I have a Sansaire (sadly, 220 VAC, so I will have to buy another) and a Bluetooth cooking thermometer with four probes (iGrill2).
Neither the Sansaire nor the the iGrill are NIST calibrated, so I don’t which one is correct, but the following experiment was pretty good for me (I should create a Google spreadsheet to share):
Boneless, skinless chicken breast of unknown weight, four pieces, starting with room temperature water of unknown volume in a tamale steaming pot of unknown volume (it’s a big pot, and the water level with the chicken was about a centimeter below the max. fill line of the Sansaire).
All four probes of the iGrill! Placement wasn’t scientific, but I tried to keep two of them deep and two of them shallow, and about 90° apart from each other.
Until the water hit 60°C the probes tended to record similar but not identical temperatures (0.1°C precision, unknown accuracy). The probe close to the outlet was consistently a degree or two above the coldest probe.
If I throw out the pre-60° data, all four probes recorded temperatures within ±0.6°C of each other. I only calculated maximum differential, but using the raw data I suppose I could perform a more detailed analysis. For my purposes (entertaining myself with the new iGrill at the time) it was good enough.
Things I would do differently if I were willing to sacrifice my expensive probes and tamale steamer:
[ul]
[li]Use known volumes of water and carefully controlled starting temperature.[/li][li]Weigh the chicken and record the starting temperature.[/li][li]Get another iGrill and four more probes, having 8 total.[/li][li]Drill holes into my tamale pot and array the probes at different depths and distances from the wall.[/li][/ul]
But, yeah, no, I probably won’t do any of that, unless… maybe… I decide to build an automated full grain home brew system.