Author Topic: 50 ohm Dummy Load With 50dB Attenuator  (Read 496 times)

0 Members and 1 Guest are viewing this topic.

Offline OnurCan

  • Newbie
  • Posts: 4
  • Country: tr
50 ohm Dummy Load With 50dB Attenuator
« on: May 27, 2020, 07:58:44 am »
Hi,
I am designing a 50 ohm 10 watts dummy load for 100-1000MHz frequency range. The signal strength should not exceed -10dBm. The load must be able to dissipate 5 watts minimum, that is +37dBm. The output to the test equipment must be no more than -10dBm. So I need an attenuation of 47dB but to make it easy lets say 50dB.

After the load as you can see in the schematic picture there is 3 different attenuators. The first one is 20dB attenuator for 5W dissipation, second one is 20dB-50mW dissipation and the last one is 10dB-50mW dissipation. You can see the values of the resistors on the schematic picture.

Specification for the PCB board:

Trace thickness : 0.036mm

Substrate height : 0.8mm

Trace Width : 1.5mm for 50ohms

Substrate Dielectric : 4.3

One thing that I couldn't figure out here is the width of the tracks when routing the attenuators as you can see some of them routed with a narrow line. Should it be as thick as 50ohms or something different considering the values, I don't know this because I have never seen it before. Long story short I need help with the circuit and routing the microstrip to have a good impedance match.


I have a problem with appliying this to a PCB board I know I should arrange R3-R6 symmetrically around the input connector, keeping the current path loops as short as physically possible. But I couldn't figure out myself. If you guys can guide me with placing the resistors and routing my problem will be solved.

I did something on my own you can check it below but I am pretty sure that is not the best way to do it. So waiting for your help.

Thank you for your time.
 

Offline EEEnthusiast

  • Frequent Contributor
  • **
  • Posts: 289
  • Country: in
  • RF boards, Precision Analog, Carpentry
    • https://www.zscircuits.in/
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #1 on: May 27, 2020, 08:55:27 am »
Few points
1. you may want to upsize those 2W resistors to something like 5W or put 8 of them in parallel, instead of 4. The 2W rating is designed for a certain rise in temperature and the value of the resistor. With the 2W resistors and 10W input, you are almost dissipating 2W in those resistors. That would heat them up to a value such that the overall attenuation is impacted significantly. Please check the temp coefficient of those resistors again.

2. Instead of routing those resistor radially, you may just place them parallel on the line itself. That will create shorter stubs.

3. I did not calculate if the individual sections are well matched to 50 Ohms but it looks like they are. In that case, the trace connecting those sections should all be 50 Ohms.

4. Also recommend to add multiple vias when connecting those resistors to the ground plane below. This would reduce the overall inductance. Also recommend to have the ground poured on the top layer as well.
Making products for IOT
https://www.zscircuits.in/
 
The following users thanked this post: OnurCan

Offline EEEnthusiast

  • Frequent Contributor
  • **
  • Posts: 289
  • Country: in
  • RF boards, Precision Analog, Carpentry
    • https://www.zscircuits.in/
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #2 on: May 27, 2020, 09:02:44 am »
Something like below, but place them closer to the SMA
Making products for IOT
https://www.zscircuits.in/
 
The following users thanked this post: OnurCan

Offline OnurCan

  • Newbie
  • Posts: 4
  • Country: tr
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #3 on: May 27, 2020, 09:39:03 am »
Doesn't the signal need to reach all of them at the same time? I mean resistor 1-4.

I am thinking a star shaped topology will solve this but I couldn't figure out how to place them
 

Offline EEEnthusiast

  • Frequent Contributor
  • **
  • Posts: 289
  • Country: in
  • RF boards, Precision Analog, Carpentry
    • https://www.zscircuits.in/
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #4 on: May 27, 2020, 11:19:05 am »
Your highest frequency is only 1GHz. If you place those resistors side by side (few mm apart), they should still be within a fraction of the wavelength.
Making products for IOT
https://www.zscircuits.in/
 
The following users thanked this post: OnurCan

Offline imo

  • Super Contributor
  • ***
  • Posts: 2621
  • Country: 00
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #5 on: May 27, 2020, 11:29:32 am »
..
I have a problem with appliying this to a PCB board I know I should arrange R3-R6 symmetrically around the input connector, keeping the current path loops as short as physically possible. But I couldn't figure out myself. If you guys can guide me with placing the resistors and routing my problem will be solved..
When the length of the traces is, say, less than 5% of the min wavelength the impact of the trace's impedance will be low.
 
The following users thanked this post: OnurCan

Offline M0HZH

  • Regular Contributor
  • *
  • Posts: 82
  • Country: gb
    • QRPblog
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #6 on: May 27, 2020, 11:51:33 am »
Keep in mind larger (higher power) resistors have larger capacitance and perform worse at VHF/UHF. On the other side, smaller resistors will warm up faster and change their resistance too much. It's quite a challenge to achieve good attenuation flatness & power handling up to GHz level with generic parts.

My medium power attenuator uses a RFP1398 flanged attenuator device (100W, -20dB, 0-1GHz), these can be found for a few dollars on eBay/Banggood/AliExpress.
 
The following users thanked this post: OnurCan

Offline OnurCan

  • Newbie
  • Posts: 4
  • Country: tr
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #7 on: May 27, 2020, 12:17:42 pm »
Keep in mind larger (higher power) resistors have larger capacitance and perform worse at VHF/UHF. On the other side, smaller resistors will warm up faster and change their resistance too much. It's quite a challenge to achieve good attenuation flatness & power handling up to GHz level with generic parts.

My medium power attenuator uses a RFP1398 flanged attenuator device (100W, -20dB, 0-1GHz), these can be found for a few dollars on eBay/Banggood/AliExpress.

Thank you for the advice but this is a finishing project for my radio frequency class so buying is not an option for me. I have to build this.
 

Offline Wolfgang

  • Super Contributor
  • ***
  • Posts: 1416
  • Country: de
  • Its great if it finally works !
    • Electronic Projects for Fun
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #8 on: May 27, 2020, 01:24:45 pm »
Hi,
I am designing a 50 ohm 10 watts dummy load for 100-1000MHz frequency range. The signal strength should not exceed -10dBm. The load must be able to dissipate 5 watts minimum, that is +37dBm. The output to the test equipment must be no more than -10dBm. So I need an attenuation of 47dB but to make it easy lets say 50dB.

After the load as you can see in the schematic picture there is 3 different attenuators. The first one is 20dB attenuator for 5W dissipation, second one is 20dB-50mW dissipation and the last one is 10dB-50mW dissipation. You can see the values of the resistors on the schematic picture.

Specification for the PCB board:

Trace thickness : 0.036mm

Substrate height : 0.8mm

Trace Width : 1.5mm for 50ohms

Substrate Dielectric : 4.3

One thing that I couldn't figure out here is the width of the tracks when routing the attenuators as you can see some of them routed with a narrow line. Should it be as thick as 50ohms or something different considering the values, I don't know this because I have never seen it before. Long story short I need help with the circuit and routing the microstrip to have a good impedance match.


I have a problem with appliying this to a PCB board I know I should arrange R3-R6 symmetrically around the input connector, keeping the current path loops as short as physically possible. But I couldn't figure out myself. If you guys can guide me with placing the resistors and routing my problem will be solved.

I did something on my own you can check it below but I am pretty sure that is not the best way to do it. So waiting for your help.

Thank you for your time.
Question:
- Why do you *design* this ? Buy a 20dB 10W attenuator, add another 20dB 1W attenuator, you are done.
With reasonable quality N connectors you end up with a few ten €, and 1GHz with usable data should be no problem at all (most go to 3GHz).
 

Offline G0MJW

  • Contributor
  • Posts: 43
  • Country: gb
  • Mike
    • G0MJW
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #9 on: June 09, 2020, 10:12:39 pm »
Seems a bit over complicated. Why not just design a 5W 50 dB pi attenuator and make the input resistor able to handle 5W? A 50 ohm resistor to ground 7.9k series and 50 ohm to ground. You can make a 5W load using a 50 ohm SMD terminating resistor - e.g. Anaren C10A50Z4 good for 10W up to 3 GHz. Make sure there is sufficient heatsinking to the ground plane - lots of vias.
« Last Edit: June 09, 2020, 10:15:12 pm by G0MJW »
Mike
 

Offline M0HZH

  • Regular Contributor
  • *
  • Posts: 82
  • Country: gb
    • QRPblog
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #10 on: June 10, 2020, 12:36:43 am »
Seems a bit over complicated. Why not just design a 5W 50 dB pi attenuator and make the input resistor able to handle 5W? ...

Because even 0.1pF of parasitic capacitance across the 7.9k resistor would introduce a 15dB variation in attenuation across the 0-1GHz range.
 

Offline Wolfgang

  • Super Contributor
  • ***
  • Posts: 1416
  • Country: de
  • Its great if it finally works !
    • Electronic Projects for Fun
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #11 on: June 10, 2020, 12:58:22 am »
Seems a bit over complicated. Why not just design a 5W 50 dB pi attenuator and make the input resistor able to handle 5W? A 50 ohm resistor to ground 7.9k series and 50 ohm to ground. You can make a 5W load using a 50 ohm SMD terminating resistor - e.g. Anaren C10A50Z4 good for 10W up to 3 GHz. Make sure there is sufficient heatsinking to the ground plane - lots of vias.

Some rules of thumb to make attenuators:

- components must match frequency ranges - GHz without SMD is hopeless, Multi-GHz needs coaxial architectures
- no components very far off 50Ohms, otherwise their parasitics dominate and flatness is gone.
  an 1Ohm resistor is an inductor, a 10K resistor is a capacitor more than just a resistor at high frequencies.
- no stages with more than 20, better 10dB. Cascade stages if you need more attenuation
- watch power handling in the input stages

Special good reason why you need to build ? No ? Buy one from a renowned source, problem gone.


 
The following users thanked this post: EEEnthusiast

Offline OnurCan

  • Newbie
  • Posts: 4
  • Country: tr
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #12 on: June 10, 2020, 07:24:47 am »


Special good reason why you need to build ? No ? Buy one from a renowned source, problem gone.




It is a school project and I have to build this to be able to pass the class that is why I have to build this. Otherwise I would have bought it.
The last design for this project is below if you are curious about it. Didn't get to chance to test it yet but in theory it looks fine and got me a passing grade.
 

Offline G0MJW

  • Contributor
  • Posts: 43
  • Country: gb
  • Mike
    • G0MJW
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #13 on: June 10, 2020, 07:29:59 am »
That depends on the resistor doesn't it? I would have thought exactly the same problem would occur with many input resistors in parallel. However, its true 8k HF resistors are rare beasts. This would be fine for HF. I saw the 1 kHz source.

Even so I would still do it this way with a big power dump broadband terminating resistor but perhaps with smaller steps to keep the series resistance low. A 15 dB attenuator can be made from a 75R terminating chip followed by 120 ohms in series and a series parallel chain to get to the desired loss.

The simplest solution is to buy a set of SMD attenuator chips... cheap, accurate, broadband... D10AA20Z4 but I guess that's cheating.


« Last Edit: June 10, 2020, 07:33:21 am by G0MJW »
Mike
 

Offline Wolfgang

  • Super Contributor
  • ***
  • Posts: 1416
  • Country: de
  • Its great if it finally works !
    • Electronic Projects for Fun
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #14 on: June 10, 2020, 09:12:23 am »


Special good reason why you need to build ? No ? Buy one from a renowned source, problem gone.




It is a school project and I have to build this to be able to pass the class that is why I have to build this. Otherwise I would have bought it.
The last design for this project is below if you are curious about it. Didn't get to chance to test it yet but in theory it looks fine and got me a passing grade.

... OK, then it makes sense. Have fun !
 

Offline geggi1

  • Regular Contributor
  • *
  • Posts: 52
Re: 50 ohm Dummy Load With 50dB Attenuator
« Reply #15 on: June 10, 2020, 11:42:29 am »
I would just get a propper attunator off E-bay or some other internet shop. Prices are typicaly form 20-50$ depending of power handleing and attenuation.
A home mede attenuator will normaly have issues with capacitance, inductance and trace with, not forgetting the propperties of the resistors.
If you are paralelling resistors each branch must have the same length and impedance as the others. all of them must branch form the excact same point.
The higher in frequency the more in accurate.

By using a already made attenuator all of these issues will be removed and your attenuator/dummyload will often have a range of 2-3Ghz meaning that you will work in a flat area of SWR.

https://www.ebay.com/sch/i.html?_from=R40&_trksid=m570.l1313&_nkw=coaxial+attenuator+n-type&_sacat=0&LH_TitleDesc=0&_sop=15&_osacat=0&_odkw=coaxial+attenuator
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf