Posts by DG9BFC

Quote from G0MJW

Peter,

Experimenting is always a good idea. A patch antenna with a single feed that is perfectly circular can't produce circular polarisation. There needs to be something else. I settled on the dual resonance method. The original POTY (it was not called that, that was Remco's idea) used an ellipse but that's hard to make, so I tried an approximation to an ellipse. A circular patch with notches, circular with tabs and the square with cut outs. The square was good enough and is easiest to make with simple tools, so that's what I ended up with. Other iterations had directors, choke rings and so on. Another way to do this dual resonance trick is with a feed and a tuning capacitor, but I found that really hard to get just right. G3RUH managed it but not with the waveguide. Achim did manage it but I don't know exactly how.

What wasn't obvious early on is were the simulations accurate? I was taking huge liberties with the meshing in the simulation and with the full version of CST we could probably design a more forgiving, broader band patch. In the end, its only when something is reproduced in hardware that you can be sure it works, or not. So, keep on cutting and trying because if it works in hardware, even without simulation, it works.

Mike

Display More

it was me who brough mike to the idea of the square with the cut corners (i have no lathe so peters feed was not easy to reproduce on the kitchen table) .. i calculated in sonnet lite and mike calculated in cst studio ... and end result was near identical

rest is history

difficult to build and tune ?? yes a bit ... but when exactly made shows a good circular pattern and is near spot on

i use big spring loaded clamps to hold the two plates pressed against the spacers (i use 3mm stainless steel rods) not only gravity

that helped a lot to have exact distance

fifth harmonic is 12ghz (5x2.4) and can desense (with lower power) or kill the lnbs first transistor stage

and 12 ghz is easy travvelling inside the waveguide (its build for 11 to 12ghz .. not?!)

an overdriven and clipping amp produces higher fifth overtone

so stay low in power and stay linear (plus a lowpass filter inside or behind amp)

that lowers the signal on 12ghz that otherwise... see above

with a better lens and a better position you can improve your rx and tx signal

either hack away from a commercial rocket lnd or print your own (drop me a mail if you need a print file)

reflector should almost touch the lnb clamp

try to avoid angled connectors (they produce more loss as half a metre cable bent in a nice wide open curve)

greetz sigi dg9bfc

and ... how does that poty work?? in principle it is a too long and a too short dipole as a cross ...both fed with the same current

now what does a too long or too short dipole represent?? a radiator with a coil and a radiator with a cap inline

and what does a cap/coil do?? they shift phase

now ... if proper designed and tuned one element shift phase plus 45 degrees and the other shifts - 45 degrees .. voila 90 degrees difference (but with equally currents!!) and we have the circular polarisation

and we do see that if the difference between the upper and lower resonances are around plus minus 2-3% (around 50 to 70 megs or 25 to 35 megs above/below working frequency)

means the mechanical dimensions should NOT be changed and the feeding point must be at the absolute exact position to feed the elements with the right phases and current (and have the 50 ohms on that spot also with not reactance added) ... not easy but it can be solved with ... build it EXACT ... and finetune in very small steps with looking on the vnas smith diagramm (not to mention use a short cable and tune the cable out with osl calibration etc)

a hint .. in several poty i made i had to slightly get the long edge near the feedingpoint a tiny bit closer to the reflector .. about half of a tenth mm only or so

greetz sigi dg9bfc

ps yes if proper tuned the circular polarity is good! and a single resonance spot does NOT mean both elements are tuned to the same single resonance .. BOTH TOGETHER form a single resonance with circular polarisation (see explanation how the poty works at the beginning of this text)

if that dip is at the working frequency AND if that is at the top of the rollercoasterloop then it is perfectly tuned

install it ...

with such a high return loss you will see the return loss will go down a bit from your 50 or 55 peak cause you see the reflections from the dish back into the feed and any slight bending will detune that very deep notch ... a bit of distance change and the reflections come back in a different phase related to the tx energy and you see that deep notch again (will vary a few db depending on distance) ... if you have made your reflector from a not too thick copperplate then you should reinforce it a bit (i glued a HPL sheet at the back for a few feeds i made for cf dish)

does the switching back and forth work on all fw versions?

using evariste 2103 here with a patch for easier datv settings

... cant wait to try it out here

dg9bfc sigi

you can add a rim to the reflector (say a patch feed driven tubular waveguide) ... that can WIDEN the angle for f/d 0.25 to 0.3

and ... as deeper the dish as more you have to look that the feed is at the phase centre ...for a patch with rim that means to place it a bit more away from the dish

i have the same problem with picture freeze after a second but audio still works

any hints what it may be?

beaon decode works fine but i see that on several ham signals its difficult to find a lock

and even if locked not always a picture is shown

only way is an lnb with external reference (gpsdo or at least on ocxo or tcxo in shack)

the ast3tq does not work with 1.8v i guess (it is a 3.3v unit) and there is NO CLEAN 3.3v insidfe the pluto

but you can now get 40 megs 1.8v with 0.5 ppm from epson

TG2520SMN 40.0000M-ECGNNM

the astx13/asvrtx13 is now out of stock but the epson works identical

sidenote DO NOT USE THE 1.8v at the front (close to usb connector) ... use the solder lug between fpga and trx chip named 1p8v_clk!!!

the 1.8v at the front is a bit "dirty" while the 11.8v clock voltage is superclean

there is a reason why AD mounted a special low noise voltage chip in front of the xo (now tcxo) and that power chip only powers the clock and nothing else ... wh9le the 1.8v at front powers different things (and so is a bit noisy)

If you use a new tcxo inside he pluto the external ref command does not work ... cause new tcxo has no enable pin (to switch it off) .. i rerouted that switching transistor from enable pin of xo ... to enable pin of the power supply ic for the clock (u12) .. now i can have both .. a stable internal clock and still have the option to feed external 40 megs in

(feel free to ask for details)

the pluto has a frequency "nailed in stone" in uboot-extra-env.dfu.

example:

VíÞserial=1044734c9605000cf9ff120010c84f79ec productkey=d9f04f074f5e6992d04b879a176410570a3ed77a ad936x_ext_refclk=<39999843> ÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿÿ

now the pluto "thinks" on boot he has a ref clock tha is157 hz low

xo correction = 40 000 000 plus setting also 40 megs in sdrconsole and all is well

in pluto options calibration set the checkbox to "apply corrction when connecting to pluto" and the 40 megs is sent to pluto on boot

LET THE TX OFFSET AT 8089.5!! and set XIT to 0 !!!

if done well you should be spot on (except the doppler shift of around 150hz)

and bewteen input and output normally the xtal is found

(and input needs lower ref level to inject ... that is where a tcxo should be injected)

lnb adaptor shopuld be made excentric

... that gains you a bit of height

Quote from SV1BDS

Hi,

Maybe R5 near TCXO that missing is for disabling internal TCXO.

disabling internal tcxo (now) is done via enable pin and q2 and not via r5!!

r5 can be installed if you solder in a tuneable tcxo (and want to tune it via auxdac voltage!) ... if you replace tcxo with any other (as example astx13/asvtx13) with no enable pin then switching to external reference does not work anymore ...but i had a clever idea to solve that

solution is to rewire Q2 to pin 4 of U12 (enable pin of 1p8v_clk regulator) ... so instead of switching enable pin of the tcxo ... you switch enable pin of the power regulator ic that powers the tcxo

(time for a revision e ?!?)

why the hell did they not install in minimum the 0.5ppm tcxo from factory?!? grmbl grmbl

how is rewiring done?? lift pin 4 of u12 ... and rewire that enable pin to R67 (cut connection to enable pin of tcxo)

r101 ... disconnect from 1p8v_clk and rewire to solder lug pg_1p3v

now selecting external reference switches power of tcxo instead enable pin of tcxo

that pin is now free ... so a tuneable tcxo can be mounted (with trimmer or with installing r5 and using ad9363_auxdac)

happy modding DG9BFC (big fat chicken, beat frequency counter, bastler für charly?)

you have to use the boot frm from the 31 version ... then also the "hack" works

if you use boot frm from 32 then the hack does not work

it was me ... just to show HOW CLEAN AND NARROW such a signal from the keyer software is (the news reader has the same setable rise and fall time settings as the "normal keyer" ... a bit longer rise time ... and signal is VERY narrow)

its a nice tool to spread out news (or other urgent messages) ...

AND IT WAS JUST AN EXPERIMENT!!

SORRY IF SOMEONE FEELS UPSET ABOUT THAT!!

i wanted to show how narrow (no keyklicks!!) such a signal can sound over the air ... much better as switching on sdr console to tx (permanent on with tone !!!)) and key the amp on and off to send cw signs

yes i have seen and heard such signals on qo100 ... UGGLYYY WIDE SPLATTER

it was just to show a superb cw quality (remember it was just to SHOW IT IN LIVE)

so at least we know it works fine (and i switched the news reader off for now)

the keyer soft produces a superb cw signal quality ... and thats all i wanted to show with the newsreader ...

greetz sigi DG9BFC

best 73 and cu on the bird

IT IS THE NEW TRANSPONDER ... tune sdr console to 11.970 and see the stripes

(if you have a different lnb frequency is a different story!)

in some setups 2 gig is a few db lower and noise can not be seen (or just barely) depending on how your setup is done ... length of cables... bias t used etc etc

many things influence the losses on 2 gig while on 740 megs all the attenuation factors are a lot lower ... and that is the reason why some do see those streipes and some do not ...

the stripes are NOT from lte (even if an lte filter helps in this case)

Nearly all sdr have a problem that they also receive the 3 times and 5 times (and above) frequency

means if you have your rx tuned to 740 megs (rounded up a bit) ... it also receives on 2.22 gig

since a few days a new tv transponder on the sat was switched on on 11.970 ...

minus 9.75 equals 2.220 ... voila we have noise on 740 megs

several ways to solve it

you can add a narrow 740 megs filter in your setup (there are saw filters available for that range)

you can add an "lte-filter" in the lnb cable (a 800 megs lowpass filter with dc feed trough)

you can add a 2.2 gig notch inside your bias t (just an inch of coax need as an open end stub)

i now have two of the above solutions combined (the lte filter PLUS the notch for 2.2 gig in bias t)

normally one of the three solutions should be enough but i got the lte filter after i added the notch so i left it in

spectrum now is super clear again

greetz dg9bfc sigi