|
Important: At this time, April 2008, this article is being updated frequently. If you have questions or corrections you can contact me at kmyersefo@mac.com
Return or Go To the Electric Flyers Only of Southeastern Michigan (EFO) Homepage By Ken Myers Originated on December 04, 2007 Last Update: July 5, 2008 Introduction
Example 1: Real World Application - Dymond RC Flite 40
Watts In (power in), Watts Out (power out) and Efficiency Cubic Wing Loading: An Explanation A Performance Factor Typrical Onboard Radio System (ORS) Components K2 Energy 26650 cells - another lithium iron phosphate cell How I Have Been Zip Charging a 3S1P ANR26650M1 (2300mAh) A123 System, Inc. Pack HXT 42-60/06 Motor Review - now known as the Turnigy TR 42-60C Lithium Face Off: A Head-to-Head Comparison of Li-Po, M1/A123 & Emoli - also includes the effects of ambient temperature on these cells Timing Test - demonstrates the effects of different timings on power system output TowerPro 3520-6 #1 Motor Review - This is part of the review for the Sportsman Aviation Sport Stik 40 ARF Low Wing Review on RC Groups. TowerPro 3520-6 #1 (second review) Motor Review - this is the second review of the same motor now being used in the Sports Aviation Ryan STA 40 ARF TowerPro 3520-6 #2 part of the reivew of the Sportsman Aviation Sonic 500 25-46 ARF on RC Groups. Also contains a review of the Jeti Spin 44 ESC and Spinbox programmer. TowerPro 3520-7 Motor Review All motors and cell chemistries can be used in a myriad of ways to power model aircraft. No one way is more correct than another, as long as it works as desired in the application!
By Ken Myers February 2008 There are standards in Lithium based cylindrical cell sizing that are important to know about. There are many different Lithium based rechargeable (secondary) chemistries used in cylindrical cells.
I have personally used the ANR26650M1 cells produced for A123 Systems, Inc. for about eighteen months. Besides my personal experience, I owe a HUGE thanks to Charles of Haralson County, GA who is known as everydayflyer on RC Groups. He got me interested in these cells and has been a great contributor to the knowledge base about these cells on RC Groups.
A123 Systems Inc.
and A123 Racing (appears to be a division of the above)
APR18650M1 (1100mAh)
ANR26650M1; A123 cells, M1 cells, DEWALT 36V cells [DC9360 10-cell], DEWALT 28V [DC9280 8-cell], DEWALT Lithium cells, DEWALT 18V [DC9180 5-cell, probably] APR18650M1; Black & Decker VPX, VPX cells, smaller A123's Summarized, paraphrased and annotated from: www.a123racing.com/SpecSheets/A123FAQ.pdf This information applies specifically to the ANR26650M1 2300mAh cells but can be generalized to the APR18650M1 1100mAh cells. 1.) The cell is cylindrical in an aluminum canister. It has a nominal voltage of 3.3V and a charge voltage of 3.6V. It has a capacity of 2300mAh, and is capable of 30C (69A) continuous discharges and 60C (138A) pulse (10 second) discharges. Each cell weighs 70 grams (2.47 oz). (See specific notes about the APR18650M1) 2.) A special electronic speed controller (ESC) is not needed to run these batteries. The low voltage cutoff should be set to 2.0V per cell or it can be turned off. (KM Note: It is best to fly timed flights with these cells. There is no knee to warn of lowering power. It's a "cliff" and it drops off "right now!") 3.) There are no special instructions for protection during use or charging. Treat it as you would any other battery. 4.) Balancing is an important precaution when using Lithium batteries. Batteries made up of these cells are not as prone to as much individual cell voltage variance as other batteries, but balancing keeps the pack in good health and ensures maximum cycles. 5.) Voltage sag is how much the voltage drops during the course of a discharge. A Nickel Metal battery's voltage sags throughout the complete discharge. A Nickel Metal battery is only operating at full performance for part of the discharge. The batteries being produced by A123 Systems, Inc. show very little voltage sag during the discharge. (KM Note: until the "bottom drops out" at the end) 6.) These batteries have the fastest charge time for any RC battery. They can be charged to full capacity in 15 minutes or less with a charger capable of providing the input amperage and voltage. Charging at these high rates seems to have no effect on the cycle life of the pack. 7.) These batteries are very safe, and abuse tolerant. They have many safety advantages over Lithium Polymer batteries.
8.) These batteries can be stored at any state of charge for short periods of time 3-5 days. They can also be stored safely for long periods of time. At a 50% to 100% state of charge, the batteries can be stored for 6 months. They can be stored for up to 24 months if they are charged to 100% state of charge beforehand. 9.) Up to 1000 cycles can be expected before reaching 75% capacity. In an average RC application, expect to see over 300 cycles before noticing any change in the battery pack. (KM Note: This has been independently confirmed by several individuals.) 10.) The battery can be charged immediately after use. I am extremely pleased with the safety, performance, ease of care and charging, fast field charging time and longevity of the ANR26650M1 (2300mAh) cells, previously manufactured by China BAK for A123 Systems, Inc. It appears that Enerland Co., Ltd., which is now a division of A123 Systems, Inc., is now the manufacturer.
At this time, April 2008, the cells are available from many sources.
Updated June 26, 2008 For a while the prices on individual cells was approaching what the cells were "costing" in DEWALT DC9360 packs on ebay. It appears that the supply of DEWALT DC9360 packs is drying up on ebay, and they are now pretty much in the $160 to $170 range with shipping. Meanwhile, A123 Racing, the supplier of the cells to our suppliers has threatened our suppliers with a notice to cut off their supply if they do not use MAP (manufacturer's advertised pricing). In other words, price fixing is in affect on these cells. It is NOT our suppliers' fault! Sources for individual cells and packs are listed below. |
|
Unlike Li-Po and Emoli cells (another type of "cased" Lithium based cell), which have a termination voltage of 4.2v, the cells from A123 Racing have a termination voltage of 3.6v or 3.7v and need to be charged to that termination voltage. If you already have a NiCad/NiMH or Li-Po charger, Sid Kaufman has an adapter to be used with those types of chargers for charging these cells. He calls it the "Dapter". To find out if your charger will work with the "Dapter", visit this page - Newly Improved! Dapter (a.k.a. LiPoDapter+). If you do not already have a charger, I recommend the Astro Flight Lithium Ion Charger for A123 Cells or the Tejera Microsystems Engineering, Inc./TME Xtrema. The Xtrema charger also has the capability to be an in-line power meter. Both chargers can charge up to 10-cell packs from a 12v DC source like a Marine/RV deep cycle battery or power supply. If you already have an Astro Flight 109 Li-Po charger, you may purchase a new chip from Astro Flight. It is called the upgrade micro chip for model 123 Lithium Ion Charger. Replacing the original Li-Po chip with the new chip modifies the charger for ANR26650M1 (2300mAh) and APR18650M1 (1100mAh) cell use only. If you wish to be able to charge both Li-Po and these types of cells with your AF 109 then you can use this thread to modify your AF 109 - AF109 hardware hack to charge A123 cells by Pat Mackenzie. The Hyperion EOS 1210i revision A is said to be able to do 12 of these cells. The power limit for this charger is 180W so it might be able to charge a 12S pack of these cells at about 4 amps and a 10S ANR26650M1 2300mAh pack at not quite 5 amps. IMPORTANT NOTE! The following methods are only for those who know what they are doing. Misuse can result in personal and property damage! These cells may also be charged using a power supply such as the MASTECH HY3020E or MASTECH HY5020E.
For cell combinations where the pack can be broken into three cells groups for charging (i.e. 3, 6, 9, 12 etc.), a Marine/RV deep cycle battery or two in parallel, to increase the capacity NOT the voltage, may be used to directly charge the 3-cell pack(s) from the Marine/RV deep cycle. A How-To for a 3S pack may be found here. If you understand this set up, then even greater numbers of these "3-cell" packs can be done.
One of the big advantages of these cells is that they are very robust and seem to withstand over-charging and over-discharging quite well. They do not seem to need as much balancing as Li-Po packs, but it appears to be a good idea to have balancing leads on them and balance the pack(s) on occasion using the Astro Flight "Blinky" Battery Balancer for A123 Cells. Here is a thread on RC Groups about balancing these packs - Yet another A123 Thread / Balancing, is it necessary? As mentioned before, everydayflyer is an excellent resource for information about these cells. Here is a post that contains links to just about everything you might want to know about these cells. David Theunissen, of the UK, has more useful information here. Recommended background reading on the lithium iron phosphate battery (LiFePO4) for the type these cells belong in. Quoted and paraphrased from the above source: 1.) LiFePO4 was developed by John Goodenough's research group at the University of Texas in 1997. 2.) In 2002, Yet-Ming Chiang and his coworkers at MIT (Massachusetts Institute of Technology) reported that they had successfully doped the cathode with appropriate cations1 - such as aluminum, niobium, and zirconium allowing development to move forward. Products using the doped nanophosphate materials developed by Prof. Chiang are now in high volume mass production by A123Systems and are in use in industrial volumes by major corporations including Black and Decker, DeWALT, General Motors, Daimler, Cessna and BAE Systems among others. 1. Cations (cat-eye-ons) are positively charged ions. Cations have fewer electrons than protons. There has been ongoing litigation between the University of Texas, MIT and A123 Systems, Inc. about patent issues regarding these cells. Rick Page of Victoria, BC Canada posted some interesting information on RC Groups. "A123 says they are producing batteries at their own plants now, but because of the continuing legal actions they are not being more specific. BAK indicated that they may use their A123 tooling with Phostech electrodes to make an A123 substitute but only time will tell. All of which should be very worrisome for A123 investors. The industry seems to have decided that their patent may be invalid. The reason for the A123 high current advantage that was stated in the patent has now turned out to be incorrect and their patent may infringe on the one originally issued to U of Texas and now held by Phostech. Rick." "This is some of what BAK disclosed for their reasons to terminate their contract with A123. BAK is 'the Company'. Quote:
Rick" Also: On February 6, 2008, I was made aware that A123 Systems, Inc. had purchased Enerland Co., Ltd of Korea. Enerland, the manufacturer of extremely high quality Li-Po cells used in Polyquest batteries, FlightPower EVO batteries and more, is now a division of A123 Systems, Inc. and is sharing marketing with the A123 Racing division. My research on the Internet showed that the acquisition was completed in August of 2007. This most likely explains who has been manufacturing the cells since China BAK backed out their deal with A123 Systems, Inc.
What this all means to us, I am not sure at this time. It seems that the DeWALT packs have been increasing in price on ebay, which may indicate that the supply is getting shorter or that the manufacturing costs have risen. What is actually happening at this time is unknown. March 16 - GE buys into A123 Systems, Inc.
CHRIS MORRISON | MARCH 5TH, 2008 In two connected investments, General Electric has put $4 million into Think, a Norwegian electric car manufacturer, and $20 million into A123 Systems, which manufactures batteries that are used in the cars. Both companies are already well funded. Think has taken over $80 million to date and A123 has topped $150 million (past coverage here). The new investment by GE makes it the single largest shareholder in the latter company. A123 is also planning an IPO sometime in the next year, and Think is preparing to roll out the Think City in Europe. 1.) Duration: Flight times for the 2300mAh cell are good (~ 7 minutes), and about equivalent to a 20C 2700mAh Li-Po, which they no longer seem to produce. Most manufacturers are now jumping from a 20C 2500mAh to a 20C 3300mAh Li-Po cell. 2.) Weight: A 6-cell pack made up of ANR26650M1 2300mAh cells weighs a bit over 17 oz. with wiring and connectors. A 5-cell 2500mAh 20C Li-Po (~$148) (equivalent voltage to a 6-cell pack made up of ANR26650M1 2300mAh cells) weighs about 11.3 oz. and flies for about the same duration. A 5S1P 20C 3300mAh Li-Po (~$178) weighs about 14.8 oz. but has about 440 usable mAh, or about a minute and a half longer flight time based on the 35-amp static draw that I have recommended. All flight times will vary, as the pilot's skill, aircraft's task and RTF weight varies. 3.) Form factor: There are times when it is much easier to get the basically low-profile brick type form factor of a Li-Po pack to fit well into a given plane. 4.) Easy Availability of Li-Po packs: The availability of pre-made Li-Po packs, to power our electrically powered models, is extremely high with many mAh capacities, power ratings and physical form factor choices. They come ready to use. 5.) Ease of charging: With the very large number of Li-Po packs being sold, finding a charger is much easier for Li-Po cells, as well as Nickel type packs. 6.) Ease of motor selection: A majority of power systems, recommended for todayÕs aircraft, are based on Li-Po use and require some serious rethinking when using the cells from A123 Systems, Inc. 7.) Cell voltage difference: Because the single cell voltage of LiFePO4 is quite different from Nickel based cells and Li-Po cells, direct conversion in existing systems is sometimes difficult. Using 100 watts in per cell makes figuring the completed, ready to fly (RTF) target weight very easy. A two pound (32 oz./900g) plane, 2 cells, a three pound (48 oz./1350g) plane 3 cells, etc.
Table 1 shows the anticipated battery weight including balance leads and plugs, power leads and connectors, shrink-wrap or equivalent tape and Velcro for two-cell through ten-cell packs of ANR26650M1 2300mAh cells. It is not really necessary to know the weight of the pack when selecting the appropriate power system components (motor/prop/ESC) and airframe, but the table gives a realistic idea of how much these cells weigh when configured into packs. A single ANR26650M1 2300mAh cell has a diameter of 1.045 inches (26.5mm) and length of 2.6 inches (66mm). In the USA, pieces of 1-inch diameter wooden dowel rod can be cut to the appropiate length and then taped together to form a dummy pack to try different cell configurations in a given project to see what pack configuration will fit best into the desired location in the plane. The Maximum Completed Airframe (MCA) weight is the key element when selecting an appropriate power system. Tables 2 and 2a show MCA weights and suggested wing area ranges for two through ten cell packs to be used for sport and sport scale planes. As always, there are a lot of exceptions, but in general, these numbers work relatively well for prop driven sport and sport scale aircraft. The Maximum Completed Airframe (MCA) weight includes everything that is not part of the onboard radio system and its installation weights and the motor and battery components and their installation weights. The suggested wing area range is based on typical cubic wing loadings (CWL) for these types of aircraft. Both somewhat larger and smaller wing areas may also be used successfully. The MCA weight, with its resulting RTF target weight, is the primary determinate in selecting the most appropriate number of cells with the wing area being a secondary consideration. If you already have an Almost Ready to Fly (ARF) model kit, it is quite easy to determine the MCA weight and the resulting RTF target weight. Weigh all of the parts that make up the airframe including the landing gear and wheels to be used. Add the parts' weights and you have a number close enough to use for the MCA weight. If you already have a builder's kit onhand, weigh all of the parts for the airframe, the plans and don't forget the landing gear (main, nose or tail) and wheels to be used. Add them together, and it should yield a reasonably close MCA weight. When doing a glow conversion, if you don't have the airframe at hand, a best guess estimate for the MCA weight would be about 60% of the highest advertised weight. If you can find a review of the plane, the weights of the actual components the reviewer used could be subtracted from the total weight to give an approximate MCA weight. Another way to estimate what the RTF target weight of a glow conversion, without using the MCA weight, might be is to increase the highest advertised weight by 15%. To do that, multiply the highest advertised RTF weight by 1.176. Use that as the RTF target weight for figuring the number of cells. The MAC weight then becomes unnecessary, but can be estimated at 1/2 the RTF target weight. There are not a lot of sport and sport scale planes designed for electric power systems. Many of those that are available are also designed for 3D aerobatics. These tend to be quite light in structure. Since Li-Po cells are usually recommended for these planes, increase the highest advertised weight by 10% (multiply by 1.111) to derive a RTF target weight for selecting the number of ANR26650M1 2300mAh cells. Many manufactures and suppilers provide the wing area, but then again, many don't. Some who do supply this information don't get it right. When the plane is available, measure and compute the wing area. If the plane is not physically present, you'll have to rely on manufacturer/supplier data. A quick look at Table 2 shows that the MCA weight is approximately 1/2 the RTF target weight. Simply double the MCA weight for the approximate RTF target weight. Use the approximate RTF target weight, in pounds, as a guide for the selecting the number of ANR26650M1 2300mAh cells to use for the project. Round up the number of required cells when the RTF target weight, in pounds, includes a decimal greater than .45. I have found that, on average, for these types of planes, the onboard radio system weight is about 12.5% of the RTF target weight. The ORS weight may include the radio receiver, switch harness, Electronic Speed Control (ESC) with or without a Battery Eliminator Circuit (BEC), servos, servo extensions, push rods, control horns, plywood used to mount the servos, onboard receiver battery or switching BEC for planes with a cell count over 3, or any parts of the radio control system. It is not necessary to know the Onboard Radio System (ORS) weight for selecting a power system, but here is a table that gives an idea of the ORS weight for reference. The actual ORS parts used depend on the individual plane. Here are some typical examples for ORS components. Sometimes, as is the case with many ARF type planes, the servo mounting plywood will be installed in the airframe. It does not make much of a difference when determining the MCA weight and resulting RTF target weight. The first step in selecting an appropriate motor and prop combination is to figure out what props match the airframe and its mission, in this case, sport and sport scale non-3D aerobatic flying. There are many props available from various manufacturers and suppliers. Again, to keep it simple, I use only APC props. I do not use the SF (slow fly) or Pylon props as they are inappropriate for this type of flying. APC props may or may not be the best for a particular application, but they are readily available and work well in most cases. For Sport and Sport Scale planes I recommend a prop disk loading (PDL) of between 75 oz./sq.ft. of prop disk area and 120 oz./sq.ft. of prop disk area. To simplify this process I have a created a table of prop diameter sizes for each ANR26650M1 2300mAh cell count. Recommended prop diameter table Larger prop diameters are more efficient. Sometimes it is necessary to limit the prop diameter because of landing gear considerations. Whenever possible, use the largest prop diameter the airframe can accommodate with a pitch and RPM combination that will allow at least the minimum recommended pitch speed to be reached. In general, typical sport/sport scale planes have a pitch speed ((RPM * pitch in inches)/1056) between 50 mph (80.5km/hr) and 70 mph (112.5km/hr). Using the stall speed and 3.5 times the stall speed, I have created tables that show possible APC props and the minimum required RPM for the appropriate pitch speed for each ANR26650M1 2300mAh cell count and matching airframe combination.
2 cell table
The manufacturers and suppliers do a terrible job at helping us to select an appropriate, suitable motor. Table 3 shows what I believe to be the appropriate "typical" weight range for brushless outrunner motors for sport and sport scale planes. It provides a starting point for selecting a motor for a given sport or sport scale project. The Weight Table was calculated using 1.5 watts in (the heaviest motor in a group) for each gram of motor weight and 3.0 watts in (the lightest motor in a group) for each gram of motor weight. I have read reviews where the reviewer has used between 3.5 and up to 4 or more, but for these types of planes, the motor ends up being too light and, it is being "worked too hard." It is not always "best" to use the lightest motor in a group of similar motors. A heavier motor will make balancing the plane easier and it will be "working" easier. The prop adapter, motor mount, prop, mounting hardware, etc. can add another 30% of the motor weight to the total installed motor weight for this type of project. Knowing the appropriate motor weight for the project is the first step in selecting an appropriate brushless outrunner motor. The second step is determining the appropriate motor Kv. I created a method to estimate the approximate appropriate Kv range. Once the motor weight and Kv have been determined, a motor can be selected that may work in the intended application. The process uses many steps to determine the Kv range. Included in the process are; RTF weight Wing Area Stall speed 3.5 times the stall speed Suggested Prop diameter The method uses suggested prop diameters based on prop disk loading and the required pitch speed. The required pitch speed is based on the stall speed to pitch speed ratio to match the pitch. Table 4 shows what I believe to be the appropriate prop diameter and Kv ranges for typical sport/sport scale planes for the number of ANR26650M1 2300mAh cells in the pack. Sometimes a slightly higher Kv or slightly lower Kv than recommended may be used in a specific pack size group. If real world testing shows that a motor is swinging an appropriate prop for a group at the required RPM, then that motor may be used in the group. One specific example is the HXT 42-60/06 (now known as the Turnigy TR 42-60C) in the 5S group, even though the Kv is lower than the recommended Kv for 5S packs. I've also identified a couple of TowerPro motors that also work "outside" the Kv group, because I have the data to make reasonably accurate predictons about their performance. The higher the Kv in a given range, the smaller the diameter the prop will have to be to be pulling only about 35 amps. It also may be necessary to use the next lower set of Kv numbers to get the larger diameter props to only pull about 35 amps, but then the RPM may be below the desired pitch speed. In the following tables, using Drive Calculator, I have suggested possible props that meet the diameter requirement and RPM required for the pitch speed when pulling about 35 amps at sea level and 17-deg C. The vast majority of motors are noted with no usable prop data available (NUPDA). This means that the manufacturer/supplier does not have sufficient data available to make a reasonable "guess" as to what prop might be used with any given number of ANR26650M1 2300mAh cells, or other cells for that matter. Caution: It is best not to choose a motor at either Kv extreme for a given cell count. If you already have one of the motors listed at the extreme, you might consider testing it, but may need a different motor. A motor with a Kv that is too low might require a prop with a diameter that will not work with the landing gear clearance for grass field take offs. A motor with a Kv that is too high might not be able to turn the minimum recommended prop diameter with the required pitch to meet the minimum pitch speed without going way over the 35 amp desired amp draw for this application. For an example of how I would go about selecting a motor for a proposed application, read Example 2 - A Proposed Application. Showing 2-stroke & 4-stroke Glow Equivalents For Comparison 2S Motor Suggestions 3S Motor Suggestions 4S Motor Suggestions 5S Motor Suggestions 6S Motor Suggestions 7S Motor Suggestions 8S Motor Suggestions 9S Motor Suggestions 10S Motor Suggestions By Ken Myers (A Fully Detailed Review can be found on the RC Groups thread: www.rcgroups.com/forums/showthread.php?t=735972 KM)
Step 1: Determine the MAC weight - All of the ARF kit parts that might be used in the conversion weighed a total of 1295.2g or 45.69 oz.
I originally flew this plane with a TowerPro 3520-7 using an APC 12x7 sport prop and a 6S ANR26650M1 2300mAh pack. It was okay, but that motor is really "best" used with a 5S ANR26650M1 2300mAh pack. I wanted to go with a larger diameter prop for better prop efficiency. I didn't want to spend much. I chose the HXT 42-60/06 (now known as the Turnigy TR 42-60C) because it was the cheapest and I wanted to know if it was "adequate". It worked okay and allowed me to compare it to my AXI 4120/18, which I run wiith a 6S ANR26650M1 2300mAh pack in my Fusion sport plane. By Ken Myers - 11/24/07
The proposed model, by design, has 416.5 sq.in. of wing area and will be about 3 pounds or less, making it suitable for a 3S ANR26650M1 2300mAh pack. Since it is being specifically designed to use a 3S pack, it has ground clearance to use an 11-inch diameter prop. It will have a CWL of 9.75 oz./cu.ft. or less.
The same process is used. Using the table, determine 10-inch prop appropitate motors by their Kv numbers including any NUPDA noted motors that fall within the Kv range.
If and when this plane gets off the drawing board, what am I really going to use? I already have a Hyperion Z3019-10 that is sitting in a drawer, therefore, I am creating this plane to use that motor. Ideally, I would use the Scorpion S3020-12, if I were purchasing a new motor for the project.
I have gathered a lot of information about typical electrically powered sport and sport scale planes and have found the average CWL for sport/sport scale planes to be about 8.5 oz./cu.ft. and about 11.5 oz./cu.ft. for Advanced sport/sport scale planes. Those two numbers were used for the recommended wing area ranges in Table 2 and Table 2a.
The following are some examples of typical ORS system components and their measured weights. The components listed are for example only. The builder of the model must determine which components are "correct" for the specific use and will be SAFE to use. DO NOT skimp on servo power!
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
