LIstening post -update

[VK3EB]

A few pix from todays work,
First impressions currently listening to 80m noise floor is excellent. My USB 5v supply, part of my battery box turned out to be a noise generator. Denys unit to the rescue. Tune to 3330kHz CHU time signal from Canada -very impressive!
Great location, thanks Bill.

 

[Josh]

https://listen.sadarc.org/

now points to the temp listening post.

 

[VK3ZYZ]

I've been listening to the SDR this morning and have a couple of observations.
The background noise is higher than that on the Ironstone SDR. That may be my switch mode power supply producing noise. It may be worth while making a linear supply to see if that helps. VK3EB has a higher current reg I gave him a couple od fays ago, or I could put one together.
Also, the frequency seems to wander slightly as the SSB audio wanders up and down a bit.

 

[Josh]

I don't know if it is relevant at all.
If I read right the SDR can have a GPS sync for precision control of the ADC clock, maybe just maybe its is trying to do that loop in the background without a gps sync which might, might cause the ADC clock to wander a little.
I did swap out ntpd for chronyd but i did not take the time to set it up for GPS PPS sync, as I was and am not sure if the open web rx software needs to control any of that.

This is ntp sync status:

210 Number of sources = 4
MS Name/IP address         Stratum Poll Reach LastRx Last sample              
===============================================================================
^+ 159-196-3-239.9fc403.mel>     2  10   377   180  +1231us[+1231us] +/-   34ms
^* time.cloudflare.com           3  10   377   775  -4621us[-6698us] +/-   14ms
^+ bitburger.simonrumble.com     1  10   277    14  -2577us[-2577us] +/-   21ms
^+ 14-202-65-230.tpgi.com.au     1  10   377   308  -3799us[-3799us] +/-   26ms

Reference ID    : A29FC87B (time.cloudflare.com)
Stratum         : 4
Ref time (UTC)  : Sat May 17 01:10:44 2025
System time     : 0.001697727 seconds slow of NTP time
Last offset     : -0.002077176 seconds
RMS offset      : 0.002063389 seconds
Frequency       : 6.657 ppm fast
Residual freq   : +0.001 ppm
Skew            : 0.111 ppm
Root delay      : 0.025742332 seconds
Root dispersion : 0.003906695 seconds
Update interval : 1040.1 seconds
Leap status     : Normal

As you can chrony picked the stratum 3 source not the 1's.



also a note on bandwidth. the water fall seems to an average take 100KiB/s / per user.

and to date we have used 1.13GiB upload and 159MiB download with respect to the modem.

here is a pic of the MW band last night. I didn't know that there were still that many stations. I've never heard 621 as good.

 

[VK3YNV]

I have a spare LM1085ISX-5 Fixed 5V 3A linear regulator that could be used for experiments if you want to have a play.

 

[BillC]

Thanks Ray . The Listening Post noise :- There is a very strong spur, fundamental frequency about 268 khz the level as measured is S9 + 20 dbm. This generates harmonics all the way up the spectrum. The noise level is higher at the lower range of the spectrum. When listening with a sniffer probe in and outside of the cabinet there is a very strong "modulated (audio garbage) tone" at 268 khz. To get the best out of the low noise site we will investigate the power supplies. The noise floor seems to be about S5 in reality this site has about S1 normally. However as it stands the reception is quite good.The system might go off for short periods while we check out power supply noise etc. I might try a trap filter in the power leads could a kind person do the calculations for a trap on 268 khz.

 

[Josh]

Thanks Ray . The Listening Post noise :- There is a very strong spur, fundamental frequency about 268 khz the level as measured is S9 + 20 dbm. This generates harmonics all the way up the spectrum. The noise level is higher at the lower range of the spectrum. When listening with a sniffer probe in and outside of the cabinet there is a very strong "modulated (audio garbage) tone" at 268 khz. To get the best out of the low noise site we will investigate the power supplies. The noise floor seems to be about S5 in reality this site has about S1 normally. However as it stands the reception is quite good.The system might go off for short periods while we check out power supply noise etc. I might try a trap filter in the power leads could a kind person do the calculations for a trap on 268 khz.

That frequency is suspiciously in the range of the switch mode controller(225kHz-280kHz) that is inside Denys 5V power supply. It could be just a coincidence, but I agree it is definitely worth looking at. see attached page from datasheet. If its just conducted a linear post reg might give the best of both worlds.

 

[VK3YNV]

I might try a trap filter in the power leads could a kind person do the calculations for a trap on 268 khz.

Hi Bill, you want minimal DC resistance in line, so just a ferrite in series and a low esr cap in parallel might help.
Denys just told me there is no filtering on the input, so clamp on ferrites on both on the input DC and output DC cables since the 268 kHz could be radiated as well as directly conducted.

 

[BillC]

Hi Bill, you want minimal DC resistance in line, so just a ferrite in series and a low esr cap in parallel might help.
Denys just told me there is no filtering on the input, so clamp on ferrites on both on the input DC and output DC cables since the 268 kHz could be radiated as well as directly conducted.

Thanks for the help Ray and Josh. I will try a few ferrite clip on,s and see how that works, but we might change the power supply. By the way the battery voltage "dived" overnight / 24 hrs dropped by .29 of a volt. I think we can handle that discharge rate easily.

 

[BillC]

I've been listening to the SDR this morning and have a couple of observations.
The background noise is higher than that on the Ironstone SDR. That may be my switch mode power supply producing noise. It may be worth while making a linear supply to see if that helps. VK3EB has a higher current reg I gave him a couple od fays ago, or I could put one together.
Also, the frequency seems to wander slightly as the SSB audio wanders up and down a bit.

Hi Denys .Thanks for the report. The noise could be from the power supply. I will try to check it out soon. Thanks for the use of the power supply it certainly got the system up and running quickly, plug and play etc.

 

[BillC]

The Listening post battery is now sitting at 12.15 volts. I will now run a lead out from the house for a "quiet charge" 2.30 pm 18/5.

 

[BillC]

I have just checked out the 5 volt power supply with the scope . The pictures tell the story, the square wave image shows waveform across the input terminals and the spiked waveform is across the output terminals. I will build up a linear type 5 volt regulator and we will see how that works, hopefully no noise.

 

[VK3YNV]

Good work Bill, 267.9 KHz must be the switching frequency.

 

[Josh]

indeed it is.

 

[BillC]

indeed it is.
View attachment 3237

Cheers .Thanks for the replies. Ray and Josh. I am constructing a regulator using an LM 338k, it is taking some time to organize the hardware. At this stage I intend to use fixed resistors to set the output voltage to about 5.05 volts, we will see how it holds up under load and wiring loom and contact resistance. More to follow.

 

[BillC]

Changed the regulator at 5 pm no spur now. Also I will build a 31 meg low pass filter for the antenna input as there is a bit of FM breakthrough around 30 megs. I hope we have obtained a lower noise result.

 

[VK3YNV]

I listened a bit on FT8 earlier, lots of US and European Stations on 40M, this is about 30 minutes of signals on the PSK reporter map.

Looking pretty good Bill, I think the noise is lower with the linear regulator, but it was already pretty good anyway.

 

[VK3ZYZ]

It looks to be playing up this AM (8:30 Tues 20th)

 

[VK3ZYZ]

I think I'll make another regulator but this time, use the correct input electros. The original (2019) has a couple of 47uf caps but I only found 18uf as used on the o/p. Also I do not remember if the switcher chip was insulated from the case or not. It is in this current build. A large electro added to the i/p terminals will not hurt.
The linear reg is definitely better noise wise. Maybe the switcher can be used as a pre regulator and mount it right at the battery.
And thanks Bill, for finding a loose terminal so it is working well again

 

[BillC]

It looks to be playing up this AM (8:30 Tues 20th)

Thanks Denys. I have been out and tightened all the power wires on the linear regulator connector strips and the Rcvr seems to be working. I think that long term we will use your regulator with a bit more filtering. It is of course much more efficient in terms of power consumption.Cheers.