Needs Pictures: 0
Picture(s) thanks: 0
Results 31 to 40 of 40
Thread: DC temperature/pressure meter
-
13th April 2019, 10:21 PM #31SENIOR MEMBER
- Join Date
- Dec 2017
- Location
- Aldinga Beach
- Posts
- 478
Hi Bob...any progress on this project??
-
13th April 2019 10:21 PM # ADSGoogle Adsense Advertisement
- Join Date
- Always
- Location
- Advertising world
- Posts
- Many
-
14th April 2019, 11:27 AM #32.
- Join Date
- Feb 2006
- Location
- Perth
- Posts
- 27,807
I am still waiting on the supplier of the DC filter - not holding my breath though..
I've also had a chat with a friend who is a bit of a guru on low noise DC stuff and he says it's definitely the VFD and I really need to clean that side up of things.
In the interim I have a reasonably fair work around.
When I want to use the meter I power it via a USB battery. The radiated noise does occasionally throw a wild temperature reading or two every now and then but they are easy easy to see (-127ºC!).
The pressure reading seems unaffected.
The meter only draws 35mA so it will run continuously for days on a singe battery charge but I only use it when I have the DC running so a charge lasts for more than a week. I have the battery setup so that I can charge the battery at the flip of a switch as needed.
PowerSupply.jpg
Based on the limited number of measurements I have made at 50Hz the DC motor temperature rarely rises more than a few degrees above ambient.
This is probably because of the gale howling past the DC motor.
I haven't tested it at 60Hz.
-
27th April 2019, 02:05 AM #33New Member
- Join Date
- May 2015
- Location
- Australia
- Posts
- 2
Hi BobL -
I just caught-up with this thread now - interesting stuff.
I wonder if you tried to use something less susceptible to noise, like a Resistive Temperature Device (RTD), whose resistance changes with temp? These are good within the temp range you're interested in. The RTD comes in a range of configurations that would suit your application.
Analog Devices makes the AD693, which would allow you to convert the resistance to a 4-20 mA loop-powered signal that should be (I hope!) more impervious to noise. One the Arduino, you'd simply terminate the signal across a precision 250 Ohm resistor to give you a 0-5 volt input. You may need to modify the resistor to get to 3.3 volts, but that's pretty straightforward. I used to have a circuit schematic for this AD, but it's long gone.
Hope this helps.
- Phil
(Edit- I read some of the above link I sent and it looks like that AD693 can also amplify a wide range of thermocouples. It's a pretty handy integrated circuit.)
-
27th April 2019, 09:39 AM #34.
- Join Date
- Feb 2006
- Location
- Perth
- Posts
- 27,807
Thanks Phil, I am aware of these and will give these a go for another project.
If you read all of the thread you will see I subsequently worked out that that most of the noise is not coming via the sensor but via the PS so I now have it running off a rechargeable battery.
I have been running the DC for periods of up to half an hour on 50Hz during the last week and motoring the temperature and have yet to see the motor temp go >4ºC above the DC air temperature. I haven't tried it on higher speeds yet.
-
27th April 2019, 02:57 PM #35New Member
- Join Date
- May 2015
- Location
- Australia
- Posts
- 2
Hi Bob -
Yes, saw that you used the battery pack, which is an effective solution to be sure.
I'm confused, though - were the power supplies you trialled inherently poor or were they susceptible to the EMF rubbish leaving the three phrase speed controller?
- Phil
-
27th April 2019, 03:56 PM #36.
- Join Date
- Feb 2006
- Location
- Perth
- Posts
- 27,807
Strange/noise signals only occur when the VFD is driving the motor.
I did not conduct exhaustive tests but all of the 20 or so wall warts and USB adapters ( that were able to drive the Arduino) that I tested let noise through. But there's still more to it. EG using anything other than a short (EG 200 mm long) USB cable between the battery and Arduino lets noise in and the longer that cable the longer the noise.
-
28th April 2019, 09:49 AM #37Senior Member
- Join Date
- Mar 2012
- Location
- South Gippsland
- Age
- 58
- Posts
- 103
Hi Bob,
I’ve used DS18B20s extensively for a long time, mostly beer brewing machines. I’ve not had trouble with noise but found them far more resilient than analog solutions. Reading-127 is a timing issue rather than noise. You can use code to filter these out or alternatively lower the read resolution.
Sent from my iPhone using Tapatalk
-
28th April 2019, 10:23 AM #38.
- Join Date
- Feb 2006
- Location
- Perth
- Posts
- 27,807
-
28th April 2019, 11:59 AM #39Senior Member
- Join Date
- Mar 2012
- Location
- South Gippsland
- Age
- 58
- Posts
- 103
I found reducing resolution made readings far more reliable. Try 9 bit and see how that goes (assuming you are using higher). 9 bit returns 0.5 degree steps.
Sent from my iPhone using Tapatalk
-
28th April 2019, 04:31 PM #40Senior Member
- Join Date
- Mar 2012
- Location
- South Gippsland
- Age
- 58
- Posts
- 103
When I say timing issue, I meant issue of synchronising the reading of data/bits returned from the sensor. It may be noise from the VFD contributes to the problem in some way, but I've had read fails (maybe up to 1 in 10) using 12 bit mode with arduino without interference externally. If you are getting mostly good reads, you can eliminate the -127 degree reads in code, or make it more reliable by lowering resolution. The higher the resolution, the longer it takes for a DS18B20 to return the value. The longer the time, the more problems with timing. As I understand it, those probes work asynchronously ...
It may be that your 250ms read requests are shorter than the return time for data. I think it may take up to 500ms to return info at 12 bits.
In any case, '-127' represents a read failure which is essentially a digital error, not analog.Last edited by arniew; 28th April 2019 at 04:32 PM. Reason: Adding more info
Similar Threads
-
Rising temperature
By Allan at Wallan in forum WOODIES JOKESReplies: 1Last Post: 15th January 2013, 06:35 PM -
Problems with temperature
By Allan at Wallan in forum WOODIES JOKESReplies: 2Last Post: 5th November 2008, 10:36 PM