

Water Temperature Sensor Calibration

Lieber Besucher, herzlich willkommen bei: Aqua Computer Forum. Falls dies Ihr erster Besuch auf dieser Seite ist, lesen Sie sich bitte die Hilfe durch. Dort wird Ihnen die Bedienung dieser Seite näher erläutert. Darüber hinaus sollten Sie sich registrieren, um alle Funktionen dieser Seite nutzen zu können. Benutzen Sie das Registrierungsformular, um sich zu registrieren oder informieren Sie sich ausführlich über den Registrierungsvorgang. Falls Sie sich bereits zu einem früheren Zeitpunkt registriert haben, können Sie sich hier anmelden.

Caldorian

Junior Member

Beiträge: 42

Water Temperature Sensor Calibration

Montag, 6. November 2023, 16:48

In my waterloop, I've got the following components in order:
  • Aqua D5 Next
  • EKWB Velocity 2 AM4 (Ryzen 5900X)
  • EKWB Vector2 w/ ABP (EVGA 3080 Ti FTW3)
  • Aqua HighFlow Next
  • 3 x Alphacool NexXxoS ST30 360mm
  • EPDM Tubing between all connections


My flow rates (as identified by the HF Next) are around 138L/h. The one part that interests/concerns me is the water temperature readings coming from the D5 Next vs. the HF Next. Under minimal load, the D5 Next reads ~30C where as the HF Next is about 33.5C, with an Ambient Temperature of about 25C. Under load, the temperatures increase proportionally, with the HF Next staying around 3.5C higher then the D5 Next, reaching some 38C on the HF Next.

What I'm curious about is whether my fluid temperature is truly rising/dropping that 3.5C between the 2 sensors vs whether the 2 sensors just aren't properly calibrated and the fluid temperature is more normalized between the 2. Intellectually, is would make sense for there to be a temperature difference between the 2, since one is right after all the heat components vs. the other being after all the heat dissipation. However, I'm not sure if *that much* of a variance is proper, especially when it's not really under load and the same variance is seen as the load increases (with increased load, I would have expected increased variance).

Assuming that the water sensors are off and I would need to set some offsets to properly calibrate them, anyone have good suggestions on how I'd go about finding the "true" temperature?

Taubenhaucher

God

Beiträge: 2 119

Montag, 6. November 2023, 17:18

The only possibility I can think of would be to hold a thermometer in the circuit when the power supply unit is jumpered, both high flow NEXT and D5 NEXT have a display to show you the measured temperatures.
The temperatures are then only dependent on the heat generated by the respective device.
This is probably as close as you will get.

But my question is, why do you need such an accurate measurement as long as your components have a good temperature?
Es gibt keinen Ausweg, den ein Mensch nicht beschreitet, um die tatsächliche Arbeit des Denkens zu vermeiden.
Thomas Alva Edison (1847-1931), amerik. Erfinder

Speedy-VI

Senior Member

Beiträge: 784

Dienstag, 7. November 2023, 04:04

With a flow rate of 138L/H, I think the coolant is moving through the loop too fast for there to be that much of a temp difference between the 2 points in the loop. I think the temp sensors in the D5N and HFN are just 10K Thermistors that are not super accurate, although 3.3°C is a pretty big spread. I suggest the following procedure, which will require a 3rd temp sensor as a reference, preferably one with just the flat sensor like #53026.
  1. Connect the reference temp sensor to a temp header on one of your devices and stick the end in a glass of ice water. If it does not read 0°C, add an offset until it does. Assuming the temp sensor accuracy is the same between 0°C and the max loop temp, this sensor should now be calibrated.
  2. Power down and let the coolant reach ambient temp.
  3. Stick the reference temp sensor in the reservoir so it is immersed in the coolant.
  4. Boot up, start Aquasuite, and immediately note the temp of the reference temp sensor. It should be the same as ambient temp. Also note the temps reported by the D5N and HFN.
  5. Add whatever offsets are required to make the D5N and HFN temp sensors match the reference temp sensor immersed in the coolant.

Caldorian

Junior Member

Beiträge: 42

Freitag, 10. November 2023, 21:00

Thanks Speedy, that's a good idea, and I'll look into doing just that.

As for the question of why I need things that accurate? I don't. But if I can, why not :D

mditsch

Senior Member

Beiträge: 586

Samstag, 11. November 2023, 18:00

Zitat von »Caldorian«

As for the question of why I need things that accurate? I don't. But if I can, why not :D
That's the reason, why doing watercooling ... :thumbsup:

Dieser Beitrag wurde bereits 1 mal editiert, zuletzt von »mditsch« (17. November 2023, 22:40)

cptninc

Full Member

Beiträge: 198

Freitag, 17. November 2023, 21:33

I ran a single point calibration on mine when I was able to leave my machine turned off for a day.

What I did was this:
  1. Power down the machine so that no additional heat is being put into the loop
  2. Keep the loop itself running with the pump circulating coolant and the fans blowing air through the radiators
  3. Keep this going until the coolant reaches ambient air temp. The time this will take is sort of fuzzy. I had to take care of other things, so I left mine running for about 24hrs like this but that is far longer than it actually takes.
  4. Record the temps shown on the different sensors
  5. If you have a calibrated air temp sensor, record that as well (this is what I did)
  6. Boot up the machine, load up Aquasuite, enter the offsets for each sensor


If you're just using these for power dissipation measurement, then you don't need to worry about the air temp sensor. You just need the two sensors to get the same reading off of a single reference.

Since performing this calibration, the calculated power dissipation in my loop (bottom left in my sig) matches the summed TDP of the cooled components (bottom right in my sig) almost exactly.

Dieser Beitrag wurde bereits 1 mal editiert, zuletzt von »cptninc« (18. November 2023, 05:16)

demilleroh

Junior Member

Beiträge: 14

Single Point Calibration Procedure

Samstag, 18. November 2023, 04:48

cptninc,

Thanks for posting your single point calibration procedure ! I just added two Calitemps to my loop to join two uncalibrated Calitemps
located in pita locations next to my radiators. Fortunately, I did calibrate the two most recently added ones.

I've been dreading removing the two uncalibrated Calitemps just to calibrate them. Now, thanks to you, I can leave them in the loop and still
calibrate them !

Ähnliche Themen