Jobo ATL2200 temperature issue

I recently tried to calibrate the temperature of my Jobo ATL2200, but I found a problem.
When I set the process temperature to 38℃, the ATL screen showed the water temperature at 38.1℃, but the actual measurement was about 38.9℃ (I confirmed that my thermometer is accurate). The indication of the temperature of the chemical in the bottle is correct, and the difference of 0.1℃ is a normal.
I can't calibrate using the options in the menu (I have added +0.5 error temperature calibration, which is the limit of the menu), so I have some questions about this.
Is this a feature of the JOBO design? For example, automatically increase the water temperature to compensate for temperature loss.
Or, do I need to have it repaired?

LomoSnap said:

(I confirmed that my thermometer is accurate

Click to expand...

How?

LomoSnap said:

Or, do I need to have it repaired?

Click to expand...

Another option: just set the temperature to 37.1C or so, which will make the real temperature (assuming your separate measurement is correct) end up close enough to 38C.

koraks said:

How?

Click to expand...

I have a reference mercury thermometer with a calibration report, which I use to verify and calibrate my electronic thermometers (because mercury thermometers are fragile and slow to respond, so I only use them as a "calibration reference" rather than "daily work")

koraks said:

Another option: just set the temperature to 37.1C or so, which will make the real temperature (assuming your separate measurement is correct) end up close enough to 38C.

Click to expand...

This is indeed one approach, but I'm here to get the thoughts of other ATL2x00 users to verify whether this is a feature of JOBO or a device malfunction.
I searched for the temperature calibration instructions for JOBO ATL2+ (an older generation of Autolab) :
"There is a purposeful offset designed into the calibration of the water bath temperature sensor. The offset (the higher temperature at the "waterfall") is critical for proper temperature control. Please note that the 0.4ºC offset only applies to a calibration temperature of 32.0ºC."
I'm not sure if 1 degree of deviation is normal for the design of the ATL2200. I can't find any more information about temperature calibration for the ATL2200.

OK, good job on the calibration.

In your place, I wouldn't worry about this. The temperature probe on the ATL2200 is likely a simple NTC which has OK linearity, but the analog amplification circuit needs to be properly calibrated. There may be a hardware calibration (typically a potmeter) somewhere in this circuit in addition to the software calibration you've already found.

But, as you also indicated - it's likely that Jobo included this measure based on their own measurements and aimed at correct temperature of the actual developer. I think I've read something similar about other Jobo processors, and it does indeed make sense. Given that the target temperature of the developer appears to be correct, I wouldn't worry about any of this.

Btw, at the risk of driving you crazy - what matters is not just the temperature of the developer in the bottle, but more so the developer temperature inside the tank as development takes place. Technically speaking, that is what you should be measuring and controlling for instead.

koraks said:

Btw, at the risk of driving you crazy - what matters is not just the temperature of the developer in the bottle, but more so the developer temperature inside the tank as development takes place. Technically speaking, that is what you should be measuring and controlling for instead.

Click to expand...

As you said, I have already done this.

Before I used ATL, I used JOBO CPE2, which is a manual lift. The temperature loss of the liquid when it is manually poured into the developer tank is very large (the highest temperature may even drop by 1 degree).
This is also one of the reasons why I bought ATL. The liquid of ATL is transferred to the developer tank in a closed pipeline, and the temperature loss is much less (the test shows a temperature drop of about 0.2-0.3 degrees). Therefore, for the standard temperature of Fuji Hunt C41 Developer of 37.8 degrees, I usually set it to 38 degrees to compensate for this loss. This is a known and confirmed thing.

However, since the water temperature display of ATL is inaccurate, I cannot feel safe to use it.
Maybe I can get the thoughts of other ATL2x00 users to confirm that "this machine is like this, there is no malfunction"

OK, let's hope some other ATL2000/2200 users pop up. So far I can't see anything that looks like a malfunction. I'd suggest using the machine with confidence.

The ATL2000 is a older design than the JOBO ATL's 2200 / 2300 / 2400 / 2500