Homeroasters - Home Roasting Coffee Community » DATALOGGERS - CONTROLLERS - RATE OF RISE METERS » Dataloggers/Controllers/Rate of Rise Meters
Who is here? 1 guest(s)
 Print Thread
Just finished Thermocouple and htc/tc4c install - temp problems
yukoncornelius
Top  #1 Print Post
Posted on 04/18/2016 11:41 PM
Hey guys... so all weekend I undertook the task of installing thermocouples into my hottop, along with htc/tc4c. I followed Barry's excellent guide with parts listed (brass lamp fittings to hold the thermocouples). Unfrotunately I ran into problems with being off on my measurements (major brain fart - luckily some washers from lowes saved me and I'm pretty satisfied with the install). The thermocouples are sitting through silicone tubing, inside the brass nipple. They are the same thermocouples from Omega which Barry recommended - HTTC18-K-18G-1.75-GG

THe bigger headache honestly was racking my brain over mounting the pcb boards - I don't know if my factory wiring was unique (in how it was zip tied with really no slack) but I did NOT have the leeway or space that Randy shows in his guides and pictures. I cringed every time I moved the board with the power/control panel cables attached because I was just fearing a wire coming loose - there's just such little room to work with. I ended up drilling holes on the right side of the hotop and letting the nylon screws rest through those holes. So I FINALLY got everything put back together tonight. The hottop seems to be properly controlled via the arduino controller now...

HOWEVER..and a very frustrating however - my temp readings are way off at room temperature. Before I even start the roast program, they're sitting at 40F. They seem to read at about HALF of what the actual temperature is. I really do think my connections are right, and I have fiberglass sleeving surrounding the thermocouple wiring. Quickly searching another thread on here - I saw someone had a thermocouple from omega (very similar to the ones I bought), and the temp was actually being read midway down the probe, rather than at the temp (which could be plausible in my case as its sitting in the silicone sleeving). On the otherhand even sitting in the silicone tube and reading from the middle of the probe, I should get a proper room/starting temp I'd think? The temps also seem relatively slow to respond, although thats hard to tell. I have a cheap thermocouple and data logger from my old freshroast sr500 days and that one is reading properly and responds very quickly. Another quick question - is the temp displayed on the hottop control panel still accurate now (or rather, as accurate as it was before)?

Any ideas or places to start trouble shooting first?

If I did this all over again, I'm not sure I would've bought the htc/tc4c - I would've cut the job in half and saved myself alot of anxiety if I was just routing the thermocouples to an external datalogger.

Thanks for your help everyone
-Joe
 
yukoncornelius
Top  #2 Print Post
Posted on 04/19/2016 8:34 AM
Ok so this is going to be rather silly.... I want to ensure that the readings I'm seeing in roastlogger are actually FAHRENHEIT and NOT CELSIUS. When I first turned the hottop on before I put the rear cover and everything else back together - I roughly remember seeing readouts of 20ish - if thats celsius, it would be around room temperature in F!!!!

Also last night, as the hottop was cooling down, 40C would be around 100ish F which seems roughly accurate also!

I changed no settings on the tc4c or within roastlogger other than selecting "restore defaults to Fahrenheit" from the arduino window (after I started communication). Is there an additional step on either the tc4c or within roastlogger which I need to do to show fahrenheit?
 
yukoncornelius
Top  #3 Print Post
Posted on 04/19/2016 10:30 AM
Emailed JimG asking him if my version of the htc/tc4c needed a jumper in place to read out in F. His answer was no so he told me to try changing axes setting in roast logger to Fahrenheit. I did that and it works!!!!!!!!!! YES!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

I did a quick preheat and the only thing I notice right now is these omega thermocouples seem relatively slow to respond compared to the exposed wire I have hooked up to my cheap datalogger. During the short preheat, when I cut heat, the temperature continued to rise over the next minute or so on the omega thermocouples (from 130 to 140ish). That seems really slow? I guess the only thing to do is test with an actual roast.


Would it be fair to compare the temperature read out of the stock hottop thermocouple (control panel) to what I see from my omega readouts in roastlogger? How should those correlate to each other as far as responsiveness?

Thanks everyone!!!!
Edited by yukoncornelius on 04/19/2016 10:36 AM
 
yukoncornelius
Top  #4 Print Post
Posted on 04/20/2016 8:05 AM
Well no one is responding so I'll just keep talking to myself lol....

Did my first roast last night, whoa - this is cool... and whoa - I'm going to need to relearn when to make adjustments, temperature points, etc....

Surprisingly my BT thermocouple and the stock hottop K thermocouple were not too different from each other leading up to first crack, however after FC - wayyyyy more data coming from the new thermocouples - this is going to be fantastic for actually seeing the real change in temperature. I think my hottop display temp was showing 399 vs my BT probe showing 434 at roast end - that is crazy.

I think this first roast of some columbian huila beans I had may have turned out surprisingly well although I went through end of FC to start of SC pretty quickly. There was a little over 3 minute gap from start of FC to start of SC. RAO ratio was 26% (love that rao calculation).


Can anyone give some basic tips or guidelines? I don't even know what RoR metric I should be shooting for. after the first crack I was at a RoR of 15 which seems very quick - I feel like I remember people saying to shoot for 7 - thats what I was always trying to achieve with the control panel display temp - although now I see that temp was totally false. But now that I see the RAO ratio - I'd think that would result in a very high RAO %
 
Jump to Forum: