Balanced Tx echo/reverb

[K6JR]

Since doing the software and firmware upgrades I have been using the TRS balanced input. The audio sounds great but I have been told that there is an echo/reverb effect. I noticed that when I switched to the TRS input the gain was very much higher than the regular mic inputs. I had to lower my mic gain way down. Is it possible that something is getting over driven in the balanced audio stage. I am using Thetis when this happens. When I switch back to the mic input the echo goes away. Any ideas?

[w-u-2-o]

99.999% of the time echo/reverb is an obvious sign of RFI getting onto the mic input. The reason that echo/reverb is an obvious symptom is because of the delay (latency) inherent from the mic input to the antenna output. That delay will be anywhere from around 30ms with all audio processing turned off (no leveler, EQ, etc.) to over 100ms with a bunch of audio processing stages activated. Thus the RF getting back onto the mic input sounds exactly like reverb or echo.

Try throwing a few ferrites on the mic cable at the input to the G2. Also review how you are (or aren't ) doing grounding and bonding of your station.

Note also that the TRS balanced mic input is fed into a legit, fully balanced op amp. That amp provides additional gain, and thus less mic gain is needed.

Just out of curiosity, what microphone are you using?

[AG5CK]

I use external audio gear and had this issue. Its always been RFI.

I've had good luck using AM and an oscilloscope to troubleshoot and verify it's been fixed. Monitor the scope while increasing the carrier. You'll see the waveform get fuzzy when the RF starts to creep in. Swap cables, add ferrite, rinse and repeat.

73

[K6JR]

Well thanks to both of you. I have never had RFI but i do have some clip on type 31s that I will give a try. The op amp is probably more sensitive to rf. My microphone is a Neumann TLM103MT with a 1 tube mic preamp/phantom voltage.

[w-u-2-o]

That's a hell of a microphone!

The reason you have to turn down the mic gain so much is because you are feeding the output of a mic preamp into another mic preamp.

You might consider just using a 48V phantom injector by itself. If you are hard over on using the tube preamp then consider using a direct box (transformer) to feed the line in input.

[w9ac]

...My microphone is a Neumann TLM103MT with a 1 tube mic preamp/phantom voltage.

In time, perhaps reconsider the real need/want of a tube preamp and replace it with an ASIO USB audio interface device. Some benefits:

1) Minimal noise and distortion through a well-designed mic preamp stage with optimized gain structure that's commensurate with the performance of your Neumann mic;
2) A truly balanced pro-grade 3-stage instrumentation mic op-amp to minimize common-mode noise and RFI at the mic input;
3) With VAC engaged, internal ANAN analog mic circuity is bypassed, simplifying the audio path with fewer points of RF ingress;
4) +48V phantom power supplied by the sound interface device;
5) Depending on the interface used, monitor latency can be minimized with a USB 3.0 connection and optimizing Thetis' DSP filter type. But, I suspect just going to ASIO drivers will be enough.

With a forum search, you'll find several recommendations for USB audio devices if you decide to forgo the tube preamp.

Paul, W9AC

[K6JR]

Thanks for the help. The problem turned out to be a lose ground at the back of the radio. It sounds great again.
73, Jim K6JR

[Trucker]

Jim,
Was the loose ground wire inside the radio or the microphone connector?
James
WD5GWY

[K6JR]

The lose wire was a braided ground wire outside of the radio going to the ground bolt on the back of the radio.

[TonyBlight]

371536121_1080001779652937_5319061247641788725_n.jpg (8.99 KiB) Viewed 2114 times

Reminder those that have the G2 with face you need to exit piHSDR before you can do the Audio test using the App.This rig has a very small card and computer and can get bogged down, if you load up too much it can freeze.My audio was perfect on the test but a small echo on transmitting , Ive tried new shielded Mike leads with chokes and checked many leads and unplugged to locate rf but no luck yet.Not using amp until sorted.The upgrade firmware and P2 App went well.Anyone out there have the G2 with face,I would like to hear from you as to your experience.I believe no one in Vk has one with face,the few with out face are using Thesis,I decided to explore the stand alone capabilities first,Its got great receive.

[laurencebarker]

You've established that good quality audio is getting into the codec inside the radio by using the audiotest app; that's a great start.

Another topic on the forum sounds like the same issue; it was resolved by grounding. https://community.apache-labs.com/viewtopic.php?f=48&t=4784

I don't yet know how the shield for the 6.3mm jack socket on the radio is grounded. It looks like it has a nylon body so it may not be grounded at the rear panel. I know that it is eventually grounded on a PCB inside the radio, but I don't know how far it travels. Ideally your XLR cable needs a solid connection to ground at the rear panel. Are you able to connect a ground strap from the radio's rear ground post to your 6.3mm jack plug ground and see if that makes a difference?

(Also I don't think this question is posted in the right place - it won't get noticed here. You are using piHPSDR so it could be posted in a folder for that but this should really be posted in the "G2 operating system and applications" folder to get most attention. )

[laurencebarker]

There is at least one other report of an echo when using an XLR input.

In case it helps others can you clarify exactly what the braided ground wire connects please? To what cable or device? I'd like if possible to get the mystery fully resolved and the solution documented.

[kc2rgw]

There is at least one other report of an echo when using an XLR input.

In case it helps others can you clarify exactly what the braided ground wire connects please? To what cable or device? I'd like if possible to get the mystery fully resolved and the solution documented.

He’s talking about his station ground.

[TonyBlight]

You've established that good quality audio is getting into the codec inside the radio by using the audiotest app; that's a great start.

Another topic on the forum sounds like the same issue; it was resolved by grounding. https://community.apache-labs.com/viewtopic.php?f=48&t=4784

I don't yet know how the shield for the 6.3mm jack socket on the radio is grounded. It looks like it has a nylon body so it may not be grounded at the rear panel. I know that it is eventually grounded on a PCB inside the radio, but I don't know how far it travels. Ideally your XLR cable needs a solid connection to ground at the rear panel. Are you able to connect a ground strap from the radio's rear ground post to your 6.3mm jack plug ground and see if that makes a difference?

(Also I don't think this question is posted in the right place - it won't get noticed here. You are using piHPSDR so it could be posted in a folder for that but this should really be posted in the "G2 operating system and applications" folder to get most attention. )

Ok Laurence,The 6.5mm.plug is metal but the total grounding is on pin 1 of the XLR plug.The manufacturer of the plug and lead say try to wrap a wire around the metal outer cover on the 6.5mm.plug and then to the earth of the radio screw, which is what you suggested. I will try and let you know if any change in the echo during TX.

[w-u-2-o]

I'd like if possible to get the mystery fully resolved and the solution documented.

There's no mystery here. This is a classic RFI best practices example where stray RF (conducted or radiated) was not being managed. Echo (or reverb) continues to be the number one indicator of RFI on the mic input for all SDRs that have latency measured in tens of milliseconds or more.

However, you do bring up a very interesting point which is the possibility that the balanced mic TRS connector on the rear of the G2 is not a chassis mount connector that makes metal to metal contact with the rear panel. If this is truly the case it's not too surprising. Apache has always preferred "poke through" connector mechanical design because it's less expensive to manufacture. Unfortunately it is not as good from a shielding perspective.

BREAK

Tony: a standard balanced mic cable TRS connector is wired tip mic +, ring mic -, shield ground. At the XLR end it's pin 2 mic +, pin 3 mic -, pin 1 ground.

Edited to clarify which mic TRS input is being discussed.

[BPerk]

I'm running 1300 watts on a G2-100 using v2.10.1 with all the recent updates installed. Using a condenser mic through a phantom 48 volt power supply, I've run on the air tests 80-10 meters, and am experiencing no echo or reverb on the transmit audio. The mic is run directly to the radio on the back trs connector. I have not tried the XLR port. I do use a handful of mix31 ferrite filters applied several places along the microphone cable to the radio.

73
Bob

[Dennis]

ANAN G2 no face, XLR TRS mic input using Balanced audio,
Original RESPONSE 1530 Pro cable
Balanced XLR into PYLE PHE-400 Dual Eliminator
XLR to TRS wired Balanced to G2
No issues, using either TRS Balanced in,
or MIC Tip INPUT

as Scott mentions, Starts with Basic Station RFI grounding,
NO DAISY CHAINS,
one Daisy Chain and all gone to moo poo.
Check . Daisy Chain Station earthing

image showing Incorrect Daisy Chain used to be in EVERY HAM Radio manual years ago.

NOT one single piece of Ferrite in use here, never has been, most VK's know me,
Grounding a ham station is not as simple as it sounds to do 100% correctly,
I reckon 20 + + in every 100 are incorrectly grounded, first thing I look for when asked for assistance to go QRO,


https://www.google.com/search?client=fi ... =965&dpr=1