1420MHz cantenna advice...

[Steve Olney - VK2XV]

I have started to look at 21cm antennas for HI observations and the first attempt is to build a cantenna to do basic tests (not observations).

I thought from the articles on the 'net that this would be simple but I am stuck in 'analysis paralysis'.  The questions I have are:

1.  How far should the probe be from the back wall ?  Using the standard equations for 1420.4MHz and a 154mm diameter can (6.1" dia.) the lambda waveguide (λg) comes to 354mm, so the probe should be about 89mm (3.5") from the back wall.   On the 'net the distance given for 6" dia. cans ranges from 3" to 3.5".
   Are the values closer to 3" for impedance matching purposes ?
   
2. Some references mention to be sure and coat all surfaces with a clear lacquer (except connectors...:-) to prevent corrosion.   Other references say to be sure and remove lacquer from the inside of the can.   The inside of my tin can ($3 piggy bank can from a reject shop - can't afford to buy coffee tins...) is coated with a thin layer of something to prevent corrosion.  Is this going to be a problem ?

Cheers

Steve

[John Mannone]

I'll answer part of this.

The waveguide must be conducting, so I would not coat it with an
insulator like lacquer. However, I don't know how to deal with the
corrosion problem unless you use a sacrificial electrode, maybe zinc if
your waveguide is aluminum. I don't think that's the typical solution.
maybe I wasn't as much help as I thought I would be when I started to
answer :-(

In any case, good luck.

John

On Feb 11, 2014, at 8:10 PM, Steve Olney - VK2XV wrote:

> I have started to look at 21cm antennas for HI observations and the
> first attempt is to build a cantenna to do basic tests (not
> observations).
>
> I thought from the articles on the 'net that this would be simple but
> I am stuck in 'analysis paralysis'. The questions I have are:
>

> 1 How far should the probe be from the back wall ? Using the

> standard equations for 1420.4MHz and a 154mm diameter can (6.1" dia.)
> the lambda waveguide (λg) comes to 354mm, so the probe should be about
> 89mm (3.5") from the back wall. On the 'net the distance given for
> 6" dia. cans ranges from 3" to 3.5".
> Are the values closer to 3" for impedance matching purposes ?

> 2 Some references mention to be sure and coat all surfaces with a

> clear lacquer (except connectors...:-) to prevent corrosion. Other
> references say to be sure and remove lacquer from the inside of the
> can. The inside of my tin can ($3 piggy bank can from a reject shop
> - can't afford to buy coffee tins...) is coated with a thin layer of
> something to prevent corrosion. Is this going to be a problem ?
> Cheers
>
> Steve
>

> --
> --
> 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.
> For more options, visit https://groups.google.com/groups/opt_out.

[Marcus D. Leech]

On 02/11/2014 08:17 PM, John Mannone wrote:

I'll answer part of this.

The waveguide must be conducting, so I would not coat it with an insulator like lacquer. However, I don't know how to deal with the corrosion problem unless you use a sacrificial electrode, maybe zinc if your waveguide is aluminum. I don't think that's the typical solution. maybe I wasn't as much help as I thought I would be when I started to answer :-(

In any case, good luck.

John

I've never had a problem coating waveguide feedhorns with a thin protective layer of lacquer.  Surface conductivity does affect efficiency somewhat,
  but having a rusty interior surface is worse.

The coffee-can feed I used last summer for my 21cm back-yard radio telescope was painted.  I was still able to get great results.

On Feb 11, 2014, at 8:10 PM, Steve Olney - VK2XV wrote:

I have started to look at 21cm antennas for HI observations and the first attempt is to build a cantenna to do basic tests (not observations).

I thought from the articles on the 'net that this would be simple but I am stuck in 'analysis paralysis'. The questions I have are:

1 How far should the probe be from the back wall ? Using the standard equations for 1420.4MHz and a 154mm diameter can (6.1" dia.) the lambda waveguide (

λg) comes to 354mm, so the probe should be about 89mm (3.5") from the back wall. On the 'net the distance given for 6" dia. cans ranges from 3" to 3.5".
Are the values closer to 3" for impedance matching purposes ?
2 Some references mention to be sure and coat all surfaces with a clear lacquer (except connectors...:-) to prevent corrosion. Other references say to be sure and remove lacquer from the inside of the can. The inside of my tin can ($3 piggy bank can from a reject shop - can't afford to buy coffee tins...) is coated with a thin layer of something to prevent corrosion. Is this going to be a problem ?
Cheers

Steve

--
--
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.
For more options, visit https://groups.google.com/groups/opt_out.

-- 
Marcus Leech
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org

[Steve Olney - VK2XV]

Marcus and John,

On Wednesday, February 12, 2014 12:21:53 PM UTC+11, Marcus wrote:

I've never had a problem coating waveguide feedhorns with a thin protective layer of lacquer.  Surface conductivity does affect efficiency somewhat,
  but having a rusty interior surface is worse.

The coffee-can feed I used last summer for my 21cm back-yard radio telescope was painted.  I was still able to get great results.

This makes sense to me.  I have nil experience of microwaves and waveguides, but my general RF experience had me assuming that a dielectric coating would introduce minimal loss (tenths of dBs) over the length of the cantenna.  Given that I have seen feed horns mounted on PVC pipes (and so firing through them) I had taken that view.  It was only when I saw a few references to removing inside coatings that I was concerned.  I have since found a reference to the extra loss at 60GHz found by drawing a polythene tube through a circular waveguide.  It gives a measured extra loss over an unlined 14m of about 0.33dB.  I take from this that a lacquer or other thin coating will not be significant for the cantenna.

Marcus - I see your article specifies 3" as the distance from the back wall for the probe.  Was this calculated or determined empirically from gain or impedance measurements ?

Cheers

Steve

[Marcus D. Leech]

--

Based on a 6" waveguide, and given the operating frequency, that distance is "about right".  Ideally, you want a slider that allows you to slide
  the feed probe back-and-forth for ideal power transfer--getting the mechanicals and electricals right at the same time on that is a challenge....


The original dimensions I got from the SETI League article on circular waveguide feeds for 21cm.  Also the VE4MA circular-waveguide
  feed articles.

If you're using cofee-cans or equiv, you can afford to experiment a bit.  And you can also just cover up the "dead" holes with aluminum
  tape or similar.

[Steve Olney - VK2XV]

Marcus,

On Wednesday, February 12, 2014 1:38:07 PM UTC+11, Marcus wrote:

Based on a 6" waveguide, and given the operating frequency, that distance is "about right".  Ideally, you want a slider that allows you to slide
  the feed probe back-and-forth for ideal power transfer--getting the mechanicals and electricals right at the same time on that is a challenge....

The original dimensions I got from the SETI League article on circular waveguide feeds for 21cm.  Also the VE4MA circular-waveguide
  feed articles.

OK - I see the SETI details give a distance of 3.5" (http://www.setileague.org/hardware/feedchok.htm and http://www.setileague.org/hardware/probe.gif)  which agrees with my calcs here from the standard equations - so I will go with that.

I can probably devise a semi-permanent slider to allow a couple of increments either way.

Thanks for your help - much appreciated.

Cheers

Steve

[Paul Oxley]

Marcus

The tuning is best made by having the probe constructed out of telescoping tubing. You can find this kind of tubing at a model airplane store where they use it for fuel feeds. The tuning should be done to get the best return loss (VSWR) on the antenna port to match the coax. It is tuned by sliding telescoping tubing to change the probe length. When it is tuned, you can lock the setting with finger nail polish.

I agree with the distances shown in the SETI tool.

Paul Oxley

From: Steve Olney - VK2XV <st...@joataman.net>
To: sara...@googlegroups.com
Sent: Tuesday, February 11, 2014 10:06 PM
Subject: Re: [SARA] 1420MHz cantenna advice...

--

[Steve Olney - VK2XV]

Paul,

I get that the fuel feed tubing is hollow, but how is it telescoping ?

Cheers

Steve

[Jan Lustrup]

Sreve,

For model airplan copper/brass tubing you use two sizes, so that they slide into each other, and lock it with a dab of superglue,  lacker or nail polish.

 

I  use the SETI spacing for my  probe made of “thin copper tubing”.

My probe is soldered to the N-conector witch is bolted to the horn side.

My probe tip has a brass screw (25mm long, 2mm diameter) with brass nut so I can screw in or out of the prob as to adjust of best return loss (SWR) and tighten the nut to keep it in place.

This works for me over years in a harsh saltwater environment.

 

Earlier I tried a coffe can type feedhorn and it became usless due to rust (I did not paint it). So I would give it a thin coating of something to make it last in weather.

 

Good luck,

 

Jan Lustrup LA3EQ

 

 

 

From: Steve Olney - VK2XV

Sent: Wednesday, February 12, 2014 5:46 AM

To: sara...@googlegroups.com

Cc: Paul Oxley ; oxl...@att.net

Subject: Re: [SARA] 1420MHz cantenna advice...

 

--

[Steve Olney - VK2XV]

G'day Jan,

I have a copy of the SARA article (July 2012) about your experiments with small antennas - very interesting.

On Wednesday, February 12, 2014 8:08:37 PM UTC+11, JAN LUSTRUP - LA3EQ wrote:

For model airplane copper/brass tubing you use two sizes, so that they slide into each other, and lock it with a dab of superglue,  lacker or nail polish.

OK - I understand now.  Funny how I start looking for photons from distant hydrogen clouds and along the way learn about model airplane fuel feeds... :-) 

I  use the SETI spacing for my  probe made of “thin copper tubing”.

My probe is soldered to the N-connector which is bolted to the horn side.

My probe tip has a brass screw (25mm long, 2mm diameter) with brass nut so I can screw in or out of the probe as to adjust of best return loss (SWR) and tighten the nut to keep it in place.

Understood.   I have also seen arrangements where the probe tip is made short and tuned by a screw entering from the other side of the waveguide.

What gear do you use to measure return loss at 1420MHz ?   Is there something affordable ready-made I can get ?

 

Earlier I tried a coffee can type feedhorn and it became useless due to rust (I did not paint it). So I would give it a thin coating of something to make it last in weather.

OK - good idea.   Marcus has done the same. My tin has a covering but I have scratched it in places to solder to (I soldered along the longitudinal seam and around the base) so I will give it a spray to cover those spots.

Thanks for the info.

Cheers

Steve

[Jan Lustrup]

Steve,

 

One should use a VNA to tune for the best return loss, but they are very, very expensive.

I used my spectrum analyzer, SHF signal sweeper and a SHF directional coupler to find the “sweet spot” on the probe.

But I don’t know about any cheap VNA’s that cover 1420MHz. There is the “Mini VNA” that goes from 100kHz to 200MHz and with extender all the way up to 1200MHz, but that’s short of the 21cm band.

You could do it realy cheap by just doing alot of “ground noise/cold sky” measurments until you find the best sky/ground “Y”  noise ratio. It’s a lot of work, but it should get you close to good return loss for free! Be aware that the LNA output noise level will change with changing impeadances as you adjust the pobe length so you do not want to tune for max noise! You are looking for the biggest difference between sky noise and groud noise- Just be sure you point the antenna to quiet sky direction.

I use my ham radio with a 1420MHz converter, in AM mode with AGC turned off, the audio is feed to the PC running RadioSkypipe chart program. Then it should be easy to see the difference between ground noise and skynoise.

 

Have fun..

 

73’s

Jan Lustup LA3EQ

 

 

 

From: Steve Olney - VK2XV

Sent: Wednesday, February 12, 2014 1:59 PM

To: sara...@googlegroups.com

Cc: Jan Lustrup

Subject: Re: [SARA] 1420MHz cantenna advice...

[Paul Oxley]

Steve

It appears that your question in this e-mail was answered by others.

However, in answer to a different question in one of the e-mails:

To tune the probe, I used a Wiltron SWR Autotester Model SP2369 together with a signal generator (At 1420 MHz) and a Digital Voltmeter. In addition, I found that the adjustment was very difficult inside the waveguide. Therefore, I used a bolt and two nuts as a measurement device. The distance between the bolt head and nuts was set with a digital micrometer outside the waveguide. I developed a table of probe length vs voltage as displayed on the digital meter.  During the test, I had the waveguide feed mounted between the legs on a 12 foot ladder pointed down at the floor of my basement. I experimented a bit and found that the VSWR didn't change with small changes in the height above the floor. Thus I concluded that I wasn't seeing significant returned signal from the antenna aperture. You want to find the minimum voltage and set the probe to that distance.

The Wiltron unit is spec'd for 2 to 12 GHz. However, I found out that it worked OK at 1420 MHz.

Many other SWR or return loss measurement test equipment would work. Most of them are much more expensive than the Wiltron unit.

If you don't have the ability to tune for best SWR, you might be able to peak the reading on some source. However, you wouldn't know exactly what frequency you were optimizing.

Paul Oxley

From: Steve Olney - VK2XV <st...@joataman.net>
To: sara...@googlegroups.com

Cc: Paul Oxley <oxl...@att.net>; oxl...@att.net
Sent: Tuesday, February 11, 2014 11:46 PM

Subject: Re: [SARA] 1420MHz cantenna advice...

[Steve Olney - VK2XV]

Jan,

On Thursday, February 13, 2014 3:21:17 AM UTC+11, JAN LUSTRUP - LA3EQ wrote:

One should use a VNA to tune for the best return loss, but they are very, very expensive.

Yes - they are a little 'over budget'. 

You could do it really cheap by just doing a lot of “ground noise/cold sky” measurements until you find the best sky/ground “Y”  noise ratio.

OK - understood.

One question - why cannot a comparison be made between the environment with and without a blocking screen.  I saw this video which shows the difference between a screened and unscreened cantenna...

http://www.youtube.com/watch?v=4jWVbjuOj8Q

[Marcus D. Leech]

Shorting out the feedhorn produces unpredictable results, because it wildly changes the impedance.

A better test is:  point at ground, point at celestial pole.  You're looking to maximize the difference--looking for the difference in detected power between roughly 10K of sky brightness temperature, and roughly 297K (+/- ambient) ground brightness temperature.

[Steve Olney - VK2XV]

Paul,

Thanks for the info.

On Thursday, February 13, 2014 3:35:31 AM UTC+11, Paul Oxley wrote:

To tune the probe, I used a Wiltron SWR Autotester Model SP2369 together with a signal generator (At 1420 MHz) and a Digital Voltmeter.  

 

The Wiltron unit is spec'd for 2 to 12 GHz. However, I found out that it worked OK at 1420 MHz.

What type of connector is that on the DUT port on that unit.  I don't recognise it.

Also, what is the drive level required ?

Cheers

Steve

 

[Steve Olney - VK2XV]

Marcus,

On Thursday, February 13, 2014 7:20:34 AM UTC+11, Marcus wrote:

Shorting out the feedhorn produces unpredictable results, because it wildly changes the impedance.

OK - I bow to your experience here. I confess I would have thought that this would be more predictable than pointing at the ground.  Time for me to experiment instead of talking to increase my understanding.

BTW - I am handicapped here as there is nowhere on our block where the 90 degrees beamwidth of the cantenna does not include lots of trees... :-(

Cheers

Steve

[Steve Olney - VK2XV]

Doh !!!  When I said...

"I confess I would have thought that this would be more predictable than pointing at the ground."   I meant...

"I confess I would have thought that this would be more repeatable than pointing at the ground."

Steve

[Steve Olney - VK2XV]

Paul,

OK - I have tracked it done, it's an APC-7 connector.   Genderless - never seen one before.

Steve

[Wolfgang]

Steve,

I believe this is a APC7 connector.

Regards

Wolfgang

 

 

---

Von: sara...@googlegroups.com [mailto:sara...@googlegroups.com] Im Auftrag von Steve Olney - VK2XV
Gesendet: Mittwoch, 12. Februar 2014 21:21
An: sara...@googlegroups.com
Cc: Steve Olney - VK2XV; Paul Oxley; oxl...@att.net
Betreff: Re: [SARA] 1420MHz cantenna advice...

[Paul Oxley]

Wolfgang

You are correct. My Wiltron came with an adapter to a type N connector.

Paul

From: Wolfgang <fasle...@googlemail.com>
To: sara...@googlegroups.com
Cc: 'Steve Olney - VK2XV' <st...@joataman.net>
Sent: Wednesday, February 12, 2014 4:24 PM
Subject: AW: [SARA] 1420MHz cantenna advice...

Steve,

I believe this is a APC7 connector.

Regards

Wolfgang

 

 

Von: sara...@googlegroups.com [mailto: sara...@googlegroups.com ] Im Auftrag von Steve Olney - VK2XV
Gesendet: Mittwoch, 12. Februar 2014 21:21
An: sara...@googlegroups.com
Cc: Steve Olney - VK2XV; Paul Oxley; oxl...@att.net
Betreff: Re: [SARA] 1420MHz cantenna advice...

 

 

What type of connector is that on the DUT port on that unit.  I don't recognise it.