help me on filter vswr!!!!!!!!!!!!!!!

[qiang]

i am doing coaxal cavity filters. in the room temperature,my filter is good
but in low temperture (-25),the vswr is bad,why??
i kown the passband have some movement with temprature,but in my experrence,in the new passband,the vswr can't be too bad.
why,
thank you

 

[vfone]

Check for humidity point in the Temperature Chamber.
High level of humidity can increase the VSWR.
regards

 

[qiang]

thanks,i am strange that in high temperature the vswr is good
but in the low temperature some is good and others is bad the most is bad. then how to explane this

 

[qiang]

now the vswr in the high temperature and low temperature is all bad
why?????

 

[xtasa]

[Giang,

It is normal that VSWR change with temperature.It depend from numerous things
1.Design-very narrow and sharp filters are very sensitive to temperature change.
2.You have to calculate temperature coeficient for used materials and made design with all to be close zero.
3. If you use copper rod with temparature it will have positive coeficient. Copper for use in wide temperature range have to be put in termic chamber to lost all tensions collect during production and after that you can adjust filter.
3. How you made tap for input/output port. In many cases if it is mechanical construction it is source of problems with bad VSWR with temperature change.
..
Please specify you filter

XTASA 8)

 

[qiang]

my copper line is directly tapped
and after the high and low temperture ,when back to the room temperture ,the VSWR is good

so it is not the machanical tenstions, i think
and my work bandwidth is smaller to my bandwidth of the filter, so it is not the question of bandwidth
and ,the most suprising me is when in high temperture the VSWR is good,and in the low temperature it is bad

 

[flatulent]

Try enclosing the whole thing in a plastic bag to keep the condensation of water vapor down. Even if this fixes the problem, you will still have to encase it for field use.

Another thing to try is measure the line with an ohm meter at the low temperature. A contact you think is firmly soldered might open at that temperature.

I once had a switching power supply that worked on the bench but not when plugged into the receiver it powered. It turned out that the C cores of the transformer were pulled apart by gravity when oriented on edge in the receiver and they came back together when the card was laid flat on the test bench.

 

[qiang]

let me try again

now i am disapointed about the filter because of the high/low temperature

 

[caocao]

HI
If you are design coaxial filter and using the dielectric material which DK (relativity dieclcetric constant )will change with the temperature changing .Please ask the DK cure with the temperature and design again consider it .
Caocao

 

[Debeli]

Which material you use for the filter housing and which for the screws? Generally, I prefer brass better than aluminium for wider temperature range. Do you use silver plating for the housing resonators and screws. Different materials can give totaly different results. Try to find article "Temperature Characteristics of Combline Resonators and Filters" from Atia and Yao published in 2001. I don't have an electronic version of it.
Have you checked connector specs and mounting?
Maybe you can try to measure the response with smaller temeperature steps in longer intervals. If the changes are slow than you have problem with material strain. If it is abrupt you must check your mechanical construction.

Hope I helped a litlle

 

[g579]

I think your problem is originating from simple dimensional changes of your cavity or cavities.
In sensitive cases consider using invar material instead of copper.

 

[VSWR]

qiang, in order to help you:

Can you give us some filter data such as center frequency, 3 dB bandwidth and no of poles? And measured results of VSWR in room temp and cold temp? Are you talking about a bandpass or lowpass filter?

 

[lguancho]

Hi. At low temp, water condensation will occur, and this will impose attenuation to RF / microwave signal or result in poor VSWR. Ever drop water onto printed microstrip bandpass filter and look at the poor passband response?

Usual, commercial filters (military grade) are manufactured with hermetic packaging.

 

[qiang]

the center frequency is 830MH,and the 2dB bandwidth is 10MH,and in the 2.5MH offside hte loss is larger than 23dB??
and in room temperature the return loss is larger than -20