CRYSTAL FAIL

[ErwinsCat]

I'm working on a device which has no sign of (digital) life. Unfortunately, I have zero clues on how digital systems including CPU, RAM, USB, and all shebang actually operate but as Dave would say "thou Shal check all the voltages", I checked all power rails and all seems to work correctly, after that I probed order places and haven't seen any digital pulse, there is 3 xtal in the circuit, one isolate with USB connection and its seems working correctly, the other ones somehow connected to CPU and the one connected to clock-in isolate but the amplitude is only about 4 mv, is this about the right amplitude or something a miss? and what is the chance of xtal going faulty? is it common or very rare? thanks

[srb1954]

I'm working on a device which has no sign of (digital) life. Unfortunately, I have zero clues on how digital systems including CPU, RAM, USB, and all shebang actually operate but as Dave would say "thou Shal check all the voltages", I checked all power rails and all seems to work correctly, after that I probed order places and haven't seen any digital pulse, there is 3 xtal in the circuit, one isolate with USB connection and its seems working correctly, the other ones somehow connected to CPU and the one connected to clock-in isolate but the amplitude is only about 4 mv, is this about the right amplitude or something a miss? and what is the chance of xtal going faulty? is it common or very rare? thanks

Is the crystal you are probing a 2-pin device i.e. just plain quartz resonator or an integrated crystal oscillator, which will have 4 pins or more?

If it is a 2-pin device there will be an amplifier inside the chip used to form the oscillator together with the crystal. These circuits are usually very sensitive to disturbances and and if you are probing directly on the  crystal the loading of the scope probe will quite often stop the oscillator. You either need to use a high-resistance, low-capacitance active probe to minimise the loading or find another clock output further downstream of the clock oscillator to monitor the signal.

Integrated crystal oscillators are much more tolerant of loading and can be successfully probed with a normal scope probe.

[Wallace Gasiewicz]

4 mV is very small and may just be from something else on the board.
Xtals do fail, I have had a few. but these were old xtals, newer xtals on digital boards do not commonly fail.
Perhaps check the voltages on the pins of the xtals that work and the one that does not work.

srb1954 is right on: Probing with a regular scope probe right on the xtal will change the reactance of the circuit and frequently shut down an xtal oscillator, otherwise it will change the freq.
Sometimes you can check an xtal by putting a loop on a probe (loop probe tip to probe gnd) and putting the loop next to the xtal or oscillator. Check if freq is proper xtal freq. This puts less reactance into the circuit. The 4 pin devices are not just xtals but also contain the oscillator, so you are not directly probing the xtal.

Sometimes chips have some way of turning on an external clock which is your xtal oscillator, sometimes the oscillator is in the chip and it uses the ext xtal and sometimes the entire oscillator and xtal are external to the chip.

Perhaps you could identify your circuit or post pics.

[twospoons]

You can use a standard scope probe on a crystal if you put a 1p capacitor in series with the probe tip. You just have to remember that you are creating a capacitive divider in combination with the probe so your scope measurement will be lower eg 10p probe + 1p series cap = 11:1 divider.
The additional loading seen by the crystal oscillator will be <1p, which should not stop oscillation.

[bdunham7]

I'm working on a device which has no sign of (digital) life. Unfortunately, I have zero clues on how digital systems including CPU, RAM, USB, and all shebang actually operate but as Dave would say "thou Shal check all the voltages", I checked all power rails and all seems to work correctly, after that I probed order places and haven't seen any digital pulse, there is 3 xtal in the circuit, one isolate with USB connection and its seems working correctly, the other ones somehow connected to CPU and the one connected to clock-in isolate but the amplitude is only about 4 mv, is this about the right amplitude or something a miss? and what is the chance of xtal going faulty? is it common or very rare? thanks

It's not unheard of for a crystal to go bad.  What is the device and what are the specifics of the crystals? 

[ErwinsCat]

it is a guitar effects board (line6 podhd400). it wasn't the xtal but I learn about the measuring and xtal oscillators, thanks for the replies anyway.
I realized the voltage on the bios chip was 1.2 instead of 3.3 and there was some sort of corrosion on the ic pins so I re-flow the ic and it starts booting up.
after that, the signal issue shows up with the input from the guitar preamp was choppy and fading and cutting off. realized the issue was the leaked out audio input caps which weren't conducting and also messed up with ic pins creating conductive green salt between opamp pins. my question is how do input caps without much voltage and not much heat go bad so often as tuned out this was a known issue with the preamp section?

 https://i.ibb.co/0rBXBJP/PXL-20220610-134135998.jpg

lin6 support forums:
https://line6.com/support/topic/33436-pod-hd-300-problem/page/2/

[twospoons]

Its possible they are just crappy capacitors - some no-name junk brand used to keep costs down.  Or there may be a topology fault that results in the capacitor getting a reverse bias sometimes.