Radio Killed The Studio Mix

[sr1200]

Make sure the screws that connect the XLR shells to the chassis are tight and make good metal-to-metal contact. Most "modern" XLR's have a chassis ground connection via the mounting screws, assuming yours do, you want to make sure that's a solid connection. ALL screws loosen up after assembly.

XLR grounds are soldered directly to the board. The xlr connectors are held in with small pointy screws through the chassis and plastic of the connector. (2 per XLR, there 8 total).

 

[Ike Zimbel]

XLR grounds are soldered directly to the board. The xlr connectors are held in with small pointy screws through the chassis and plastic of the connector. (2 per XLR, there 8 total).

If you look carefully, you should (hopefully) see the glint of a couple of pieces of metal inside one of the two holes. That's supposed to be the chassis connection. Optimistic? You bet! Ideal? Far from it! Enough to keep 10kw of samba out of your device? Maybe not...
The only other thing I can suggest at this point would be to solder some flying leads to the Pin-1's on the XLR cables, run those out through the strain relief (same path as the cable) and then try screwing those to the chassis via the XLR screws...or maybe connecting the Pin-1 and shell contacts inside the ends of the XLR's that mate with the RTA. The idea being to ground the cable shields firmly to the chassis so the RF goes AROUND your sensitive electronics and not through them.

 

[ccaudle]

The only other thing I can suggest at this point would be to solder some flying leads to the Pin-1's on the XLR cables

Assuming chassis is reasonably conductive I would cut the pin 1 contacts near the pcb surface, bend back and solder to the connector shell or chassis tab, and find someplace else to reference circuit ground node to chassis.
That is assuming suspicions of pin 1 problem can be verified. Seems weird that the designers would be knowledgeable enough to add RF filters to input and output but connect pin 1 into circuit reference node, so would be nice to get confirmation of the layout.

 

[radeng]

The station's engineer may be bilingual, or more likely is a contractor that usually isn't at the station. You could maybe find out by asking around at the local SBE chapter meeting about who is taking care of the RF end of things there. Most engineers will try to help with interference issues caused by their station. They don't want a FCC complaint.

 

[sr1200]

So a faraday cage does the job!!
If the mesh on the back is making contact with a ground or xlr housing im silent. If that disconnects, the signal comes back.
If anyone finds themself in this situation, im using some art copper mesh. Very inexpensive and just plain copper (it comes as a sleeve). Thankfully i was able to cut the front out without breaking the effectiveness of the cage which leads me to believe that theres something towards the back thats the culprit. In any case, a solution that works for my situation.

 

[ccaudle]

Is the case actually conductive? Is there paint in the areas where the case seams overlap? Looks like a metal case, but obviously not doing a good job of shielding.

 

[sr1200]

Is the case actually conductive? Is there paint in the areas where the case seams overlap? Looks like a metal case, but obviously not doing a good job of shielding.

case is painted all the way around. There is a 'star' ground inside that is connected to the chassis. (i say "star" since theres only 2 wires connected to it)

 

[ccaudle]

case is painted all the way around

Meaning even along the joints where the top cover joins to the rest of the case?

There is a 'star' ground inside that is connected to the chassis.

Did they at least leave an unpainted spot there? I could not find internal pictures of DriveRack Studio, but I found pictures of DriveRack Venue, and the case on that model was not painted on the inside, so the internal chassis connection point for the wire looked good. It looked like the inside of the chassis was masked, but not the edges, so the contact point between the top cover and the bottom part of the chassis would have an insulating layer of paint. Is DriveRack Studio the same way? I would assume a lot of the chassis parts were reused between the different DriveRack variants.

 

[sr1200]

Meaning even along the joints where the top cover joins to the rest of the case?



Did they at least leave an unpainted spot there? I could not find internal pictures of DriveRack Studio, but I found pictures of DriveRack Venue, and the case on that model was not painted on the inside, so the internal chassis connection point for the wire looked good. It looked like the inside of the chassis was masked, but not the edges, so the contact point between the top cover and the bottom part of the chassis would have an insulating layer of paint. Is DriveRack Studio the same way? I would assume a lot of the chassis parts were reused between the different DriveRack variants.

yeah the inside is bare metal, but the surround on the case itself (under the lid) is all black, i guess i should check to see if theres continuity... that could be an issue.

 

[sodderboy]

1. is the apogee DAC wall wart powered? Any warts and proper grounding is the first place I would look. Something like a mini-me might want a better chassis connection to earth than mic cable shields (in this specific, new RFI situation)
2. Check the settings tree in the DRack, there could be a setting in there messing with pinout, polarity, impedance, *, etc. The box is deep. RF filter? Those joints are used in very RFI intense places.
3. It's clumsy with monitor eqpt not on patch, but you can use known good mic cables to connect the DAC-dbx into a different amp or powered monitors, try analog into the dbx or try the DAC-dbx chain in any other line level chain for the same RFI. I carry 600 and 2k8 ohm inline xlr slugs to change line impedance when things don't play well together. You can try it with a mic cable. Your runs are prolly short, so this should be less of an issue. I am assuming that all equipment is balanced. That said, does unbalancing the amp input change things?

The male operatic, 0:35 on this track is RFI that was recorded during tracking and flown-in from 1/4".
Mike
* perhaps it has a setting where it can feed a stereo amp with normal cables to use as a mono amp, sending hot to the LHS and cold to the RHS. We used to have to do this manually with 80's Crowns and Brystons. Just thinkin'

 

[ccaudle]

Any warts and proper grounding is the first place I would look.

Sorry, have to point out that incorrect statement. RF shielding in airplanes and automobiles works perfectly well, there is no "proper" grounding required.
Third wire protective conductor in power connections is for electrocution prevention with conductive chassis, and connection of protective conductor and neutral to earth is for lightning safety in buildings. The requirement for RF protection is a conductive covering which is continuous around all circuitry which needs to be protected. Gaps in the continuous covering can cause problems, but the wavelength of AM radio is so long that the small gaps you get from things like XLR connectors are not a problem (although can potentially be a problem at much higher frequencies).

 

[ccaudle]

the surround on the case itself (under the lid) is all black

This would be a little ugly, but you could probably go at the paint with a power sander or rotary sanding drum (e.g. dremel tool with sanding drum) to expose the bare metal. That might be enough to get the chassis pieces connecting, or you could even sand back on the outside as well, and put a strip of copper tape with conductive adhesive between the two pieces, making sure the areas where the tape adheres is bare metal. Might not look so nice, but probably nicer than a copper net wrapped around your gear. You could probably even paint the outer surface of the tape black after it was in place and verified effective if you want to make it a little less obtrusive looking.

 

[sodderboy]

Sorry, have to point out that incorrect statement. RF shielding in airplanes and automobiles works perfectly well, there is no "proper" grounding required.

Vehicles? C'mon, man! Even a location van will bond to gaia at location. An outlier at best in this forum.
If the dbx needs a cage, something is wrong with the chassis bonding properly to earth. Even a perfect bond would still leave the plastic display area open for RF penetration, and the eqpt is tested under all sorts of external EMI conditions, especially a dbx product destined for probable performance venue use, so it does not seem to be working 100% as designed.
Faraday shield seems like it's treating a symptom and not a cause.
Mike

 

[ccaudle]

Even a location van will bond to gaia at location

Sure, but you misunderstand the physics of why.

If the dbx needs a cage, something is wrong with the chassis bonding properly to earth

Unrelated. I am amazed that so many audio people just brush off the example of aircraft systems. They have some of the most stringent EMI/RFI requirements in industry, but a lot of people can't figure out how to fit that into their incorrect understanding of how earth connections are used, so they just brush it off rather than figuring out why everything works just fine even with no possible earth connection.

Faraday shield seems like it's treating a symptom and not a cause.

The purpose of a conductive chassis is exactly because it provides a Faraday shield to protect against electromagnetic interference in and out. The symptom is noisy audio, the cause is electromagnetic interference from a radio station, and shielding is what you use to protect against electromagnetic interference.

 

[sr1200]

1. is the apogee DAC wall wart powered? Any warts and proper grounding is the first place I would look. Something like a mini-me might want a better chassis connection to earth than mic cable shields (in this specific, new RFI situation)
2. Check the settings tree in the DRack, there could be a setting in there messing with pinout, polarity, impedance, *, etc. The box is deep. RF filter? Those joints are used in very RFI intense places.
3. It's clumsy with monitor eqpt not on patch, but you can use known good mic cables to connect the DAC-dbx into a different amp or powered monitors, try analog into the dbx or try the DAC-dbx chain in any other line level chain for the same RFI. I carry 600 and 2k8 ohm inline xlr slugs to change line impedance when things don't play well together. You can try it with a mic cable. Your runs are prolly short, so this should be less of an issue. I am assuming that all equipment is balanced. That said, does unbalancing the amp input change things?

The male operatic, 0:35 on this track is RFI that was recorded during tracking and flown-in from 1/4".
Mike
* perhaps it has a setting where it can feed a stereo amp with normal cables to use as a mono amp, sending hot to the LHS and cold to the RHS. We used to have to do this manually with 80's Crowns and Brystons. Just thinkin'

So here's what i originally tried:

• Changed cables from DAC to dbx = no change
• Changed cables from dbx to power amp = no change
• plugged dac directly into power amp = no radio interference (tells me the DBX is to blame)
• nothing plugged INTO the dbx, but the dbx out plugged into the power amp = no change
• plugged dbx into power amp using different cables (not using the installed cables in the wall) = no change
• ferrite clips on power and signal cables = no change
• physically moving the dbx to another room = no change
• ungrounding the xlr cables going to the power amp = made interference way worse
• ungrounding the xlr cables from the DAC = no change, just added hum
• resoldered all XLR's on the board and re-flowed solder on some potential cold solder points = no change
• changing power source to isolated BATTERY power on DAC and dbx = no change

then:

• Faraday cage = problem solved

Yes everything is balanced. The DAC (which doesn't seem to be related here since the problem still exists when its taken out of the equation) DOES have a wall wart.
There are no settings for polarity or anything i can see menu diving on the unit.