6
$\begingroup$

I have a Daiwa CS-201A two position switch. The single side is connected to by quarter wave vertical (40m). The A and B sides are connected to an IC-7300 and TS-830S.

The switch is claimed to have 60dB of isolation. It also shorts the unused port to ground.

If I transmit 100W (50dBm) from the TS-830S to the antenna, am I correct in that about -10dBm will leak to the other side into the IC-7300?

I tried it and saw about +40 over S9 on the 7300. The service manual for the 7300 says never give it a signal over 20dBm (0.1W).

Am I safe doing this? How can I tell if it’s too much?

UPDATE: I did some testing with an Elecraft XG3 signal source. It produces signals at -107dBm (1uV), -73dBm (S9), -33dBm (S9+40), and 0dBm.

Following the instructions using the 1uV signal and a DMM on the speaker output, the MDS is around -135dBm and -142dBm with the preamp off and set to 1, respectively.

Also, the -73dBm signal is showing just shy of S9 on the S meter. That worried me a little, but I saw in the August 2016 QST magazine article that the ARRL test showed an S9 with a 70uV signal. So there's an error of about 3dB (which is also the S unit increment on this rig's meter).

So I think my receiver is still okay.

I also tested the isolation of the switch by hooking the 7300 to port A, the signal generator to port B, and a dummy load to the common port. With switch in the B position (allowing a direct path from the signal generator to the dummy load), I set the signal output to 0dBm and noted the S meter deflection on the 7300. It was around S5.

I don't know exactly what signal level that is, but it's less that what it read when I fed -73dBm directly into it. So even if we overestimate the received signal at -73dBm, that's still at least 73dB of isolation- if I'm thinking this through correctly. This was done on at a frequency of 7.020MHz. I think the switch claims an isolation of 60dB at 600MHz. It appears the isolation improves as you go down in frequency.

Added a bounty for feedback on the tests and the data. Is it really possible for the switch to have that good of isolation? Is there anything else I should look into? Do the tests seem valid? Am I in the clear running two rigs on this switch?

Improve this question
$\endgroup$
6
  • $\begingroup$ Manual says no. What makes you think it's wrong? $\endgroup$
    – Zeiss Ikon
    Mar 26, 2019 at 11:49
  • 1
    $\begingroup$ Respectfully, manual doesn't say that. $\endgroup$
    – Aleksander Alekseev - R2AUK
    Mar 26, 2019 at 12:33
  • $\begingroup$ @AleksanderAlekseev-R2AUK "The service manual for the 7300 says never give it a signal over 20dBm (0.1W)." Seems to me +40 over S9 exceeds the manual limitation as quoted by the questioner. $\endgroup$
    – Zeiss Ikon
    Mar 26, 2019 at 13:20
  • 6
    $\begingroup$ S9 is -73dBm. +40 over S9 is therefore -73 + 40 = -33dBm. The manual says nothing over 20dBm. That's like +93 over S9. Can you imagine if a receiver couldn't handle anything over S9+20? All it would take is someone in the next zip code over to turn on their amplifier. They could remotely let the smoke out. $\endgroup$
    – user3191449
    Mar 26, 2019 at 14:09
  • 1
    $\begingroup$ For the bounty, could you perhaps say a little more about what you're looking for, especially as there is currently an accepted answer? $\endgroup$
    – Kevin Reid AG6YO ♦
    Apr 16, 2019 at 20:04

1 Answer 1

Reset to default
4
$\begingroup$

S9 corresponds to 50 microvolts or -73 dBm. Thus S9+40 is -33 dBm. In theory, since it's much less then 20dBm your 7300 should be fine. HOWEVER I believe that +40 is just a maximum level your transceiver can show. Given that the attenuation of your switch is 60 dB and you are using 100W (50 dBm), -10 dBm will leak. It's much more than -33 dBm but you are still in the safe range.

From personal experience I used a DIY switch with measured 58.7 dB attenuation in the same manner with a 100W transceiver and an RTL-SDR, which maximum input is 10 dBm. RTL-SDR showed a large signal level as well, but worked just fine.

This being said I wouldn't recommend to trust the manufacturers claim regarding the attenuation level of the switch. It's always a good idea to measure it using a low-power signal generator (if you don't have one you can solder a simple Clapp oscillator) and an oscilloscope (or your transceiver).

Improve this answer
$\endgroup$
3
  • $\begingroup$ Yeah -33dBm suggests even better isolation. I doubt that. Maybe the S meter isn't all that accurate (are they ever?). Also the max scale on this radio's S meter is +60. But maybe it's software capped to 40? Is there any way to tell if there's damage to the front end? $\endgroup$
    – user3191449
    Mar 26, 2019 at 13:04
  • 1
    $\begingroup$ It is not unusual to have a poorly calibrated / designed S meter. For instance, it may show an accurate S9 value but everything above and below is not very accurate. You can easily check it by generating X dBm value (generated / measured using some device you trust) and comparing it with S meter reading. If you radio receives signals and sounds OK I wouldn't worry that it actually suffered the damage. Just in case you can examine the look of the components to make sure nothing is damaged. $\endgroup$
    – Aleksander Alekseev - R2AUK
    Mar 27, 2019 at 7:03
  • $\begingroup$ I updated my question with some details on tests I performed with the rig. $\endgroup$
    – user3191449
    Apr 15, 2019 at 15:40

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .