Mic input issues on G2 with display panel

[rbanasik]

I am hoping for suggestions on how to address the following issues with my New G2 with front display.
• Using PTT to key up on LSB, others have noticed an initial carrier that diminishes gradually during the first 2-3 seconds of each transmission. The dynamic MXL DB-1 mic is connected to the front panel connector and the PTT line connected to the rear ⅜” connector.
• Using the 20dB Mic boost with mic on front panel connector seems to cause a 1sec to 2sec delay in transmitted audio resulting in the first word or two clipped from the transmission. Unchecking the 20dB mic Boost eliminates this issue.
• When my dynamic mic is connected to the rear connector labeled “XLR”, there is an echo on the transmitted audio resembling a 100ms-200ms delay of a second audio signal after PTT key up. The PTT is connected to the rear ⅜” connector. If I pull out the PTT line and use MOX, there is no audio at all. The XLR-labeled input is therefore not currently usable.
I am running Thetis 2.10.1 and have updated the G2 as follows:
FPGA BIT file date code = 23082023
Product: Saturn; version = 2
FPGA firmware loaded: Saturn prototype, full function; FW version = 8

p2app client app software Version:13 Build date: 18 Aug 2023 18:53:07
I would appreciate anyone’s suggestions on possible strategies to address these three anomalies.
Bob – Ki1n

[Trucker]

Bob, I believe there are some updates to PiHPSDR that may address the problems you are having. Look in the PiHPSDR thread as that is where I remember reading about similar issues being resolved.
James
WD5GWY

[rbanasik]

Thanks James.
I did just update the firmware, but I'll go check to see if there is anything newer.
-- Bob - Ki1n

[w-u-2-o]

The echo problem with the rear balanced mic input (somewhat erroneously labeled "XLR") has now been reported by numerous people. This is indicative of a significant RFI problem with the design and implementation of that input.

The problems you are describing with the front panel input are indicative of a wiring error in your mic cable. The front panel input is an unbalanced input and you are using a balanced microphone. This requires you to bond the XLR pin 1 (shield) and pin 3 (Mic -) to the shield of the TRS connector. Pin 2 of the XLR (Mic +) goes to TRS tip or ring depending on what you choose in the Thetis setup. PTT, and only PTT, must go to the opposite of that choice.

Finally, do not confuse the front panel mic input pinout with the rear panel "XLR" pinout, they are completely different.

[rbanasik]

Hi Scott.
Thank you for that valuable information. That certainly helps address my confusion!
Have you any thoughts on when and if Apache might address the balanced input issue?
-- Bob - Ki1n

[w-u-2-o]

The possible RFI issue with the balanced input is by no means confirmed, but there certainly are numerous reports. Even if it is formally confirmed, I have no idea what Apache may or may not do with respect to that.

[rbanasik]

Good morning Scott.
Understood, thanks. I suppose another workaround might be to employ audio input hardware or software to the PC and use VAC, right?
-- Bob - Ki1n

[w-u-2-o]

Both the VAC and the front panel input have been proven to be pretty robust. Either of those should work fine.

[rbanasik]

Thanks Scott. I appreciate your input.
-- Bob - Ki1n

[rbanasik]

Thanks Scott. I appreciate your input.
-- Bob - Ki1n

I have made numerous attempts to use a balanced input dynamic mic using VAC and a Presonus Audiobox GO, but cannot raise the input level without causing some type of "runaway" screeching or oscillation. I can only achieve about 5 watts output on the G2 and beyond that the apparent feedback takes off. I've noticed that I have to change Thetis settings from "mic" input to "line" input to get a decent audio level, but bringing up power results in lots of echo—perhaps rf feedback?
Echo seems to be an issue with other mics as well (both balanced and unbalanced input) using direct connection to the front or rear panel input. The XLR-labelled input has been a bust. Any thoughts on how I might proceed in addressing these issues?
-- Bob - Ki1n

[w-u-2-o]

Just to clarify: are you using piHPSDR or Thetis?

[rbanasik]

I'm using Thetis.
-- Bob - Ki1n

[w-u-2-o]

I have made numerous attempts to use a balanced input dynamic mic using VAC and a Presonus Audiobox GO, but cannot raise the input level without causing some type of "runaway" screeching or oscillation.

This seems like a problem with the Presonus setup that is causing feedback. The Presonus software is quite complex and it's possible you've got some setting wrong. Can you listen to yourself speaking into the mic (plugged into the Presonus) by using your computer speakers, just to make sure that sound is getting in and out of the Presonus OK?

99% of the time any echo (or reverb) means you do have an RFI problem. This is because mic-to-ant latency is right on par with reverb or echo timing, thus any RF getting back into the mic input will cause a reverb or echo effect. If you have echo then it's time to look at improving your station's grounding, bonding and shielding. A few ferrites here or there, often on the mic cable, can be helpful.

[laurencebarker]

Have you tried the audiotest app, to test the microphone with no RF present at all? That's what it's there for!

[rbanasik]

Hi Scott.
Thanks for that information. I have tried using the Presonus alone using the mic XLR input and headphone output and that seems to work normally with clean and full audio. The only software I'm using for the Presonus is the Asio drivers for the Win 11 I/O.
I agree that this sounds like an RF problem because I get an echo using a condenser mic in the front panel input that becomes worse if I increase drive power above 45 watts, and even more prominent if I engage a linear amp. Running barefoot the symptoms subside on 80 meters as opposed to 40, but are still detectable. Using direct condenser mic input or using dynamic mic through VAC both have echo and cascading feedback. I can start transmitting and get a "hello" out but by the time I can start counting the audio has cascaded into full oscillation.
All station equipment have individual bonding straps connected to a common ground line going to a dedicated 8' ground rod 15' away. I have ferrite mix 31 beads on all coax lines, mic lines, usb lines, Keyboard and mouse lines, power line, headphone line, and Ethernet line directly connecting i7 4.7Ghz 6 core computer to G2 with no switch or router, As suggested by others, I've tried isolating the power supply from the house neutral ground by using a 3-prong to 2-prong adapter plug on a dedicated home run 110 VAC line to no effect. I've also tried lifting the ground strap off the G2 to no effect. I've used field strength meters to sniff around the shack for rf and have not seen any indication near the G2 and the linear, but have seen moderate needle deflection (gain at max) a foot or two above the linear.
I have anecdotally heard that others have also experienced problems that might be similar to what I am experiencing; is it possible that this illustrates an issue that is internal to the G2?
Thanks very much for your help, I would appreciate any suggestions you might have.


Best regards,

Bob - Ki1n

[K4IBC]

Curious as to what you are using for audio cables and connectors?

[rbanasik]

I'm using a shielded XLR cable for the MXL BD-1 dynamic studio mic for VAC, and the condenser mic on the front panel TRS input is a D-104 with supplied cable and shielded TRS plug. Both mic cables have multiple ferrites. AudioBox GO is connected to the i7 via usb cable with ferrites.
-- Bob - Ki1n

[K6JR]

I have never been able to get the XLR 1/4" connector to work. I use mix 31 and have a good ground. I have tried running a ground to the shield of the connector and that didn't make it work either. I have used a Neumann 103 with a tube preamp and also a MXL BCD-1 direct to the XLR jack with the same RF distortion. Even with a dummy load and low power it doesn't work. The other microphone inputs work fine with 1500 watts out.

[w-u-2-o]

Bob -- that does beg the question, have you tried operating into a dummy load? This would help narrow the RFI problem. If this works then it's probably a conventional, radiated RFI problem. If the dummy load does not work, then it's either a conducted RFI problem or a problem internal to the G2.

[rbanasik]

I have never been able to get the XLR 1/4" connector to work. I use mix 31 and have a good ground. I have tried running a ground to the shield of the connector and that didn't make it work either. I have used a Neumann 103 with a tube preamp and also a MXL BCD-1 direct to the XLR jack with the same RF distortion. Even with a dummy load and low power it doesn't work. The other microphone inputs work fine with 1500 watts out.

Agreed, I have also not been able to use the the TR¼" connector. The XLR connection I do successfully use is on the Presonus AudioBox.
-- Bob - Ki1n

[rbanasik]

Bob -- that does beg the question, have you tried operating into a dummy load? This would help narrow the RFI problem. If this works then it's probably a conventional, radiated RFI problem. If the dummy load does not work, then it's either a conducted RFI problem or a problem internal to the G2.

Hi Scott. Thank you very much for that suggestion. I ordered a new dummy load and have been using it the past couple of days. The problem seems to disappear when the G2 is connected directly to the dummy load, so now I'm trying to figure out the source of the external RFI. Field strength meters don't seem helpful in pointing to any particular location. I've checked the antenna in question with a Nano VNA and it shows resonant on its operational bands and 49 ohm input impedance on the transmission line.
I've taken all peripherals such as the linear amp, rf sampler, external wattmeter, and antenna switch out of line but the problem remains when I increase audio drive on the Presonus Audiobox Go unit. Lower drive seems to work moderately well but some artifacts still show up intermittently such as an occasional clicking sound.
I have seen others such as K6JR who have also been unable to resolve this but don't know if that is significant here. Any thoughts on how I might be able to nail this down? Again, thanks very much for your help Scottt.
-- Bob - Ki1n