The ES9038Pro DAC Nightmare (Did My Dedicated Oscillator Mod Kill my DAC? NLOCK)

[mansr]

3 hours ago, HallmanLabs said:

I was told that NLOCK means the DAC can't lock onto a signal and this points at the PLL (Phase Locked Loop). My buddy who said this, told me he thinks the PLL circuit is embedded in the ES9038, so to fix it I would need a new IC. Can anyone confirm either part of his statement?

ESS DAC chips use a PLL so the system clock doesn't need to be synchronised to the bit clock. The system clock does, however, need to be at least 192x the sample rate with a maximum of 100 MHz. If the system clock is too low, failure to lock onto the input is a likely outcome.

[mansr]

14 minutes ago, HallmanLabs said:

I'll be using the Digilent Analog Discovery2 first as that is the easiest to capture images/video with.

That scope has a bandwidth of 30 MHz. You'll need something better if you want to capture the digital signals here.

 

Quote

It's going to be tough, probing a chip this tiny. I will probably just put one of the breadboard type jumper wires in the probe's mouth/clip and then just hold that one wire instead of the entire probe going in there. Shorting a pin is the most dangerous thing about doing this type of diagnostics.

Briefly shorting two adjacent pins on the DAC is unlikely to damage it. The worst case would be shorting an output pin to ground or a supply rail. The pinout makes this difficult, and it can probably withstand the condition briefly should it nonetheless arise.

[mansr]

55 minutes ago, HallmanLabs said:

Yeah, that is exactly what I already said on both the website and post here in the OP was possible that I sent the power pin to the output pin with the oscillator. Even after I said, nobody said this was for sure even dangerous to the ES9038. ESS hasn't given out the real datasheet on the IC yet, I asked them already. It is possible this damaged it with 3.3V vs. 0.5V or possible, it did not.

All input pins will withstand a voltage no higher than the supply voltage to the chip, if datasheets for other ESS chips are any indication. Unfortunately, I don't have the 9038 datasheet.

[mansr]

56 minutes ago, pkane2001 said:

I was thinking more of the little SMD components, like resistors and capacitors that may not have survived the larger voltage/current you may have sent through them.

Those things are surprisingly resilient, at least for short periods of torture. If nothing is obviously blackened, it's probably OK.

[mansr]

8 hours ago, HallmanLabs said:

ESS sent me two docs when I requested all ES9038 documentation and one was a glorified sales ad/paper. The other was just the basic 64 pins assignments. By this point I expected the ES9038 whitesheet or atleast the full datasheet.

You won't get it that easily. They'll only supply it to customers. Maybe buying a sample chip is enough. The ESS datasheets I have were obtained through unofficial channels.

 

8 hours ago, HallmanLabs said:

If there is little danger with short exposure to non-ideal conditions, then this thing might come back to life with the right troubleshooting.

Have you tried restoring it to a once working configuration?

[mansr]

3 minutes ago, HallmanLabs said:

The only odd thing I noticed was the intensity of the sine wave (how "bright" the waveform is) was lower than usual, but still easily visible.

The trace brightness on an analogue scope reduces as the horizontal sweep rate increases.

[mansr]

6 minutes ago, HallmanLabs said:

Yeah, right around 0.5V

 

Does this lead us back in the direction of the PLL circuit then? I don't think I ever heard anyone say for sure where the PLL circuit is. If it is on the ES9038 itself or not.

The PLL is part of the ES9038 chip. The "no lock" indication just means the chip fails to assert the lock signal. This doesn't mean the damage is necessarily to the PLL circuit. Not that it matters. Whatever part is broken, it's useless.

 

6 minutes ago, HallmanLabs said:

If this does come down to a ruined ES9038, I will just stick my spare ES9028Pro on this board.

Are they pin compatible?