View Full Version : Yaesu SSB Filter for 817/857/897

03-07-2014, 11:15 AM
I've had this around for a while. I had two optional filters in my 857, a 2KHz INRAD being the other, and honestly I can't tell enough difference between them to keep both. The stock ceramic Murata is slightly more full sounding and less distinct than either to my ears and both seem to do the job of sharpening your transmitted signal about equally. Ultimately I'd like to fill the second slot with a narrow filter (300 or 500Hz) but this has got to go first.

Anyone interested? I was going to put it on QRZ with an asking of $100 and see where it goes from there.

Seldom Seen
03-07-2014, 03:53 PM
2.3 kHz, right? Yaesu or W4RT? Wanna make a trade?

03-07-2014, 04:04 PM
Yes, 2.3KHz, 10-pole. It's the Yaesu XF-119SN/YF-122SN.

The response (from the INRAD #717, which along with the W4RT are all the same guts AFAIK).


Seldom Seen
03-07-2014, 06:36 PM
From my research the Yaesu and W4RT are identical right down to the label and similar part numbers, the Inrad OTOH has sharper skirts and is the slightly more desirable filter, either one will get the job done though.

If you came up with an alternative to the push up pole I'll it that back. I've been thinking of buying another one but this time I was gonna order the sections al-a-cart to make a 38' more rigid mast. I could use the original mast and just order the parts to increase the rigidity. With the $ saved I'll get the inrad filter.

03-07-2014, 07:27 PM
I understood the three filters are identical inside, a Rockwell-Collins 526-8739-010/-020. That is probably what is meant by them all sharing the same markings, it's the Collins part number. The difference I believe is that INRAD makes their own, thus a different shield and PCB, while the Yaesu and W4RT are made by Collins directly.

Collins Torsional Mechanical Filters (PDF) (https://www.rockwellcollins.com/Capabilities_and_Markets/More/Precision_Manufacturing/~/media/BAFB93A7B1BE40EA8FBF15B4F5218268.ashx)

Center: 455.00KHz
Nom BW 2.26KHz@-6dB
Max BW: 4.60KHz@-66dB
Max Loss: -5dB
Number of Resonators: 10
Case Style: PL

Rockwell Collins Filter Products (https://www.rockwellcollins.com/Capabilities_and_Markets/More/Precision_Manufacturing/Rockwell_Collins_Filters.aspx)

This is what they look like if you get them directly from Rockwell-Collins.


While INRAD's looks like this.


Seldom Seen
03-07-2014, 09:41 PM
I don't think I'd buy from W4RT directly (2nd hand maybe) based on their poor customer service reviews. That leaves Yaesu and Inrad. Reviews on the 2.3 kHz are mixed, for both, ranging from no difference from the stock 2.4 ceramic filter to overwhelming positive reviews. Interestingly there are very few negative reviews for either.

For $100 I'm still considering it, Thats a $60-$80 savings (60 for the Inrad - 80 for the Yeasu).

How does it compare to the stock filter for adjacent QRM?

03-08-2014, 09:39 AM
Here's a comparison between the stock 2.4KHz Murata and the INRAD #720 still installed in my FT-857. This is a 2.0KHz Collins 8 pole, which has slightly narrower band pass with slightly less steep skirts. The difference in response is subtle between the two but hopefully gives you some feel how a ceramic filter sounds compared to a mechanical Collins. The 2.3KHz Collins generally performs slightly better with close-in QRM due to the increased skirt slope but has a wider bandpass. There are reasons for each, mainly which side of RX or TX the performance is preferred. Narrowing your bandpass on TX helps range and this filter usually goes in the FT-817 to eek out every bit of QRP phone performance (which is marginal to be sure).

Antenna is a 44' doublet oriented east-west (broadside is very close to directly N-S) at about 25' AGL with an Icom AH-4 tuner at the feed point.

Test 1

The first 10 seconds are the stock CFIL, the next 14 seconds are the narrower mechanical and the final 10 are the stock again. This is with two S9+ signals on 20m with DNR DSP turned off.

Ceramic vs Mechanical IF filter (MP3) (http://www.armbrusterweb.com/public/music/Murata-vs-INRAD.mp3)

Test 2

Next is a sweep across two strong local 40m stations, one S8 and the other S9+. Each sample is 35 seconds long, starting at -3KHz from their center and stepping 1KHz each 5 seconds. So they were rag chewing at 7.166.00MHz and I tuned from below, past and stopped above them. The listener will have to imagine the interference if you were trying to get a station within the 3KHz band. This is with DNR DSP turned on. You can hear that fixed IF filters are not a substitute for good IF DSP, but within +/-2KHz a strong station does change what is demodulated depended on the filter selected. I apologize that the bands were very quiet this morning so I don't have a good example.

T+0s @ 7.163 (-3KHz)
T+5s @ 7.164 (-2KHz)
T+10s @ 7.165 (-1KHz)
T+15s @ 7.166
T+20s @ 7.167 (+1KHz)
T+25s @ 7.168 (+2 KHz)
T+30s @ 7.169 (+3KHz)
T+35s audio ends

Stock Yaesu Murata Sweep (MP3) (http://www.armbrusterweb.com/public/music/cfil_sweep.mp3)

Optional Collins INRAD Sweep (MP3) (http://www.armbrusterweb.com/public/music/inrad_sweep.mp3)

03-08-2014, 09:58 AM
A basic history of mechanical filters. It seems odd, but when we say mechanical we really mean mechanical in that there is a transducer that vibrates, a little like a microphone or speaker but at higher frequencies than audio. The same equations drive the characteristics as with purely electrical filters, physics is cool that way. Just like electric voltage and current are analogous to water pressure and flow, so are nickel and iron similarly bound by RF phenomenon as ceramic and copper.



03-08-2014, 10:13 AM
To the question of whether it's worth it or not. I prefer the response curve of mechanical filters a little more when communication is the point. They sound a bit more harsh to my ear but the audio in the bandpass is clearer to me. Ceramic filters are smoother and tend to upset worse. I compare a ceramic filter to a vinyl turntable and mechanical filters to CDs. It's odd because I MUCH prefer to listen to vinyl... :-) But HF to me is about contacts, so rag chews are best done via FM.

Seldom Seen
04-11-2014, 11:26 AM
Still got it? I'll Be up in that neck of the woods on Sat. I could pick up the filter and fiberglass pole.

04-11-2014, 11:44 AM
Totally neglected to remove the post, the Yaesu filter sold on QRZ. :-/