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 : Electronics : Just what to log
Page 1 of 7 | |||||
Author | Message | ||||
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Many readers are interested in RE, and once they have a system, then questions arise as to how is it performing. There are questions often asked, of how to size a system. It usually turns out that most DIY, are cash strapped, and so start off small, and then keep adding to it. A convenient system incorporates an RE source, a battery and an inverter supplying loads. We will ignore the RE safety management aspects. In this system we need to measure the loading, and the performance of the battery. The trouble is that to measure the loading would require measurement of either the 240VAC power, after the inverter [low currrent high voltage], or the battery current [high current, low voltage]. The battery current can be complicated, if the inverter is a charger as well. Downwind has made a very useful current sensor using a CT. [current transformer] This is an electrically isolated unit, and is suitable, with say a clip on type CT probe, to measure the 240VAC current. The 240VAC current measurement, combined with measuring battery voltage can be used to check performance of an RE setup. In combination with a logging system, a true load profile can be recorded, and the battery performance to provide the load can be compared. The RE recharging can be checked in the same manner. The question is asked periodically, "How much power is the house using in a day?" This question can be easily answered. "How is the Battery handling the loading?" This can be answered. "Is the battery being recharged properly?" This will be visually evident. These big picture questions will no longer be a problem in answering. Downwind has a suitable hardware component, and I am working on the Human interface. It is expected that this will be available in kit form. The hardware could also offer an alternative windmill logging unit. I will be testing a prototype in coming weeks. This will be a build and use system, with no access to a CRO or hardware trimming needed. Any calibration will be in software. It would be helpful for some feedback from members who think this may be useful for their own RE systems. Gordon. become more energy aware |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
A little more on the electronics. The electronics involved is a rather simple system and will not require a high degree of electronic knowledge to construct. It is designed to work on the AC wiring from a mill or mains wiring. The line voltage is not important and will work on any AC wiring. It will also read the rpm of the mill or the Hz of the inverter. It is a simple 4 wire device....two wires to the battery or a plug pack for power and two wires to the current transformer. The current transformer is a device that you simply pass 1 phase wire through a hole in it. There is no connection to the phase wire it self at all involved. All the data generated is fed to a computer via a serial cable or a USB to serial adaptor cable for logging. The data is recorded into a file and Gordon is working on a software system for simple interface of viewing. The system will work with Gizmos Piclog viewer but has some limitations which causes the odd error from time to time. I have allowed for a wireless transmitter module to be fitted to the board for wireless data transmittion but this has not been tested as yet and will be an optional extra. Wireless transmitters have limitations of distance and are subject to interferance from other devices causing coruption of the data. It also adds exspence to the project as it then requires a receiver board aswell. For the mean time it will be a simple logger direct to the computer. All feed back from members is welcome as it will assist in the finer developments of this project. Any expressions of interest is also welcome. Pete. Sometimes it just works |
||||
powerednut Senior Member Joined: 09/12/2009 Location: AustraliaPosts: 221 |
Interesting concept. I'm working on something along similair lines for my own use - not as a kit. I'm using UGN3503 hall effect sensors in clamps (kind of like in this kit: Altronics K2582 AC/DC current clamp meter ). They'll be hooked up to an AVR microcontroller, which will also be logging voltage and handling battery charge control and dump load control. I'm planning on logging all the data to SD cards - they are cheap and easy to interface to with a microcontroller and it means you don't need to leave a computer on to log the data. SD Card logging (instead of wireless) might be a worthwhile option to consider for the kits. I'll be interested to see your progress - I'm only in the early planning and experimenting phase at this point. I'm an electronics beginner, so a lot of it is "uhh, will this work?" testing and attempting to find reputable supplies of the magic smoke that I let out of things occasionally. |
||||
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I am logging 30 parameters, with 6 applications, using my LAN file server, using about 12W from my 24V RE battery system. This is a DC system. Many of us have some computer running most of the time. The SD cards. I have just viewed a program for writing data to an SD card on a 28X2. This was about 8 pages of code, and it was apparently not an easy task. Did nothing else but check for a card, and write data. There are many problems with file systems etc, and particular card needs. You may want to share your experience, having claimed that it is a simple task. I don't deny that the cards are cheap. I am looking at a carrier unit from Dontronics, but this is a fair way off. Gordon. PS edit: the altronics kit has 1mV/A sensitivity. This would need a significant boost to make it useful with most micro controllers. There is probably a problem with noise, or offset, for the output to be so low. become more energy aware |
||||
davef Guru Joined: 14/05/2006 Location: New ZealandPosts: 499 |
You can save data to a SD card without using a file system. Search here Couldn't find it a first crack. There is some method using a single file, maybe a pre-formatted SD card with a single file placed on it by a Windows box. |
||||
BjBlaster Regular Member Joined: 04/04/2008 Location: AustraliaPosts: 55 |
Personally I like the idea that my home linux server (which runs all the time) can do the logging. This means I can check the output instantly with a web browser anywhere in the world live. Also I don't have to worry about an SD card filling up as 1TB should be enough for the logger! I just wrote a basic serial app in gambas that interfaces several PICAXEs doing temperature, CT's for in/out 240V and my 12V RE network via a RS232 serial bus. It's open source, I just haven't put it on my site yet as I'm lazy Check out my projects here in: Bj's Shed |
||||
Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5074 |
This sounds like a interresting project. The space for a RF module is a very good idea. Having used a Piclog on several occasions, next time I would like a system where I dont need a PC running full time to record the data. I know you can get cheap laptops, low power PC's etc, but my laptop was expensive, to me at least, and my PC uses a megawatt just to run the cooling fans. And I dont want to buy another computer just yet, I'm still annoyed my rarely used celeron 400 died last week, I dont buy computer hardware if there is something here that might work OK at the job. So, getting back to the logger. Gordons right, adding a memory SD or usb memory stick to a PicAxe is a big job with a even bigger software overhead. But how about a custom made add on memory board using 24LC256 chips? The late Gill used the 24LC256 chips in his wind speed logger http://www.thebackshed.com/Windmill/articles/GillsDataLogger .asp, to store 2M of data on the board. The problem with this sort of setup is downloading the data, it can take over and hour to download from the logger to a PC. So how about two memory add on boards? Swap them over at the end of the loggin interval, then use another PicAxe chip to download the data to the PC. Just an idea, shoot it down as you please. Also, has anyone ever looked at what can be done with a mobile phone? A mobile phone has a built in file system and will run for many days. Is there a serial port in that connector? What about the scripting language used to write mobile phone applications? I'm totaly ignorant of mobile phone OS, could be way off the mark. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
The worse part of logging is getting the data into a form that can be easily viewed. Gizmo's original concept works really well, when run from the one computer. The single file per day system, with a date coded file name for each day produces a searchable directory of files. This is great to browse through, just to see what happened. I see problems occurring with having to download parts of a days, logged data, and splicing this into previously saved files. I have tried this and would not recommend it without an automated process. No doubt, systems will improve over time, but at the moment, this will be a PC type application. Gordon. PS: With port forwarding enabled on the router, and a fixed IP, I think any server could be viewed from the web, not just a Linux one. If the onboard SD card was used, then there would need to be a RTC available, to time and date stamp the data record. become more energy aware |
||||
powerednut Senior Member Joined: 09/12/2009 Location: AustraliaPosts: 221 |
I'll certainly post the schematic and code when its all up an running. The interface doesn't appear all that difficult on AVRs - maybe its harder on PICAXEs, i'm not familair with those. For example code see: Atmega with SD card + FAT32 If you don't bother to use a filesystem the code comes out a bit smaller (still about 7 pages for the SD card interface code alone though). The avrs at least have lots of space/code execution capacity left over for doing other tasks such as data logging. I don't really see the code size as a problem for anything other than maintainability. I'm aware the altronics kit is unsuitable as is. It uses a couple of op-amps to scale the output from the Hall Effect sensor into an appropriate range for multimeters (apparantly most have the mV scale as their most accurate). I was merely posting it as an example of using a hall effect sensor to measure current, and because the way they have the sensor in a torrid core is the way I'm hoping to do it. Your reasons for not using the SD card are good - I was offering it as a suggestion, because at the moment my RE system is too small to run a computer 24/7 (currently a single 100 Watt panel). Other than that your design does what I want, and i'd be inclined to buy a kit rather than design my own. To be honest I'm still fairly inclined to do it anyway, for when my need increases (RSN) |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
Hi powerednut There is one draw back with what i am doing in circuit compared to how you intend to use a hall sensor. With a hall sensor you can measure the DC amps and is better suited to the current input on a solar system. With what i have done with the use of a current transformer (CT) is only suitable to measure from a AC source, as a DC source will produce no voltage within the CT. The logger design was only intended to be used with a mill or off mains wiring (including inverter). There is no need to monitor DC Amps with a mill as the AC will give better results and the mill RPM is also extracted from the mill AC. My thoughts was to design a simple little circuit that was easy enough for a novice to electronics to be able to construct and get some real life data back from their mill or monitor their power consumption. I never set out to build a bells and whistles logger as this makes it too complicated for a novice and requires a Cro etc to be able to set the circuit up. All that is really needed is to be able to record the mill RPM, the AMPS to the batteries and the battery Voltage, then the Watts and Watt/Hours can be worked out in software. This is all that this logger will do. There is an input left to allow an aneomometer to be added for wind speed if one desires. You might say a base model logger. I do agree with the SD card storage or usb stick but this makes it outside the novice level. I do have plans to design a logger of this type but it adds $$$ to the project fast and requires a much higher skill level in electronics to construct and get working. A good example of this is Gizmo's voltage controller it is basic and simple and has been constructed by many because it is within their ability. My intension is to offer a simple kit to make the dream of logging ones mill a reality. In the long run it will help to make for a better forum with realistic data taken over a period of time and not just peaks and troughs of data. At this point it is still a prototype and the guru Gordon is testing the unit out and is compiling the computer software to go with it. I must add a Thankyou to Gordon for doing this and i know he to would like the rest of you to be able to enjoy looking at actual data from your own creations. Heres a photo of the prototype board. Pete. Sometimes it just works |
||||
powerednut Senior Member Joined: 09/12/2009 Location: AustraliaPosts: 221 |
Hi Pete, Yup, currently my set up is solar only. I do plan on having a wind turbine in the near future though. The hall effect sensor will still work for AC. The prototype certainly looks simple enough - congratulations on a very elegant design Actually I had an idea about the SD card thing last night. Rather than implement logging to SD as part of the kit why not just implement a seperate, stand alone, logger? It would just read the serial output from your device (or any other serial output) and write the data to an SD card. I.e it plugs into your device where the computer normally is and writes the data (with time stamp) to the SD card. That way it could also easily be used with other projects Anyway, if enough people are interested I'll take a crack at designing and prototyping a serial data logger. |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
powerednut If you worry about what the interest of others is, you will not get a reason to have a crack at the logger design. I see it a little different and do these thing in the interest of others. I for one would be interested in what you design and if its a suitable add on with SD than it would be a worth while project i think, as there is many using a pc that could just plug a sd adaptor into the D9 socket instead. A word of caution with using the hall sensor to monitor the DC current. Over a little time the core will retain some magnetism and throw all your readings out of wack. To use this system for quick periodic tests is ok but to have it connected 24/7 i think you will get large errors in your data. Glenn tried a simiular system and found the same problem i think, and resorted back to using a shunt. This was the main reason i choose not to go this way myself. Pete. Sometimes it just works |
||||
powerednut Senior Member Joined: 09/12/2009 Location: AustraliaPosts: 221 |
Thanks for the tip Pete about the hall effect sensors - i'll keep it in mind. Actually i've pretty much decided to do the SD card adaptor. I'll move it off to a different thread though, rather than hijacking yours (further ). |
||||
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi powerednut, I will be trying a commercial serial SD adapter already available, for about $30 or so. Is this along what you are looking at? Gordon. become more energy aware |
||||
vasi Guru Joined: 23/03/2007 Location: RomaniaPosts: 1697 |
I will be trying a commercial serial SD adapter already available, for about $30 or so. Is this along what you are looking at? Gordon. That is the best solution for a PICAXE system, involving only one pin from the logger. Is cheaper than doing all circuit by yourself (making a Serial SD with a Microchip or Atmel microcontroller) and is available to anyone. So, the kit can be made without this serial SD, but having the option to connect to it. If anyone want to have a SD, then can buy that separate Serial SD module. Is the best solution in long therm, if the alternative is to keep a PC running all the time. It is expensive looking at it as a separate component but is the best solution for RE. I vote for it. Vasi Hobbit name: Togo Toadfoot of Frogmorton Elvish name: Mablung Miriel Beyound Arduino Lang |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
The SD boards i looked into some time back was these SD & USB Boards I have not tried one as yet but am very tempted now i have a working basic logger up and running. I have no interest to compete with others in this quest and am happy to review the findings of those that choose to go before me. With years of tinkering in electronics and desiging many projects i think it is far from cost effective to try to construct and write the software for a home brew SD/USB interface board when you can buy one for around $30-50.00 considering what is involved for the home user to construct one. The whole basic idear behind most projects i do is for others to be able to repeat my circuits with minimal electronic knowledge and abilities, and an of the shelf board is often the way to go for some purposes. Although in saying that i would still like to have a go at constucting one someday, just to learn how the little suckers actualy work. Pete. Sometimes it just works |
||||
SSW_squall Senior Member Joined: 20/03/2010 Location: AustraliaPosts: 111 |
The monitoring of RE system performance is good question that needs answering... I'd be happy to give some input on the hardware side of things: Given the current transformer is measuring AC, how is the power factor accounted for?? The AC current waveform coming from a F&P PM alternator is going to be very spiky and non-sinusoidal, making a measurement of the RMS power delivered quite difficult. Also the inductance of the alternator will cause extra phase shift. As a side-show, has anyone seen or used the Silicon chip RE battery monitoring kit?? SC USB battery monitor kit - Altronics It's got USB that can be used to log data and can meter the net power into and out of a battery AB Einstein: Everything should be made as simple as possible, but not one bit simpler |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
SSW_squall To answer your question...bugger if i know! I precision rectiifer the ac input and get a linear signal output,as well as generate a square wave portional to the input frequency for Hz or rpm. The phase shift is relevant to both signals and in software the signal readings is sync together. If you would like to be involved in desiging an advanced logger with bells and whistles than i and a few others could be tempted, but i have tried several times to get a online project running but hit dead ends every time. The one thing i have learnt on the forum is not many others have the abilitys a few of us take for granted and the need to keep it simple so others can construct a working circuit is paramount. If it requres some trade off's to keep it simple than a basic system in use is far better than no logging system at all. Pete. Sometimes it just works |
||||
KarlJ Guru Joined: 19/05/2008 Location: AustraliaPosts: 1178 |
I have no valuable input here.... one of those electronics challenged members all sounds better than the $500 piece of kit I use to log windspeed and temp. It'll do heaps more but all the add ons cost a fortune thus i think this project is a winner Luck favours the well prepared |
||||
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
I am toying with changing the board to take a higher quality transceiver to give stronger wireless data transfer should the option of a wireless system be choosen. What is the thoughts of others on this as it would add an instant $50.00 to the project for the 2 transceivers alone. Then there is still the cost of a decoder board at the PC end regardless of what wireless modules are used. This all makes a cheap logger start to get up in price and complicates things a little more. Should i just keep it as a simple lower cost kit that plugs direct into a computer via serial or USB or go the wireless route. Pete. Sometimes it just works |
||||
Page 1 of 7 |
Print this page |