Is measuring square wave on spdif cable possible?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Jump to Latest
G

Gobble

Member
Joined 2010
Greetings,
I plugged this device into a friend's oscilloscope when it stopped working some time last year. I am not into electronics theory, just handy with a solder gun and some diy kit building. So I don't know what to make of this spdif signal. I just wanted to see if there was any signal at all when it failed to work.
Can you folks help me understand what is wrong with that spdif output in the attached pics?

An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.


Thanks
G0bble


Sent from my GT-N7000 using Tapatalk
 
S

stratus46

Member
Joined 2009
I work in network broadcast video. The serial transmission rarely looks as good as your scope display. You have a slight equalization / termination problem - nothing needs to be done about it.

I've been reviving std def Sony Digital BetaCam machines and the serial digital video output NEVER EVER looks that good. OTOH, the data transmission has no problems at all. The AES output (3 megabit vs 270 megabit video) looks a lot like your scope photo.

G²
 
G

Gobble

Member
Joined 2010
Thanks guys. So that signal fed into a DAC should have made music? I wonder why the sound stopped flowing then. The device has since been replaced. I wonder if it was a loose contact in my Dac power socket/plug then- it doesn't have a power l.e.d to tell...

The spike on they leading edge is likely the reflection from the connector of the scope I guess.

Marcie, if you can elaborate a bit more about scope probe loading, I would appreciate it.



Sent from my GT-N7000 using Tapatalk
 
peufeu

peufeu

Member
Joined 2001
audio.peufeu.com
Scope probe and cable has capacitance which smooths out the signal... also scope has limited bandwidth... this turns nasty-looking spikes into innocent looking bumps.

Also scope probe capacitance can ring with the circuit, make it crash, or just eat all the signal if you probe a very high impedance node... Always use 10x mode...

Example : probe on reset lead of microcontroller. System works, nothing suspicious on scope. Probe removed, microcontroller crashes. Turns out reset pin was not connected and probe capacitance filtered out the noise just enough for it to look like it worked...

> the waveshape is fine (more than fine, really), but that doesn't mean that the transmitted code isn't garbled.

Well, if the chip inside the CDP is transmitting stuff that looks like SPDIF, chances are that it works.

You can try playing a CD. You'll see the bits move, use a slower timebase on the scope so the bits are a few mm wide on screen. If the bits don't move, CDP has a problem. Maybe MUTE or digital volume control set to 0 ? or broken I2S data line, etc...

Or just connect it to the SPDIF input of your PC and check if it works.
 
Last edited:
M

marce

Member
Joined 2007
peufeu said:
Example : probe on reset lead of microcontroller. System works, nothing suspicious on scope. Probe removed, microcontroller crashes. Turns out reset pin was not connected and probe capacitance filtered out the noise just enough for it to look like it worked...


.
Click to expand...

Same, with a JTAG interface, it was great fun watching the engineer ripping his hair out. Scope on, circuit works, scope off, circuit stops working!!! Engineer starts steaming:D Best Lab day ever. It was the JTAG clock distribution to five devices, daisy chain didn't work so used a clock driver with multiple outputs and controlled each line.

A bit of info, all the scope probe manufacturers have papers and guides on modelling their probes and loading effect, so you can add them to simulations.
http://www.dfad.com.au/links/THE SECRET WORLD OF PROBES OCt09.pdf
 
sesebe

sesebe

Member
Joined 2011
Hello,

I still waiting an explanation about how jitter on digital signal influence the reconstructed audio signal. You can consider the jitter in any place you whant from ADC output until DAC imput (SPDIF, optical, I2S...).

I need clear and well documented explanation and no explanation based on feeling because based on my feelig there are no diference untill the jitter it is exagerated high.
 
D

DF96

R.I.P.
Joined 2007
myweb.tiscali.co.uk/g8hqp/
sesebe said:
I need clear and well documented explanation and no explanation based on feeling because based on my feelig there are no diference untill the jitter it is exagerated high.
Click to expand...
Fortunately in engineering our feelings don't count.

The only jitter which actually matters is jitter at the DAC output, but some of this can arise from jitter earlier in the chain. Jitter means the timing is wrong. If the timing is wrong then the reproduced waveform coming out of the reconstruction filter will not be identical to that coming out of the anti-aliasing filter before the ADC.

How much jitter is audible is something people argue about. It depends on the statistics of the jitter, and whether it is signal-correlated or not. It would be nice to think that the argument is caused by lack of real experimental data, but I suspect that no amount of data would satisfy those whose livelihood (or ego) depends on them claiming sensitivity to tiny amounts of jitter or tolerance of large amounts of jitter.

Some of the things DIYers do to reduce jitter (e.g. add external clocks, play with DIY cables) are quite likely to increase jitter. If they genuinely hear a change then there is no guarantee that the change is actually for the better, but the DIYer is satisfied.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Top Bottom