Graphs - Showing Temperature & Set Point Readings Against Different Temp Ranges

Plugins & Plugin Development Unmaintained
1

Hi Chris

I’d like to compare the set point value on a thermostate with the actual temperature in the room, just so see how quickly a room takes to heat up or cool down

As things stand when two Temp readings are choosen the Plugin will show you only one temperature range on the left hand side, but because set and temp are different variable it gives me two separate ranges… (See screen shot)… which makes for a tricky comparrison

Is there a why I could compare the two against a single temp range ?

2

This is due to the data types being different. Currently there’s no way to change this (it’s on my todo list).

However, there’s a “feature” you could take advantage of. If you save this graph, then click on it from the saved graphs area, I think everything comes up on the same graph. Really this is a bug, but I need to sort out the whole axis selection issue so I’ve not done anything about it as yet…

Chris

(ps. this post is in the DSC alarm forum).

3

moved :wink:

4

moved ;)[/quote]

I have no idea how that happened ! - Sorry @Chris and thanks @Guessed for moving it - I’ll blame on too much alcohol but I don’t drink :slight_smile:

5

Hi Chris

As you saw from the original post, I try to compare the set point against the current temp to see how quickly heat escapes (and boy does it go fast) :frowning: . If you look at the attached, due to them being different devices types they do not use the same side value, therefore how do I know which one relates to which?

Is there a rule the tool uses? E.g.

First one chosen is the first line?
Top in the legend, is the first line?

PS . why does the graph decide to do a curvy line sometimes, rather that a straight line between the graph points?

6

I agree - it’s not easy to see. I keep meaning to change the way axis are chosen - currently when graphing channels directly (ie not using saved graphs) it is based on the variable type. Unfortunately, while setpoints are temperatures, they are given a different type, so they end up on a different axis. I have a modification on my list which is to change the axis selection to use the units, so if you set units to be the same, then things would also end up on the same axis. There may also be a middle option which uses a combination of both…

As an aside, if this graph is something you display a lot, did you try saving it as a saved graph as per my previous message. This allows you to specify the axis for each channel, so this wouldn’t be an issue.

[quote=“parkerc, post:5, topic:174279”]Is there a rule the tool uses? E.g.

First one chosen is the first line?
Top in the legend, is the first line?[/quote]
Probably, but this is hidden in the depths of the graphing library so I’m not really sure…

It shouldn’t - it should always be based on the graph type.

Cheers
Chris

7

Thanks Chris

It shouldn’t - it should always be based on the graph type.[/quote]

Can you see from my graph, how some of the lines sweep and curve? Is this the graph type thing? Which type would be straight, point to point?

Oh and one more to add to your list and that is the ability to open the graph in full writing a new tab/window. The space on an iPad screen is pretty limited.

8

Ah - sorry, I misunderstood. This is just the algorithm the graph package uses to smooth the lines. I think it produces a kind of sine waveform between two points, but as it always draws the lines through the points, if two points are close together, you get a straight line. It’s not the best spline algorithm that I’ve seen, but I can’t do much about that…

I have a similar thing on my list - not quite a new tab, although I may look at that, but I’d like the channel list to be able to “hide” to the left of the screen on mobile devices. A month or so back I was playing with a compiled version of the app which could install as a native Android/iOS/etc app, and this buys you back some screen real-estate by getting rid of the browser address bar… This is still sort of work (not) in progress though at the moment…

Cheers
Chris