www.elektron.si

[Spajky]

@ HotPepper : Hi & welcome to the main slovene WLAN site forum!

Damn, if you really used that web based translate service (any of them give a lot of times very funny translations results), you really must be very interested in my experiments, since doing that surely was a tremendous job & effort to finish it; I couldn´t even imagine someone to start a job like that & thanks for the compliments. I invited only few times foreigners just to take peak only at the pics and numbers & get some fun, if there on the usenet was any debate about fast DIY homemade active wifi antenna stuff; never thought that someone would try to translate sentence by sentence the whole article ...

Yes, I´ve seen about a year ago & checked all those links you gave us & studied that stuff also a bit that time ago too.

Lets go on my stuff: (replicable for dummies on 2,4GHz stuff)

- Enh. B : your thinking is correct & that is a 185g tuna fish tin can (d8,5xh4,2cm external), has a volume about 220m3 & just squeezes EM Wave not to propagate where it shouldn´t & is practically some kind of reflector with all-around side-wing
- Enh. E : internal volume is approx. 850m3, shorted one is just half of that & yes, putting the wifiKey thru the bottom (3,5cm wifiKey inside can = 3cm chip antenna distance from the bottom; new article is coming soon) on full size can is better for at least 0,5dB & makes it strictly vertical polarized with H/V ratio of 12dB (biQuad as an example, has a minimum of 20dB ratio at 90° polarity turn) ...
- Enh. F : internal volume is approx. 1750m3, well, it works very well, but has only max.5dBi if antenna/wifiKey is pushed 3cm (1/4WL) thru the middle hole of that cookie box (like on enh.D), if is pushed 6cm (1/2WL) than has around 8dBi, which is less than setup on that picture, so I presented only the best version & works like stated if mounted exactly like on that picture! With a can so big, you just wanna maximize the gain ... It has nothing to do with the waveguide theory !!! , so forgot about that on all that mine "antennas" !
- Enh. G : Yes that mine also wasn´t a good parabola & is quite critical to match the proper "focal" point at this kind of paraboloid stuff, so I gave for that mine exact numbers of diameter/depth & focus distance in case someone has very similar one; if not, than everybody is on its own with own measurements of gain & focus; thanks for the "dish" focus formula for dummies ...
- Enh. L : it is just a half of the standard d10cm-h11,5cm tin can (like enh.E) w/o the upper & the bottom cover, so w/o that support, it spreads a bit curved sides. The wifiKey edge should be aligned visually with the side curved edges of the tin, so the internal antenna is approx. 4,5cm away from the tin at the hole passing the wifiKey, which is so inserted approx.5cm deep in the paraboloid space. I showed that just for the principle & to measure gain, for finishing & mounting I let people to do it by themselves to strengthen the stuff (using some wood, plastic materials etc.)
- Enh. M : its 30x30x30cm corner (always outside Alu foil-metal measures!); the inner antenna of a wifiKey should be approx.1/2WL away from L & R sides or a little bit more, & also other measures you´ve got it right. All corner enhanced antennas of this type irradiate main signal at the vertical elevation of approx.45°, so, the back corner of the whole antenna has to be enough elevated for the horizontal beam, like is seen on the left pics! Yes the only difference of placing a wifiKey vertically on lower two pics, is to have exclusively a vertical polarized antenna & this type of antenna is a very efficient one, but is not also a smallest one either! Do some googling about 3D corner antennas on the Net!
- Enh. A1 : you got it right, the distance to the wifiKey is 2,5cm, the can is common size like enh.E (will do the job also a cat-Food 400g can, but bigger can is better IMHO). Practically it prevents EM signal to spread to the the back, if there is no need to ...
- Enh. O : adding a reflector to a wifiKey makes it a proper "dish Feed", using enh.B for that instead, maximizes gain, further feed enhancements are non-productive in that umbrella case for proper illumination of the "dish". Using a proper 80cm dish instead, would be even better idea naturally, since its parabola is better! Well, other have used a metal net with more or less greater holes in it mounted on umbrella metal spikes, but I used a solid kitchen Alu-foil, which is better for reflections (acts as a solid metal)...
- Enh. P : yes, the foil goes all around the cone; follow the link for other CorNET antennas, especially direct cable-connectorless v.3x ones - really 11dBi, I have to correct a mistake there ...

As I´ve seen from your post, you´ve been digging into Cantenna design quite deeply (so did I time ago & done some testing/measurements also, but with connectorless version - you gain approx.0,5dB). There is a lot of available info on the Net, how to do it & about waveguide principle, so we´ll not discuss it here again. But I must tell you, that a standard (no funnel, 8,5cm dia.& less than 20cm long) proper made Cantenna has no more than 10dBi gain, usually even a bit less! So is not a really good antenna, but is not a bad one either & its easy to make. Do not believe all the numbers of magic gain about that type of antenna you´ve seen on the internet! With its length, the gain can be improved for a dB or two, but not much. Best size/gain improvement is a 6cm (1/2WL) deep 45° horn-funnel (min.+3dB) added to a standard size cantenna. Look at the results of (no-connector version) cantenna measurements I´ve got (8,3cm inner diameter can) :

- Cantenna ~ 50cm long - 11,5dbi
- Cantenna ~ 40cm long - 11dbi
- Cantenna ~ 28cm long - 10,5dbi
- Cantenna ~ 19cm long - 10dbi (standard one)
- Cantenna ~ 19cm long - 10dbi (standard one) + upper mentioned horn added ~ 13,5dBi

Please, do not doubt about about my gain measurements, they are really properly done !!! I use a commercial public AP 35dBi EIRP pointed to my home almost 5km away to do it & have checked & rechecked my hardware/software computer equipment to be sure about its proper functioning/showing me data!

I am gonna give You also some idiot proof formula for cantenna length:
1000cm3 (one liter) / surface of a can opening in cm2 = can length ... no high mathematics needed ... than place the 1/4WL exiter at the 1/3 of the can length & you´ve done it !

Now lets talk about your idea to place an Usb wifiKey instead of that standard cantenna exiter place : theoretically you could do it, but it won´t work properly (metal cage of a RF part & other metal parts on a wifiKey disrupts standing waves), so you CAN´T !!!
How can I be so sure ? ´cause I´ve tried it & no big joy !!! The Usb wifiKey chip antenna has to be in the middle & approx.1/4WL away from the bottom/back reflector, best if pushed thru the hole in the bottom. And you can not apply cantenna principles to my antennas! Will explain you why.

There are two main type of exiters: Electro-Magnetic & Magneto-Electric, depending which part of EM energy they mainly use. 1st one is typical 1/4WL antenna (or open dipole), the other are all closed loop ones like quad or delta-loop or closed dipole. The 3rd ones are combined ones. Typical factory ones are those ceramic chip antennas used in cell phones and on Usb wifiKeys & are better than 1/4WL antenna, but worse than dipole & mainly are more or less omni directional. They are a bit better than PCB antennas on wifi stuff! Also they act almost as an isotropic radiator - the point type of antenna and are damn small, so are less affected of metal sides/tube proximity of the cans; so are perfect for a job! Usually are mounted on wifiKey PCB so, that the more magnetic part of EM energy is radiated perpendicular up to the PCB. Since that enhancer F helps more on magnetic part of EM wave, that wifiKey has to be oriented like on that picture for best results! Any magnetic type of antenna put in the middle inside that cookie box works like a charm; for example, a standard no-side-wings biQuad mounted in, gets a 14dBi gain - believe it or not!

There are also two main type of theory involved, when it comes to microwaves: the resonant cavity one & the waveguide one. But ... I "developed" my own one after a bunch of experiments & I called it the "Spajky´s New Wifi Tin Can Theory", which is applicable for EM wave "squeezing" tin cans with Usb wifiKeys. Its a hybrid one & it has some common points with waveguide, but more with the other theory & could be placed in-between the two main ones. The basics consists of resonant volume of an one side open metal cube with sides of 1WL (12-12,5cm) for 2,4GHz WLAN. Thats something between 1,7 & 1,95 L (~1.800cm3). The shape of the space could be different, but dimensions (at least one) closer to a multiple 1/2WL (or 1/4WL) are desired. Useful volumes are also around -/+ ~6% margin max.; better to take smaller one for higher wifi channel resonance :

- 2/1 -too big can to use ...
- 1+1/2 -(15,5x15cm BigTinCan) or better 3x length (jumbo 1/4m) enhancer E [15dBi]
- 1/1 -the wonderful Cookie box
- 2/3 -1200g dog-Food d10cm-h18cm can [13dBi]
- 1/2 -850g most common food can, Enh.E !!! [11dBi]
- 1/3 -SLOT-ed WG antenna 1WL sections, 250g coffee can 1/2WL miniCantenna
- 1/4 -Enhancer E half shorted 9dBi
- 1/8 -Enhancer B 6dBi (or STminiTenna w/ MiniDoubleDeltaDipole exiter)

These are the facts confirmed with my practical experiments using also paper+Alu foil round tube "telescopic can extender"; when the diameter of a can is 10cm (to big for proper Cantenna design), than that my theory kicks in!

About that overheating of those Wlan UsbKeys : yes that can be true, so I advised once to cut a bunch of small holes on the surrounding plastic cover on a wifiKey (loosing warranty) & not to cover the cans front for better ventilation. Also mounting finished such an active antenna in a shade & preventing rain entering even if windy, is a good idea. I also developed for this Usb antennas special USB cable prolongue:

http://users.volja.net/tatipa/HP/UsbEnh/UsbLongCable.gif
http://users.volja.net/tatipa/HP/UsbEnh/UsbLongCable2.gif

Those two upper linked schematics could be used also for other Usb stuff ...


=======================================================================================

That translating site is momentary best, but is more mentioned to translations from Eng2slo than vice-versa & the problem is also that I did not write in proper Slovene "by the book" (it would come out better Eng result). Well, I was planning to write all those articles (one new shortly coming too, hope before Christmas) about my antenna experiments shortly in ENG to put in to my own site down in signature, which BTW needs serious updating, but since I am (& will be for a year period) quite busy (I depend for everything in my life exclusively on myself & a bit incapacitated by my health conditions, yes, I am older guy, will be 50 in few years!) these times & also with moving myself to another place to live, I do not aspect to do that before next summer for sure. But at least when I draw certain schematic plans, all comments are directly in Eng not to draw/make pics twice for the same stuff. And I have from my 35y hobbies carrier archive a bunch of stuff to publish on that my site still waiting for some time too ...

About some stuff from that NZ site that did not work for you : sometimes some people get lucky & things work for them, but for some others don´t ; I also encountered sometimes myself situations in real life when certain stuff should work, but it didn´t & vice-versa regarding the theory, happens & thats why I made so much practical experiments ...

Enh. L : if is done like that, it surely works as stated ...

Enh. M : it is a normal 3D corner enhancer antenna with replaced on base vertical 3/4WL & front director with an vertical UsbKey glued with its dock on the base (so the internal antenna is on the same position like the top point of 3/4WL exciter on this type normal design) & shorted sides diagonally to half (takes less space, but looses some dBi), nothing special; than complete antenna is with its back corner elevated for 45° to propagate EM wave horizontally ... The distances from "metal" are quite critical (few mm -/+) for best & proper function & gain within -/+ 0,5dBi; if you miss too much, you can loose 2-3dBi. Not to loose more than 1dBi, you have some kind of a virtual ball type best space "focus" of 1cm diam. around best place for internal chip antenna positioning ...

Enh. A1 : any vertically flat back reflector does "flattening" the vertical radiation pattern to an exciter antenna, its logical ...

About a proper "dish Feed" you are right about not wasting EM wave around & to point it all to a dish; general rule of -10dB from feed radiating pattern hitting the dish edge should be implemented for best illumination of the dish. That is not the case with UsbKey w/o a kind of reflector (&/or further than a focal point away from dish), but its easier to do it for experiment (but you loose gain/waste EM energy that way!) & if the reflector according to dish makes too big shadow to a dish if implemented (case of my small Enh.J). Thats IMHO why you´ve seen on the Net rarely implemented a reflector to an UsbKey. Feed shadow if a dish (like that mine 80cm umbrella one) is not small, is almost negligible. The optimum solution for feed can be discussed with UsbKey enhancers & different kind of "dishes" after you make some proper practical experiments ...

The case of Cassegrain folded-optics dish type antenna I will not discuss, since I wrote that article just for Dummies & thats too complicated for them. The best feed for that umbrella dish would be for dummies an Enh.B (instead of only round reflector), which some day (just for fun) I will try it with 1,25m umbrella (with wooden holder, I already have) for long distance link ... I aspect gain around 28dBi or maybe a dBi less ...

That my gain of a such umbrella antenna is normal! The others on the Net made mistakes while measuring! As I saw, people use quite short distances with normal RF powers of their wifi equipment for measuring, which is IMHO wrong. When the signals are still too strong, the internal RF chip adjusts gain (AGC), so the measurements are spoiled & compromised !!! It has to be properly done with weakest possible signals far far away!

Enh. P : (CorNET ones) you don´t need translations there of that article; there on plans is everything clear & marked in ENG, check carefully again ...

About cantenna designs : yes 1/2m one is a good & better antenna (approx.12dBi) even w/o funnel, but is already big (more than twice longer than more common "standard" one), if even longer, than surely looks like a bazooka ... About added horns: angle of them larger than the bare cantenna beam-width (-3dBi) angle is no good like is no good narrower than 60% of that. So if 50cm 12dBi cantenna has a beam-width of approx.55° that is also a max. angle of horn added; the 45° is Ok & maybe 30° for the same gain antenna is a bit too small, but could be used if longer, not to loose additional gain ...

About my gain measurements: check only the pics (forget translating!) here , well Fresnel zone is not so perfect ... I have on my main PC another lucky combination of PCI card, software & a way (another AP 1km away with dd-wrt2.3 software too) to to check properly gains & signals & some other already "calibrated" antennas to compare with my 2nd spare PC with that my version of ZyDAS software. So I calculated & rechecked values & compared to that another one & also some other hardware & the result came out that on that ZyDAS configuration tool is approx. 1,5% = 1dBm, which I already wrote once , so I can determine quite properly the gain of some enhancer; the software just has a threshold for graphic bar & % show set too low IMHO ... Yes I know how to orient antennas for proper functioning !

I am NOT using an "open air" WL for cantennas, you got it wrong & about the gains stated over the Net for cantennas (w/o horns & with around 8,5cm diam.) & even with much of "factory" made other type of antennas : there is a tendency of adding some dBi unfortunately lately! Thats why I made so much of experiments & measurements for myself, to cut the crap! I did not say, that my "dummy" formula for most common size cantenna is better; it is just simpler!

If you plan to use an UsbKey inside an already made cantenna pushed like I advised thru the bottom of a can, you need to shorten the tube/cantenna (a bit more than 3cm) to match that my theory criteria of a bit less than 900cm3 of internal volume to work it properly! But it would be better for you something closer like you are planning to do to have still an "active" antenna : connect a proper made short (with connectors included 1WL electrically long to match Z !!!) antenna pigtail from UsbKey to standard cantenna; yes you need to properly mod/solder the cable to the UsbKey & know how to make a proper pigtail too!

The "problem" with enh.F : it is not that does not work otherwise, but works worse (to too much worse) than oriented like on that slide according to the size of that box; no matter how you turn the antenna box ...

I know that you are perplexed by my "TinCan" theory for Dummies (some others were too!), which suddenly popped in my mind AFTER writing that article & additionally compared results & volumes between different antennas made that way. I few times rechecked stuff & IT really WORKS, believe it or not! But WG theory for cantennas & other similar stuff differs & should be used for that only, not for my TinCan ones !!!

I had on occasional visit time ago right on testing that my miniTena w/ butterfly dipole (mentioned & linked slide somewhere up before) some highly graduated professional guy in radio links/antennas/electronics matter; he was also perplexed & could not believe that that one could actually work for wifi more than few 10m, if even could ... So I explained him a bit briefly, but he was also not convinced yet, so I actually demonstrated him a few km away basic link working with that antenna. The guy was shocked & astonished & could not believe his own eyes, he remained speechless ... I have fun shocking people a lot of times, its my kind of amusement ... Sometimes being too "deep-in" in some matter theoretically, can prevent you to see/imagine things simplified & easy; happened to me also few time in my life ...

Yes, a Cantenna can be build also with 10cm can, but is too large for dominant mode, so it looses efficiency & you get even a lower gain than with a so called "proper" diameters ... You mentioned my "open metal cube" idea & yours recommended value being about 0.7WL for rectangular waveguide antenna. If you do the math for volume of one such a section , you will got, believe it or not, 0,6L for wifi on 2,45GHz. That is one of common points of that "my theory" (funny & for dummies only!) with WG one (1/3 of my basic volume if you check can volumes on my previous post)

Regarding my custom USB cable schematics : (works well, tested on WLAN link w/ my UsbKey antenna on Usb1.1 interface & 802.11b wifi @ 11Mbit/s)

Sometimes I have a feeling, that you have problem with understanding/missing some basic stuff, but digging anyway to more complex things (maybe thats why you needed to translate my article & did not figured out some principles just from slides & drawings like some others did, but I could be wrong)

You forgot about the Ohm law & voltage drop on longer lines with greater currents; Usb stuff needs at least 4,75V for proper functioning. Also because of the power & data lines capacity/resistance/inductivity load is better to use such a coax cables than a Cat5e Lan cable for Usb cable replacement for extra long line ranges. That Zener diode is just a protection in case if the voltage regulator "dies in wrong way", not really needed; you did not see the remarks on that schematics ...

Data rate is not equal to frequency/clock (example: AGP clock frequency is always 66MHz, data rates are 1x - 8x; p4 Cpu data bus is 200MHz clock, data rate is as 800 but quad pumped & still on real 200MHz). Basic clock for USB chips on motherboards & Usb serial line is 48MHz (packet frame clock) which determines all other "multiplied" ones. That is a simplified explanation. Drivers take care of data processing & data speed actual transfer according to capabilities of the Usb line & devices involved if data packets are getting corrupted.

Yes, I am quite capable also in electronics, computers, acoustics, music, RF tech etc. (for decades as my hobby activity), but not an expert or specialized, but I can arrange myself with stuff mods quite successfully; & why things should be complicated, if can be DIY simple & cheap to make it ...

If you are interested, (hope you won´t try to translate again, see pics & drawings with Eng remarks & try to understand) here you have my another article of some my actually never really finished project with a pot cover dish (that needed to be returned to the owner), just tested, measured [& also calculated] stuff ...

I am quite busy these days & few months further, so probably I will not be present on the Net so often & for longer periods absent in closer future, especially for longer explanations like this ones here, since also I have to write some other articles planned & announced on this forum time ago (ultimate & big enh.E style antennas up to 20dBi also with horn added, maybe before Christmas) if & when the time & my health will allow me to do it. Playing with wifi stuff time to time it used to be now already almost a year long period my hobby activity & I need to move on to other more urgent matters when having a spare time to spend ...

But anyway, you could still post that other Cantenna stuff your planning to; maybe someone else will get involved here in this debate about that matter instead of me (alone) or being absent; IMHO some others on this forum could also be interested & are IMHO enough capable in ENG to do it & maybe also have a will to participate in a debate about cantennas ...

You´ve got so a bunch of ideas, so maybe is time for yourself to experiment in practice further by your own; many times different theories need practical experiences too & vice-versa; play with stuff, its FUN (well time consuming too) ... Bye & Merry Christmas & Happy New Year in advance if I will be rarely on this forum!

=======================================================================================