 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Microcontroller and PC projects : Silicon Chip USB Datalogger Kit
Page 1 of 3   |
|||||
| Author | Message | ||||
norcold Guru  Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Have just finished building kit from Altronics. Had a small smd that gave my shakes and eyesight a bit of a test, but it is operational ie. the computer found it and loaded device driver etc. Now I`ve got to try and get my head around writing scripts etc. Have no experience at all here thus expect a fair bit of pain. Will get right into it after 2 weeks of salt air. Built it to log output of 3 wind turbines, hoping to adapt 75Mv shunts(current LCD ammeters) as sensors but that too is unexplored teritory. Thought I`d start this new thread as follow up to a post I recal from you Glen back a few months ago. We come from the land downunder. Vic |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi norcold I will be interested to see how you go with it as I do not have much luck with small things but would like a multi channel device. All the best Bob Foolin Around |
||||
shoots Newbie  Joined: 31/07/2011 Location: AustraliaPosts: 32 |
Hi Guys, some interesting project going on here. Just finished building the USB logger. Very handy and now trying out the scripts. Had a few issues- watch out for these: the PCB from the Altronics kit has some places where the solder would not take well. Pin 5 of the SMD was one that looked perfect to the eye but was not getting through. Despite applying flux before hand. Had to scape a few places to be sure it was ok. Also after putting it together but before plugging in the IC I checked all the voltages and no earth on pin 19 and while I did not have a PCB track diagram in front of me and the USB socket was already on- it does not seem there is provision to get the earth to that pin. Maybe there is a path through the IC. I put a link undet the PCB to ensure 19 was at OV. The push button needs a good press to ensure it pulls the input down- I was to easy on it at first and wasted some time checking. I have used windows 7 and the only way I can get the computor to recognise the device is to disconnect the battery, reconnect it and a few seconds later plug into the computor. A restore of defaults was also needed then and once I had to remove and replace the SD card for it to be recognised. I doubt it is a problem with solder joints as fexing the board does not cause it to fail in operation. until I resolve these I can log away and when I want the data ensure logging has stopped and remove SD card to get the data Twice I have left the logger overnight and despite there being no logging taking place have found the batteries dead flat in the morning so another thing to track down. When sleeping it only draws 2mA, on log it can peak at 80mA Otherwise it works very well and once the above are sorted it will be a little beauty. Hope this helps others. Regards, Peter |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Peter, Note the firmware has been updated on SC site to remove the problem re. computer not recognising whilst batteries in unit. Have not tried it as yet as found a work around by removing batteries as you did. Seems also SC considers we should purchase the issue with the firmware update before we can download the update. Had no problem re. solder taking on PCB but a dry joint may show up with use later on so will keep an eye on this. As yet have not put mine in service as am having trouble getting my head around scripts and sensors. Concept is all alien to me as yet as am basically a paint by numbers kit constructor with no computer programming skills thus will follow with much interest your posts. We come from the land downunder. Vic |
||||
| shoots Newbie Joined: 31/07/2011 Location: AustraliaPosts: 32 |
Hi Vic, Good to hear it went together well. I have not had the time to do any more with it since I put up the post. I did run it overnight with a PC based logger checking the battery voltage and it has been rock solid. The current draw was around 1-1.2mA so very economical. I can only suspect that both times I had the batteries go flat they were poor to start with. I read the note in the May Silicon Chip (Page 64) where the issue of the firmware is mentioned with respect to the batteries being drained flat: this is where the processor shutdown (due to low battery) is initiated but the processor resets itself and is no longer in sleep mode and tries to wake up, can't do this properly, and ends up in "limbo" until the battery goes totally dead (in my case one battery at 0 volts and the other 1.2, the next time they were both around .4 volts). So it sounds like my experience was exacly as the new firmware was supposed to address. Except my unit has version 9.92 which is the latest on the S Chip download site. Maybe the issue persists. When I get time I will drop SChip a line about this and also the need to remove the batteries to get the PC to recognise it. Sounds like some fun ahead learning the scripting. I Had a go monitoring the output of a heart rate monitor and it logged the frequency of that OK. I have found that after logging it goes to sleep (if the time between logs is enough) and when I called on it to do a sum after waking it was erratic (sometimes the sum was wrong or had not been performed). I made sure it had something of no consequence to do first and it seemed OK. Could be just that I have a bit to learn before I get it sorted out. I had it monitoring two low frequency inputs and that went well. Once I have something reliable to post I will put it up- but don't want to put up anything misleading while I am still getting the hang of things! Regards, Peter _________________________________ so many projects, so little time! |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Getting head around scripts(think!!), have not found a method to viabbly use existing shunts as sensors thus will have to source hall effect sensors. But I`m hanging in and will update when like Peter have something reliable to post. But am posting now to keep thread alive and current. If as some jokingly quote "We don`t cark it with unfinished projects", think I`ll live forever.  We come from the land downunder. Vic |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Am progressing with scripts ok, working with Win7 on desktop,but cannot get the software to run under XP on notebook. Success with Panucatt current sensors, set up scripts to handle these. Device driver no drama on XP but no go with datalogger.exe bit of reverse to the norm. XP seems to run every bit of software I have but this. Means without long cable runs I cannot test scripts and monitor easily. Tried the combatability modes in XP every combination. Just get error message that program needs to be reinstalled as configuration is incorrect or incomplete. Bloody frustrating because I feel there is an easy solution just out of grasp. We come from the land downunder. Vic |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Surrounded by logic but can't get to it. 
Is this like you? 
The way you feel. 
My solution. Bob Foolin Around |
||||
| VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539 |
Hi Vic I checked out my files and found the solution to your problem, mind you it isn't 100% but it works most times. 
All the best Bob Foolin Around |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Bob, I think its got a mind of its own, but seriously looks like XP has not got the supported files reqd to operate Visual C programs, currently downloading such. 99.9% of time problem would be with 7 but this time trying to run a new application under an old operating system seems to be the problem. Perhaps I`ll get a bigger hammer, and some hair restorer. We come from the land downunder. Vic |
||||
| shoots Newbie Joined: 31/07/2011 Location: AustraliaPosts: 32 |
I think you are right with the issue on XP. I have had no luck with running other programs on XP: eg a logic analyser tries to install and freezes. Also with vista but on windows7 is no problem. I ended up going to Dick Smith Electronics and the sales guys let me test installing the software on a small cheap (<$500) notebook (an HP) with windows 7 and OK. So bought it! Runs the logger software no problems. Back then I thought maybe the brand was an issue and acer would not play ball as it failed to install on two acers. Its small footprint is a bonus as bench space is limited. Sometimes wish for a larger screen or two screens.... The time these software issues waste is incredible and if you add up the value of the time lost it would be scary.....unless you write it off as a (expensive) lesson.. regards, Pete p.s. have not had the time to play with the logger since my last post. |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Up and running on XP, downloaded the visual C files from MS, still no go, went to Update MS downloaded .NET framework 1 & 3. Tis a goer. Now progress(I hope) after 1.5days in jungle. Have got calibration on a sensor(current) OK and script reading and logging date stamped. Requires a very stable external power supply to operate precisely otherwise readings are way out.(have not installed batteries as yet,perhaps that will stabilise voltage) Now to expand the script for additional sensors and new log file each day perhaps with maths functions to compute daily Whrs. Hair falling off on ground, think I`m a millionhair. We come from the land downunder. Vic |
||||
| apberg Newbie Joined: 22/09/2011 Location: NetherlandsPosts: 2 |
Hello Shoots & Norcold, Thanks for the info, I have also made the logger, pcb made in SprintLayout, original pcb was hard to get here in the Netherland. Thanks to a friend and Silcon Chip, have now also 2 oiginal pcb's. So give it a tray again in the next weeks.... ( http://www.geotech1.com/forums/showthread.php?t=17895 ) So your info was very usefull.. thanks Ap |
||||
Gizmo Admin Group  Joined: 05/06/2004 Location: AustraliaPosts: 5078 |
Hi Ap Could you check your email address in the Settings screen please. When I set up your membership the email bounced back. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| apberg Newbie Joined: 22/09/2011 Location: NetherlandsPosts: 2 |
Hello Glenn, Think it must be working good .... Best regards. Ap |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Have had no joy getting logger to operate correctly with batteries installed or with external power. Seems to lose its time unless run through USB with computer on. Latest firmware update has not addressed this problem, I`ve either stuffed up in construction or it has a problem. Very happy with its operation whilst connected to computer. Am finding some logic in scripts, really like the softwares error checking that advises of errors when compiling fails, it is a hard slog but enjoyable for this old mind. Once I am happy with script and it logs my 3 mills output will post script if it is allowed to share such.(Will contact SC on this) We come from the land downunder. Vic |
||||
| garthwimbush Newbie Joined: 12/04/2012 Location: AustraliaPosts: 7 |
I too have been having problems with getting my SC USB data logger to work properly. I have been experimenting with using a DS18B20 temperature sensor, which is a Dallas one-wire device. Everything is fine when connected to Windows XP; it logs successfully to the memory card. If I make the sleep period a few seconds in the logging loop, then I am able to disconnect the data logger from the computer, and it continues to log. If however I make the sleep period longer than a few seconds (say a minute) then disconnect, the logger manages to talk to the sensor once, then fails from then on. It all seems to be related to whether the data logger goes to sleep or not. Upon waking for the next cycle, it fails to contact the sensor. From my observations with a logic analyser, the data logger seems to go into some sort of slow-mode, where the logic is taking twice as long to execute - almost as if the clock were running slower after wakeup. The following picture shows the trace at the DS18B20 sensor data pin, when the device is initially reset in the script, using the @@resetOneWire() function. The data logger is connected to the computer, and working properly: 
- Note that logic levels are inverted. There is a (roughly) 480us pulldown (which resets all the devices on the one-wire bus), followed by a presence pulse from the DS18B20. This picture is the same case after data logger wakeup, where the pulldown is more than twice as long (over 1ms), and the @@resetOneWire() function returns a failure, even though the presence pulse from the DS18B20 is still there. 
Here is the script I have been using to do the testing: header DS18B20PingHeader { } script DS18B20PingScript { @@openOneWire(#oneWireUsingIO, 0); WHILE(1) { $RESULT=@@resetOneWire(); PRINT "Connecting to One Wire Sensor: "; IF($RESULT) { PRINT "Ok.", NEWLINE; } ELSE { PRINT "Unable to access sensor.", NEWLINE; } SLEEP(30); } } I guess this could be a hardware issue, although I dont see how - it is more likely to be a problem with how the micro-controller is coming out of sleep mode. Has anyone had similar experiences, or could offer guidance? |
||||
| garthwimbush Newbie Joined: 12/04/2012 Location: AustraliaPosts: 7 |
I just did some more sleuthing - if the data logger is running on battery power, and it goes to sleep, it wakes up only a certain number of times before apparently going to sleep for good. Pressing the Start/Stop button seems to wake it up again for a while. It runs without problems under battery power with a < 5 second sleep (meaning that the microcontroller doesnīt actually go to sleep). |
||||
| norcold Guru Joined: 06/02/2011 Location: AustraliaPosts: 670 |
Yes I ended up having it run on USB continually using a powered hub and a 15m usb extension lead to computer(elebay china cheapie), took the batteries out of it also. Works without glitches this way, plus have the advantage of having a stable power supply. I considered the problem was one I introduced in the making of the kit, perhaps a dry solder joint etc etc., but it is starting to appear there is a problem with the device when under its own battery power ie. your device reactions under battery are the same as mine. Other than that it performs well. We come from the land downunder. Vic |
||||
| garthwimbush Newbie Joined: 12/04/2012 Location: AustraliaPosts: 7 |
Wow thatīs a long cable. As long as you dont sleep more than about 5 seconds in your script, it should work on battery power. Of course not sleeping will drain the battery faster  . I havent found where in the data logger code sleeps and wake-ups are handled, but I reckon the problem is in there somewhere. |
||||
| Page 1 of 3 |
|||||
 Print this page |
