HeaterMeter Hardware v4.3.3, v4.3.4, v4.3.5, v4.3.6


 
What HeaterMeter firmware (AVR) are you on? EDIT: And how many probes are plugged in and all that? And can you test and see if it works any better without any probes plugged in?
 
Last edited:
Thanks, I'll have to look into it. I thought when I made the changes last year to the button read code that everything was perfect and it never missed a button press. However, when I was using my 4.2 HeaterMeter this week I also had a problem where it didn't go into Lid Detect immediately after I hit the button. I thought I must have been drunk because clearly that shouldn't happen but if you're reporting it too it warrants a second look.

I think it might be due to the fact that there is a lockout from any button event until the next button event can occur, which prevents double button presses the can occur every ~27ms. Under the hood there are internal button events that control scrolling through the probes among other things and that might cause a conflict with a user interaction. That's just a wild guess so I don't want to write any code until I can reproduce it reliably. This button has a 10k resistor on it right? I am going to build a test rig because the LCD/Button board is a little to inaccurate when the scope is hooked up due to the height of the full stack.
 
I'd bet it's the internal button events that are causing this issue, because the button response really did improve when you did the update last year. They seem to work perfect for me since then, but I don't enter lid mode manually very often.
 
Whee this was fun, involving oscilloscopes, digital timing, seriously staring at code, lots of AVR flashing and 2x mile-long thinkin' walks. Can you do an AVR update from the "online repository" and get the hex file from today? I think I got it ironed out, and I can even trigger lid mode and turn it back off in under one PID loop (1 second) which was almost impossible before due to all the timing overlaps. I can get it to not detect a button press sometimes if I release the button as fast as I can as soon as it clicks, but watching the oscilloscope waveform the analog line isn't at 0 for about 35ms but really has only stabilized for about 10ms so there's almost no way to detect that without adding more circuitry or not sampling the probes as much.

Something to note is that if you are on the No Probe home screen, there won't be a lid mode indicator LED light up for up to a second (because the LEDs update only once per second, where the LCD text would update immediately but you can't see it because it says No Pit Probe).

HeaterMeter 20160415B is the version number
 
Initial testing shows a much more improved response to button pushes. Really appreciate the time and effort you spent on figuring this out!
 
Glad to hear it, and thanks for reporting it. It's one of those things I think about when I press a button and it didn't work then completely forget about fixing. It was pretty bad too, causing a button lockout for 25% of the time on the homescreen if there was no food probe plugged in. The rest of the menus it worked fine, but most of the time we're not going through there (apart from setpoint).

As a side effect of all this pondering and walking, I think I have a way to implement long presses that will work with the existing event menu system which means we can add some extra functionality like:
  • Long press to start a timer of pre-defined duration(?) Would replace the food probe line I think and add new alarm type. Actually could probably long press to enter timer-set then use up/down to set the duration, left to cancel, right to start?
  • Long press to toggle on/off. Wouldn't it be nice to have a button press to just turn off output rather than just turning down the setpoint and having to turn it back up to cooking temperature next time?

EDIT: Forgot to say this is now standard in bundled firmware in the snapshot builds
 
Maybe it's just me, but the button oddness is back. I've confirmed that I'm on 20170415B.

Took 3 presses of the button to enable lid mode, waiting a couple of seconds between presses for the LED to come on. Took another 7 presses to disable lid mode. The next button press made the LED come on and then it went off about half a second later. Then it took 2 more button presses to get lid mode enabled again.

Very strange.
 
That's odd and definitely something I can't reproduce. I can do it a hundred times and it works every time unless I specifically try to tap the button as fast and then it only fails maybe one in ten times. Is there a possibility that because your button isn't a tactile switchie kind that it isn't actually activating or is so noisy that it is hard to read? You could try putting an RC filter on the output with something like a 47k resistor followed by a 0.1u capacitor which would give it an ~8ms rise time to 400mV which is the ballpark of switching noise I saw.
 
I think you're right about the button not fully activating. If I keep my thumb right on top of the button, it acts normally.
 
Has the Rpi 3 Model B been certified to work with the latest build? I just ordered a 2nd HM kit and I happen to have a Model 3 laying around Id like to put to use and avoid buying another B+.

thanks!
 
Hi Guys,

Just getting back on here after taking quite a while being away. I am still using my HM 4.0 w software v 20130604B. I know it's not fancy but the dang thing just works. I have been wanting to step up and get a damper but I am not sure my board would support it. Does the new v4.3.3 support the 4 line lcd display? I used those on the few 4.0 builds I did and love the display. I wouldn't mind building a new one based on the 4.3.3 board
 
Hi there Troy,

I can confirm that the latest version of HM hardware and firmware do support the 4 line LCD, it has the option in the configuration - I have one on my setup (https://tvwbb.com/showthread.php?70981-My-HM-build-for-UDS)

I used ribbon cable to wire up my LCD as you would see on the pictures in my post. The 4-Line unit I got from China only have connections on one side of the PCB unlike the LCD that Bryan supplies with the HM kit. The 4 line only set me back $7 Australian, postage included. I received it only 2 weeks after ordering.
 
Thanks VicusJ,

Glad to know the 4 line still works.

Hey Bryan, are you printing up any 4 line 4.3 cases? I bought my last 4.0 case from Tom but I just did a search and it looks like his last post was back in January. I didn't see a 4 line option for a 4.3 case in the Heatermeter store.

Thanks,
Troy
 
Troy Gamm said:
Thanks VicusJ,

Glad to know the 4 line still works.

Hey Bryan, are you printing up any 4 line 4.3 cases? I bought my last 4.0 case from Tom but I just did a search and it looks like his last post was back in January. I didn't see a 4 line option for a 4.3 case in the Heatermeter store.

Thanks,
Troy
Click to expand...

I believe someone asked and Byran said he wasnt going to make one, I could be wrong though.
 
You must log in or register to reply here.

 

Back
Top