Date   
Re: Help Please FS2-2A-112 Relay

Richard GD8EXI
 

Mike
            My limited experience with PCB mounted RF relays is that insertion loss and isolation has as much to do with how it is mounted on the PCB as that of the relay itself.
 
 If you have one and can terminate it with N Type plugs I would be happy to measure both the insertion loss and the isolation up to 4GHz.
 
73s
Richard
GD8EXI



On 20/02/2019, 22:22, "Mike Webb" <gd6icr@...> wrote:

Got sent spec sheet from UK supplier

Looks good but no detaled RF spec

Thanks all

Mike


Re: Help Please FS2-2A-112 Relay

Richard <richard@...>
 

Well you can calculate cross talk, isolation, and with the contact current handling you can get an idea of max freq
Richard


On Wed, 2019-02-20 at 14:22 -0800, Mike Webb wrote:
Got sent spec sheet from UK supplier

Looks good but no detaled RF spec

Thanks all

Mike
  •  

Re: Help Please FS2-2A-112 Relay

Mike Webb
 

Got sent spec sheet from UK supplier

Looks good but no detaled RF spec

Thanks all

Mike
  •  

Re: Es'Hail PA?

John E. Beech
 

In the past I've used two separate Yagis at right angles to each other, with their feed impedances adjusted to 100 ohms each and then connected to a splitter with equal length cables. One Yagi was mounted a quarter wave in air, in front of the other to produce the required phase shift. Much easier to manufacturer than trying to fiddleabout with cables of uncertain velocity factor. The only drawback is that you can't switch from LHC to RHC easily.

de John G8SEQ

-------Original Message-------
From: on4cjq@...
To: UKMicrowaves <UKMicrowaves@groups.io>
Subject: Re: [UKMicrowaves] Es'Hail PA?
Sent: Feb 20 '19 16:44

I think it's tricky to build a 13cm crossyagi on 1 boom due to
spacings. Maybe 2 seperate yagis and 1/4L shifted.On 23cm i made it
both ways, but used mostly the 23cm crossed yagi, tuned at 1268 RHCP
and worked FB on sat, much better then the 30t helix.The difficulties
at the crossed yagi:the feedpoints. RL was 26dB, but spent a lot of
time in getting it right.

On Eshail in use here at the moment (after 1 week of swapping stuff):
2w into dualband feed on 95cm offset, which gives me a signal 25db
above transponder noise, so gonna use less. RX is lnb octagon-twin
with little less than 26MHz input, to IF 439MHz, 19dB attenuator
between lnb and receiver, no use of SDR etc, and 30cm offset dish.
After next weeks contest i will do a final mod at he station to make
it portabel with 1 dish(75,65 or 30cm), 1 trx,1 feed on a small tripod
or car-roof magnet.

CU on the bird...or next week-contest

Jerry

-------------------------

VAN: "willis mj" <willis.mj@...>
AAN: "UKMicrowaves" <UKMicrowaves@groups.io>
VERZONDEN: Woensdag 20 februari 2019 17:08:52
ONDERWERP: Re: [UKMicrowaves] Es'Hail PA?

You may be better off building two yagis, elements at 90 degrees to
each other on one boom and phasing them for circular. Easier to make.
--
Mike G0MJW

Re: Es'Hail PA?

on4cjq@...
 

I think it's tricky to build a 13cm crossyagi on 1 boom due to spacings. Maybe 2 seperate yagis and 1/4L shifted.On 23cm i made it both ways, but used mostly the 23cm crossed yagi, tuned at 1268 RHCP and worked FB on sat, much better then the 30t helix.The difficulties at the crossed yagi:the feedpoints. RL was 26dB, but spent a lot of time in getting it right.

On Eshail in use here at the moment (after 1 week of swapping stuff): 2w into dualband feed on 95cm offset, which gives me a signal 25db above transponder noise, so gonna use less. RX is lnb octagon-twin with little less than 26MHz input, to IF 439MHz, 19dB attenuator between lnb and receiver, no use of SDR etc, and 30cm offset dish. After next weeks contest i will do a final mod at he station to make it portabel with 1 dish(75,65 or 30cm), 1 trx,1 feed on a small tripod or car-roof magnet.


CU on the bird...or next week-contest

Jerry


Van: "willis mj" <willis.mj@...>
Aan: "UKMicrowaves" <UKMicrowaves@groups.io>
Verzonden: Woensdag 20 februari 2019 17:08:52
Onderwerp: Re: [UKMicrowaves] Es'Hail PA?

You may be better off building two yagis, elements at 90 degrees to each other on one boom and phasing them for circular. Easier to make. 
--
Mike G0MJW

Re: Es'Hail PA?

Mike Willis
 

You may be better off building two yagis, elements at 90 degrees to each other on one boom and phasing them for circular. Easier to make. 
--
Mike G0MJW

Re: Es'Hail PA?

militaryoperator
 


Have a look at patch antennas

HTH
Richard


Ok, thanks Richard. I have had a look at Patch but will do so again. Also Bowtie? looks interesting. 

cheers, Ben.

Re: Es'Hail PA?

Neil Smith G4DBN
 

Helix antennas are *very* tricky indeed to get right. Have a read of Graham's article at g8haj.uk/?page_id=477 about a backfire helix feed.

Neil G4DBN

On 20/02/2019 10:27, militaryoperator via Groups.Io wrote:
Looking at this web site for building a helix antenna


Once wound and fitted to socket on back plate could I slide plastic tube over the helix and glued to back plate so I could hold it in the lnb mount on the dish?

As a feed to the dish how many turns, 4 maybe?


Ben.


Re: Es'Hail PA?

Richard <richard@...>
 

Ben Have a look at Kents comments on helicals
like this one posted earlier:-
I think we can say ham build Helix antennas have a very poor history of performance.
(Include ones built from website calculators and never actually tested)
In theory you need at least 6 turns to generate CP.  So those 2 or 3 turn ones commonly used
as dish feeds are ellipticaly polarized, not really circular.
The biggest problem is experimenters who build the helix by wrapping the helix around
a plastic pipe using free space calculations.    With the traditional wire helix, the waves
travel down the wire at about 95% the speed of light.    When the helix is wrapped around
a plastic tube, the waves travel down the wire at about about 60% the speed of light.
(Much like a transmission line on a PCB)
What the heck did that do to your turns ratio?     Yes, in theory this can be part of the
design, but what is the dielectric constant of the plastic tube you got a Pound Land 
when measured at 2.4 GHz???
Historically only about 10% of the ham built helix antenna had forward gain and had an axial
ratio of 3dB or better.      Sam noticed another problem with ham built helix antennas.
One technique is to use just one rod and attach the helix to the rod at each turn.   Typically
this rod is along the top of the coil.    Well, the rod acts much like a dielectic lens and bends the beam.
So with a top rod, the beam is bent upwards.   Your beam is not where you think it is
and it totally destroys the gain of really long helix antennas.     The classical way of a
central post and spokes going out to the helix wire is the best way to build them, unless you
have test facilities.       Oh yes, the classic Krause formulas for helix antennas are about 3 dB optimistic.
Kent   G8EMY/2E0VAA/WA5VJB



Have a look at patch antennas

HTH
Richard

On Wed, 2019-02-20 at 10:27 +0000, militaryoperator via Groups.Io wrote:
Looking at this web site for building a helix antenna

https://www.changpuak.ch/electronics/calc_12b.php

Once wound and fitted to socket on back plate could I slide plastic tube over the helix and glued to back plate so I could hold it in the lnb mount on the dish?

As a feed to the dish how many turns, 4 maybe?


Ben.


Re: Es'Hail PA?

militaryoperator
 

Looking at this web site for building a helix antenna

https://www.changpuak.ch/electronics/calc_12b.php

Once wound and fitted to socket on back plate could I slide plastic tube over the helix and glued to back plate so I could hold it in the lnb mount on the dish?

As a feed to the dish how many turns, 4 maybe?


Ben.


Re: Es'Hail PA?

Richard <richard@...>
 


I just got 2 black .png images

Richard

On Tue, 2019-02-19 at 22:09 +0000, geoffrey pike via Groups.Io wrote:
Hi Paul,
Probably will be fine for DATV at about 1/4  of the power
You will soon find out!
Geoff
GI0GDP


On Tuesday, 19 February 2019, 20:22:38 GMT, Paul Bicknell <paul@...> wrote:


Hi  thank you  for the link to the SG Laboratory Ltd. Power amp / linear  It looks’ just the job for SSB  I wonder if it will be linear enough for video

 

 


From: UKMicrowaves@groups.io [mailto: UKMicrowaves@groups.io ] On Behalf Of geoffrey pike via Groups.Io
Sent: 19 February 2019 10:46
To: UKMicrowaves@groups.io
Subject: Re: [UKMicrowaves] Es'Hail PA?

 

Just supposing it does 8 watts (very unlikely) and you get your money back its all a hassle

 

Pyro Joe is i imagine the route to go ( not cheap but works)

 

But i see that this will be available soon

 

 

 


 

SG Laboratory Ltd.

Peter Gramatikov

13 cm transverter

 It really revolves around your attitude to risk! and if you wish to use both narrow and wide modes.

cheers

Geoff

GI0GDP

 

My question would be what 8 watt device would this amplifier use? and yes 37 dBm mmmh (5watts)

 

On Tuesday, 19 February 2019, 09:33:13 GMT, Chris G8BKE via Groups.Io <eightgoes@...> wrote:

 

 

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.8048 / Virus Database: 4793/15884 - Release Date: 08/14/18
Internal Virus Database is out of date.


--
Paul Bicknell G8KFW   South Coast UK

Re: Es'Hail PA?

geoffrey pike
 

Hi Paul,
Probably will be fine for DATV at about 1/4  of the power
You will soon find out!
Geoff
GI0GDP


On Tuesday, 19 February 2019, 20:22:38 GMT, Paul Bicknell <paul@...> wrote:


Hi  thank you  for the link to the SG Laboratory Ltd. Power amp / linear  It looks’ just the job for SSB  I wonder if it will be linear enough for video

 

 


From: UKMicrowaves@groups.io [mailto: UKMicrowaves@groups.io ] On Behalf Of geoffrey pike via Groups.Io
Sent: 19 February 2019 10:46
To: UKMicrowaves@groups.io
Subject: Re: [UKMicrowaves] Es'Hail PA?

 

Just supposing it does 8 watts (very unlikely) and you get your money back its all a hassle

 

Pyro Joe is i imagine the route to go ( not cheap but works)

 

But i see that this will be available soon

 

 

 


 

SG Laboratory Ltd.

Peter Gramatikov

13 cm transverter

 It really revolves around your attitude to risk! and if you wish to use both narrow and wide modes.

cheers

Geoff

GI0GDP

 

My question would be what 8 watt device would this amplifier use? and yes 37 dBm mmmh (5watts)

 

On Tuesday, 19 February 2019, 09:33:13 GMT, Chris G8BKE via Groups.Io <eightgoes@...> wrote:

 

 

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.8048 / Virus Database: 4793/15884 - Release Date: 08/14/18
Internal Virus Database is out of date.


--
Paul Bicknell G8KFW   South Coast UK

Re: Es'Hail PA?

Paul G8KFW
 

Hi  thank you  for the link to the SG Laboratory Ltd. Power amp / linear  It looks’ just the job for SSB  I wonder if it will be linear enough for video

 

 


From: UKMicrowaves@groups.io [mailto:UKMicrowaves@groups.io] On Behalf Of geoffrey pike via Groups.Io
Sent: 19 February 2019 10:46
To: UKMicrowaves@groups.io
Subject: Re: [UKMicrowaves] Es'Hail PA?

 

Just supposing it does 8 watts (very unlikely) and you get your money back its all a hassle

 

Pyro Joe is i imagine the route to go ( not cheap but works)

 

But i see that this will be available soon

 

 

 


 

SG Laboratory Ltd.

Peter Gramatikov

13 cm transverter

 It really revolves around your attitude to risk! and if you wish to use both narrow and wide modes.

cheers

Geoff

GI0GDP

 

My question would be what 8 watt device would this amplifier use? and yes 37 dBm mmmh (5watts)

 

On Tuesday, 19 February 2019, 09:33:13 GMT, Chris G8BKE via Groups.Io <eightgoes@...> wrote:

 

 

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2016.0.8048 / Virus Database: 4793/15884 - Release Date: 08/14/18
Internal Virus Database is out of date.


--
Paul Bicknell G8KFW   South Coast UK

Re: IC heatsinkung ........

Richard <richard@...>
 

I've uploaded two application notes on heat sinking plastic body RF
transistors and RFICs.
Worth reading
Richard

On Tue, 2019-02-19 at 16:56 +0000, Neil Smith G4DBN wrote:
The thermal performance of the SE5004L is totally invisible on any
of
the data sheets I've looked at. Never a good sign.

I'm running it at at 30.5dBm output, and the heat spreader temp
settles
at about 42C after ten minutes key-down, so I'm making an assumption
based on calculation of the thermal conductivity of the solder
layer,
under-pad vias plus fill, transfer compound and aluminium spreader,
that
the case temp at top and bottom will be no more than 8C above the
spreader and bridge temp, so around 50C.

Some moderately useful info here:
http://www.skyworksinc.com/uploads/documents/Thermal_Design_Considerations_for_AWB_Products_204919B_AN.pdf

Some other 5mm Skyworks devices have a thermal resistance of 20C/W,
so
on that basis, at 30.5dBm out, with 700mA and 5V in, that is 2.5W
dissipation in the chip, so perhaps a 50C rise, taking the junction
temperature to around 100C. That seems comfortable to me. Pushing
it
to 33dBm out, the current in rises to 1A, so perhaps 3.5W
dissipation,
and 120C junction temp. Not so comfy, but probably still just about
OK.
Quiescent dissipation is 1.5W, but I turn it off during receive.
Whether
that makes any sense (temperature cycling?) is another question,
also,
that 20C/W is based entirely on guesswork. Not ideal. In the
finished
setup, there will be much more cooling, and a max of about 45% duty
cycle, so the spreader is likely to stay under 40C.

In the DF9IC 23cm PAs, I solder the MRFE6S9160 FETs down to the big
copper spreader, and mount the spreader on a watercooling plate, so
the
FET package base doesn't rise to much over 27C after 10 minutes key-
down
at full chat. Those devices have a thermal resistance junction to
case
of 0.33C/W. At 280W out, 30V and 22A in, the dissipation per FET is
180W, so the junctions should be at about 63C more than the case, a
very
comfy 90C. The case tops only reach about 48C after 10 mins key
down,
so I haven't bothered with any cooling bridges over the FETs.

Neil G4DBN


On 19/02/2019 16:04, Chris Bartram G4DGU wrote:
Mark's note and Gordon's comment both, in different ways, make a
particularly important point. Good thermal design is absolutely
crucial to the reliability of power amplifiers. Ignore it at the
devices, and your reputation's peril! For professional standard
reliability it's usual to design the the cooling system to arrive
at a
temperature in the channel of the FET (or at the junction of a
BJT)
not exceeding about 120C. In my former life, I wouldn't have
looked
twice at any PA device which didn't have a data sheet showing the
thermal resistance from the channel/junction to the heatsink
flange.

Because modern modulation schemes often have a peak-to-mean power
ratio of ten or more, a lot of devices are being deliberately
designed
with excellent peak power handling capability, but which due to
package thermal limitations are not capable of sustained operation
at
close to that point.

It's absolutely necessary to fully read and understand data sheets
before choosing devices!

New file uploaded to UKMicrowaves@groups.io

UKMicrowaves@groups.io Notification <UKMicrowaves+notification@...>
 

Hello,

This email message is a notification to let you know that a file has been uploaded to the Files area of the UKMicrowaves@groups.io group.

File: AN3263.pdf

Uploaded By: Richard

Description:
Bolt Down Mounting Method for High Power RF Transistors and RFICs in Over - Molded Plastic Packages

You can access this file at the URL:
https://groups.io/g/UKMicrowaves/files/Application%20Notes/AN3263.pdf

Cheers,
The Groups.io Team

Re: IC heatsinkung ........

Neil Smith G4DBN
 

The thermal performance of the SE5004L is totally invisible on any of the data sheets I've looked at.  Never a good sign.

I'm running it at at 30.5dBm output, and the heat spreader temp settles at about 42C after ten minutes key-down, so I'm making an assumption based on calculation of the thermal conductivity of the solder layer, under-pad vias plus fill, transfer compound and aluminium spreader, that the case temp at top and bottom will be no more than 8C above the spreader and bridge temp, so around 50C.

Some moderately useful info here: http://www.skyworksinc.com/uploads/documents/Thermal_Design_Considerations_for_AWB_Products_204919B_AN.pdf

Some other 5mm Skyworks devices have a thermal resistance of 20C/W, so on that basis, at 30.5dBm out, with 700mA and 5V in, that is 2.5W dissipation in the chip, so perhaps a 50C rise, taking the junction temperature to around 100C.  That seems comfortable to me.  Pushing it to 33dBm out, the current in rises to 1A, so perhaps 3.5W dissipation, and 120C junction temp.  Not so comfy, but probably still just about OK. Quiescent dissipation is 1.5W, but I turn it off during receive. Whether that makes any sense (temperature cycling?) is another question, also, that 20C/W is based entirely on guesswork.  Not ideal.  In the finished setup, there will be much more cooling, and a max of about 45% duty cycle, so the spreader is likely to stay under 40C.

In the DF9IC 23cm PAs, I solder the MRFE6S9160 FETs down to the big copper spreader, and mount the spreader on a watercooling plate, so the FET package base doesn't rise to much over 27C after 10 minutes key-down at full chat.  Those devices have a thermal resistance junction to case of 0.33C/W.  At 280W out, 30V and 22A in, the dissipation per FET is 180W, so the junctions should be at about 63C more than the case, a very comfy 90C.  The case tops only reach about 48C after 10 mins key down, so I haven't bothered with any cooling bridges over the FETs.

Neil G4DBN

On 19/02/2019 16:04, Chris Bartram G4DGU wrote:
Mark's note and Gordon's comment both, in different ways, make a particularly important point. Good thermal design is absolutely crucial to the reliability of power amplifiers. Ignore it at the devices, and your reputation's peril! For professional standard reliability it's usual to design the the cooling system to arrive at a temperature in the channel of the FET (or at the junction of a BJT) not exceeding about 120C. In my former life, I wouldn't have looked twice at any PA device which didn't have a data sheet showing the thermal resistance from the channel/junction to the heatsink flange.

Because modern modulation schemes often have a peak-to-mean power ratio of ten or more, a lot of devices are being deliberately designed with excellent peak power handling capability, but which due to package thermal limitations are not capable of sustained operation at close to that point.

It's absolutely necessary to fully read and understand data sheets before choosing devices!

New file uploaded to UKMicrowaves@groups.io

UKMicrowaves@groups.io Notification <UKMicrowaves+notification@...>
 

Hello,

This email message is a notification to let you know that a file has been uploaded to the Files area of the UKMicrowaves@groups.io group.

File: AN3789.pdf

Uploaded By: Richard

Description:
Clamping devices for max heat transfer

You can access this file at the URL:
https://groups.io/g/UKMicrowaves/files/Application%20Notes/AN3789.pdf

Cheers,
The Groups.io Team

Re: Es'Hail PA?

Chris G8BKE
 

Interesting comments from various posts on these. On balance it seems you get what you pay for and if your prepared to do a bit of rework, they might survive,  but I'll stick to my Spectrian!
I never do trust suppliers who say the item is located in London but the seller is in China. eBay vowed to stamp this out but don't seem to have succeeded. ;)


73 de Chris G8BKE
 
Website:  www.g8bke.co.uk

Re: Es'Hail ?

Dex W4DEX
 



On Tue, Feb 19, 2019 at 9:55 AM KENT BRITAIN <WA5VJB@...> wrote:
No @#$!#! way that little 17 element Yagi has 25 dBi gain!

I have a very similar one I purchased several years ago, it barely did 10 dBi gain.
And that was only after I took those rubber tips off the elements.

Unfortunately I bought two of those Yagis for a point to point link.  One I sold on a $1 hamfest table.  I felt guilty for selling the one so the other in on the aluminum scrap pile.
A pair of homebrew biquads made the link well.  With the Yagis the link was unuseable.  

Dex


Re: IC heatsinkung ........

Chris Bartram G4DGU
 

Mark's note and Gordon's comment both, in different ways, make a particularly important point. Good thermal design is absolutely crucial to the reliability of power amplifiers. Ignore it at the devices, and your reputation's peril! For professional standard reliability it's usual to design the the cooling system to arrive at a temperature in the channel of the FET (or at the junction of a BJT) not exceeding about 120C. In my former life, I wouldn't have looked twice at any PA device which didn't have a data sheet showing the thermal resistance from the channel/junction to the heatsink flange.

Because modern modulation schemes often have a peak-to-mean power ratio of ten or more, a lot of devices are being deliberately designed with excellent peak power handling capability, but which due to package thermal limitations are not capable of sustained operation at close to that point.

It's absolutely necessary to fully read and understand data sheets before choosing devices!

73

Chris G4DGU

--
Chris Bartram
Luxulyan, Bodmin, Cornwall
chris@...