Vali DC offset issue

On the topic of measurements, here's a PSA:

Absolutely do not leave your headphones plugged in when you turn the Vali 2+ on. In fact, I wouldn't even plug them in for several minutes.

My new Vali 2+ developed a static scratch sound when adjusting the volume. Schiit wanted me to send it in and they replaced the volume pot. However, the problem still remained in my repaired unit. So, since DC on the pot can cause a scratchy sound, I tested my Vali 2+ quite extensively.

Upon turn on, and even after the mute relay has disengaged, the Vali 2+ throws over 3 v (not 3 mV, but 3 whole volts) of DC offset regardless of tube used. It takes about 10 minutes to get the DC offset under 2 mV. Since my Vali 2+ was serviced and tested, I am to assume mine isn't defective and this is how the circuit works.

 

I fired off an email this morning. We'll see what they say. I'm not the slightest bit concerned about the persistent 1-2 mV DC being fed to the headphones, but the brief surges of 3 v sure make me a little nervous.

 

Is that offset value only when completely unloaded? Try hooking up a resistor on the output that matches your headphone impedance and see what the offset peaks at. With a path to ground, you might find that the peak offset is much lower.

 

This is unloaded. With the volume all the way down, I connected the meter to a headphone cable. When I first powered up, there was no DC. As soon as the muting relay disengaged, it would spike to 50 mV and then surge as high as 3 V. I don't have any resisters, but I emailed Schiit as a follow up.

 

It's already been serviced once, so I'm assuming whatever values they tested may be normal. We'll see what they say.

I tested again just a few minutes ago without any sources connected or active. This was unloaded. Before the muting relay disengaged, it was 0.0 mV. As soon as it disengaged, it went from 50 mV to 1.8 V in a matter of seconds, went down to 1 v in a few seconds, and stayed about 20 mV for several minutes. I hope I didn't damage my headphones from this spike of DC in the range of volts. It's a beautiful sounding amp and I've love to use it without this worry.

 

OK. Just for context/refetence, my Crack will spike to 12VDC unloaded, but only about 1V loaded with HD650. That's not headphone frying levels but it's still a lot, so I do disconnect on turn on/off just to be safe.

Long story short, I'd bet the Schiit is a lot lower than it seems when that voltage can bleed through the headphones. But it might also be a fault, so definitely still wait to see what they say.

[EDIT] In response to your new post, most headphones have power ratings of 1W. Let's use half that for a safety factor. 300R would pull 0.5W at 12.2V. For 30R it's 3.9V. You won't have damaged anything.

 

Thank you, that's good to hear. I already had my LCD-2 repaired once due to their infamous membrane tears. They sound glorious and I wanted to get several good years out of them.

 

Schiit is having me send the amp in again. It wasn't specified if they will be testing/repairing it, or if they'll send me a new one off the shelf.

 

My unit also produces a loud thump when the relays kick in, FWIW. Will try to measure it when I get the chance (not anytime soon though).

 

I'm aware of the DC offset on the Bottlehead Crack, but this should not be an issue for the Vali. Especially with a relay.

There is definitely something going on and you should send it back.

@Beefy - I recall on the Bottlehead forums someone mentioned a modification where you solder a resistor somewhere on the neutrik plug to ground and this reduces the voltage a lot faster.

 

The default Crack build/schematic has a (from memory) 2800R resistor from each output to ground at the headphone jack. Even with this, you still get the 8-12V spike on turn on. An unavoidable consequence of the time it takes the output capacitors to charge after the output tube starts conducting.

A builder can use a smaller resistor to bleed the voltage faster, but you dont really want to do that because then it begins to impact utput voltage swing.

 

Last night I recorded a video of a DC surge of nearly 5 v (!!!) immediately after the muting relay disengages when using the stock Canada tube. I sent it to Schiit for them to evaluate.

 

Good suggestion. I am including the stock tube when I send it in. However, it seems like no matter which tube I use, whether lower mu or higher mu, or 300 mA or 600 mA heater, they all throw off at least 2-3+ volts DC after the rely disengages. Also, it seems the scratchy pot happens no matter which tube I use. Interestingly, the position on the pot where the scratch happens changes with tubes as well.

 

Source connected or not makes no difference in the DC. I have no idea what's going on that is causing these issues.

 

Final update to the story:

I sent back the Vali again to Schiit, and they must have agreed with my assessment of something not being right because they sent me a new one off the shelf. To further confirm there was something wrong, the new one sounds a bit different. It has more bass, more overall detail, and it stages differently. I didn't dislike the sound of the old one, but the new one is better.