Problems with TX

General discussion and support for the Peaberry SDR V2.

Re: Problems with TX

Postby hansvi » Wed May 29, 2013 3:08 pm

Also, are you sure the measurement is correct? If you measure pin 3 on U14, the voltage is correct? (3.3V p2p square wave 56.32MHz). It is my understanding that esp. 1x probes are not well suited for high frequencies, and start to attenuate.
73,
Hans - ON8VQ
hansvi
 
Posts: 70
Joined: Wed Apr 24, 2013 3:37 am

Re: Problems with TX

Postby lelazary » Fri May 31, 2013 5:37 am

Thanks again for all the suggestions.

I have looked over all the solder connections, but I could not find anything that looks bad. So I am thinking it might be a bad transformer, but I am trying to find out which one. I originally took this project to learn more about rf and how the signals propagate, so I would like to learn how the circuits works and how to measure the various components to see which one is bad (I prefer to find the bad component, then to just rewind all the transformers). I have also bought a VNWA from sdr-kits.net and will try to use it on the filters to measure their response, but I am still learning on how to use it.

Here are a few things I found out with my oscilloscope.

First, I noticed that under FM I was getting more power (6V pk-pk or 90mW) then on CW or AM. I noticed that under CW, the mic gain will increase the power, as well as increasing the bandwidth of the filter in HDSDR. So all the measurements mentioned bellow are on CW with a wide bandwidth and full mic gain and full power output. I have been using a 100MHz Agilent 54622D oscilloscope, an Agilent 10:1 probe (10074C http://www.home.agilent.com/en/pd-10000 ... =US&lc=eng ) and a 50 ohm dummy load from here (http://www.qrpkits.com/dummyload.html). To measure the final power, I have been connecting the probe to the RF and GND output on the dummy load.

I measured pin 3 on U14 and I am getting 5.75V p2p somewhat square wave at 56.3MHz (pin 6 on U9 Si570 is 3.41V). So it looks like I might need to calibrate the scope (I will need to read more about how to do that). I have always used this scope for digital, and never had the need to calibrate it. So the values bellow might not be correct.


On 10 meter (28MHz) I am getting 6V pk-pk using a 50 ohm dummy load (90mW). I measured the output on T5 pin 1 (the output before the band-pass) and found a 4V pk-pk wave on it. From there I measured T1 pin 4 (after the band-pass) and found a 4.38V pk-pk wave. The output of T1 pin 1 connected to the antenna had the final 6V pk-pk wave on it.

On 15 meter (21MHz) I am measuring a final wave output of 7V pk-pk. However, the output of T5 pin 1 has a wave of 12.12V pk-pk (367mW) but on T1 pin 4 (after the band-pass) I am only getting 8.5V pk-pk (180mW, 3db drop). On T1 pin 1 I was getting 6V pk-pk (90mW another 3db drop).

-What kind of power should I expect at these junctions for these bands (T5 and T1)?
-Since on 15 meter I get a drop of 3db inside the band-pass, does that mean that my band pass is bad for 15 meter?
-Also since I am further getting a 3db drop between the input and output of T1, does that mean that T1 is bad?
-Since it still looks like I am not quite getting over 500mW, should I investigate the push-pull amp? What should I measure to test the amp and what should I get? What should be the power going into the amp?
-The heat sink is getting very hot, but I can still touch it.

Thanks again for all the suggestions. I am trying to both build the kit and learn how it works, so excuse my excessive questions.

Thanks, Lior
lelazary
 
Posts: 10
Joined: Mon May 27, 2013 4:37 pm

Re: Problems with TX

Postby AE9RB » Fri May 31, 2013 8:22 am

When this type of problem happens you have to start pulling parts. I don't know of any way to diagnose with test equipment. A bad transformer is either open or short on half of a balanced circuit. Shorts downstream are generally indistinguishable from opens upstream and vice versa.

Also, you can't validate the BPFs with an oscilloscope based on the output of a faulty class A-B amp.
73 David AE9RB
AE9RB
Site Admin
 
Posts: 511
Joined: Thu Nov 15, 2012 5:27 am
Location: Portland, Oregon, USA (CN85ql)

Re: Problems with TX

Postby lelazary » Fri May 31, 2013 1:21 pm

Thanks for the response.

I am going to try and use my VNWA to test the band-pass filter. I think this can be done easily without removing anything, since the relay is insulating them.
Here is the thought.
Connect the TX port of the VNWA to pin 5 of the relay and RX port of the VNWA to pin 4 of the relay. Since the relay is not energized, pin 4 and pin 5 are not connected to anything. This will allow me to test if one of the band-pass filter is working. I can then energize the relay, and using the same process test the other band-pass.


I would like to determine at least if T3 or T4 and T5 are bad. If I remove C44, do you know what kind power I should expect at that point (for a CW at full power). What kind of power should I expect at the output of the preamp (Q9 collector)?

Thanks,
Lior.
lelazary
 
Posts: 10
Joined: Mon May 27, 2013 4:37 pm

Re: Problems with TX

Postby AE9RB » Fri May 31, 2013 4:20 pm

The next step is to replace the transformers.
73 David AE9RB
AE9RB
Site Admin
 
Posts: 511
Joined: Thu Nov 15, 2012 5:27 am
Location: Portland, Oregon, USA (CN85ql)

Re: Problems with TX

Postby hansvi » Sat Jun 01, 2013 2:34 am

lelazary wrote:Thanks again for all the suggestions.
I measured pin 3 on U14 and I am getting 5.75V p2p somewhat square wave at 56.3MHz (pin 6 on U9 Si570 is 3.41V). So it looks like I might need to calibrate the scope (I will need to read more about how to do that). I have always used this scope for digital, and never had the need to calibrate it. So the values bellow might not be correct.


I don't really understand how you can read such a voltage there. If it's a calibration problem, then I wouldn't expect an error that great (I mean, Agilent?). If it's a probe problem, I would expect an attenuation due to the high frequency, not an amplification.

Then again, I'm self-taught on the field of hardware so I might be missing something utterly basic that I have overlooked all this time. I think I make sense, but I'm always prepared to learn otherwise :-).

Can you measure that voltage on pin 3 with a multimeter, as explained in the manual?

My thought here is that at the very least, you should get meaningful measures that you can rely on. If that condition isn't met, then all the rest is rather moot.

PS: Are you reading that 5.75 from a scope measurement, or directly from the image? Could the 5.75V be from ringing? I.e. if you count the divisions, do you get the proper value?

PPS: Do you have an RF power meter that you can use to measure the TX power? Again, I'm focusing on the reliability of the measurements. Dave is probably already a couple of steps ahead of me, but I don't think it does harm to focus on the measuring itself for a moment.

PPPS: If you really, really, really want to make sure, take a scope measurement of the 5V and 3.3V regulators. What voltage do you measure there? If you're measuring .5V and .33V there, it may have something to do with the 10x attenuation of the probe ;-).

Maybe Dave can chime in and tell me if I'm talking nonsense? (Once again, my confidence in hardware related topics isn't the best, but it seems to be slowly growing - this kit is a great help on that part, btw).
73,
Hans - ON8VQ
hansvi
 
Posts: 70
Joined: Wed Apr 24, 2013 3:37 am

Re: Problems with TX

Postby lelazary » Sat Jun 01, 2013 3:26 pm

It looks like I am getting some low frequency noise on the signal, which makes pin 3 on U14 modulate below GND and above 3.3V. If I measure the actual waveform then I am getting 3.250V p2p. My digital multimeter reads 2.454V on this pin, so its also seeing this low frequency noise (I have checked the multimeter against a precise power supply that I have). I also performed the user calibration on this scope and calibrated the probe with a square wave reference signal.

Here are the readings from the regulators:

5V regulator: scope reads 5.12V (from max internal measurement, looks like some HF noise on the signal), digital multimeter reads 5.030V
3.3V regulator: scope reads 3.28V digital multimeter reads 3.293V
So it looks like the reading is correct from the scope/multimeter and the regulators are putting out the correct voltages (I also checked that the scope is seeing that its a 10X probe).

It looks like I am getting two different wave forms depending if I am in run mode on the scope or a single mode (the run mode continuously triggers while the single mode gets one trigger). The trigger is set to auto mode on edge. In single mode, it looks like is separating the wave forms correctly, while in run mode its combining the waveforms into one deformed waveform. I think this has to do with the sampling rate between the modes, but I am not sure.

I have always been using this scope to measure digital data, and don’t have much experience myself with measuring AC signals. So I am very happy to learn.
I am attaching some images to help explain what is going on. They are all from pin 3 on U14. In the single mode, the images are from the same trigger, but looking at different scales in frequencies. The measurements using the single mode are made manually using the cursors, while in the run mode they are using the scope automatic measurements.

Thanks again for all the help so far.

Lior.
Attachments
U14P3_scope.zip
Scope measurements of U14 pin3.
(244.38 KiB) Downloaded 185 times
lelazary
 
Posts: 10
Joined: Mon May 27, 2013 4:37 pm

Re: Problems with TX

Postby hansvi » Sat Jun 01, 2013 7:01 pm

Hm, with a 100MHz scope and a 150MHz probe, shouldn't you be able to see the square wave (the Si570 outputs a square wave as far as I know)? Maybe you can check the "bandwidth limit" setting of the channel? Although I doubt that's the problem, as it limits the bandwidth to 20MHz and this signal is close to 40MHz. Maybe the sample rate has been reduced?

Unfortunately I'm currently stuck myself, there is a new scope coming my way (a Siglent 1102CNL - at 300€ I'm quite curious about its performance) but at the moment I can't measure anything to verify over here.
73,
Hans - ON8VQ
hansvi
 
Posts: 70
Joined: Wed Apr 24, 2013 3:37 am

Re: Problems with TX

Postby lelazary » Sat Jun 01, 2013 10:25 pm

I know my scope has never been able to see a square wave over 30mhz. Even if i plug the scope to a 50mhz can oscillator i see deformed sine wave. I always thought that you needed at least up to the fifth harmonic to see a square wave. Since the wave is over 50mhz , I would need a 250mhz bandwidth to see a good wave. So my 100mhz scope can only see up to the second harmonic for a 50mhz wave before attenuating the rest of the harmonics.

Let me know if you think I am wrong, and could do something to see the wave.


Thanks,
Lior
lelazary
 
Posts: 10
Joined: Mon May 27, 2013 4:37 pm

Re: Problems with TX

Postby AE9RB » Mon Jun 03, 2013 3:24 am

It's cool if you want to play around with test equipment but I really don't have anything to add to that discussion. If you want to work on fixing the power output then the next step is to replace T4, T5 , and T6 since the problem is likely one of those. There is no practical advice I can give for diagnosing which one using test equipment.
73 David AE9RB
AE9RB
Site Admin
 
Posts: 511
Joined: Thu Nov 15, 2012 5:27 am
Location: Portland, Oregon, USA (CN85ql)

PreviousNext

Return to Peaberry SDR V2

Who is online

Users browsing this forum: No registered users and 2 guests

cron