ok. I am using a scope for levels although I tried a meter in earlier tests. I've been round the board looking for bad joints or bridges and can't find any. I have access to spectrum analyser so will try that at the weekend to look at lo and see if there is any modulation. Anybody got any plots of what it should look like on the ouput end? I know roughly what shape I'm after but any idea of levels. Is the CQ TV article out there yet.
Andrew
Digilite Jumpers
Re: Digilite Jumpers
http://www.m0dts.co.uk/datv_old.htm has some spectrum analyser screen shots. The top 2 are the lower and upper sideband tests. There will be more lines visible on a wider plot, due to harmonics. Bottom left is No Mod and bottom right is normal data. The pseudo random output from the serialiser doesn't give a very good pattern, so it's best giving it real data from the PC for the normal data test.
The X-Y plot lower down is the constellation. QPSK has 4 phases, so the 4 states are at the corners. The better the filtering, the tighter the convergence at the corners. The two scope probes need to be on the top ends of L2 and L4 for the constellation.
I took my DigiLite to G3UVR to have a look at it on his analyser. I don't have a MAR-6 on it yet as I wanted to see the barefoot output from the AD8346. The data sheet suggests 100uW, but I was only getting 10uW. This would give 1mW or so after the MAR-6.
I could get 35-40dB of attenuation on the side bands and local oscillator, which is about right. I measured the levels with a scope afterwards at the vias I mentioned in an earlier post and the dc level was about 1.1v and the pk-pk on the In Phase test was 750mV.
I sent you a PM, but it seems to have got stuck. I'll try it again.
Brian
The X-Y plot lower down is the constellation. QPSK has 4 phases, so the 4 states are at the corners. The better the filtering, the tighter the convergence at the corners. The two scope probes need to be on the top ends of L2 and L4 for the constellation.
I took my DigiLite to G3UVR to have a look at it on his analyser. I don't have a MAR-6 on it yet as I wanted to see the barefoot output from the AD8346. The data sheet suggests 100uW, but I was only getting 10uW. This would give 1mW or so after the MAR-6.
I could get 35-40dB of attenuation on the side bands and local oscillator, which is about right. I measured the levels with a scope afterwards at the vias I mentioned in an earlier post and the dc level was about 1.1v and the pk-pk on the In Phase test was 750mV.
I sent you a PM, but it seems to have got stuck. I'll try it again.
Brian
Re: Digilite Jumpers
i am having a look with a scope. with the inphase test I still see a data stream round the board, not a sine wave. I'm still stuck and not rreally getting anywhere just now. could really do with more measurements round the board or scope pictures to give me a reference. Sorry to be so thick and demanding! maybe this is all in the cq article? Just on my previous message am I looking 0.4V or 4V??
Thanks
A
Thanks
A
Re: Digilite Jumpers
MM0MGB, did you get my PM?
Can you email me please: my callsign @mem.batc.org.uk
Brian
Can you email me please: my callsign @mem.batc.org.uk
Brian
Re: Digilite Jumpers
Hi Brian,
Yes got the PM once I found out how to get it and have replied. I think my issue may be poor LO suffering from phase noise,
Andrew
Yes got the PM once I found out how to get it and have replied. I think my issue may be poor LO suffering from phase noise,
Andrew
Re: Digilite Jumpers
I have been involved in getting a couple of boards running locally for guys without Spectrum Analysers . In all cases when the dc levels appear to be correct and wiring/soldering errors have been eliminated, the fault is TOO MUCH LOCAL OSCILLATOR signal, sometimes induced into the board from the poorly screened local oscillator can and cable. Keep it short, keep it grounded and seal the screening can on the oscillator as if it was gas rather than signal that was trying to escape.
If you are not using the Ultram VCO board make sure that the phase noise figure is good enough for digital modulation.
If you are using the Ultram, take the local oscillator output from the input to the MMIC on the VCO board.
There is sufficient signal from the VCO chip to operate the modulator chip via the balun but may need the three attenuator resistors on the input of DigiLite board removed and R27 (82ohm) replaced with a 0 ohm resistor to couple the local oscillator signal to the balun.
Hook the output of the DigiLite board to the aerial socket of the satellite receiver (watch the dc line to the lnb!) and aim to get at least 60% signal level.
If you wish you might try in-line attenuators on the local oscillator feed. Dont use homemade attenuator boards for the reasons given above.Here is my preferred method of inital setup:
Set all four pots to mid-way.
Gently and slowly edge one of the 'level' pots watching the satellite box info display. It takes a moment or two for the quality reading to kick in.Once it has, move to the other pot and aim to raise the level of the quality reading until the level drops back down again.
If no Quality reading leave the 'level' pots at centre travel and adjust the two bias pots a small nudge clockwise and repeat the settings described above of the level pots.
The dc readings on the level pots are also affected by the dc coming out of the 74AC574 IC1. Check the 3v line is within tolerance. Check voltages on the output pins of IC1.Always read IC voltages on the IC pins not on nearby components.
Good luck,
Dave G8AJN
If you are not using the Ultram VCO board make sure that the phase noise figure is good enough for digital modulation.
If you are using the Ultram, take the local oscillator output from the input to the MMIC on the VCO board.
There is sufficient signal from the VCO chip to operate the modulator chip via the balun but may need the three attenuator resistors on the input of DigiLite board removed and R27 (82ohm) replaced with a 0 ohm resistor to couple the local oscillator signal to the balun.
Hook the output of the DigiLite board to the aerial socket of the satellite receiver (watch the dc line to the lnb!) and aim to get at least 60% signal level.
If you wish you might try in-line attenuators on the local oscillator feed. Dont use homemade attenuator boards for the reasons given above.Here is my preferred method of inital setup:
Set all four pots to mid-way.
Gently and slowly edge one of the 'level' pots watching the satellite box info display. It takes a moment or two for the quality reading to kick in.Once it has, move to the other pot and aim to raise the level of the quality reading until the level drops back down again.
If no Quality reading leave the 'level' pots at centre travel and adjust the two bias pots a small nudge clockwise and repeat the settings described above of the level pots.
The dc readings on the level pots are also affected by the dc coming out of the 74AC574 IC1. Check the 3v line is within tolerance. Check voltages on the output pins of IC1.Always read IC voltages on the IC pins not on nearby components.
Good luck,
Dave G8AJN