Narrow bandwidth 1420Mhz filters
Andrew Thornett
Marcus D. Leech
--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/LNXP123MB36586FA9374DDA9F0DA31A0FFFAEA%40LNXP123MB3658.GBRP123.PROD.OUTLOOK.COM.
Lester Veenstra
Curious; What is the demonstrated problem with using all the bandwidth that comes out of the SAWBird?
Lester B Veenstra K1YCM MØYCM W8YCM 6Y6Y W8YCM/6Y 6Y8LV (Reformed USNSG CTM1)
452 Stable Ln
Keyser WV 26726 USA
GPS: 39.336826 N 78.982287 W (Google)
GPS: 39.33682 N 78.9823741 W (GPSDO)
Telephones:
Home: +1-304-289-6057
US cell +1-304-790-9192
Jamaica cell: +1-876-456-8898
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/71e39bf4-e6b1-4275-8339-380ff833fa2e%40gmail.com.
Marcus D. Leech
Curious; What is the demonstrated problem with using all the bandwidth that comes out of the SAWBird?
about 5% fractional bandwidth, and for amateur-scale baselines, it likely won't be much of a problem. RFI is more likely
a problem.
Of course, if you're using an SDR for interferometry, you can arbitrarily trim the bandwidth prior to correlation, without
needing analog filters. Similarly, you can make yourself an FX correlator and side-step both issues at the same time...
But I think the question is driven by the recent resurgence of interest in hardware-based solutions like the AD8302
phase-detector chip, whose input bandwidth is "like a barn door". Jan Lustrup has been using it, but with
3MHz-wide filters in front of it--partially due to the bandwidth decorrelation issue, but also, I think, because that's
what he had :)
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/062401da13ef%24ff2cd290%24fd8677b0%24%40com.
Andrew Thornett
Sent: Friday, November 10, 2023 4:14:49 PM
To: sara...@googlegroups.com <sara...@googlegroups.com>
Lester Veenstra
OK. I had not followed the discussion on the use of AD8302 phase-detector chip,, but I would guess (which means I have not tried it) that the same process could be done in software. Suggest looking at the fine work done at Wolfgang’s group.
The real practical issue with a wider noise bandwidth, wider than just the SOI, is getting sufficient signal into the A/D to be above the conversion noise floor, and simultaneously not generating significant Percent Time In Clip (a/d overload)
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/c385ee07-da5b-4e25-882c-08fc80395ea8%40gmail.com.
Lester Veenstra
But why not pre-fix with post A/D, software BW reduction?
Lester B Veenstra K1YCM MØYCM W8YCM 6Y6Y W8YCM/6Y 6Y8LV (Reformed USNSG CTM1)
452 Stable Ln
Keyser WV 26726 USA
GPS: 39.336826 N 78.982287 W (Google)
GPS: 39.33682 N 78.9823741 W (GPSDO)
Telephones:
Home: +1-304-289-6057
US cell +1-304-790-9192
Jamaica cell: +1-876-456-8898
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/LNXP123MB3658D615BABC122FFEFAA77AFFAEA%40LNXP123MB3658.GBRP123.PROD.OUTLOOK.COM.
Marcus D. Leech
OK. I had not followed the discussion on the use of AD8302 phase-detector chip,, but I would guess (which means I have not tried it) that the same process could be done in software. Suggest looking at the fine work done at Wolfgang’s group.
integrate and decimate to taste. Fringes. But the AD8302 eval board is about 18dB$ cheaper from outlets like
AliExpress than something like a USRP B210, and about 13dB$ cheaper than a KrakenSDR. So, there's some attraction to
this approach, particularly for "getting ones feet wet".
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/064601da13f2%24ba518910%242ef49b30%24%40com.
Marcus D. Leech
But why not pre-fix with post A/D, software BW reduction?
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/066001da13f3%2450ec3910%24f2c4ab30%24%40com.
Lester Veenstra
No post A/D, Pre-Correlation
Assuming digital correlation of course
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/497a3d99-9c51-4516-99a5-252b10052017%40gmail.com.
Marcus D. Leech
No post A/D, Pre-Correlation
Assuming digital correlation of course
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/06a201da13f9%249c6b19a0%24d5414ce0%24%40com.
fasleitung3
Marcus D. Leech
If my math is correct, 5% fractional bandwidth at 21cm and a 10m baseline will already show noticable degradation of the fringes at fairly small angles from the boresight.I find this very educating: https://www.atnf.csiro.au/research/radio-school/2010/lectures/Interferometry-Practicalities_R_Perley.pdfWolfgang
Basically, when sinTheta = c/(B*dV), there's the first NULL in the sinc function. Where Theta is the offset angle
from the meridian. I couldn't make their simple example (2MHz, 35km baseline) come up with the 27 degrees they
quote. So, I'm missing something somewhere....
Am Freitag, den 10.11.2023, 11:14 -0500 schrieb Marcus D. Leech:On 10/11/2023 11:07, 'Lester Veenstra' via Society of Amateur Radio Astronomers wrote:
You can end up with bandwidth-decorrelation issues when doing interferometry, although the SAWBIRD+ bandwidth is onlyCurious; What is the demonstrated problem with using all the bandwidth that comes out of the SAWBird?
about 5% fractional bandwidth, and for amateur-scale baselines, it likely won't be much of a problem. RFI is more likely
a problem.
--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/276f7348a23aa06f86050b33260c989c2e1bb73a.camel%40googlemail.com.
Marcus D. Leech
If my math is correct, 5% fractional bandwidth at 21cm and a 10m baseline will already show noticable degradation of the fringes at fairly small angles from the boresight.I find this very educating: https://www.atnf.csiro.au/research/radio-school/2010/lectures/Interferometry-Practicalities_R_Perley.pdfWolfgang
equation is linear in both bandwidth and baseline, we can scale that result, and the first NULL for 65MHz bandwidth and
0.01km baseline is about 3500 degrees. If you're willing to tolerate attenuation up to perhaps 25% of the first null, then
you still have about 875 degrees before things start to head towards the first NULL in earnest. For typical
non-tracked interferometer used by an amateur, with a beam-width of, let's say 10deg, the primary beam-shape will
be the limit, and not bandwidth fringe attenuation?
Am Freitag, den 10.11.2023, 11:14 -0500 schrieb Marcus D. Leech:On 10/11/2023 11:07, 'Lester Veenstra' via Society of Amateur Radio Astronomers wrote:
You can end up with bandwidth-decorrelation issues when doing interferometry, although the SAWBIRD+ bandwidth is onlyCurious; What is the demonstrated problem with using all the bandwidth that comes out of the SAWBird?
about 5% fractional bandwidth, and for amateur-scale baselines, it likely won't be much of a problem. RFI is more likely
a problem.
--
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/276f7348a23aa06f86050b33260c989c2e1bb73a.camel%40googlemail.com.
Andrew Thornett
Subject: Re: [SARA] Narrow bandwidth 1420Mhz filters
fasleitung3
Marcus D. Leech
I did a little spreadsheet using my parameters: 1420 MHz with 10 MHz bandwidth, 10m baseline.I now end up with only a negliable degradation over any direction. So I must have done something wrong when I did the calculation on a pocket calculator.If you go up to 5% fractional bandwidth you loose ~50% of the signal at around 50° off meridian.
interferometer at our previous site--100m baseline, 611MHz Fc, 6Mhz bandwidth. Someone "out there" had warned me
to expect significant fringe attenuation from the bandwidth X baseline product. So, I went back to the books to convince
myself that it would be negligible, and confirmed with a radio astronomer friend of mine.
50deg off-meridian would be inconvenient for an array with wide-field or tracking antennas, to be sure. Which is why
"real" observatories will insert delays during observations. But for your typical amateur "getting started" it's probably
not a huge issue. My main concern with 65MHz bandwidth into a hardware correlator is RFI in any practical
amateur deployment. But, the proof of the pudding, etc. Hmmm, thinking about it, does an FX correlator eliminate
the need for delay insertion--because each FFT bin is narrow-band, and thus the first NULL can be pushed out
quite a bit? Never having spent deep technical time at a synthesis array, I don't know.
I'll note that *spatially-fixed* interferers will come out of the correlator as a fixed DC offset, which can be eliminated
with an appropriate fringe filter--either in hardware (as Jan has done), or in the software you use to process the
correlator low-rate ADC output.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/8a3b7112f0130c659cd37fe30fd4e89def6aa334.camel%40googlemail.com.
Andrew Thornett
Subject: Re: [SARA] Narrow bandwidth 1420Mhz filters
Marcus D. Leech
Dies that mean that for antennae in range 80-100cm, it is significantly to have narrower bandwidth 1420MHz filter for interferometry? If answer to that is yes then what is most cost effective filter bandwidth and where would an amateur buy such a filter?Andy
meridan-offset "zone of disappointment" starting about 50deg from meridian center for a 5% (65MHz at 1420) fractional
bandwidth.
So, good to go, and then if things end up being a little bit wonky off-meridian (more than is implied by your
inherent antenna beamwidth), THEN consider tighter filtering.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CWXP123MB365369B35206A964FB09EE35FFAEA%40CWXP123MB3653.GBRP123.PROD.OUTLOOK.COM.
Andrew Thornett
jan Lustrup-LA3EQ
Question: Can you do without any extra 3MHz narrow band pass filters and only use the wide Sawbirds built-in SAW filters when doing interferometry with the AD8302 phase detector?
That is the test I’ve done with my S.S.I. (Super Simple Interferometer) setup.
I made two drift scan interferometer observations with the same setup,
One yesterday without any extra band pass filters in each RF input port of the AD8302 phase detector, (using only the wide Sawbird+H1 built in saw filters).
And today same setup but this time with a narrow 3MHz interdigital band pass filter centered @ 1424MHz in front of each of the RF inputs of the AD8302 phase detector.
The results speak for them self. You can do without any extra filters and still see fringes…but the quality get much better and larger uniform fringes with using extra filters. Maybe some desencing going on due to possible strong signal interference close by when not using the extra filters?
Setup: two 1.9m dish with stove pipe feed and Sawbird+H1 LNA’s , spaced 17.3m on an east/west baseline, AD8302 Phase detector, Lab Jack U3 12bit A/D converter, Radio Skypipe software.
Image 1 Raw data results of Cygnus A & Cassiopeia A without extra filters, no post detection averaging.
Image 2 Raw data results of Cygnus A & Cassiopeia A with extra filters, no post detection averaging.
Image 3: AD8302 interferometry test setup drawing.
-Jan Lustrup LA3EQ- Norway
.
From: sara...@googlegroups.com [mailto:sara...@googlegroups.com] On Behalf Of Marcus D. Leech
Sent: fredag 10. november 2023 20:06
To: sara...@googlegroups.com
Subject: Re: [SARA] Narrow bandwidth 1420Mhz filters
On 10/11/2023 13:15, 'fasleitung3' via Society of Amateur Radio Astronomers wrote:
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/00005c4f-7ac9-480c-901a-72d0bff74dd5%40gmail.com.
Andrew Thornett
Sent: Sunday, November 12, 2023 10:54:06 PM
To: sara...@googlegroups.com <sara...@googlegroups.com>
Subject: RE: [SARA] Narrow bandwidth 1420Mhz filters
B & MR Randall
Notice the fringe rate on Cyg A about 1.5 x Cas A, as expected.
Cas A declination = 60 degrees. Cyg A declination = 40 degrees.
Cos(40) / Cos(60) = about 1.5.
Careful examination of fringe rate can get the declination of the object.
RadioEyes has a fringe rate calculator in one of the menus. Uses declination, frequency and antenna spacing
Bruce Randall
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/000c01da15bb%24254382e0%246fca88a0%24%40no.
jan Lustrup-LA3EQ
Hi Andy.
I found my filters on e-Bay…
You can build them yourself if you are experienced in metal work.
I know of one person that builds filter on order. be...@pe1rki.com
These types by Bert Modderman PE1RKI are for 23cm but he can custom build for 21cm. Be patient as he is very slow on answering e-mails!
Good Luck,
Jan Lustrup LA3EQ-Norway
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CWXP123MB3653354F368E2FCBDD27482FFFACA%40CWXP123MB3653.GBRP123.PROD.OUTLOOK.COM.
Andrew Thornett
Andrew Thornett
Sent: Monday, November 13, 2023 10:06:27 AM
To: sara...@googlegroups.com <sara...@googlegroups.com>
Andrew Thornett
jan Lustrup-LA3EQ
From: 'Andrew Thornett' via Society of Amateur Radio Astronomers [mailto:sara...@googlegroups.com]
Sent: mandag 13. november 2023 11:14
To: sara...@googlegroups.com
Subject: Re: [SARA] Narrow bandwidth 1420Mhz filters
Thanks Jan and others,
I'd like to have a go at making one - found some instructions on the Internet.
What I can work out is some of details:
(I) I assume outer braid of coax both input and output are connected to metal casing?
yes
(2) Am I right in thinking that the rods are electrically isolated from the casing?
No they are connected to the case at one end with a screw.
(3) Is the inner central wire on coax input soldered onto the first rod and the inner wire on coax output soldered onto the last rod?
Yes
(4) Most confusing of all is the rods = I am assume they are can moved in and out to tune - is that all three or just one that can be moved? BUT alos what about other sides of rods- they appear to be connected to casing both sides which I don't understand - can someone explain in really simple language for non-electronics expert like me?
Each of all rods is tuned to resonance by its tuning screw. The rods them self’s are securely screwed to the casing on one side, the other side is where the tuning screw moves in or out to find resonance. In interdigital filters every rod is mounted upside down ever other time.
(5) Does it matter what rods made of? For example can I use a thin steel bolt?
Low loss metal is best as it will have lower RF loss but steel could be used in a pinch. Brass or copper tubing is best. Does not have to be a solid rod as RF flows only on the outside of a conductor so tubing is easier to use…..
Thanks in advance!
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CWXP123MB36533ACDBA8E473DFD052D21FFB3A%40CWXP123MB3653.GBRP123.PROD.OUTLOOK.COM.
Andrew Thornett
Marcus D. Leech
Hi Andy.
I found my filters on e-Bay…
You can build them yourself if you are experienced in metal work.
I know of one person that builds filter on order. be...@pe1rki.com
These types by Bert Modderman PE1RKI are for 23cm but he can custom build for 21cm. Be patient as he is very slow on answering e-mails!
Good Luck,
Jan Lustrup LA3EQ-Norway
(at a cost of considerable bandwidth). They are superbly made, and I cannot recommend Bert's work strongly enough...
Jeff Kruth
Andrew Thornett
Sent: Monday, November 13, 2023 4:19:51 PM
To: sara...@googlegroups.com <sara...@googlegroups.com>
Jeff Kruth
Jeff Kruth
James Abshier
Andy - Here is an example of a home-made interdigital bandpass filter in a wideband receiver that I put together about a year ago. The sides of the filter are aluminum 1x1/4 inch hardware stock drilled and tapped to allow for assembly. The elements inside are brass rods from a hobby store. I would take the lid off to show you, but right now the receiver is presently in use. The receiver has a bandwidth of 25 MHz and is being used in a 1418 MHz phase switched interferometer. The filter elements are bolted to one side of the enclosure and a tuning screw is installed on the opposite side for adjustment. Adjustment is best done with a nano-VNA.
Jim Abshier
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/396130586.2980932.1699894402808%40mail.yahoo.com.
jan Lustrup-LA3EQ
Look at this:
The Utah ATV page - Putting together an interdigital filter (utaharc.org)
This will give you an idea how to do the job.
Jan
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/CWXP123MB36538F49ADDC422661FD3CE8FFB3A%40CWXP123MB3653.GBRP123.PROD.OUTLOOK.COM.
andrew....@googlemail.com
Thanks
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/9d987d9a-1deb-4be5-8fc0-70fa26131b79%40ameritech.net.
andrew....@googlemail.com
Ahhh! This link gives me exactly what I need – thanks!
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/005d01da1660%24bf7e06f0%243e7a14d0%24%40no.
Marcus D. Leech
THere's this as well:Ahhh! This link gives me exactly what I need – thanks!'
https://www.changpuak.ch/electronics/Coaxial_Tank_VHF_Filter_Designer.php
I've found their online designers to be a hit-or-miss type of deal, though.
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/014301da166f%2401ad3380%2405079a80%24%40googlemail.com.
andrew....@googlemail.com
Thanks!
To view this discussion on the web visit https://groups.google.com/d/msgid/sara-list/8c453d3a-89c5-4cca-a005-e41fe831ff84%40gmail.com.
fasleitung3
http://s53mv.s5tech.net/cavity/cavity.html
We made a couple of these for both 1420 and 1612 MHz and they work
great with typical insertion loss of 0.4 dB.
What I like about this design is that they can be made from standard
parts with the need for milling.
Wolfgang
Andrew Thornett
Sent: Tuesday, November 14, 2023 8:23:05 AM
To: sara...@googlegroups.com <sara...@googlegroups.com>
Subject: Re: [SARA] Narrow bandwidth 1420Mhz filters
--
You received this message because you are subscribed to the Google
Groups "Society of Amateur Radio Astronomers" group.
To post to this group, send email to sara...@googlegroups.com
To unsubscribe from this group, send email to
sara-list-...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
fasleitung3
Wolfgang
Am Dienstag, den 14.11.2023, 08:36 +0000 schrieb 'Andrew Thornett' via
Society of Amateur Radio Astronomers:
>
> Sent from Outlook for Android
andrew....@googlemail.com
-----Original Message-----
From: 'fasleitung3' via Society of Amateur Radio Astronomers <sara...@googlegroups.com>
Sent: 14 November 2023 08:50
To: sara...@googlegroups.com
Subject: Re: [SARA] Narrow bandwidth 1420Mhz filters
Actually it should read "WITHOUT the need for milling"
Wolfgang
Am Dienstag, den 14.11.2023, 08:36 +0000 schrieb 'Andrew Thornett' via Society of Amateur Radio Astronomers:
> Thanks!
>
> Sent from Outlook for Android
> From: 'fasleitung3' via Society of Amateur Radio Astronomers <
> sara...@googlegroups.com>
> Sent: Tuesday, November 14, 2023 8:23:05 AM
> To: sara...@googlegroups.com <sara...@googlegroups.com>
> Subject: Re: [SARA] Narrow bandwidth 1420Mhz filters
>
> I can also recommend the design by Matjaz Vidmar:
> http://s53mv.s5tech.net/cavity/cavity.html
>
> We made a couple of these for both 1420 and 1612 MHz and they work
> great with typical insertion loss of 0.4 dB.
> What I like about this design is that they can be made from standard
> parts with the need for milling.
> Wolfgang
>
>
>
> --
--
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
For more options, visit this group at
http://groups.google.com/group/sara-list?hl=en
---
You received this message because you are subscribed to the Google Groups "Society of Amateur Radio Astronomers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sara-list+...@googlegroups.com.
djl
the design calculators out there.
For the US I did find:
https://www.polycase.com/aluminum-enclosures
for various sized boxes
and: https://ksmetals.com/collections/aluminum
for rod and tubing.
also use for aluminum/aluminum and aluminum/copper interfaces, a little
Noalox-Anti-Oxidant-Compound which keeps the Al oxide at bay.
Thanks again
Don
On 2023-11-14 01:23, 'fasleitung3' via Society of Amateur Radio
------------
"It's always something."
Roseanne Rosannadanna
----------------------"
Dr. Don Latham AJ7LL
PO Box 404, Frenchtown, MT, 59834
VOX: 406-626-4304
fasleitung3
Rectangular aluminium profile 2x1.5 inch, 1/8 wall thickness. See for
example
https://www.metalsdepot.com/aluminum-products/aluminum-rectangle-tube
stock # T3211218
Round aluminium bar 5/16 such as their stock # R3516
As a side note: The metal end caps which are depicted in Matjaz design
are just dust protection. They have no influence on the electrical
paramters as long as the profile is long enough. We are using 3-D
printed dust caps instead.
Wolfgang
Marcus D. Leech
> For US material I believe this comes closest:
> Rectangular aluminium profile 2x1.5 inch, 1/8 wall thickness. See for
> example
> https://www.metalsdepot.com/aluminum-products/aluminum-rectangle-tube
> stock # T3211218
> Round aluminium bar 5/16 such as their stock # R3516
> As a side note: The metal end caps which are depicted in Matjaz design
> are just dust protection. They have no influence on the electrical
> paramters as long as the profile is long enough. We are using 3-D
> printed dust caps instead.
>
> Wolfgang
0.125" 6061 and/or 6063 aluminum
tubing, which I think is closer to the size Matjaz used for his 1420
and 1296 filters.
Cheers
Marcus D. Leech
> For US material I believe this comes closest:
> Rectangular aluminium profile 2x1.5 inch, 1/8 wall thickness. See for
> example
> https://www.metalsdepot.com/aluminum-products/aluminum-rectangle-tube
> stock # T3211218
> Round aluminium bar 5/16 such as their stock # R3516
> paramters as long as the profile is long enough. We are using 3-D
> printed dust caps instead.
>
> Wolfgang
https://www.microchip.com/en-us/product/TFS1429A-SAW-Filter
Part of their passband is in the actual radio astronomy band.
Outfits like PCBWay will do small-run manufacturing of designs these
days pretty easily.