daleyd
Member
- Messages
- 10,401
- Location
- Wrexham, North Wales
- Thread starter knighty
- Start date
hotponyshoes
Member
- Messages
- 7,815
- Location
- Somerset. Uk
What is the physical mounting location of the probe tip like?
I had a similar problem, the sensor was mounted on the top of the unit and after a few hours of operation the heat soak/rise would cm shut it all down.
It was actually functioning correctly, ie, the probe was reading the correct temperature ie, the top part was 60c but the body was only 55c and was fine running at that.
My solution was to stick a heatsink over the top of the probe with some thermal paste. That passively cooled the probe enough to let everything function as I wanted.
You could always use a heatsink with a fan if you needed more reduction.
I had a similar problem, the sensor was mounted on the top of the unit and after a few hours of operation the heat soak/rise would cm shut it all down.
It was actually functioning correctly, ie, the probe was reading the correct temperature ie, the top part was 60c but the body was only 55c and was fine running at that.
My solution was to stick a heatsink over the top of the probe with some thermal paste. That passively cooled the probe enough to let everything function as I wanted.
You could always use a heatsink with a fan if you needed more reduction.
daleyd
Member
- Messages
- 10,401
- Location
- Wrexham, North Wales
Changing the spreadsheet a little and sticking some figures in gives this, which seems to be a reasonable compromise between having it accurate at the setpoint and tweaking the resistance at higher temps. The trade off is that at lower temperatures the readings are way out - I assume the controller doesn't need the lower temps to be accurate?
So the resistance reading at 80 is (almost) the same but at 90 is reading higher resistance, so should fool the controller into thinking its actually lower temperature. If you made the series resistance adjustable you could tweak the 80 degree setting to be accurate.
So the resistance reading at 80 is (almost) the same but at 90 is reading higher resistance, so should fool the controller into thinking its actually lower temperature. If you made the series resistance adjustable you could tweak the 80 degree setting to be accurate.
daleyd
Member
- Messages
- 10,401
- Location
- Wrexham, North Wales
Sorry these figures, not the previous ones....you can see as you change the values the accuracy really goes wonky at lower temps.
Changing the series value changes the position of the curve vertically and changing the parallel value alters the "rotation" or shape of the curve - I think what you need is the base and combined curves to "cross" at your running setpoint, which means it will be accurate at that temperature, and the combined curve to be above as the temp goes higher. The trade off is that it is way off at lower temps.
that looks good to me!
I'll give it a try and report back.... it'll probably be at least a week before I get a chance to play with it
there's 2 temp sensors (in temp and out temp) - I can take the load off it so there's no temp difference between in and out, then add resistors to one temp sensor at a time so can compare real/adjusted temps, mark sure they're roughly where they need to be.
I'll give it a try and report back.... it'll probably be at least a week before I get a chance to play with it
there's 2 temp sensors (in temp and out temp) - I can take the load off it so there's no temp difference between in and out, then add resistors to one temp sensor at a time so can compare real/adjusted temps, mark sure they're roughly where they need to be.
