HM Build - Troubleshooting LCD and Thermocouple. HELP!!

[JWywal]

Hi all!

I just completed my first build and am encountering two issues.

1) Thermocouple / probes

After many years away from soldering components I decided to try my hand at doing my own TC build. First attempt displayed "Pit: 22*" with no probe connected, so I knew I had an issue somewhere. Hit all little bits with the iron again and ran a braid across the AD849X to clean up some of the globs i left behind initially.

Second attempt showed the 'no pit probe' message on the LCD as hoped, but now when I connect a TC probe, the HM doesn't detect it at all (web page just says 'off' for the pit probe when connected). Have I cooked the AD849X?

Interestingly, probes 1 and 2 work perfectly, but probe 3 shows a temp about 15* cooler than the other two. Before reflowing all three probes showed the same temps.

http://i.imgur.com/WIEIScc.jpg

2) LCD

When I power up the unit (5v USB to rPiB), the display comes up with -no pit probe- as desired, but after 10-15 seconds the display starts to go haywire. As time goes on, the display gets more and more garbled:

http://i.imgur.com/2nyE1TX.jpg

When I pull the power and fire it back up, I just get solid blocks on the top row. If I unplug and leave it for a while, then plug it back in it goes through the same process again (from 'no probe- to garbled characters).

Functionally the HM seems to be ok otherwise. I can connect to it via the web interface, probes 1-3 display on the graph (although probe 3 shows a lower temp than the other 2), admin page works fine, etc. While I could conceivably operate without the display if need be, I'd like to fix it, but more importantly I need to get my probes sorted out!

I don't even have a clue where to start troubleshooting these two issues, so I'm hoping one of the pros can help me out with this.

Thanks!
Jason

 

[Bryan Mayland]

1) Did you change the probe type to Thermocouple for the pit probe? By default I think it reads ~21 degrees because it is set to type thermistor. Did you check the voltage output from the thermocouple amp?

2) Do any of the components on the HeaterMeter get warm? If so something is amiss. Check for stuff getting hot without the RPi attached.

 

[JWywal]

1) Did you change the probe type to Thermocouple for the pit probe? By default I think it reads ~21 degrees because it is set to type thermistor. Did you check the voltage output from the thermocouple amp?

2) Do any of the components on the HeaterMeter get warm? If so something is amiss. Check for stuff getting hot without the RPi attached.

1) Good point - ever since I changed it to thermocouple in the settings the 22* went away and displayed the no probe message. I'll confirm the voltage output this evening. I was afraid to apply a voltage to re-test because now all the rest of the components are on the board now and didn't want to fry anything.

Any thoughts on Probe 3 showing a different temp as compared to 1 and 2? They're all set to the same thermoworks type and I'm swapping the same probe between the jacks.

2) I'll check for temp things this evening and post back.

Thanks Bryan!

 

[John Bostwick]

The two different probe temps between jacks regardless of probe being used, could be from having a wrong resister, instead of the 10k.

Check the resistance of each jack, they all should read the same

 

[Bryan Mayland]

Good point John, check the resistance from the (empty) VCC pin of the FTDI connector to the back pin of each probe jack. It should read within 1% of 10k ohms. If it doesn't then there's something else in contact with it or pulling the line down. Do this without the pi connected, probes inserted or power applied.

 

[JWywal]

Thanks guys

Familial responsibility kept me from playing around with the HM last night. Should get on there tonight to troubleshoot.

 

[JWywal]

1) Did you change the probe type to Thermocouple for the pit probe? By default I think it reads ~21 degrees because it is set to type thermistor. Did you check the voltage output from the thermocouple amp?

2) Do any of the components on the HeaterMeter get warm? If so something is amiss. Check for stuff getting hot without the RPi attached.

Good point John, check the resistance from the (empty) VCC pin of the FTDI connector to the back pin of each probe jack. It should read within 1% of 10k ohms. If it doesn't then there's something else in contact with it or pulling the line down. Do this without the pi connected, probes inserted or power applied.

The two different probe temps between jacks regardless of probe being used, could be from having a wrong resister, instead of the 10k.

Check the resistance of each jack, they all should read the same

Probe 1: 9.95
Probe 2: 9.95
Probe 3: 9.52
Pit (TC): n/a



I scanned the board with my IR gun:

Starting temp: 21*C

Temps only went up around the BS170 nearest the 0 ohm jumper (24*C before unplugging)
Anywhere else on the board went up marginally, if at all.
This was after being plugged in for about a minute or so. Didn't want to leave it any longer for fear of cooking something.

Plugged it back in and checked the voltages of the LCD pins:

pin 1 = gnd
pin 2 = 4.96
pin 3 = 0.58
pin 4 = 1.96
pin 5 = gnd
pin 6 = 0.90
pin 11 = 0.02
pin 12 = 0.02
pin 13 = 0.02
pin 14 = 0.41
pin 15 = 4.95
pin 16 = 2.66

Interestingly, after I finished testing the voltages I heard a little sizzle and noticed that the area around pin 15/16 started to smoke. IMMEDIATELY unplugged the power. Looks like there was a short between the LCD and the HM. Some solder must have bled across pins between the sandwich? SADFACE

Decided to call it a night there so I haven't checked the voltages around the TC yet to troubleshoot THAT issue.

I'm starting to feel like I should just order a (working) kit from you, Bryan lol. At least I'll know the TC is working that way!