Date
21 - 40 of 64
QCX-mini constant 700Hz tone #qcxmini
|
Richard - N1RBD
On Mon, Dec 28, 2020 at 07:10 PM, geoff M0ORE wrote:
RMS Multimeter. I did try a factory reset with no change. The background tone did become significantly more apparent when I changed the reference frequency to 25 instead of the default since I have the TCXO. |
|
|
|
That's a pretty strong signal radiating out the antenna!
73, Randy, KS4L |
|
|
|
Richard - N1RBD
Well, I replaced a bad Q5 on my board, but that did nothing for this problem. I am getting 4w at the BNC now, though! :)
|
|
|
|
g4edg
Hi Mont
I wonder if you ever solved the 700Hz tone thing? I have the same problem on my 18MHz QCX mini (I also have a original 18MHz QCX and have no trouble) I've noticed that if the rig is in split mode or the RIT is on, the tone can be tuned up and down. The manual says the synth produces a separate output for transmit and receive, and that the TX output is switched off during receive and the RX output is switched off during transmit. I can hear the RX synth output on another rig and of course the TX synth output when I key the QCX. Neither output seems to be there when it shouldn't. So where is the QCX mini hearing this low level signal from? Any ideas anyone? Any subtle changes in the firmware 1.07a? Steve G4EDG |
|
|
|
Gregg Myers
Hi Steve, Mont, I also have this low level ~700Hz tone. I also agree this is something I hear on my QCX-mini that I do not hear on my regular QCX at all. I have not been able to solve it. I can only say what hasn't worked for me so far: It's not the LCD module because it is present without the LCD in place. It disappears completely when you are doing factory reset (before you select the band) so I think it only happens when the Si5351a is sending a signal or being told to send a signal by the Atmega328. I tried all sorts of attempts at shielding various components with little copper tape 'houses' and no luck yet. I tried a 0.1uf decoupling cap on the VDD line of the Si5351a (as recommended on the datasheet) to no avail. I resoldered the bypass/filter caps/inductors to the Atmega328 supply lines, no difference. I can see the low level signal tone on a scope if I place the scope on the audio output jack. If I put my finger on the Si5351a it is very slightly louder. In summary, I have had a very difficult time tracking down the source. As low level as this constant tone is, I find it very fatiguing to listen to. I'm hoping someone else can suggest further testing as I'm running out of ideas. 73, Gregg W7GRM Hi Mont |
|
|
|
g4edg
Hi Gregg
Thanks for the list of things you have tried... Pretty comprehensive! I had mine apart yet again and adjusted the 3 trimmer pots and managed to reduce the 700Hz tone level a bit. It's still there, but quiet enough to ignore. What sort of level are you getting? Would like to hear Han's take on this. Steve G4EDG 73 Steve |
|
|
|
Mont Pierce KM6WT
Hi Steve, Gregg:
toggle quoted message
Show quoted text
Nope, have not been able to solve it. I've been waiting/hoping someone would... It also did not go away with a factory reset. I thought it had at first, then I realized I forgot to reset the Ref Frequency back to 25MHz for the TCXO... After changing the Ref Frequency back to 25MHz, the 700Hz background tone was the same as before Factory Reset. On Tue, Jan 5, 2021 at 10:02 AM, Gregg Myers wrote:
Mine isn't very loud, and with my Tinnitus, I don't notice it too much, Except... When tuning in moderate to weak signals they beat with the 700Hz background tone and are very difficult to copy... Hope someone solves this... 73 km6wt |
|
|
|
I still think that sounds like the oscillator leaking through, and on all the time, which it usually is not unless you are in the menu system. That's especially a problem if it's making it's way through the transmit path. It's controlled by the microprocessor, so something is triggering it to generate the signal, which might be the NAND gate, IC3. Check the voltages again. Here is a link to the trouble shooting of that section. https://www.qrp-labs.com/qcx/qcxtrouble.html#transmit
-- 73, Dan - W2DLC |
|
|
|
Gregg Myers
Hi Steve, Mont, I re-aligned the BPF, IQ balance, Low and High and all seems quite reasonable. BPF peaks high 8, almost 9, I guess, but the IQ, Low, and High are volume dependent so it is hard to compare. I did note a slight level difference between low and high alignment minimization, so I'll have to see if this is normal or abnormal by comparing it to my regular QCX. I can say it is also not the firmware version, nor a rogue noisy processor as I replaced the processor from a standard QCX (different firmware version) and the low level signal is the same. The noise starts at the moment the Si5351A signal shows up on my scope. I also can make the noise disappear if I turn on split mode on the VFO (A and B on different frequencies), but when I adjust both A, B VFO to be the same frequency, the noise also returns. I can only conclude that it is noise from the Si5351A and all my attempts to minimize it so far have failed. If I do a factory reset and choose 40m instead of 17m, I don't hear the noise, by the way. Maybe I could rebuild for 40m, but I really want 17m to work in this case. Since all our signals go through a 700Hz CW filter, it is not that surprising I guess that what we end up hearing as noise can end up somewhere around the 700Hz beat. Hi Dan, Yes, I agree it is noise/leakage, but not from Clk2 (I think that is what you were saying since you refer to IC3). The Clk2 signal is not on when not transmitting as I can monitor this with my scope, say pin 4 of IC3. But of course we have an oscillator on, even during receive, with Clk0 and Clk1, otherwise our quadrature sampling detector wouldn't be able to do its thing. But yes, I agree it is leakage/noise from somewhere, likely around the Si5351a, but this is very elusive so far and could even originate in signal coupling on the board traces. Otherwise everything on the mini works fine so far as I can see (after I had to replace a dead OPA2277U, but that was another issue). 73, Gregg W7GRM On Wed, Jan 6, 2021 at 9:00 AM Daniel Conklin <danconklin2@...> wrote:
|
|
|
|
Today i worked on a QCX Mini for someone who reported a 700 Hz tone on all the time. I probed IC1 and CLK2 was off during receive as expected. I checked the tone with an oscilloscope and it was actually varying between 780 and 812 Hz. That sounded more like an oscillation. I checked voltages on all the op amps and all were good. I traced the source of the oscillation back to IC8B, the first stage Lowpass filter in the CW bandpass filter circuit. It was oscillating and putting out a 12 volts P-P tone to the following stages. Since the following stages finish up a 200 Hz wide filter centered at 700 Hz there was some attenuation of the 800 Hz signal but it was very loud in the headphones. Since IC8 voltages checked good, I checked all the parts around it as well as the expected connections between pins and all looked good, so I replaced the IC and the oscillation went away and the receiver settled down and aligned properly. I know symptoms and results vary, but I hope this helps some find their own problems regarding this particular phenomenon.
Ron |
|
|
|
Gregg Myers
Hi Ron, Interesting information and good troubleshooting on your part, too. In my case, the tone is very soft, so more of a subtle background tone and I don't have a large signal as seen on my scope on either IC8 or IC9 on any of the inputs/outputs. Well, back to the drawing board for me, but I'm glad you helped someone else get their mini fixed. Maybe I'm being too picky. It's like my mini has a subtle case of tinnitus. 73, Gregg W7GRM On Wed, Jan 6, 2021 at 6:27 PM Ronald Taylor <wa7gil@...> wrote: Today i worked on a QCX Mini for someone who reported a 700 Hz tone on all the time. I probed IC1 and CLK2 was off during receive as expected. I checked the tone with an oscilloscope and it was actually varying between 780 and 812 Hz. That sounded more like an oscillation. I checked voltages on all the op amps and all were good. I traced the source of the oscillation back to IC8B, the first stage Lowpass filter in the CW bandpass filter circuit. It was oscillating and putting out a 12 volts P-P tone to the following stages. Since the following stages finish up a 200 Hz wide filter centered at 700 Hz there was some attenuation of the 800 Hz signal but it was very loud in the headphones. Since IC8 voltages checked good, I checked all the parts around it as well as the expected connections between pins and all looked good, so I replaced the IC and the oscillation went away and the receiver settled down and aligned properly. I know symptoms and results vary, but I hope this helps some find their own problems regarding this particular phenomenon. |
|
|
|
Daniel Walter
Good troubleshooting Ron. Another bad op amp problem. Gotta be something from the PCB factory.
-- 73, Dan NM3A |
|
|
|
Steven Dick, K1RF
Interesting. The LM4562 used in the QCX and QCX+ has a unity gain-bandwidth spec of 55 MHz. The NE5532/5532A used as s substitute chip has a gain-bandwidth spec of 10 MHz.
The OPA2277 used in the QCX mini has a gain-bandwidth spec of 1 MHz. The filter design is based on the 4 states QRP Hy-Per-Mite filter design by NM0S which uses a TLC274. That part has a gain-bandwith 1.7-2.2 MHz. It's a pretty standard type of circuit, but without simulating it, I'm thinking the unity gain-bandwidth for the OPA2277 may be too low in this application. -Steve K1RF ------ Original Message ------
From: "Ronald Taylor" <wa7gil@...>
Sent: 1/6/2021 8:27:16 PM
Subject: Re: [QRPLabs] QCX-mini constant 700Hz tone #qcxmini Today i worked on a QCX Mini for someone who reported a 700 Hz tone on all the time. I probed IC1 and CLK2 was off during receive as expected. I checked the tone with an oscilloscope and it was actually varying between 780 and 812 Hz. That sounded more like an oscillation. I checked voltages on all the op amps and all were good. I traced the source of the oscillation back to IC8B, the first stage Lowpass filter in the CW bandpass filter circuit. It was oscillating and putting out a 12 volts P-P tone to the following stages. Since the following stages finish up a 200 Hz wide filter centered at 700 Hz there was some attenuation of the 800 Hz signal but it was very loud in the headphones. Since IC8 voltages checked good, I checked all the parts around it as well as the expected connections between pins and all looked good, so I replaced the IC and the oscillation went away and the receiver settled down and aligned properly. I know symptoms and results vary, but I hope this helps some find their own problems regarding this particular phenomenon. |
|
|
|
Hans Summers
Hi Steve
That's interesting but have you any theoretical framework around that, other than observing the difference in gain-bandwidth product specifications for the op-amps? Why would 1MHz gain-bandwidth spec not be enough for a 700Hz filter with low gain? It seems to me that the difference in gain-bandwidth specification is not relevant for this application, but if you have any explanation I'd love to hear it. 73 Hans G0UPL http://qrp-labs.com |
|
|
|
Steven Dick, K1RF
I will look further into it and report back. If related, it is a function of what the open loop response looks like with the component values and configuration of those components used in the feedback paths. A typical op amp has a single pole in the open loop response. If the feedback components add an additional pole, in the region as the amplifier open loop gain plus feedback components attenuation is such that gain is more than zero at the crossover point, there could be two poles in the open loop response with gain greater than one which can cause instability. A problem could still happen even with audio frequency type filters. I don't know if this is a problem here or not yet.
-Steve K1RF ------ Original Message ------
From: "Hans Summers" <hans.summers@...>
To: "QRPLabs@groups.io Notification" <QRPLabs@groups.io>
Sent: 1/7/2021 7:35:22 AM
Subject: Re: [QRPLabs] QCX-mini constant 700Hz tone #qcxmini
|
|
|
|
Steven Dick, K1RF
In summary, the circuit itself looks fine. I did a quick AC analysis and transient analysis of the first stage of the filter, using a few different op amps with different gain-bandwidths. The AC analysis shows a normal looking bandpass shape. The transient analysis does show quite a bit of ringing at the output when fed with a square wave, indicating a relatively high "Q". But this is a consequence of the filter design and is normal, pretty independent of the op amp gain-bandwidth spec. So I don't see anything obvious. Bad op amp? Don't know.
-Steve K1RF ------ Original Message ------
From: "Steven Dick" <sbdick@...>
To: "QRPLabs@groups.io" <QRPLabs@groups.io>
Sent: 1/7/2021 8:55:17 AM
Subject: Re: [QRPLabs] QCX-mini constant 700Hz tone #qcxmini
|
|
|
|
Rod Smith
Hi all
I'd be grateful for any updates on this constant tone issue please. I have a 20m QCX mini that is "nearly" fixed. It now just has symptoms similar to those described by Gregg, Steve, Mont, and Richard: a low volume tone that beats with weak received cw in the passband. In the absence of signals the background noise peaks at the same frequency and it's as though the audio filter stage is at or on the verge of oscillating. When I change the cw offset (menu 3.8), the low volume note also changes frequency (without my changing the frequency of the sig gen) and the beat rate remains the same. A factory reset didn't help. Voltages on all ICs look OK. Compared to my 40m "regular" QCX my mini sounds noisier and the minimum detectable signal is around 8dB worse - quite probably because the noise is so narrow band. Comparatively it sounds 'orrible. A quick history of the mini: it had a faulty IC7, in fixing that I blew the voltage regulator (with the ~7.5V dc supply thereby provided to the 5V rail); that blew the FST3253, which I replaced together with the voltage regulator. It's fitted with a 25MHz TCXO. On my 40m regular QCX I changed the sidetone, offset, and audio filter to something around 550Hz (from memory). It doesn't look straightforward to do that on the mini but that's a different issue isn't it. Thanks for any and all help with this. And a big thanks to those who directly or indirectly have helped me along the way so far. Best wishes Rod G0VKX |
|
|
|
Mont Pierce KM6WT
On Mon, Feb 1, 2021 at 08:05 AM, Rod Smith wrote:
I'd be grateful for any updates on this constant tone issue please.After my original post, I've also looked for this on my 40M QCX+ and found it there too, although volume of the tone is so low it blends in with the background noise and is barely noticeable. In addition (which this tone might be also), I've discovered that all 3 of my QCXs have several birdies that can be heard if using Split mode or RIT. I believe these birdies are inherent to internals of the Si5351a chip. I've report this to Hans and hopefully he can reproduce it and look for a solution... To repeat what I found, try the following:
I also found that tuning RIT up/down several hundred hertz has the same birdies as split mode. Please let me know if I'm the only one seeing this? I rarely use split, or tune RIT that far off frequency. But it would be nice if they worked without the birdies... Also, if the constant 700Hz background tone is related, hopefully possibly a change in firmware can avoid internal Si5351a issues related to these birdies... tnx & 73 |
|
|
|
Rod, What other things have you checked in the 5v line? Diodes D1 and D2 may be damaged, as well as IC1 and IC3.
-- 73, Dan - W2DLC |
|
|
|
Gregg Myers
On my mini, I have this same observation as Mont: "the volume of the tone is so low it blends in with the background noise and is barely noticeable". So I can live with it, but it would be nice to be able to fix it or reduce it further. In actual listening with actual atmospheric noise, I start to not notice it at all. With a dummy load, it is very noticeable (but again, albeit at a very low volume). It changes tone with any setting changes to the si5351a (for example, split mode, cw offset, reverse mode). As for my build, I do not have the TXCO option on my mini. I don't think of the tone I hear as a birdie as such (which I think of as appearing as a fictitious signal at isolated fixed frequencies) as the low tone does not depend on what frequency I am listening to but is constant across the band frequencies. I applied some shielding to the various ICs in the circuit without any improvement at all. I did check all the bypass/decouple caps to the power to the various ICs to no avail also.
My current thinking is that it is either leakage within the si5351a itself, or crosstalk due to the device layouts (compared to a regular QCX re-arranging the components is, of course, quite unavoidable due to the small size of the mini!). If it is either one of those, I don't think I can fix it in any reasonable way and I have also run out of other ideas. If someone else is able to reduce it, I would like to hear about how they managed it! 73, Gregg W7GRM On Mon, Feb 1, 2021 at 9:30 AM Daniel Conklin <danconklin2@...> wrote: Rod, What other things have you checked in the 5v line? Diodes D1 and D2 may be damaged, as well as IC1 and IC3. |
|
|
