Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › How to fix 60 cycle hum in a DAC?
New Posts  All Forums:Forum Nav:

How to fix 60 cycle hum in a DAC?

post #1 of 25
Thread Starter 

Greetings,

 

This isn't really a DIY question, but I thought this would be the most knowledgeable group for this question.

 

I have two Frontier Tango 24 DA/AD Convertors that I want to use for mixing audio with analog summing.  I did a bunch of measurements on them using RMAA, and they are excellent performers overall, with about 105 dB of dynamic range, flat frequency response, and 0.002% THD.

 

They have 1 problem though.  They have 60 cycle hum at -98 dBFS on the DAC side.  They use an external AC/AC power supply, 10V @ 4 Amps, line lump style.  I'm thinking that the 60 cycle hum has to be coming from the power supply.

 

Does anyone know of a 10V AC power supply that has excellent performance?  Budget is maybe $100.  If it could run two DACs (8 amps I guess), that would be even better.

 

Also, will using a better power supply fix the problem?  Or is it inherent in the design of the DAC?  Is there a simple mod that may fix it?

 

EDIT: The 60 cycle hum also has harmonics.

 

Thanks,

 

Harley.


Edited by barleyguy - 12/10/12 at 7:22pm
post #2 of 25

It has to be a problem with the DAC, as you are feeding it AC power, you can't do anything to get rid of the 60hz because thats what AC is

 

Your best bet would be look at the dac's schematics and see if you can improve the filtering

 

cheers

FRED

post #3 of 25
Thread Starter 

Thanks.  I'm wondering if changing the filter design isn't going to do any good, because I may not do better even if I mess with it.  Here's the graph from RMAA:

 

700

 

It's not audible as a single track, but I want to do multitrack recording to an analog SSL mixer with this, and I'm afraid if I mix 16 tracks together, the hum is eventually going to become audible.  Maybe it's not a problem.

post #4 of 25

If it takes 10 V AC, chances are the DAC has a bridge rectifier as the first stage. If that's the case, then feeding it with DC should not be a problem. You just have to study the power circuit to see if that's the case. Because if it uses a switching PSU, I don't know if you could feed it DC... You could potentially destroy it by feeding it DC.

 

But if it's a vanilla linear supply, then you could probably just get 12 V DC power supplies to replace the AC ones.

post #5 of 25
Quote:
Originally Posted by KimLaroux View Post

If it takes 10 V AC, chances are the DAC has a bridge rectifier as the first stage. If that's the case, then feeding it with DC should not be a problem. You just have to study the power circuit to see if that's the case. Because if it uses a switching PSU, I don't know if you could feed it DC... You could potentially destroy it by feeding it DC.

 

But if it's a vanilla linear supply, then you could probably just get 12 V DC power supplies to replace the AC ones.

Not if it uses half-wave rectification to get a +/- supply though. Worth finding out though, as it would likely be a good solution if it can be used.


Edited by DingoSmuggler - 12/11/12 at 6:09pm
post #6 of 25

Ah... yeah I didn't think about this. It would explain why it's fed AC and not DC. 

 

I thought about it as a single-ended device, so I forgot about this. It makes reverse-engineering the power supply more complicated, as there's many ways to create a dual supply. It could still use a full wave rectifier and then create a virtual ground. Though if it used a full wave, the hum would be 120 Hz. Or at least, the 120Hz element would be louder than the 60 Hz one.

post #7 of 25
Thread Starter 
Quote:
Originally Posted by KimLaroux View Post

Ah... yeah I didn't think about this. It would explain why it's fed AC and not DC. 

 

I thought about it as a single-ended device, so I forgot about this. It makes reverse-engineering the power supply more complicated, as there's many ways to create a dual supply. It could still use a full wave rectifier and then create a virtual ground. Though if it used a full wave, the hum would be 120 Hz. Or at least, the 120Hz element would be louder than the 60 Hz one.

 

It's a pro-audio DAC, and from looking at it, it's balanced all the way through.  I think the specs even state "electronically balanced".

 

If you look at the graph above, the 120 Hz is softer than the 60 Hz tone, and the 180 is higher than the 120.

 

Would posting a picture of the board help?

 

Thanks,

 

Harley.


Edited by barleyguy - 12/12/12 at 1:11pm
post #8 of 25

It would help. But if you don't have the skills to figure out the topology of the power supply, maybe you should just forget about it. I don't want to be responsible for damage caused to your equipment. What I said still stands as a possibility, but not facts. If you doubt, have it verified by a professional. (which I am not, by the way)

 

Balanced operation does not mean it has a bipolar power supply. Those are two different things, and you can have one without the other. What's more, having one or both of these features does not disprove what I theorized. It may still be possible, even if your equipment is balanced and uses bipolar supplies, to feed it DC instead of AC. Just don't do it before making sure it can. And don't blame me if you get magic smoke. I warned you.

post #9 of 25
Quote:
Originally Posted by barleyguy View Post

They have 60 cycle hum at -98 dBFS on the DAC side.  They use an external AC/AC power supply, 10V @ 4 Amps, line lump style.  I'm thinking that the 60 cycle hum has to be coming from the power supply.

 

98 dB down is quite a long way down. Below the noise floor of a CD long way down.

 

It's rare in a domestic environment that there isn't a certain amount of mains hum picked up from the surrounding wiring by most any piece of equipment. Depending on the exact nature of your surroundings and installation you may not get an improvement by changing your power supply. It's not absolutely clear exactly how it's getting into the system, you'd have to cross test or take special precautions to be sure that it's on the DAC output.

 

I'd just push ahead with what you want to do and not worry too much about these numbers unless the hum actually turns out to be audible in the finished product. Depending on the exact process the hum may not be correlated (may cancel).

 

w

post #10 of 25

Here's my first thoughts and suggestions...

 

-How close are the transformers to the DAC's?  Try moving them further away.

 

-Are you using shielded cables for your interconnects?  Are the shields connected on both ends?  If they are, try placing a ground lift on the output of the sources (output from the dac, output of the source feeding adc).

 

-Are these balanced or unbalanced connections?

 

-What is the case of gear made of?  metal/plastic?

 

-Where are they located in relation to other pieces of gear?

 

-Are there power cords for other gear laying on top of the dac's, or running parallel to your signal cables?

 

-Is RMAA running on a desktop or a laptop?  If it's a laptop, have you tried unplugging the power supply before doing these tests?

 

-Are you using a power conditioner?

 

There could be many ways 60Hz can find its way into your signal path, I'd try to eliminate all outside possibilities before attempting to rebuild a piece of gear.

post #11 of 25

+1 on the last two posts. There are a lot of other ways the hum can get to the signal besides the PSU itself. It may just be a problem of cable management or the placement of the different equipment relative to each other.

 

I just have a tendency to complicate problems... to give me reasons to hack stuff. rolleyes.gif

post #12 of 25
Thread Starter 
Quote:
Originally Posted by wakibaki View Post

 

98 dB down is quite a long way down. Below the noise floor of a CD long way down.

 

It's rare in a domestic environment that there isn't a certain amount of mains hum picked up from the surrounding wiring by most any piece of equipment. Depending on the exact nature of your surroundings and installation you may not get an improvement by changing your power supply. It's not absolutely clear exactly how it's getting into the system, you'd have to cross test or take special precautions to be sure that it's on the DAC output.

 

I'd just push ahead with what you want to do and not worry too much about these numbers unless the hum actually turns out to be audible in the finished product. Depending on the exact process the hum may not be correlated (may cancel).

 

w

 

Quote:
Originally Posted by samsquanch View Post

Here's my first thoughts and suggestions...

 

-How close are the transformers to the DAC's?  Try moving them further away.

 

-Are you using shielded cables for your interconnects?  Are the shields connected on both ends?  If they are, try placing a ground lift on the output of the sources (output from the dac, output of the source feeding adc).

 

-Are these balanced or unbalanced connections?

 

-What is the case of gear made of?  metal/plastic?

 

-Where are they located in relation to other pieces of gear?

 

-Are there power cords for other gear laying on top of the dac's, or running parallel to your signal cables?

 

-Is RMAA running on a desktop or a laptop?  If it's a laptop, have you tried unplugging the power supply before doing these tests?

 

-Are you using a power conditioner?

 

There could be many ways 60Hz can find its way into your signal path, I'd try to eliminate all outside possibilities before attempting to rebuild a piece of gear.

 

A couple of answers to the questions posed above:

 

- I am running RMAA on a desktop.  The sound card is a ProFire 2626, connected via firewire to the computer, and then optical to the DAC.  I also have a laptop with Firewire, and could test that way as well.  I don't suspect that it will make a difference though, since the DAC is connected with optical and the problem is isolated to the outputs.

 

- There is no hum in the graphs for the input.  It is only in the DAC output of both Tangos.  (I have run RMAA on the just the input and just the output, using the built in ProFire IO for the other side.) The level varies slightly depending on the port.  It varies from -98 to -102 for the 60 hz peak.

 

- I am not using a power conditioner.  I'm using a surge suppressor, connected to a thick 10-foot extension cord and then the wall.  I did try plugging the Tango power supply into another outlet and placing it in a different spot, and the results were roughly the same.  I can try a power conditioner as well.  I have one on the AV system in my living room that I can temporarily move.

 

- The case of the DAC is made of metal.  It's an ungrounded power supply though.  Should I try grounding the case?

 

Overall, as noted above, I won't know if the hum is even a tangible problem until I've tried summing multiple channels from the DACs.

 

Thanks,

 

Harley.

post #13 of 25

I doubt a power conditioner would change anything. These things are designed to filter noise out of AC. The problem here is the AC power, not the noise it contains.

 

And yes, an ungrounded metal enclose is pretty useless as a shield. Grounding it would make a difference in blocking EMI, but unless the enclosure is made of ferrous metal (as in, a magnet will stick to it), it won't shield magnetic fields. 60 Hz hum are usually magnetic fields induced from power transformer or mains wiring.

 

Though if the metal enclosure is electrically connected to the signal ground, grounding it to earth may make things worse.

post #14 of 25

OK, from re-reading your original post, the DA and AD in each unit share the same PSU, is that correct?

 

That would mean that the PSRR of the DA is poorer than that of the AD, if the cause is ripple on the PSU.

 

So you have to wonder why that would be the case... and take a look at the schematic and the PCB to see if there's anything that might provide an explanation, or that might be down on performance (electrolytic cap?) that you might reasonably replace in the hope that it would improve the situation.

 

w

post #15 of 25

I'm starting to get confused about your set up, you say you're connecting the profire and tango through an optical cable, so how are you measuring the analog outputs of the tango?  Are you looping them into it's own inputs?  Can I see a picture of your setup?

New Posts  All Forums:Forum Nav:
  Return Home
Head-Fi.org › Forums › Misc.-Category Forums › DIY (Do-It-Yourself) Discussions › How to fix 60 cycle hum in a DAC?