AudioCircle
Industry Circles => Tortuga Audio => Topic started by: tortugaranger on 19 Jul 2015, 03:29 pm
-
After a bit of struggle we finally solved a hum problem that only a few of our balanced preamp customers were having with our LDRxB models so we wrote up a brief article on the subject of the dreaded "pin 1 problem". Enjoy!
http://www.tortugaaudio.com/humming-along-to-the-pin-1-problem/ (http://www.tortugaaudio.com/humming-along-to-the-pin-1-problem/)
-
It sounds like that was not actually a pin 1 problem, but a missing ground...?
BTW you have a typo - "but connecting pin 1 the right avoids hum"
-
It sounds like that was not actually a pin 1 problem, but a missing ground...?
BTW you have a typo - "but connecting pin 1 the right avoids hum"
Thanks for the heads up on the typo. It's not clear to me why some balanced amps hum simply by connecting to a preamp without a grounded pin 1. What is clear is that shouldn't happen if the amp is designed properly. I completely accept that we could have avoided any hum problems by grounding pin from the beginning but then we wouldn't have anything to write about. :duh:
-
To augment the discussion in the 1st post:
http://www.diyaudio.com/forums/diyaudio-com-articles/163575-audio-component-grounding-interconnection.html
A very nice and thorough read.
Best,
Anand.
-
Thanks Morten for your informative post. As you know, one of the amps I'm using has that "pip' problem. It's good to know the solution is straightforward.
-
To augment the discussion in the 1st post:
http://www.diyaudio.com/forums/diyaudio-com-articles/163575-audio-component-grounding-interconnection.html (http://www.diyaudio.com/forums/diyaudio-com-articles/163575-audio-component-grounding-interconnection.html)
A very nice and thorough read.
Best,
Anand.
That's an excellent article on the subject of interconnection and grouding. He could have made the diagrams less difficult to understand but the narrative is the best I've read on the topic. Does a very nice job on star grounding as well. I'm big on star grounding. Hasn't failed me yet.
-
Although many of the links need to be updated, the pin1problem.com page has tons of information.
The John Windt "Hummer" is the best test procedure. But little information on doing the 'Hummer Test' is available online.
Bill Whitlock has an updated Hummer on page 122 of:
"An Overview of Audio System Grounding and Interfacing"
by
Bill Whitlock, President
Jensen Transformers, Inc.
http://centralindianaaes.files.wordpress.com/2012/09/indy-aes-2012-seminar-w-notes-v1-0.pdf
***********************************
The 1995 John Windt paper:
"An Easily Implemented Procedure for Identifying Potential Electromagnetic Compatibility Problems in New Equipment and Existing Systems:
The Hummer Test"
This paper along with 6 other good papers is available for the AES for only $15 including shipping.
SHIELDS AND GROUNDS: SAFETY, POWER MAINS, STUDIO, CABLE AND EQUIPMENT,
The June 1995 issue of the Journal was a definitive and comprehensive collection of information on this important topic. The seven papers by Neil Muncy and other experts in the field have been reprinted into a convenient guide for designers and practitioners.
http://www.aes.org/publications/specialpubs/journal_issues.cfm
-
There is the AES 48 / 2005 standard for this. If devices followed it, it would not be a problem. Well, I said "if" :)
This is a preview:
http://www.aes.org/tmpFiles/aessc/20150722/aes48-2005-r2010-i.pdf
-
The AES48 standard is a very short summary of the 1995 Neil Muncy paper:
"Noise Susceptibility in Analog and Digital Signal Processing Systems"
A search might find the complete paper.
The paper is also part of the June 1995 AES Journal that I listed above.
*********************************
The John Windt 'Hummer Tester' will quickly find poorly designed audio components.
-
I do not have access to the aforementioned 1995 paper. If the proposed scheme is like this one based on star grounding
http://home.comcast.net/~zenkai-audio/gndtss/pages/s3_star.htm#top
I think there is one difference in the AES-48: the input and output shields are connected to the closest point in the chassis, instead of the star ground, as shown below.
(http://i19.servimg.com/u/f19/15/22/78/37/aes4810.jpg)
-
More modern thinking is:
a] AC power Safety Ground/Protective Earth connects to the chassis near where the AC power entries the chassis.
b] Both the DC power supply common and the signal circuit common near the audio input chassis connectors.
-
Speedskater, assuming a stereo component, and we follow your "more modern thinking", then do we have two ground wires from signal boards, one to ground point "near" (or at, XLR jack mounting screws are convenient) to each XLR jack?
I guess it might depend if the component in question is a true dual mono design in single chassis with separate R and L power supply grounds (0 volts)?
Perhaps this distinction is not important?
My DAC has common power supply 0 VDC for both channels and XLR output only-I run a single single G wire to one of the XLR jack's mounting screws, and things are silent.
But my amp is true dual mono in one chassis, here I must run a signal G wire for each channel to each XLR jack (terminated to mounting screw/chassis)
In both cases pin 1 goes to chassis via shortest possible wire on both XLR jacks.
-
I see point a].
As for point b], is shield current entering the signal in this way? I see a potential danger here :)
I am referring to XLR input chassis connector, which is suggested to be left unconnected in the Rane white papers, e.g., http://www.rane.com/note151.html and http://www.rane.com/note110.html . I mean, only pins 1, 2, 3 connected, not the chassis tab on the connector.
Thanks
Roberto
More modern thinking is:
a] AC power Safety Ground/Protective Earth connects to the chassis near where the AC power entries the chassis.
b] Both the DC power supply common and the signal circuit common near the audio input chassis connectors.
-
About an XLR balanced analog audio interconnect system:
a] There is no 'ground' connection in the interconnect system.
b] Pin #1 is the shield/screen (not the audio circuit common)
c] Pin #1 may also be a microphone phantom power common.
d] Each XLR chassis connector should have its Pin #1 connected to the chassis at the connector.
(less than 1 inch/2cm)
e] The audio circuit common/power supply common should have its single chassis connection to the chassis in the vicinity of all the audio input connectors.
f] Not all XLR interconnects have the shield connected at the receive end. Some may have the received end connected through a capacitor (hybrid). Different experts have different view-points on this subject.
g] The AES as a standard that the XLR shell/tab be not connected to anything. They worry that in a large system the shell could come in contact with metal at a different potential. But in many hi-fi systems the rule can be ignored.
Did I miss anything?
-
I generally agree with the last few posts which are mostly saying the same thing with minor variations. In my view the absolute best approach is a fanatical adherence to using a single star ground point. The star ground point can be the chassis itself for an all metal enclosure but I actually prefer a isolated star ground point - an isolated bus bar if you like. Everything goes to the star ground point, including the chassis. And in the case of the pin 1's, they should also go to the star ground point although that's not shown in this graphic. This way there are no ground currents running through chassis or anything else except on its way to star ground and thus everything sees only a single ground potential. The star ground point has one connection back to earth ground via the incoming AC mains if present or the DC supply negative/ground. I've yet to run into any grounding issues by following this approach. When I do run into hum (on things I design/build) I've always been able to trace it back to failure to star ground everything.
(http://www.tortugaaudio.com/images/ldr3x.v2.audio_power_grounding_balanced_092014.jpg)
-
I definitely agree with you "Star Ground Point" system.
The small changes that I would make:
1] Move the "Star Ground Point" chassis connection to near the input connectors (it may already be that way).
2] Remove the AC Safety Ground from the "Star Ground Point". Connect it to the chassis near AC J3.
3] Remove all the XLR Pin #1 jumpers and connect each one to the chassis at the XLR connector.
-
I definitely agree with you "Star Ground Point" system.
The small changes that I would make:
1] Move the "Star Ground Point" chassis connection to near the input connectors (it may already be that way).
2] Remove the AC Safety Ground from the "Star Ground Point". Connect it to the chassis near AC J3.
3] Remove all the XLR Pin #1 jumpers and connect each one to the chassis at the XLR connector.
1) I'm ok with this but in my view that's really a nit.
2) This moves away from a star ground centric approach and invites the possibility of the star ground being at a different potential than the chassis. I realize the "earth to chassis" has been a traditional approach. I just don't agree with it. In my view "earth to star" and "chassis to star" (everything to star) preserves both the integrity of the star philosophy while also being safe.
3) Not sure what this accomplishes. Everything to star all the time - no hum.
-
Bruno Putzeys has mentioned that while using a star ground will work at low frequencies, it can give rise to noise problems at higher (RF) frequencies. So if we are talking DACs (or anything with some processor noise) or Class D amplifiers, using a star ground scheme may not be the best idea.
Generally, I consider Mr. Putzeys to be a pretty accomplished engineer. He has mentioned that the star ground will have too much impedance for high frequency noise, hence the recommendation to connect pin 1 direct to chassis with the shortest possible wire, right at the input point.
This is an interesting distinction to me, as I am sure a well implemented star ground will eliminate typical audible hum problems, I find that there can be more subtle sonic degradation caused by RF noise issues, and eliminating these lesser effects is of great interest to me, especially in the world of contemporary components where there are all kinds of sources of RF pollution.
-
Frequency-dependent Pin 1 lifting schemes are used in a variety of professional equipment. Generally, a bypass jumper can enable/disable the function to create optimum conditions for either input and output connectors to the external equipment.
There's not a hard and fast solution to all situations regarding the Pin 1 problem.
Dave.
-
Frequency-dependent Pin 1 lifting schemes are used in a variety of professional equipment. Generally, a bypass jumper can enable/disable the function to create optimum conditions for either input and output connectors to the external equipment.
There's not a hard and fast solution to all situations regarding the Pin 1 problem.
Dave.
Those are poor methods and invite the camel into the tent. The Pin #1 should be connected directly to the chassis with well less than 1 inch of wire (2cm). Longer wires inside the chassis will act as interference antennas. The better way is to modify the cable (not chassis) connector. It should always be the input cable connector (never the output) that is modified.
-
1) I'm ok with this but in my view that's really a nit.
It's about 'good engineering practices'.
The chassis connection should be near the most sensitive circuits.
2) This moves away from a star ground centric approach and invites the possibility of the star ground being at a different potential than the chassis. I realize the "earth to chassis" has been a traditional approach. I just don't agree with it. In my view "earth to star" and "chassis to star" (everything to star) preserves both the integrity of the star philosophy while also being safe.
In line with what barrows wrote. A star ground is far from a be- all and end-all idea. The AC Safety Ground should not be part of the DC supply, audio circuit chassis connection point. Besides a long Safety Ground wire inside the chassis, just acts as an interference antenna.
3) Not sure what this accomplishes. Everything to star all the time - no hum.
It's the main part of AES48.
-
It's about 'good engineering practices'.
The chassis connection should be near the most sensitive circuits.
In line with what barrows wrote. A star ground is far from a be- all and end-all idea. The AC Safety Ground should not be part of the DC supply, audio circuit chassis connection point. Besides a long Safety Ground wire inside the chassis, just acts as an interference antenna.
It's the main part of AES48.
I think that when you distill all of these last few posts down there's very little substantive difference. If you locate the star grounding point on the chassis in close proximity to the input near where pin 1 is landed right by where the earth safety gets connected that it becomes for all practical purposes a single star ground point regardless of what you call it. To say the AC ground should not be part of the DC supply kind of defies gravity since unless you're designing a floating +/- split DC supply your supply ground will most definitely connect to the AC earth ground. Whether that connection is a few inches over here or over there are differences without material distinction....in my opinion.
Having said all that, let's not turn this into a debate about NEC, AES48 or best engineering practices. We can quibble endlessly around the margins to no good end which is something that happens in excess in audio land.
We tried to step around the pin 1 problem altogether by not connecting them at all and got bit because of that. How could not landing a cable shield (A CABLE SHIELD!!) possibly cause a hum problem? Because some equipment designer on the other end didn't follow good engineering practices and landed pin 1 where it shouldn't be connected. What a PITA! :thumb:
-
2.5 words: jensen iso-max :D