 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 : Turbine Performance
Page 1 of 4   |
|||||
| Author | Message | ||||
dereksoftstuff Newbie  Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Just found your discussion forum on how to measure the performance of a wind turbine. Well, I've made a system to do just that. It's easy to build, cheap, and the software is free to download. performance Feel free to add value ... |
||||
Gizmo Admin Group  Joined: 05/06/2004 Location: AustraliaPosts: 5019 |
H Derek Welcome to the forum. Great post, thanks for sharing so much detail and making it available for others. Have you seen the Piclog logger? http://www.thebackshed.com/windmill/PicLog.asp It uses a very basic microcontroller, most of the hard work and loggin is done at the PC. You system has a many advantages over the PicLog, especially the SD card data storage and ethernet data transfer. 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 |
Hi Derek, It is good to see others on a similar track. I have my own version I call a windmill analyzer. It seems that you also see the benefits of real time graph updating of windmill performance. I have only seen the video, but if the hardware is not too complicated, is available and will work reliably then it should be of use to many. Gordon. become more energy aware |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Derek, I have just had a bit more of a look at what you have, and this is an evaluation board. I have not had much luck with sourcing info for the unit you mention. It does seem to be current technology. I have programmed similar PLC micro processors using ethernet COMMs. The latest was a BR brand, and had ADC, COMMs, digital IO, CF storage etc. This was a commercial product, as used to control 1000's of IO. These units have support software products in place, and are reasonably straightforward in setting up. I will not comment too much on the cheap price aspect, as I have not been able to source a supplier, or pricing. This evaluation board seems to require C-source code programming. The BR unit that I programmed had a choice with Automated runtime, or C coding. Gordon. become more energy aware |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Hello Gordon Here's a link to the microcontroller : lm3s6965 The eval kit is very nice, ready to go, just plug it in. It's a good all rounder 32 bit 50MHz C/C++ - free development tools Codesourcery. If you use it for development, there's 'driver' support s/w from Luminary micro/LMI (now part of TI). There's the usual set of peripherals - UART/SPI/I2C/GPIO/ADC/Ethernet/PWM/QEI, 256k flash, 64k ram. |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Hello Glenn Had a look at the PicLog - looks good. Couldn't find the results section (performance data) for the users of the kit. Could you point me to the metrics? |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5019 |
Hi Derek Online data is here - http://www.thebackshed.com/piclog/. There are a few forum members using it on this site, from time to time, and I know of another using it on his own server for logging a remote solar powered radio relay site. You can see my data here, windmill currently off line, solar only. http://www.thebackshed.com/piclog/Report1.asp?LogID=1 I have a updated version of the PicLog in the pipeline, few more features, like daily peaks and trend lines. Will release it in a week or two. 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 |
Hi Derek, I see this type of micro controller serving a more useful task as a hub that can retrieve and collate data from multiple external controllers. There is limited on board ADC, but the unit has dual I2C channels, and many IO, PWM capacity and microSD storage. I have sourced a supplier of the unit you mentioned, and I do remember seeing this type of unit in Element14 [used to be Farnell] select catalogue. I have an account and receive mailouts. The price is $69. This type of micro is like a Picaxe on steroids, mounted on a PCB, with interface hardware supplied. Many readers might be daunted by the C-programming. Maybe a possibility for another project, but the 250mA supply current is a bit off putting. Gordon. become more energy aware |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Glenn Online data looks good - not much of it yet, but I know things are only just up and running. This is a difficult thing to do, infact I don't know if anyone else is attempting such a thing - anywhere, so this is breaking new ground. I'm not sure what other users want to see, but I'd like to be able to see the 'power curve' for a given turbine. Obviously we'd have to know the basic turbine specs, then we could compare different systems. Is this the kind of direction you are going ... |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Gordon Yes it's a powerful little thing. The system I've developed is ready to go, you don't have to do any programming. Just use as is. For those who like to develop - software etc, then it's a good platform, and completely free toolchain - no licenses or purchases. C is universal and a standard, is easy to use, and lots of free third party software to use. Learn once - use for a lifetime. I'm using the platform on several projects, and have found no limitations so far. But I know if you have invested much time on one platform, it is difficult to move to another, it's all a matter of what your project needs ... |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
Hi Derek, I have found the most useful aspect of my own windmill analyzer is in setting up a windmill MPPT. The ability to electronically change the windmill loading reigime, and see the affect on windmill output performance in real time provides a rapid development path. All user designed and built windmills will have unknown power curves, and unknown optimum loading requirements. An analyzer gives an opportunity to see how changes affect performance. Many readers have difficulty with programming. The Picaxe micro, even with BASIC programming can be challenging. The Picaxe programming editor is a free download as well. I see that the evaluation board comes with some free software, but I think that there will be a lot of C-programming to get a useful windmill logging and analyzing tool. I may be wrong. I suspect many readers would be hoping for more than just the software tools, and maybe a working turbine performance program, like on the video clip. A few of us have supplied some power curves in other threads. From my own experience, the power curve has many other external constraints. Changing rotor dimensions without being able to adapt the loading and furling etc to reoptimize the windmill may show a drop in performance, when the opposite was expected. Other factors like audible harmonic noise may come into play as well. I hope other readers are able to investigate the possibilities that tools like this can have on their own windmills optimizing process. Gordon. PS edit. I have just read that the program in the video is supplied? I can see the setup etc, but the graphing and data logging does not need any user programming. These units must have come along way. become more energy aware |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5019 |
Hi Derek Yeah I do have a lot planned for the online display of PicLog data. Its just a technology test at the moment, checking the Java applets work cross browser, making a nice looking CSS, etc. At the moment the data from the individual PicLogs is fed into a simple Access database, its all standard old ASP and SQL, easy to develop. I have a bit on my plate at the moment, but plan to get back into it before the end of this year. I would like to publish a PHP/MySQL version as well for those who dabble in the dark arts. Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Gordon 1) Yes, all software (for PC and microcontroller) is freely downloadable from my share site. So the only costs are for the microcontroller and the electronic parts for your voltage & current sensing circuits, plus rpm hall sensors. You can enter your turbines specific config via the UI on PC as shown in videos. 2) Funny you should mention MPPT, another user (different forum) wanted to develop something like that based on my performance tool. I think this is a good idea, and the current software would be easily adaptable to encompass an MPPT facility. Infact the more I think about it, surely every windmill should have an MPPT as standard. Maximising the current to the batteries is what we all want. We want a windmill that produces more power for the same wind. 3) Power curves - these are what manufacturers use to sell their stuff. They seem always to be vastly optimistic. But a real (live - historic data) power curve showing data from an extended period would tell a story ... |
||||
| Gizmo Admin Group Joined: 05/06/2004 Location: AustraliaPosts: 5019 |
I've put a page together about Dereks wind doctor software, so it doesn't get lost in the forum over time. Its just a quick run down and a few links to his YouTube page, the micro-controller, etc. http://www.thebackshed.com/Windmill/articles/DerekSoftStuff. asp Glenn The best time to plant a tree was twenty years ago, the second best time is right now. JAQ |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
There are a few posts regarding wind turbine mppt spread across the forum. Although there doesn't seem to be a common approach to a DIY solution. The diagram shows a proposed solution for a mppt system : 
where the user builds their own circuits depending on their voltage & current requirements. These circuits can then interface to the microcontroller as shown. Different control algorithms can be selected via the PC, such as maintaining constant TSR, or full mppt hill climbing etc. The algorithm output will determine the duty of the pwm signal sent to the user mppt circuit, which in turn controls the load. The microcontroller monitors the output power (V & I) so can determine when to switch to a diversion load, and send a digital signal to the charge/load controller circuit to switch on/off. So the user doesn't need to do anything else but make their circuits. The software is already written, and the performance improvements can be viewed via the UI charts. Any other ideas ... |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
This philosophy is all well and good at low power levels. As soon as the wind energy starts to get exciting, [highly dynamic windmill behaviour] control systems seem to break things. The block diag you have presented is typical of what a MPPTracking regulating component is required to do. My own testing, and involvement in fixing commercial, small scale system installs and failures over the last 3 years has enlightened me to the power of the wind. Electronic Windmill MPPT systems seem to go hand in hand with failures. The most reliable windmill, that produces power proportional to the wind energy up to about 10m/s windspeeds is a good start. A windmill designed with a cutin of 5m/s, and blades and furling that limit power above 10m/s, coupled with a boost component that gives charging current in winds from say 2.5m/s and up, is another plus. The rotor blades are very forgiving when designed correctly to work with the power in high wind speeds. We may think that micro processors and our running control program are up to the task, but the windmill has a way of providing certain adverse power levels that let the magic smoke out of those so called within rating components. At the end, the micro algorithm is part of a complex control loop, the power electronics being probably more than DIY. There is much more to this than saying I back my opinions with real world data and observations from my original test windmill with an analogue boost MPPT and subsequent efforts on my new windmills to achieve similar results with digital processing MPPT using a buck and boost topology and capacitor doublers. Check my avatar of my own windmill in a storm. If we stick to MPPT, and not introduce power transmission loss issues, then it is possible to make a buck windmill MPPT. Reliability will be the key and the complexity and sizing of components in the unit makes for higher cost than a boost cct for the same power windmill. In the buck cct, connected after the main rectifiers, the electronic MPPT has to pass the full power all of the time. In the boost cct, the electronic MPPT works in parallel with the main rectifiers, so only has to be approx 1/8 the power handling. In this arrangement, the boost cct may be required to work most of the time. My own windmill boost cct has been working since 2006, without any down time. It is all well and good to be able to build a windmill MPPT, but unless the wind reigime dictates power levels to justify the equipment needed then full power range MPPT is unnecessary. I doubt that a windmill MPPT of the buck topology for power levels above 500W would be considered DIY. The equivalent power windmill with a boost converter would only need to handle 50-100W, and this could be considered a DIY project. Gordon. PS edit: I would opt for a MPPT algorithm to work with a boost cct, and compensate for the inertial mass of the rotor, and be able to regulate to a battery providing a battery maintenance function. become more energy aware |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Gordon Glad you like my design. I'm working on it now. You're right, the microcontroller you are using can not do the job adequately. If you get a chance - try one of the cortex-m3 arm microcontrollers. There seems to be a lot of talented DIY'ers here on the forum, so building circuits for their wind turbines is within their gift. With this design, everyone could measure the performance of their system (no mppt). Then add the mppt circuit and measure the performance again and compare results. Then try a different 'mppt' algorithm and compare results again ... So it will be very easy to see what works and what doesn't, as well as posting video evidence for everyone to see. The beauty of having a reliable performance measurement system, is that you can modify anything you want on your system and see what difference it makes. If you've had some experience with different 'mppt' algorithms (wrt buck/boost topologies), could you specify in more detail (inputs, algorithms, results) with any successes/failures. I'm interested in everyones input for things you'd like to be able to measure on your wind turbines, and ideas on displaying/comparing the results on the PC. |
||||
| GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127 |
I am sorry in not replying sooner Derek, but I had used up my daily posting limit. |
||||
| dereksoftstuff Newbie Joined: 14/11/2010 Location: United KingdomPosts: 30 |
Gordon I didn't realize there was a posting limit - that seems rather bizarre! Well, I only understand simple things, so the challenge is always making very complicated things, very simple. For extra analogue inputs - I've used a ADC108S052 - 8 channels 10 bit ADCs SPI. I think if you could condense your experiences of input/algorithm combinations with results, that would be very valuable to everyone here. |
||||
Downwind Guru  Joined: 09/09/2009 Location: AustraliaPosts: 2333 |
Hi Derek, I don't think there is many on the forum that has the skills to write algorithms or even able to design a circuit. (only a small handful) There is however quite a few members that enjoy reading about what others develop and some are even able to copy the work presented by some of us, thats why its important to give all the information and not just tell us what you have done. Please don't ask Gordon to post ALL his developments or even a condensed version of his work as you will be here till Christmas time next year wading through it. Most of Gordon's work is in the achieves already and would be a matter of searching back through if you want to find the information. As much as electronics controlling a turbine looks to be the answer, it also is the weak line in the system and many a mill has failed due to electronic controls, in reality the best and most reliable mill is the one with the least amount of electronic components other than a set of rectifier diodes. This is a well beaten path with a good record of success and also as many (if not more) failures. Pete. Sometimes it just works |
||||
| Page 1 of 4 |
|||||
 Print this page |
