Possible Bug - Lutron RadioRA2 Driver and Thermostat

[gReatAutomation]

It works but the temperature is not correct. You had mentioned a potential rounding error? Did you see the post?

[Dean Roddey]

Oh yeh, I almost forgot about that.

[gReatAutomation]

Very strange. I checked it today and it's showing the appropriate set point from the thermostat: 78

Is this indeed a rounding issue or what?

[Dean Roddey]

Rounding issues would sometimes be right, or sometimes up or down a degree, depending on how close the C temp is to the F temp. Sometimes they are just worst case and sometimes the difference is tiny.

[gReatAutomation]

Rounding issues would sometimes be right, or sometimes up or down a degree, depending on how close the C temp is to the F temp. Sometimes they are just worst case and sometimes the difference is tiny.

So how do we resolve? Should not have a rounding issue on a set point, IMHO. On current temperature? Sure, but not a set point which is set in whole degrees.

[Dean Roddey]

I need to look at how the calculation is being done, and make sure it's doing things in the right order and at the right precision and such.

[Dean Roddey]

OK, so I'm actually asking for the temp from the thermo in the scale that you tell me you want to see.

So, there's only really two scenarios:

1. When it reports them asynchronously, that it reports both of them, and the second one in this case is the C one, so that the converted one overwrites the one
2. It reports them floating point either way, so even if it only reports the one I care about, it still have to round it. I'm rounding to the nearest integral value (up or down, whichever is the least change.) Maybe the thermo is just rounding down by chopping off the fractional part.

I'll have to hook it up and see what's going on, though I may have to sit on it to get it to change the temp enough for good testing.

[Dean Roddey]

OK, so I'm correctly protecting myself against handling reports for the other scale, I only handle ones for the selected scale. So, the only thing I can think of is that the thermo is always rounding down and I'm rounding to the closest. When they disagree is the driver always the next degree up from what the thermo shows? If that's it, the driver is really sort of more correct since 78.8 is closer to 79 than to 78. But, I can always just round down if that makes it agree with the thermo.

[gReatAutomation]

Thanks for checking in to this. I was asking about the HighSetPoint and the LowSetPoint, no the actual temperature reading. I was under the assumption that the set points are round numbers (e.g., 75, 78, etc.).

[Dean Roddey]

Oh, OK. The same would apply to set points. They are also reported as floating point, and I'm also filtering for the scale that you selected. So it would be the same. If they disagree, is the driver always the next degree up?