Adjusting the Bullseye LNB for improved frequency accuracy.

Ideas, technical topics, help and discussion for ATV enthusiasts
Forum rules
This forum is run by the BATC (British Amateur Television Club), it is service made freely available to all interested parties, please do not abuse this privilege.

Thank you
radiogareth
Posts: 761
Joined: Wed Jan 06, 2016 9:46 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by radiogareth » Mon Jun 14, 2021 2:16 pm

G8EOJ wrote:
Mon Jun 14, 2021 1:58 pm
Thanks for the STL Colin, it works a treat.

Eric - G8EOJ
On my printer now (in PETG) thanks :-)

Gareth

g4eml
Posts: 446
Joined: Thu Apr 26, 2018 9:36 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by g4eml » Mon Jun 14, 2021 3:24 pm

I printed mine in PLA with 0.3mm layers and 20% infill. The dimensions were adjusted to get a tight fit. Other materials or settings might turn out differently.

Colin.

G4KLB
Posts: 233
Joined: Wed May 28, 2008 10:46 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by G4KLB » Mon Jun 14, 2021 11:37 pm

Thanks Colin :D
Potty eml.jpg
Potty eml.jpg (511.04 KiB) Viewed 691 times

g0mjw
Posts: 1650
Joined: Sat Sep 20, 2014 9:15 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by g0mjw » Tue Jun 15, 2021 7:06 am

I printed two yesterday, one TPU, one ABS. Both work but I think TPU is a little better as it is flexible and more grabby. The hardest thing was getting the red cap off the LNB. Incidentally, this could perhaps with a copper ring be used to make a C120 adapter which would be rather useful. It needs a brass disc with holes in the right places soldered to a length of 22mm pipe.
How to keep these in Position? perhaps tape around the outside?


Obvious to all of us but you don't use the LNB clamped in the original position with the POTY sticking out in front. Believe it or not I have seen people try that, thinking there was some magic to the length of pipe supplied by Hans. Then complaining the design is bad and 10 dB worse than the LNB was on its own.

The feed needs to be at the focus which is roughly where the front face of the LNB was. That probably requires a 40mm to 22mm adapter. With typical offset dishes, as the reflector is 105mm and larger than the original horn, it can foul with the feed arm. Doing it properly means modifying the feed arm. A small amount and possibly sufficient mitigation can be had by printing an adapter with the hole offset. This raises the height. That's a small error but the loss is not significant though it will change the elevation slightly.
Capture.JPG
Capture.JPG (83.57 KiB) Viewed 678 times
OffsetFeedClamp.zip
(16.92 KiB) Downloaded 26 times
I knocked this up in TinkerCad in a few minutes. Untested as I don't have that type of LNB clamp.


Mike
Last edited by g0mjw on Tue Jun 15, 2021 9:19 am, edited 1 time in total.

radiogareth
Posts: 761
Joined: Wed Jan 06, 2016 9:46 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by radiogareth » Tue Jun 15, 2021 8:07 am

Although probably insignificant (and tricky to calculate maybe) could the centre axis just for the LNB be inclined by a small amount to compensate?
Removal of the cap is difficult. To get a Rocket LNB lens off I held the plastic part in boiling water for a few minutes which made it a twist off job - with oven gloves ;-)
Source of cheap (under a £10) rocket LNBs seems to have dried up too :-(
Gareth

g4eml
Posts: 446
Joined: Thu Apr 26, 2018 9:36 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by g4eml » Tue Jun 15, 2021 10:00 am

I agree getting the cap off in one piece is quite difficult. It is only clipped on and not glued. I used a metal case opening tool (like a blunted knife blade) to lever up in one position. Inserted a small screwdriver to hold it up, then repeated the process about 30 degrees further around the rim. Eventually the cap popped off.

SY_JON
Posts: 1
Joined: Tue Jun 22, 2021 8:54 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by SY_JON » Tue Jun 22, 2021 9:40 am

g4eml wrote:
Sun Jun 13, 2021 10:01 am
Hi Mike,

I did consider adding a bias T to the board but decided to keep it simple as I suspect most people will already be powering the LNB using their normal receive setup. Likewise I considered adding a connector for the reference output but I think most people will be simply adjusting to get the correct down converted frequency of one of the QO-100 beacons.
The 25MHz reference is already DC blocked and terminated with a 75 ohm resistor on the board, so I might add a test point so that it can be extracted and measured. The level is quite low though because it is taken directly from the TCXO output and attenuated to prevent any pulling.

The LNB is using a VCTCXO with the tuning voltage supplied by a PIC10 processor’s PWM output. That limits the resolution of the adjustment to about 200Hz steps. The rest of the circuitry is using an RT320M chip the same as the Octagons.

They did actually originally design the LNB with a Trimpot for adjustment. (I have a prototype one that Jon, the UK based designer, sent me to try) However the problem was making it adjustable and still keeping it waterproof. It also only had a single turn trimpot so accurate adjustment is quite fiddly. That seems to be the main reason for switching to using the PIC.

I guess a possible upgrade might be to replace the 2ppm VCTCXO with a 0.5ppm one but the adjustment resolution might then become a problem.

I doubt there is any chance of getting a flanged version, the market is too small.

Colin.
Hi Colin/All

Issue with C120 is that its very specialised market, we are currently working on a design but its US based and limited to 12.2-12.7G and the chassis cant be used as we have narrowed the guide to help improve the image bands so 10.3G is right on the edge and performance will be very poor.
We have produced a few Universal LNB's for prospective customers using the Twin or Single chassis, I cant remember, with a C120 machined adaptor and all you need to do is cut the LNB feed tube at the correct point and insert the C120 adaptor. Bullseye uses our Twin Chassis/Design to get space and the XPOL is MUCH better than any Uni Single. I will look through my files as this was a long while ago but if you get a few machined in China or have a lathe available (PCBWAY do CNC) then the cost should be relatively cheap to manufacture.

If I can find the information I will email it to Colin as it has the correct impedance matching to keep the losses low.

BTW dipping the cap in hot water is definitely the best way as I've had several screw drivers in my fingers and hands over the years..
Also the VCTCXO in the Bullseye is 1ppm or better but the phase noise is very important and many TCXO are poor and unspecified. After SMT, Time and Mechanical relaxation/mounting a 0.5ppm TCXO will not be 0.5ppm for long.

We sis develop a full blown bias tee with 13/17, impedance transformer 50 - 75R and Reference output with diplexed with RF but Othernet didn't want to commit to the volume so we only produced a few samples. General market also always want something different, to add DC and tone, just use a 100nH inductor and a 100pF capacitor after it, the current is so small a 0402 or 0603. Impedance transformer is a capacitor to block DC and 2 resistors to form 6dB minimum loss pad - pretty low tech and can easily be dead bugged onto a PCB.

Regards

Jon

g4eml
Posts: 446
Joined: Thu Apr 26, 2018 9:36 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by g4eml » Mon Jun 28, 2021 2:56 pm

I have had a report from one user that he has successfully used the programmer with 11 metres of coax connecting the red port. So it looks like it might be possible to adjust the LNB remotely.

I have been checking my Bullseye LNB over the last few weeks. After the initial adjustment it has always been within 1KHz of the correct frequency when checked. There hasn’t been a massive temperature variation though. It I will be interesting to see how it behaves over the winter.


Colin.

g0mjw
Posts: 1650
Joined: Sat Sep 20, 2014 9:15 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by g0mjw » Fri Jul 02, 2021 5:15 pm

g4eml wrote:
Sat Jun 12, 2021 4:12 pm

I have designed a simple interface using an Arduino Nano and three push buttons which is able to adjust the LNB. Having done this my LNB is now always within 1Khz of the correct frequency at room temperature.
It is also possible to control the LNB from a PC connected to the Arduino USB port which opens up the possibility of automatic adjustment.

Full details, including the Adjustment specification, Arduino code and PCB design are in my Github at https://github.com/g4eml/Bullseye-Calibrator

Colin G4EML
KiCAD can import Eagle files so I tried that. It might be useful to make a few changes to the PCB - e.g. to include some mounting holes and arrange for it to fit in some standard box. That is, only if Colin doesn't mind.
Capture.JPG
Capture.JPG (219.15 KiB) Viewed 238 times
Mike

g4eml
Posts: 446
Joined: Thu Apr 26, 2018 9:36 am

Re: Adjusting the Bullseye LNB for improved frequency accuracy.

Post by g4eml » Fri Jul 02, 2021 8:47 pm

My thought was that it probably wouldn’t need a box as it is something that is only likely to be used occasionally.

Your image doesn’t quite show the power connector correctly. The holes are supposed to be slotted. The milling layer contains the slots. It has also shifted the component legends a bit.

Post Reply

Return to “General ATV Discussion”