|
Hi all I have been getting noise in my two receiver (TS-50 and TS-2000) when the gps is pumping (yellow led) info to the U3S, the noise is everywhere in the receivers, HF, VHF and UHF, one can find it along any band several times. Same when it transmit wspr, I can find the signal several times in the same band. I have tryed to get rid of it using a metal case on the gps that use to be just the pcb conected to an external antenna that use a RG174 cable but I had no luck. The U3S is inside a metal case too. Signal is coming by air, they don?t share power supply (U3S and Receivers). U3S is working with 2 BS170 and its own lpf and after that a linear amplifier to get 5w output and a low pass filter after it Antenas are 20m apparts. The U3S is working 20m wspr beacon using a 1/4wl vertical with 12 radial, and for receivers I use: for 40 and 15 = dipoles, for 2m =10ele beam, and for 70cm=7ele beam., all receivers antennas are at a 15m high tower. This is not knew, It has been happening since day one and I have no solution for it, every time I want to do DX (JT, EME, ETC) specially in VHF I simply turn off the U3S. Has somebody experienced something similar? any cure?
Al CX8AT
|
I've also seen this, I run the U3s on a dedicated Diamond CP-6, it's own wall wart, 3 separate regulators inside the case. There's noise from both the GPS module and (I think) the I2C commands to the Si5351 module that somehow gets out. Even when sitting in the menu and only the GPS is working it emits noise.
One clue could be on the QLG1 module, it has inductors on the suppy and ground lines but the levelshifting buffers does not have any slew rate control AFAIK. One approach could be to apply series resistors on the 1PPS and TXD lines. I also ponder the effect on the series inductor on both the gnd and vcc lines, if there's some issues when the grounding is separated with a inductor. Another thing that I've been thinking of trying is the ground path from the QCU to the chassis with a low inductance path instead of thin "long" wires. Maybe change one of the hex nut legs to brass. Or make a short braid from the CPU board directly to one of the top/bottom case pieces. I'm not exactly sure the 4 aluminium pieces are peroperly connected just with the 8 screws holding them together, being anodized aluminium and all. One experiment could be to disconnect power to the PA pin and see if the noise is amplified there.
|
Hi Alberto
Here I have a homebrew CW transceiver I have operated on 40, 20 and 17m. My QRP Labs Ultimate3S (not transmitting), VFO and Clock kits are all operational in parallel from the same QLG1 GPS. The QRP Labs kits are all in their QRP Labs aluminium enclosures. All that and the CW transceiver, are all powered from the same 12V supply. One 7805 in the CW transceiver produces 5V for it; another 7805 in the VFO kit enclosure produces 5V for the Ultimate3S, VFO and Clock kits. There's about 1m of shielded cable to the GPS. The CW transceiver antenna is a random longwire tuned by L-match ATU, and a rather long (maybe 5m) wire bolted to the metal frame inside the roof. What I'm saying is that it is a severely poor set-up...
Yet I am able to operate the CW transceiver just fine with the QRP Labs gear running; if I turn the volume up to max in a quiet part of the band I can just about hear the pulsing of the GPS (yellow LED), the "GPS pumping" as you called it. But it does not interfere with normal operations.?
You already mentioned that you have separate supplies so that is not the problem. I assume that you have used shielded cable between your U3S and your QLG1 GPS. Something somewhere in your system must be radiating that 9600 baud serial data... I would suggest to remove the YELLOW LED! Pulsing that yellow LED will certainly be putting noise onto the GPS supply lines. That would be the first thing to try I think.?
73 Hans G0UPL
On Mon, May 8, 2017 at 4:04 PM, Daniel Ekman (kng) <knegge@...> wrote: I've also seen this, I run the U3s on a dedicated Diamond CP-6, it's own wall wart, 3 separate regulators inside the case. There's noise from both the GPS module and (I think) the I2C commands to the Si5351 module that somehow gets out. Even when sitting in the menu and only the GPS is working it emits noise.
One clue could be on the QLG1 module, it has inductors on the suppy and ground lines but the levelshifting buffers does not have any slew rate control AFAIK. One approach could be to apply series resistors on the 1PPS and TXD lines. I also ponder the effect on the series inductor on both the gnd and vcc lines, if there's some issues when the grounding is separated with a inductor. Another thing that I've been thinking of trying is the ground path from the QCU to the chassis with a low inductance path instead of thin "long" wires. Maybe change one of the hex nut legs to brass. Or make a short braid from the CPU board directly to one of the top/bottom case pieces. I'm not exactly sure the 4 aluminium pieces are peroperly connected just with the 8 screws holding them together, being anodized aluminium and all. One experiment could be to disconnect power to the PA pin and see if the noise is amplified there.
|
Hi Alberto,
I see this problem too and have tried similar cures to no avail. I run separate QLGs and see severe data pulse interference on my 70Mhz U3 Rx radiated to the close by antennas. I too have tried screening the gps circuitry, series chokes and decoupling of the data and 1pps lines. I run separate 5v regulators from 12volts to the gps and screened leads. So far, the only 'work around' I have is once the gps has set the Rx VFO reference frequency, I have to disconnect the QLG1 completely (I.e powered off) and run the Rx standalone. If anyone has a cure I would like to hear it too.
Ken g4apb
Sent from my iPad
|
Hi Ken, all
Has anyone tried disconnecting the yellow LED?
73 Hans G0UPL
On Mon, May 8, 2017 at 9:21 PM, KENS 2ND ACC <lfoofui.nbz42@...> wrote: Hi Alberto,
I see this problem too and have tried similar cures to no avail. I run separate QLGs and see severe data pulse interference on my 70Mhz U3 Rx radiated to the close by antennas. I too have tried screening the gps circuitry, series chokes and decoupling of the data and 1pps lines. I run separate 5v regulators from 12volts to the gps and screened leads. So far, the only 'work around' I have is once the gps has set the Rx VFO reference frequency, I have to disconnect the QLG1 completely (I.e powered off) and run the Rx standalone. If anyone has a cure I would like to hear it too.
Ken g4apb
Sent from my iPad
|
Hi Hans
There is no change without the yellow led.
Al
CX8AT
On 08/05/2017 15:24, Hans Summers wrote:
toggle quoted message
Show quoted text
Hi Ken, all
Has anyone tried disconnecting the yellow LED?
73 Hans G0UPL
_._,_._,_
|
Hi Al
Is there any change when you have the U3S in editing-menu mode? I mean press the Left button so that it enters the configuration mode - so that nothing is being refreshed to the display when the serial data comes in from the GPS.
73 Hans G0UPL
On Mon, May 8, 2017 at 10:47 PM, Alberto Vazquez <cx8at@...> wrote:
Hi Hans
There is no change without the yellow led.
Al
CX8AT
On 08/05/2017 15:24, Hans Summers wrote:
Hi Ken, all
Has anyone tried disconnecting the yellow LED?
73 Hans G0UPL
|
Hi Hans, No difference in editing mode, nor if I set GPS mode to 0, 9600. ? 73 Ken ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of Hans Summers
Sent: 08 May 2017 21:14
To: [email protected]
Subject: Re: [QRPLabs] GPS/U3S generated Noise ? Hi Al Is there any change when you have the U3S in editing-menu mode? I mean press the Left button so that it enters the configuration mode - so that nothing is being refreshed to the display when the serial data comes in from the GPS. ? On Mon, May 8, 2017 at 10:47 PM, Alberto Vazquez <cx8at@...> wrote: Hi Hans
There is no change without the yellow led.
Al
CX8AT
On 08/05/2017 15:24, Hans Summers wrote: Has anyone tried disconnecting the yellow LED? ?
?
|
Maybe it's not important, my LCD LED's in the U3 is PWM controlled and in the RX is an 220Ohm resistor without PWM.
|
Hans, ???????? I would think that a check of grounding?might be the answer. In telecom sites making sure that you don't have any ground loops (potential differences between ground points and different ground points) and ensuring that a shielded cable is only grounded on one side might be the answer/s your looking for? This was the cause of many problems in my past engineering episodes.
Paul
|
Just throwing in my 2 cents - i have the same issue too, but it extends up into the 2m band as well.?
I'm not running my U3s a lot at the moment though, so haven't started going through the troubleshooting yet, and am also not using shielded cable for the interconnect to the GPS module - again, just in testing really until i get the filters and cases sorted out - i had assumed that once the GPS module was connected with shielded cable, and inside a metal case of its own, properly grounded, it wouldn't be an issue - maybe that isn't the case.? It certainly sounds like it's the 9600baud serial data that's causing the issues.?
|
I'm curious if folks have identified a cure for this.
I'm running my QLG1 through a fairly long cable (QLG is upstairs in the kitchen) to my VFO/SigGen (it's in the
basement).? I am, like other posters here, hearing that pulse across a broad spectrum (don't have a precise
definition of that).
I tried wrapping the cable in ferrite cores, but that didn't help.
What I found was that if I disconnected the serial data line, the problem went away.? So, I can accurately
calibrate the VFO/SigGen, and I can show the 1pps heartbeat, but I can't show any of the data (date/time,
for example) that comes in the serial data stream.
Any hints would be appreciated.
- Bruce K1FFX
|
Hi all,
I reported a couple of weeks ago about the gps noise and my qrp labs receiver, I have tried separate power supplies, shielded cable grounded one end only, metal enclosures.?
The gps is attached to my progrock and my solution is to simply disconnect the gps serial 1pps unless I check calibration of the progrock.?
I also have another, earlier, qlg1 attached to a u3s, in the same room and that one is quiet, both on the u3s and the receiver doesn't hear it. I haven't tried it on the progrock, receiver set up.
Andy
|
Hi, Andy -
Thanks for the reply.? I was interested to see that you have the noise problem with one QLG1 and not with the other.? Assuming that your two GPS units are pretty much identical, I'm wondering what's different on the other end of the line.
I've opened up the VFO/SigGen and tried (1) pulling the serial data line off on the VFO end (still heard noise) so then (2) tried wrapping various leads in the VFO case through a toroid, on the assumption that the serial data line noise was being picked up in the case ... but no luck.
Since Hans has apparently not seen this behavior, and there were just a few respondents to your post (or this one), either QLG1 users have been ignoring it or there is some tweak I can do to clean it up, i.e., something on the VFO end.
Best -
Bruce K1FFX
|
Hi Bruce
A digital signal changing at a rate of 9600 baud will generate noise. Whether or not that noise is radiated enough to be troublesome, or conducted through power supply lines or ground enough to be troublesome, depends on a lot of different factors. Shielding, grounding, location, both for the GPS and for the receiver. And how sensitive your receiver antenna system?is to local noise. So tends to be different every time.?
Two things you can try (and I have heard helped) to make it better:
1) Disconnect the yellow LED. The easiest way to try it, is to left one end of the series resistor R4.?
2) Put a 1K resistor in series with the data line between QLG1 and GPS
73 Hans G0UPL
Hi, Andy -
Thanks for the reply.? I was interested to see that you have the noise problem with one QLG1 and not with the other.? Assuming that your two GPS units are pretty much identical, I'm wondering what's different on the other end of the line.
I've opened up the VFO/SigGen and tried (1) pulling the serial data line off on the VFO end (still heard noise) so then (2) tried wrapping various leads in the VFO case through a toroid, on the assumption that the serial data line noise was being picked up in the case ... but no luck.
Since Hans has apparently not seen this behavior, and there were just a few respondents to your post (or this one), either QLG1 users have been ignoring it or there is some tweak I can do to clean it up, i.e., something on the VFO end.
Best -
Bruce K1FFX
|
Hi Bruce.
Both qlg1's were built identically, the u3s one takes its power from the u3s, which is battery powered, the progrock one is fed from an ac psu which feeds a 7805. The progrock and receiver use the same psu but have a separate 7805 and 12v from psu. All grounds feed back to psu.
The receiver, with no antenna connected hears the progrock gps but not the u3s one. If I remove power to the progrock gps the noise stops.
Both gps have 4 core shielded cable, u3s gps has shield connected to u3s case and GPS case. Progrock gps has two shielded cables, one for power, one for 1pps connection to progrock. Both shields connected to progrock gps case and then to gps ground,?both use identical active antenna.?
Only way I can remove serial burst noise from receiver is power down progrock gps and ground 1pps connection to gps through a switch to progrock ground. I just switch gps on now and then to calibrate.
Incidentally, if I remove wired connection from gps to progrock and power both from separate power sources noise is still heard.
The only major difference between both gps is the initial power supply. I'm not well at the moment but hopefully by the weekend I'll be able to swop the GPS'S over and use both a battery supply and ac supply to see if that makes any difference or whether the second gps is simply radiating more than the first, I built the first gps board over a year ago and it has been fine, I do have another on order but it hasn't arrived yet, be interesting to see how that one behaves when built.
Andy
|
Thanks a lot for your response and tips, Hans (and thanks, Andy, for your detailed follow-on).
I'm happy to report that inserting the 1K resistor in the data line (to be clear, the TxD/RxD line, not the
1pps line) seems to have solved the issue.? I've been using the VFO/SigGen as a "variable crystal" to
extend the general coverage capabilities of my vintage Drake R-4B.? Adding the 1K resistor has, at a first
listen, completely eliminated the formerly loud and pervasive serial data noise burst.
Will keep listening, but looks like the issue is solved.
Thanks!!
- Bruce K1FFX
|
Hi Bruce,
sorted, where did you place the 1k resistor?
In my set up the txd pin is not connected, only +ve, ground and 1pps.?
Andy
|
Hi, Andy -
It was the TxD/RxD line that seemed to be the major source of that noise burst.? So, I
put the 1K resistor at the QLG1 right at the TxD pin.? I'll try to attach a pic.??
So far, I don't think I've heard noise coming from the 1pps line.
Hope this helps!
K1FFX
|
Thanks for that Bruce,?
That's what I thought that you had done. I think I'll try taking a 1k resistor to ground from the txd pin as in my application the txd pin is not used. Or cut the track to the txd pin and place it in series there, like the rxd line.
73
Andy
|
Daniel Ekman SA2KNG
Stefan
M0RON