Thanks. Might end up more useful than the Pi-network approach I've used a few 
times before. I appreciate knowing of more tools that can be called upon to 
help with a design. I just wish the calculators had some way to deal with 
standard values (like TI's FilterPro). Its frustrating getting a 79pF result 
and wondering how an 82pF part works. Well, I guess that's what Spice is for...

Bob LaJeunesse



>________________________________
> From: Bob Camp <[email protected]>
>To: Discussion of precise time and frequency measurement <[email protected]> 
>Sent: Friday, August 9, 2013 6:41 PM
>Subject: Re: [time-nuts] Modified Extron DA [WAS: Rb video]
> 
>
>Hi
>
>The simplest way to design it is to do a T matching network. Two inductors in 
>the top of the T and one cap to ground. Weather it's a filter or a match, it's 
>the standard three element T lowpass. 
>
>The logic gate wants to "see" an inductor at high frequency. The T has an 
>input inductor and that keeps it happy (so would a step up L). Since it's a 
>three element match, you get to pick Z in, Z out, and Q. (with an L network 
>you just would get Z in and Z out). Simply design it for a low Q.  Q of three 
>isn't a bad number. Anything up to 5 is practical with rational parts (no 
>tuning). The narrower bandwidth of the higher Q design will increase it's 
>sensitivity to temperature. The lower Q will have a smaller coil / lower 
>impedance above cutoff. If you have 18 to 20 dbm out, you can put a 6 to 8 db 
>pad on it. That will improve the broadband match into the cable. 
>
>If you want to design it as a filter, everything still works pretty much the 
>same. It's still Zin / Zout and one other number with a three element network. 
>If you want to go to more elements, you can indeed get better filtering at the 
>cost of higher temperature sensitivity. With three elements the harmonics are 
>down > 60 db. That's plenty good enough….
>
>LC match calculators (there are many others):
>
>http://www.changpuak.ch/electronics/calc_18.php
>http://home.sandiego.edu/~ekim/e194rfs01/jwmatcher/matcher2.html
>
>Filter calculator:
>
>http://www.calculatoredge.com/electronics/bw%20tee%20low%20pass.htm
>
>If you plug the numbers into the calculators you can see what the match does 
>for you in terms of the inductor value. 
>
>Why not design a flat passband filter? You are only interested in passing 10 
>MHz. Attenuating other frequencies is not a problem and may be beneficial. The 
>bandwidth is not going to be small enough (with a low Q) to give you trouble. 
>The peaking of the filter gives you a steeper cutoff at harmonic frequencies. 
>It rolls off just like any filter, but it starts from a higher peak. 
>
>With the T you can do any Zin / Zout ratio provided the Q is high enough. If 
>you want to do low power, set it up as a 100 ohm to 50 ohm or 200 ohm to 50 
>ohm match. It's a pretty simple solution to the problem that is flexible 
>enough to get the job done. 
>
>Bob
>
>
>On Aug 9, 2013, at 5:24 PM, Robert LaJeunesse <[email protected]> 
>wrote:
>
>> Bob, I need some education. For a low-pass filter I think series L and shunt 
>> C. For two inductors that normally means 2-3 capacitors.  If you use only 
>> one shunt capacitor is the second L in series with it (as a harmonic trap)? 
>> Can you point me to a design tool (or equation set) somewhere that shows how 
>> to choose values best to match the impedances? 
>> 
>> thanks,
>> 
>> Bob LaJeunesse
>> 
>> 
>> 
>>> ________________________________
>>> From: Bob Camp <[email protected]>
>>> To: Discussion of precise time and frequency measurement 
>>> <[email protected]> 
>>> Sent: Friday, August 9, 2013 4:57 PM
>>> Subject: Re: [time-nuts] Modified Extron DA [WAS: Rb video]
>>> 
>>> 
>>> I still think that a distribution amp based on logic ic's is cheaper / 
>>> simpler / lower power / higher performance. A pair of NC7SZ125's will dump 
>>> 20 dbm into 50 ohms all day long running at 5.5 volts. Good isolation as 
>>> well. Do the lowpass filter right and the harmonics are not an issue. Two 
>>> coils / one cap plus dc blocking does it quite nicely. 
>>> 
>> _______________________________________________
>> time-nuts mailing list -- [email protected]
>> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>> and follow the instructions there.
>
>_______________________________________________
>time-nuts mailing list -- [email protected]
>To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
>and follow the instructions there.
>
>
>
_______________________________________________
time-nuts mailing list -- [email protected]
To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
and follow the instructions there.

Reply via email to