Author Topic: Ultra low insertion loss SPDT 1.4GHz RF switch  (Read 1215 times)

0 Members and 1 Guest are viewing this topic.

Online srce

  • Regular Contributor
  • *
  • Posts: 172
  • Country: gb
Ultra low insertion loss SPDT 1.4GHz RF switch
« on: October 21, 2021, 11:12:53 am »
I'm looking for an ultra low insertion loss SPDT RF switch for 1.4GHz.

Can anyone recommend anything better than an Infineon BGS12PN10? Insertion loss is given as around 0.17dB at 1.4GHz.

Input power is very low, Z0=50Ohm, can supply any voltage and cost not really important either.
 

Offline MartinL

  • Contributor
  • Posts: 31
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #1 on: October 21, 2021, 12:43:12 pm »
What are you trying to do? Whatever it is, if it depends on getting better than 0.17dB loss in a switch, then I think you probably need a different approach. That's already pretty spectacularly low. It's getting down to the range where, even if you found a part whose datasheet claimed a lower figure, I'd be surprised if you were able to reliably measure a difference between them that would be distinguishable from the noise of individual manufacturing variations between units and boards.

Here's an electromechanical one with SMA ports and a 12V solenoid drive for $127 - but it still only promises to get you down to 0.15dB: https://www.digikey.com/en/products/detail/radiall-usa-inc/R570312000/10521679

 

Online srce

  • Regular Contributor
  • *
  • Posts: 172
  • Country: gb
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #2 on: October 21, 2021, 02:32:07 pm »
What are you trying to do? Whatever it is, if it depends on getting better than 0.17dB loss in a switch, then I think you probably need a different approach.
A switch for a Dicke Switch radiometer. The switch goes in front of the LNAs so minimising loss is critical. Any ideas for a different approach?
 

Offline MartinL

  • Contributor
  • Posts: 31
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #3 on: October 21, 2021, 09:16:01 pm »
Hmm, interesting. Not a topic I'm familiar with I'm afraid.

But poking around the literature a bit, I see this paper which proposes the use of a "dual path" LNA, such that no separate switch is needed:

Looking at the circuit proposed, it's basically just two separate frontend LNAs, simultaneously connected to the input of a second stage LNA, with only one powered up at a time.

They're working on-chip at 100GHz, but I don't see why the same approach couldn't be applied in a board level design at 1.4GHz.

The presence of the inactive LNA would have some effect on the load impedance seen by the active one, and the source impedance seen by the next stage. But it should be possible to design a matching network in the middle such that everything sees the correct impedance when either one of the two LNAs is active. You would want to choose LNAs with suitable shutdown behaviour - ideally a high output impedance when off.
 

Offline Joel_Dunsmore

  • Regular Contributor
  • *
  • Posts: 81
  • Country: us
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #4 on: October 22, 2021, 05:56:09 am »
I think ADI has some MEMs switches that could work but your application should be extremely low power so there are likely lots of candidates.    I think PIN will be lower loss than FET switches, I can't tell the technology of the Infineon switch.  We have such as switch in the Fieldfox but I don't know the technology.  In fact we have it in the PNA-X but we found the improvement using it wasn't worth the trouble (you have to know the temperature very well of the load side of the switch, and our instrument drift was lower than the temperature drift). I'll ask and write back.  will this be room temperature of cryo?
 

Online srce

  • Regular Contributor
  • *
  • Posts: 172
  • Country: gb
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #5 on: October 22, 2021, 08:33:46 am »
I don't see why the same approach couldn't be applied in a board level design at 1.4GHz.
Yeah, there are some other examples where it's done at the board level with the switch after the first LNA. This obviously has lower noise, but there's perhaps more possibility for gain variation between two different ICs (which is problematic for calibration) or if there is a thermal difference between them. I may give both a try to see how they compare.
 

Online srce

  • Regular Contributor
  • *
  • Posts: 172
  • Country: gb
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #6 on: October 22, 2021, 10:57:45 am »
will this be room temperature of cryo?
I was going to try some basic cooling with a Peltier module. Don't have any experience with cryogenics. If there is an off-the-shelf cooling solution not much bigger than around 10cmx10cmx10cm I'd be interested to hear about it.
 

Online srce

  • Regular Contributor
  • *
  • Posts: 172
  • Country: gb
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #7 on: October 22, 2021, 02:21:01 pm »
I think ADI has some MEMs switches that could work but your application should be extremely low power so there are likely lots of candidates. 
Had a quick look - there's the ADGM1304 which looks like it has similar insertion loss to the Infineon part (although hard to tell precisely from the graph), however, being MEMS, it has a limit to the number of times it can switch. In the literature, Dicke Switches switch at 100-1kHz (not quite sure I'll need this frequency), so a lifetime of 1 billion switches, is only a few months.
 

Offline MartinL

  • Contributor
  • Posts: 31
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #8 on: October 22, 2021, 02:30:15 pm »
but there's perhaps more possibility for gain variation between two different ICs (which is problematic for calibration) or if there is a thermal difference between them.

One way to reduce variations between the two frontend LNAs would be to use a part with two independent LNAs on one die. Maybe GRF2078?
 

Online srce

  • Regular Contributor
  • *
  • Posts: 172
  • Country: gb
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #9 on: October 22, 2021, 04:47:17 pm »
One way to reduce variations between the two frontend LNAs would be to use a part with two independent LNAs on one die. Maybe GRF2078?
Yep - the downside is that they seem to have a higher NF than single channel parts. Will probably take a bit of experimentation to see which trade-off is the least important.
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 3627
  • Country: us
  • Retired, now restoring antique test equipment
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #10 on: October 22, 2021, 05:08:44 pm »
What are you trying to do? Whatever it is, if it depends on getting better than 0.17dB loss in a switch, then I think you probably need a different approach.
A switch for a Dicke Switch radiometer. The switch goes in front of the LNAs so minimising loss is critical. Any ideas for a different approach?
What is the noise figure of that radiometer?  The attenuation added before that input will degrade the LNA noise figure by the attenuation value.
 

Online srce

  • Regular Contributor
  • *
  • Posts: 172
  • Country: gb
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #11 on: October 22, 2021, 05:36:55 pm »
What are you trying to do? Whatever it is, if it depends on getting better than 0.17dB loss in a switch, then I think you probably need a different approach.
A switch for a Dicke Switch radiometer. The switch goes in front of the LNAs so minimising loss is critical. Any ideas for a different approach?
What is the noise figure of that radiometer?  The attenuation added before that input will degrade the LNA noise figure by the attenuation value.
The datasheet figure for the 1st LNA is about 0.2dB, so even though 0.17dB is pretty good for a switch, it's a significant degredation in the overall NF.

« Last Edit: October 22, 2021, 05:47:24 pm by srce »
 

Offline TimFox

  • Super Contributor
  • ***
  • Posts: 3627
  • Country: us
  • Retired, now restoring antique test equipment
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #12 on: October 22, 2021, 06:49:40 pm »
Remember that the NF gives the noise added to the (theoretical minimum) noise from the source impedance, not the SNR itself.
 

Offline awallin

  • Frequent Contributor
  • **
  • Posts: 694
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #13 on: October 22, 2021, 06:54:34 pm »

Can anyone recommend anything better than an Infineon BGS12PN10? Insertion loss is given as around 0.17dB at 1.4GHz.


mechanical coaxial relay from e.g. mini-circuits?
https://www.mini-circuits.com/WebStore/dashboard.html?model=RC-1SPDT-A18
rated to 18 GHz (!) - so 1.4 GHz is a low frequency and datasheet p.4 shows something under 0.05 dB for 1.4 GHz
they might have only the SPDT switch without box, psu and ethernet/usb interface also as a product.
 

Offline SQ9MT_PL

  • Contributor
  • Posts: 12
  • Country: pl
  • RF Circuit PCB designing implementing measurements
Re: Ultra low insertion loss SPDT 1.4GHz RF switch
« Reply #14 on: October 30, 2021, 08:17:57 am »
A good (expensive) mechanical relay will always perform better than a solid state one.
However, such relays are more and more often used in measuring electronic systems and it is "not from yesterday" the best if it is possible for you to present the system in which it is supposed to work.
The question is what parameters you want to achieve and choose the price-quality ratio. Well, not to mention that such a mechanical switch will be large due to its design.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf