Abrupt reduction of noise floor on receive

[JJ4SDR]

ONE more UPDATE

I am offering an up-front apology to the Moderator as I have been dubbed the "King" of being able to cram totally unrelated things into one subject so here I go again and I may have fallen prey to the same mistake . I am probably clasping at straws here and this subject doesn't belong here, but I am at my wits' end

I have been pulling my hair out over the sudden decrease in noise-floor as everyone can attest to, to the point of suspecting that I had the wrong antenna connected to the radio. Regardless of antenna and band with 4 different antennas and independent coaxial runs from the shack to the feed-point of those antennas, I have observed this issue during several days. I won't buy that the coax connectors and coaxial runs on all 4 antennas are sketchy and the culprit. Perhaps, if what I had been observing happened on only 1 or 2 bands, perhaps, but not all 4 bands. Also, I took the amp out of the circuit, connecting the radio directly to antennas, using ANT 2 port, and RX 2, swapped out the coaxial jumper that I use to connect the radio to antennas (or amp). It seemed like the receiver noise-floor was fairly steady when antenna was on ANT 2 port (2 hours of monitoring), but I am not sure how "conclusive" that is.

Lately, I am seeing problems during transmit as well, not merely receive. I was on 40 last night, rag-chewing for about 40 minutes, talking to a local fellow here in Texas. While listening to him, I noticed that the noise-floor went really low (on 40, the noise-floor is really high, evenings) and he was going to hand it over to me. From the first syllables of my talk, Pure Signal appeared to go completely out-of-whack and it wasn't correcting at all (RED flashing, and BLUE etc., finally GREEN) for around 10 s, then I saw correcting for a little while and, then the sequence repeated again. First thing that came to my mind was a defective T/R relay in the radio and this is something that others have suggested. As I had posted previously, W5WC wasn't able to reproduce what I had been observing. He did mention that he had not conducted longer term testing while transmitting. Logical as I had not observed any issues on the transmit side of things when I shipped the radio.

Now, I started looking at the Ethernet CAT8 cables plugged into the "dumb" T switch and noticed that the cable from the switch to the PC looked a little suspect. It seemed like it was somehow "askew" and when I unplugged it (yes, I didn't forget that there is a "tab" on the end that you are supposed to use to help unplug the thing) it was much more difficult than with the 3 other connectors.

Could the above have played into the intermittent issues I have been observing? I will continue running tests, to see if did play into this .

Juha

[w-u-2-o]

In this case you stayed on topic!

A loose Ethernet connector won't cause this. You'll get a lot of SEQ errors and even disconnects. BTW, CAT8 cable is a huge waste for anyone not making 10G connections.

Starting over from scratch with suggestions:

- First and foremost: disable AGC automatic noise floor compensation and disable RX1 "adjust min to noise floor" so that there can be NO confusion about where the noise floor is going and why

- Have you tried Protocol 1 at your QTH? If not, then try that. If the problem goes away you can blame the P2 firmware not working well on your 7000.

- Revisit RX2 (assigned to the second ADC). If that looks good then regardless of what Doug did or didn't find you have to suspect a bad relay on the RX1 path.

- Have you tried a completely different PC? I don't think this will fix anything, but given your inability to localize the problem it might be worth a shot.

- Is there any way you can swap/loan your 7000 to/with another 7000 user?

[Joe-W4WT]

If I had this problem and wanted to eliminate the radio, I would run a signal generator into a splitter and feed one output of the splitter to ANT 1, ANT 2, or ANT 3 (they all go to the same point in the radio) and I would feed the second output of the splitter to RX2 (NOT ANT 2). I would make sure the ADC assignments were in their default state. Then run both RX1 and RX2 simultneously on the screen and you can easily watch and see if one or both show a drop in signal. If you don't have a signal generator, put the antenna into the splitter input. A signal generator makes it easier because the output is constant unlike RX on a band plus you are completely eliminating the feedline, connectors, coax, etc.

Since you indicate that the problem frequently happens within minutes, this should be a fairly quick, easy test.

Joe, W4WT

[JJ4SDR]

Scott,

CAT8: I realize that, but I got those in a good deal so I really didn't care if they were "wasted."

On which form does the "adjust min to noise-floor" figure?

Used P1 for about a year and P2 almost a year so I doubt it the Protocol figures into this.

OK on RX2. I am not observing the drop-off/out for a couple of days now. Just hate it when the gremlins decide to throw a monkey-wrench in the works.

Other PC: I have a dual-core i5 laptop (6GB or 8GB of RAM only though) that I might try.

Loaning/swapping isn't feasible right now.

Juha

[JJ4SDR]

Joe, thanks!
Don't have a signal generator or a splitter, but what you suggest would help take a lot of guess-work out of the equation.
Juha

[w-u-2-o]

On which form does the "adjust min to noise-floor" figure?

Setup > Display > RX1, in the bottom right corner.

Used P1 for about a year and P2 almost a year so I doubt it the Protocol figures into this.

Wandering noise floors are possible when the firmware isn't making timing 100%. P1 has never, ever demonstrated this problem so it's a way to rule out firmware. Try it or not.

OK on RX2. I am not observing the drop-off/out for a couple of days now. Just hate it when the gremlins decide to throw a monkey-wrench in the works.

Just to clarify: you are saying the problem has not been observed on either RX1 or RX2 for a few days, correct? If so then there's no point in doing any further testing until the problem reappears.

[JJ4SDR]

Scott,

The "adjust min to noise-floor" box was unchecked on that form (I don't know if it is this way by default).

It is good to know what you are writing about P1.

I haven't tested RX2 as, I thought I would do that if RX1 starts exhibiting the anomaly again which it hasn't

Not that I know this to be true, but I will see if going into TX mode would somehow "trigger" the anomaly on RX mode.

Juha

[Joe-W4WT]

Juha,

If you have a bnc T-connector you can make a splitter that will work well enough for this test. If not, just parallel three pieces of small coax and put connectors on all three ends then connect an end to Ant 1, another to RX2, and the third to the antenna. Should work just fine to see if one path through the radio is failing or not..

Joe W4WT

[JJ4SDR]

Copy that Joe.

Been on RX1 for about 8 hours straight, not sitting at the radio all that time though, but during the times that I do get into the shack, the noise-floor is steady as a rock., no changes up or down.

As soon as I want to see the anomalies they deign not to show up .

Juha

[n2gq]

I just installed the latest 2.9.08 meter thetis yesterday and while listening my receive dropped as well a quick keyup brought it back up. I did hear the relay switch when it happened. This is the first time this has happened. I have protcol 2 and a 7000del mkIII. I duplicated this by actually switching the ant selection and had the same drop in noise.

Jay n2gq

I was able to record the radio in that state, but wasn't able to capture the moment it happened.
Here is a video of I was able to capture. Thetis version is 2.9.0.8 x84 running protocol 2 MW)LGE pre-release.

[ea3aqr]

You posted a private video...

[n2gq]

Will fix, now public.

[w-u-2-o]

What precipitates the drop in the noise floor? A transmit-to-receive transition? Or does it just spontaneously drop in the middle of receiving? The answer to this question will be an important clue.

Also, revisiting my previous suggestions from above:

- First and foremost: disable AGC automatic noise floor compensation and disable RX1 "adjust min to noise floor" so that there can be NO confusion about where the noise floor is going and why

- Have you tried Protocol 1 at your QTH? If not, then try that. If the problem goes away you can blame the P2 firmware not working well on your 7000.

[n2gq]

What precipitates the drop in the noise floor? A transmit-to-receive transition? Or does it just spontaneously drop in the middle of receiving?

Just during receive so far, it happened twice yesterday and nothing today. This is the third time it happened.

[w-u-2-o]

Are you going to try my suggestions? It does suck going back to P1, but it's probably a worthy experiment to see if is somehow tied to firmware.

This might be related to the other reported problem with the Cyclone V P2 firmware associated with jumps in freq. calibration.

[n2gq]

Are you going to try my suggestions? It does suck going back to P1, but it's probably a worthy experiment to see if is somehow tied to firmware.

This might be related to the other reported problem with the Cyclone V P2 firmware associated with jumps in freq. calibration.

Well, I might if it happens again it has been a while since it occurred as you can see from the above post. It happened before the smeter version. I hate to go back to P1. I do occasionally get a buzzing noise almost like a sweep.

[n2gq]

Well, it happened again this morning and I heard the relay click two times. This time I went to setup and antennas the 20mtrs selected ant 2 then ant 1 and it became normal again. Maybe has something to do with antenna selection? This happened while receiving.

[w-u-2-o]

Interesting. If you switch to an empty antenna port does the panadapter look like it does when the problem is happening?

[n2gq]

Interesting. If you switch to an empty antenna port does the panadapter look like it does when the problem is happening?

I am going to check if it does. On Saturday it seemed like the radio really freaked out a bit I was not sure what was happening. So I had a windows update in cue and took care of that. Now I will check to see if the level is the same when check an empty ant port.

[n2gq]

Interesting. If you switch to an empty antenna port does the panadapter look like it does when the problem is happening?

Okay here is a video showing receive drop problem and then a selection to an empty antenna port.



It is funny it seems to happen when I get up off my chair sometimes. Not sure it's the problem but coincidental.

[w-u-2-o]

Similar, but not identical conditions. However it does suggest a relay is either sticking in the TX position, or is being erroneously commanded there by faulty firmware.

A sticky relay is much more likely since there aren't tens or hundreds of people complaining about this problem.

In the interests of being thorough, a temporary reversion to P1 firmware is the next obvious step. If it happens with P1 then it might be time to get the unit repaired. It would not be the first faulty relay in the history of ANAN hardware.

[n2gq]

I know it sounds weird but when I get up off my chair my receive drops. I am thinking static, although I was not touch anything.

[w-u-2-o]

I know it sounds weird but when I get up off my chair my receive drops. I am thinking static, although I was not touch anything.

If it consistently and reliably does that then it certainly sounds like a static discharge of some sort. How is your station grounding and bonding?

[n2gq]

I know it sounds weird but when I get up off my chair my receive drops. I am thinking static, although I was not touch anything.

If it consistently and reliably does that then it certainly sounds like a static discharge of some sort. How is your station grounding and bonding?

The station ground and bonding is good just finished that after having a bout with bad Heil mic wiring. I have disconnected a logitech video camera I use online that uses a motion type detector. The other device I use is a flex control knob and DDutil_v4. So bit by bit I will try to narrow it down.

[N7CXI]

I know it sounds weird but when I get up off my chair my receive drops. I am thinking static, although I was not touch anything.

If it consistently and reliably does that then it certainly sounds like a static discharge of some sort. How is your station grounding and bonding?

The station ground and bonding is good just finished that after having a bout with bad Heil mic wiring. I have disconnected a logitech video camera I use online that uses a motion type detector. The other device I use is a flex control knob and DDutil_v4. So bit by bit I will try to narrow it down.

Tagging onto an old posting...
I just completed running a mic/ptt cable out from a Orion MK-III out to a new 1/4" TRS MIC/PTT jack on the front panel, and have been seeing exactly the same thing. Get up from the chair, touch the mic boom in the wrong place at the wrong time, etc.

In my case, I don't as yet have the rig ground bonded. I'll try that and see if that affects it. Thanks for bringing it to mind.

73,
Jim N7CXI

[n2gq]

If it consistently and reliably does that then it certainly sounds like a static discharge of some sort. How is your station grounding and bonding?

The station ground and bonding is good just finished that after having a bout with bad Heil mic wiring. I have disconnected a logitech video camera I use online that uses a motion type detector. The other device I use is a flex control knob and DDutil_v4. So bit by bit I will try to narrow it down.

Tagging onto an old posting...
I just completed running a mic/ptt cable out from a Orion MK-III out to a new 1/4" TRS MIC/PTT jack on the front panel, and have been seeing exactly the same thing. Get up from the chair, touch the mic boom in the wrong place at the wrong time, etc.

In my case, I don't as yet have the rig ground bonded. I'll try that and see if that affects it. Thanks for bringing it to mind.

73,
Jim N7CXI

Funny you mention this. Because yesterday it happened to me. I got off my chair and I get this drop in receive. My computer case which is new is not connected to station ground I forgot about it.