Heatermeter 4.2 -- a couple questions

[RKHanso]

I've had my heatermeter working great for a couple years, with only minimal issues that were pretty easily solved.

I'm bottle conditioning/carbing some beer bottles in my basement and put a space heater on them to warm them up over the ambient temps a bit.
I'm not using the Heatermeter to control anything, just monitor temps. I have the pit and 2 Maverick probes connected and placed around the bottles to measure temp.
The output of the Maverick probes are fluctuating much more than I've seen when they're stuck in a slab of meat. Is this normal when the probes are in the air? Or is there some problem?

Another question - It's been years since I've upgraded any firmware on the heatermeter.
Should I? Or leave well enough along?
I'm not sure how I'd upgrade it though. It seems that the link for upgrading "online" within the GUI doesn't work - I get an error: wget: bad address 'heatermeter.com' Download failed

Router Name OpenWrt
Router Model BCM2708
Firmware Version OpenWrt Attitude Adjustment 12.09 / LuCI 0.11 Branch (0.11+svn)
Kernel Version 3.3.8
Local Time Fri Sep 13 14:08:53 2019
Uptime 3h 36m 3s
Load Average

 

[RKHanso]

Recently the temp graphs changed after I changed the set temp. Odd. Making me wonder if this is a firmware/software thing.
Though I do have the HM plugged in to AC power through a couple extension cords. I'll try turning on the AC Line noise filter.
Here's the graph after changing the set-temp (but before turning on the AC noise filter)

 

[RKHanso]

Now I'm thinking it's just a browser display problem. After I turned on the AC line filter, all the graph lines looked the same again. But that got me thinking....there was a refresh of the page.
So I shut off the filter again and sure enough, the new lines added were different looking than prior lines. Then a page re-load made them all look the same again. I think the display problem is the browser. I normally use Brave browser, and that's where I'm seeing the quirkiness. If I use MS Edge, it looks a bit different. I'll chalk that up to the browser.

Back to the original 2 questions then.

 

[JKalchik]

Immerse the probe tips in a water filled pint jar, covered to prevent evaporation (and suitably treated to prevent mold, bacteria, etc.) This will cut down on temp swings caused by drafts, and is pretty commonly done in kegerators, coolers, etc.

 

[Bryan Mayland]

What you're probably seeing in that graph is that the data loaded from the database is averaged, and then as new points come in they're appended to the graph. It looks noisier because you're seeing every point. Eventually, the graph will get enough points that it is full so it pulls from the database again and it will look like all the points are smooth again. In IE and Edge, those browsers don't support streaming data so they poll the server every 10 seconds and therefore have fewer datapoints and look less noisy.

I will say those graphs do look pretty jaggedy. The Maverick probes tend to respond more slowly due to their thicker probes, so you're probably picking up some electrical noise from somewhere. If it is really bad, you'll see a tiny jagged graphic in the probe's box in either yellow (not good) or red (probably unusable).

 

[RKHanso]

Thanks for the input. I did get a couple of glasses of water to put the maverick probes into. I'll wait and see what the temp graphs show in a while.

I see that the date/time of the heatermeter graphs keep going back to 2018. I have to keep "syncing with the browser" to get it to correct. I do have the time servers set up.
Here below, was just after I synced with the browser. But before that it was back to 2018. I did the "sync with browser" yesterday, but sometime between then and now, it reverted back.
Plus, whenever I "sync with browser" - The graphs reset and appear to delete the old data and start fresh.

Might this be fixed with a firmware update? If so, I'm not able to update in the GUI because I get the error message mentioned above.

 

[RKHanso]

Well, the maverick probes don't seem to be doing any better in water. They're still fluctuating quite a bit. And I now see in the display a yellow jagged indicator in the temp box. I've never had this in the past.
What is wrong?

 

[RKHanso]

And then throw on the top of that the difference in display between the Brave and Edge browsers:

Brave (but again - when I "refresh" the page, all those quick fluctuations disappear and it's smoother - like the Edge browser):

Edge:

 

[RKHanso]

after more time, and the probes in glasses of water, the fluctuations have not changed.

 

[Bryan Mayland]

Plus, whenever I "sync with browser" - The graphs reset and appear to delete the old data and start fresh.

It resets the database because your time is jumping forward by a year so the old data falls off the graph (which only holds 24 hours). Both your problem where it can't pull firmware from the web and the incorrect time are because your HeaterMeter doesn't have internet access (outgoing). No internet access, no time synchronization and no place to download firmware from.

I explained why you see it jaggedy then hit refresh and it is not jaggedy and why some browsers are less jaggedy in my previous post.

The noise could be coming from the outlet you're plugged into, or from the space heater you're trying to measure the temperature near, or a mains-powered fan on the same circuit if you don't normally see the noise.

 

[RKHanso]

Thanks for the info. I knew there was Internet access, but couldn't figure out why it wouldn't update...which got me to look at the network diagnostics. I can Ping IP addresses, but not URL/Domain names. That told me my DNS wasn't set up correctly. I fixed that and then was able to update the software (and I suspect the time will always be correct now).