EMRFD Message Archive 14951

Message Date From Subject
14951 2018-07-27 20:08:43 rcbuckiii 2 Meter High Pass Filter
At a recent club meeting one of the guys mentioned a problem he had on Field Day. He was trying to use a 2 meter repeater to send some traffic. But the HF transmitters of the other members were overloading his 2 meter receiver. His radio is an Icom IC-2100H. If he moved 100 feet away, the overload disappeared. I told him I would build a high pass filter which would take care of the problem.

I have built two filters, a 7th order and a 9th order filter. I am using a DSA-815-TG to test the filters. With the 7th order filter, I can get the losses down to -.12 to -.13 dB from 144-148 MHz. For the 9th order filter, I can only get the losses down to -.16- to.18 dB. The loss between 3 MHz and 50 MHz is -55 dB for the 7th order filter and -60 dB for the 9th order filter.

I realize .18 dB loss is only about 2 watts and doesn't really matter. Also, I'm assuming (probably incorrectly) that the DSA-815-TG is able to resolve to that accuracy. I did normalize the TG before attaching the cables to the filter.

If I scan the filter from 50 MHz to 500 MHz, I see about 1 dB of ripple at various points between 150 MHz and 500 MHz. I also see a sharp peak at about 80 MHz where the loss is only about 1 dB. The loss then drops to -5.7 dB and then slowly rises until it is -.20 dB at 140 MHz. I'm not concerned about anything above 148 MHz so the ripple above that point doesn't matter.

Is the very small loss between 144 and 148 MHz the best I can expect for a home brew filter? I realize all the  filter calculators assume ideal components are used for construction. The filter is built on a small 1.5" x .9" PCB. It only has copper on the top side as I didn't want to have problems with the capacitors changing value due to the dielectric heating. The ground for the inductors is a .3 inch wide copper trace at the bottom edge of the board. There are 3 holes in the trace which grounds it to the 3" x 2" aluminum minibox that the filter is mounted in.The two PL-259 female jacks have about 1 inch of RG-174 between them and the circuit board at each end of the filter. The braid is grounded at the connector and at the circuit board ground trace. There are small pieces of double sided PCB material between the four inductors. Each side of the separators is soldered to the PCB ground trace.

The inductors were calculated to be 79 nH for the two outside ones and 44 nH for the center ones. I measured them with my AADE LC meter but it probably is not very accurate at such small values. The inductors were wound on a 5/32 inch drill bit and stretch adjusted to get the values correct before soldering to the circuit board. The 3 internal capacitors are 5-40 pF variables which were adjusted while watching the SA. This has been a learning experience and any comments would be appreciated. The filter was designed for a 100 MHz, -3 dB point.

Ray,
AB7HE
14952 2018-07-27 20:21:12 Ashhar Farhan Re: 2 Meter High Pass Filter
Ray,
I have been mucking around vhf and uhf filters. The best i goy was using coax as resonators. But your figures are really good. What kind of caps and inductor wire did you use? For me, AVX c0g caps and silver wire works the best. Shielding ia mandatory.

- f

14953 2018-07-28 09:09:07 alienrelics Re: 2 Meter High Pass Filter
It enhances readability if you include a leading zero when the number is less than 1.

0.18 rather than .18. Looking at it
14954 2018-07-28 09:34:00 Mike Markowski Re: Leading zero [was: Re: 2 Meter High Pass Filter]
Steve & all,

I run into the leading zero issue at work as well. Our reports go
through a technical publication review group, who point out there is an
ISO standard (sorry I've forgotten which) saying decimal values less
than 1 are always preceded by 0. Also, apparently each group of three
digits before after decimal points are separated by a half space. Since
some countries use comma and some use period, it's otherwise confusing
to one readership or the other.

Yours, with today's trivia...
Mike ab3ap

14955 2018-07-28 10:31:08 rcbuckiii Re: 2 Meter High Pass Filter
Steve and Mike,

Thanks for the suggestion. That would certainly make sure you don't miss the decimal point.

Farhan,

I am using AVX SR series (ex: SR152A820JAA) for the end caps. They are 5% C0G caps. The data sheet doesn't list Q value. The inner variable caps are 5-40 pF similar to Sprague Goodman GYC40000. The inductors are single layer air core wound with #20 gauge wire.

The top end of the inductors connect to a trace between two variable caps that is about 10 mm in length. I connect to the trace at the half way point between the caps. The traces add to the inductance so I manually adjusted the inductors while watching the SA screen.

Ray,
AB7HE

14956 2018-07-28 13:52:10 Alberto I2PHD Re: Leading zero [was: Re: 2 Meter High Pass Filter]
On 2018-07-28 18:20, Mike Markowski mike.ab3ap@gmail.com [emrfd] wrote:

I run into the leading zero issue at work as well.  Our reports go  through a technical publication review group, who point out there is an  ISO standard (sorry I've forgotten which) saying decimal values less  than 1 are always preceded by 0.

Yes. This seems to be customary in the US to omit the leading zero, starting with the
decimal point... really a BAD habit...


--
73 Alberto I2PHD
Credo Ut Intelligam


14957 2018-07-28 13:57:29 Bill Carver Re: Leading zero [was: Re: 2 Meter High Pass Filter]
Omitting a leading zero, for numbers less than one, is not "customary" any place I have ever worked. Maybe some non-technical person would not know, but every engineer/technician I knew understands the need to put a leading "0.: in front of a small number.
W7AAZ


14958 2018-07-28 15:42:51 Alberto I2PHD Re: Leading zero [was: Re: 2 Meter High Pass Filter]
On 2018-07-28 22:57, Bill Carver bcarver@safelink.net [emrfd] wrote:


Omitting a leading zero, for numbers less than one, is not "customary" any place I have ever worked. Maybe some non-technical person would not know, but every engineer/technician I knew understands the need to put a leading "0.: in front of a small number.

I was not talking about professional environments. My comment was based on forum messages where many, many times I screamed against those too lazy to write that leading zero...

--
73 Alberto I2PHD
Credo Ut Intelligam


14959 2018-07-28 18:45:46 Bill Carver Re: Leading zero [was: Re: 2 Meter High Pass Filter]
Ah yes, Alberto, outside of professional environment anything can be done!
Ciao - W7AAZ


14961 2018-07-29 14:40:57 kerrypwr Re: 2 Meter High Pass Filter
Sadly, some great companies that I otherwise admire committed this sin of omission.

From the Tek 465 manual;

[url=https://postimages.org/][img]https://s15.postimg.cc/i2jf5drvf/465.gif[/img][/url]

and from the HP3444 manual;

[url=https://postimages.org/][img]https://s15.postimg.cc/cr4ikpviz/3444a.gif[/img][/url]

Tek repeated the sin in the 465 parts list; HP redeemed themselves, at least in part, by including the leading zero in the 3444 parts list.

Omission of the leading zero sometimes makes life difficult when dealing with poor reproductions of old schematics.or parts lists.

Kerry VK2TIL.

14962 2018-07-29 14:59:50 kerrypwr Re: 2 Meter High Pass Filter
14966 2018-07-30 10:47:09 rcbuckiii Re: 2 Meter High Pass Filter
Now that the leading zero issue has been resolved, I have a question about the filter. I used 100 volt capacitors for the end capacitors. The three inside variable capacitors are also rated for 100 volts.

Are these voltage ratings high enough for a high pass filter to handle 50 watts? 50 watts into 50 ohms should be 50 volts RMS. However, the P-P voltage is about 141 volts. That is above the 100 volt rating of the end caps which are grounded through the first and last inductors. Are the peak voltages also dependent on SWR?

Ray,
AB7HE
14967 2018-07-30 11:58:50 Thomas S. Knutsen Re: 2 Meter High Pass Filter
Yes, voltage is dependent on SWR. 
100V capacitors is cutting it a bit close in my opinion, I would have used 500V capacitors, there are C0G capacitors around that are cheap and have low lead inductance that would be great at VHF. The grounding via the inductors only matter at low frequencies, when analyzing any circuit, I like to think about what will happen at DC. at extremly high frequency and then I find resonance frequencies. That way its fairly easy to have some knowledge of the circuit without investing more than 30 sec or a minute. 

73 de Thomas LA3PNA

Den man. 30. jul. 2018 kl. 20:11 skrev rcbuckiii@yahoo.com [emrfd] <emrfd@yahoogroups.com>:
 

Now that the leading zero issue has been resolved, I have a question about the filter. I used 100 volt capacitors for the end capacitors. The three inside variable capacitors are also rated for 100 volts.

Are these voltage ratings high enough for a high pass filter to handle 50 watts? 50 watts into 50 ohms should be 50 volts RMS. However, the P-P voltage is about 141 volts. That is above the 100 volt rating of the end caps which are grounded through the first and last inductors. Are the peak voltages also dependent on SWR?

Ray,
AB7HE
14968 2018-07-30 16:22:29 rcbuckiii Re: 2 Meter High Pass Filter
Thanks Thomas. I have ordered 450 volt C0G caps and will replace the end caps. I also ordered 500 volt SMD variable caps for the three inside variable caps.

Ray,
AB7HE
14969 2018-07-31 17:25:17 arfghans Re: 2 Meter High Pass Filter
Simulate your filter in Spice (ac analysis) with various mismatched loads and power levels while observing capacitor voltages and inductor currents. Depending on the topology, V and I on particular components can rise to surprising levels. It’s a real eye-opener at QRO power.

Gary NA6O
14970 2018-08-01 10:02:06 rcbuckiii Re: 2 Meter High Pass Filter
Gary,

I keep saying that one of these days I will download and learn how to use LTspice.

Ray,
AB7HE