| Forum Home | ||||
| PC World Chat | ||||
| Thread ID: 140649 | 2015-11-18 20:43:00 | Question for the electrical gurus | Tony (4941) | PC World Chat |
| Post ID | Timestamp | Content | User | ||
| 1411631 | 2015-11-23 03:21:00 | Won't it be easier to aim one of them cheap and simple infra-red temp guns for a quick reading? I think though a fiddle factor may be required, perhaps a emissivity or thermal heat capacity factor for a copper/water interface.Good idea, but not worth the hassle. I want to know the temperature when I am standing in the kitchen, which I can do by reading the display on the wall. Your solution, though probably feasible, would require me to go to the airing cupboard, possibly after washing/drying the hands, and then point the gun at the pipe. I also need to know the temperature after the hot water has been running a bit so the copper pipe has heated up to roughly the water temperature in the tank. With the display I can see the temp rise in real time (although 15° low). | Tony (4941) | ||
| 1411632 | 2015-11-23 03:51:00 | It just occurred to me to wonder whether I could actually make what I want using Arduino and/or Raspberry Pi. I've never had anything to do with either, but might that be an interesting little project - or will I be biting off more than I can chew? | Tony (4941) | ||
| 1411633 | 2015-11-23 05:22:00 | If it's a type K thermocouple, you could use an AD595 like shown here hobbybotics.com along with an arduino. (Though their board layout could probably be improved upon). A Bluetooth module to another arduino running a display could be used for the readout. Might be hard if you've never done anything like that before though. I wonder if you can recalibrate the one you have. Perhaps that is how the original was set up - to compensate for thermal loss in the water-to-pipe-to-sensor setup as was already mentioned? To verify functionality\compatibility in your current system, you could try unclamping the probe from the pipe and putting it alongside your mercury thermometer, perhaps holding both in your hand for some time so they reach the same temperature and comparing them - do you still get the 15°C discrepancy? |
Agent_24 (57) | ||
| 1411634 | 2015-11-23 05:29:00 | I don't think any recalibration is possible. Everything looks sealed to me apart from the battery access. The only calibration I thought possible was maybe altering the length of the wire, but that seems to have been shot down in flames. I'll certainly try the check vs. the mercury thermometer though. I think I am kidding myself with the arduino solution. |
Tony (4941) | ||
| 1411635 | 2015-11-23 06:12:00 | Your main problem is that the mounting method of the sensor is Mickey Mouse to say the least. There is no way that it will read water or pipe temperature with any semblance of accuracy. You can't "recalibrate" the readout with such a piss-poor mounting method. So if you don't change the way it is mounted anything else such as wifi etc to put it bluntly is a waste of time. Ideally the sensor should be inside the tank, or else in the pipe measuring water temperature directly, this could be achieved using a T piece and suitable gland fitting. You could also examine to see whether the probe could be put directly next to the top of the tank under the tank lagging. That would be better than where it is now. Alternatively a copper or brass block could be hard soldered to the pipe with a drilled hole for a snug fit for the sensor and then lagged to try to achieve a temperature close to that of the pipe. A snag is that would give a slower response time. Tweak'e suggested earlier that the probe could be a resistance thermometer, and that is a distinct possibility. They come in 2, 3, or 4 wire systems. A 3 or 4 wire system is used for best accuracy and to be independent of lead length. The photo indicates two wires. If you measure the resistance of the probe, a thermocouple will be a few ohms, whereas a resistance thermometer will be typically 100 ohms. You can easily check the readout by dunking the probe in boiling water. If it was mine I'd make up an adapter for the probe to be in the water flow. |
Terry Porritt (14) | ||
| 1411636 | 2015-11-23 06:17:00 | You can easily check the readout by dunking the probe in boiling water. I was going to suggest that, but am unsure if the probe is waterproof or not and did not want to be responsible if it isn't. |
Agent_24 (57) | ||
| 1411637 | 2015-11-23 07:58:00 | The arduino projects (a few websites), is one I researched for measuring fuel consumption and other parameters in my car, but involved hard wiring and controllers. Bit involved as Agent 24 suggested. Luckily I found an app that did all I needed and more. Terry's suggestion would be the way, I think a thermistor may be required? But maybe use a secondary pipe to fit/adapt sensor, and swap it out with what you have now - then back if need be. |
kahawai chaser (3545) | ||
| 1411638 | 2015-11-23 08:15:00 | I do really appreciate all the suggestions, and the whole discussion has been very interesting, but I need to repeat a few things: The way the probe is mounted is how it has always been, poor and inefficient as it may be. When the thing was initially installed, it was accurate, as far as I know. It only became inaccurate once the broken display unit was replaced. As far as I can see there are no adjustments possible, apart from changing the length of the wire which has not been seen as a solution. The issue is not important enough to warrant some of the more elaborate solutions suggested. I'll try the thermometer test just to see how it compares. The Arduino/Raspberry Pi project ain't never going to happen, interesting though it might be. I think I'm going to stick with what I've got and add 15 to whatever the display tells me. If I see a wireless solution that is not too complex or expensive I might give that a go. |
Tony (4941) | ||
| 1411639 | 2015-11-24 23:21:00 | A wireless weather staion maybe what you want. They range in price, can include more than 2 sensors, display/collect data, have multi alarms, and can display moving time trend plots. Though temperature range might be limited for cheaper models. Plenty on ebay... | kahawai chaser (3545) | ||
| 1411640 | 2015-11-25 21:46:00 | 1) Go back to the shop & buy a sensor that should match the readout. 2) Buy a new complete setup , reader + sensor in the package. 3) Your tank thermostat is at the bottom in a long pocket, behind that round or square cover. Why not get your plumber to fit you a smaller pocket at the top of the tank for your sensor. There will still be a slight delay when the water temp changes, but it will be surrounded by hot water. not air like you have it now 3) Your problem stems from when you fitted a new reader, so the two do not match do they. Unless, of course, the problem was always there & you never noticed it. 4) Try what has been suggested earlier & somehow wrap insulation around the sensor & pipe. See if it reads differently with your glass thermometer under the hot water. PJ |
Poppa John (284) | ||
| 1 2 3 4 5 | |||||