I guess I could....unfortunately, the fixes so far are the property of my employer and the vendor the work was done for...sigh...
I've a better idea however... How about I fix a power amplifier that had audible problems associated with the ground loop formed by the rca input, the star ground, the feedback divider grounding node, and the quality of the output?
While you won't like the amp, it worked..a swtpc tigersaurus.
The input config as designed was very messy..input ground loop currents attacked the front end input differential pair in the location I hilited earlier. The predrivers on the board used the same star ground as the input ground off the board, and this coupling caused heartaches..
Get this: The output inductor was formed by wrapping 13 turns of wire around one of the power supply caps. Believe it or not, this puppy coupled output signal into the power line ground cable, causing lots of audio signal to be pushed into the ground loop formed by the input cable, the source box, and the two line cords to the duplex..
I have already posted most of the analysis and changes I made to an existing amplifier John. For all to benefit.. this is after all, a diy site, no? That is why we are here.
Cheers, John
Yes. It also fails code to do this.John E, would the old Dynaco trick of inserting 10R resistors between the common of each input jack and circuit ground clean things up?
Since John C does this for money and you don't, you ought to charge him for the fix.
Money...hmmm...
Cheers, John
Well everyone, I don't know, but this is not going anywhere. Perhaps we should chose another subject to talk about, rather than my test methodology. I know how to get around my 'problem' so it will not interfere with my subsequent equipment tests. Generally speaking I design Class A circuits or heavy Class A-B circuits. Therefore, measurements at higher levels can be mathematically extrapolated to what should be present at lower (and higher) output levels. Almost always this is the case, except where there is some form of crossover distortion present. When we measure at lower levels, usually noise becomes a BIG factor. This is because the minimum bandwidth that I have to work with is 80KHz. This ST analyzer does not have a lower frequency rolloff, but then, because I am using a 5KHz sine wave (by preference), I need at least a 50KHz bandwidth to measure properly, anyway. Noise is annoying, BUT it can be reduced by either narrow band spectrum analysis and signal averaging. This is what brings out any distortion remnants that might be ignored, otherwise.
When we do harmonic distortion analysis, we usually have two readouts, the averaging meter, and the oscilloscope trace of the nulled residual waveform. Generally speaking, below 0.3V signal, the noise of the system overwhelms the distortion products. That is when we have to use a spectrum analyzer to remove the noise and separate the harmonics so that we can access them. Higher order harmonics 7th, 9th and 10th are VERY BAD. Thankfully they are relatively hard to generate with a circuit that is mostly linear. This is unfortunately what I was finding at VERY LOW LEVELS of 30mV, or about 1/10 the normally useful minimum level of the analyzer. Now, is this a problem with the analyzer or something else? It should be almost impossible to generate these harmonics at this level, because they are mathematically shown to always increase in value with level, EXCEPT with crossover distortion. Yet, what could be causing crossover distortion? Also, only SOME shielded cables generate this distortion, others do not, yet I can't see or measure much difference between them. I will withhold my opinion at the moment, until I find an answer.
When we do harmonic distortion analysis, we usually have two readouts, the averaging meter, and the oscilloscope trace of the nulled residual waveform. Generally speaking, below 0.3V signal, the noise of the system overwhelms the distortion products. That is when we have to use a spectrum analyzer to remove the noise and separate the harmonics so that we can access them. Higher order harmonics 7th, 9th and 10th are VERY BAD. Thankfully they are relatively hard to generate with a circuit that is mostly linear. This is unfortunately what I was finding at VERY LOW LEVELS of 30mV, or about 1/10 the normally useful minimum level of the analyzer. Now, is this a problem with the analyzer or something else? It should be almost impossible to generate these harmonics at this level, because they are mathematically shown to always increase in value with level, EXCEPT with crossover distortion. Yet, what could be causing crossover distortion? Also, only SOME shielded cables generate this distortion, others do not, yet I can't see or measure much difference between them. I will withhold my opinion at the moment, until I find an answer.
Last edited:
I assume the spectrum analyser uses FFT. The sample rate will either be locked to the signal generator or it will not. If it is locked then harmonics are already present so you need to ensure they can't leak in to what is being measured. If it is not locked then the FFT could generate spurious harmonics in the calculation because of the mismatch?
Changing a cable would not affect the latter, except randomly. It could affect the former, so how can you distinguish between harmonics generated by the cable and harmonics picked up by the cable?
Changing a cable would not affect the latter, except randomly. It could affect the former, so how can you distinguish between harmonics generated by the cable and harmonics picked up by the cable?
I believe these may be relevant to Jneutrons comments:
A Practical Interference Free Audio System (Part 2)
Designing for Interference-free Audio System Components
A Practical Interference Free Audio System (Part 2)
Designing for Interference-free Audio System Components
OK, thanks for clarifying that.john curl said:
I may appear to be asking stupid questions, but this can be a good way of solving problems. I once helped the staff of a major nuclear facility find a computer fault which had stopped part of the facility, simply by sitting in an office and asking stupid questions. After two days of answering my stupid questions they stumbled upon the solution, which had eluded them for the previous three weeks. I was forcing them to reexamine areas they thought they had already thoroughly checked, and justify assumptions which they 'knew' were sound. I was very pleased, as I had agreed that the following day I would put on a film badge and venture into a slightly 'hot' area to have a look for myself.
You are correct. Nice articles, thank you...a few notes on applicability..
Of note
art 1, section 6:""I use the term legacy equipment to mean equipment that has been designed and built with inadequate or incorrect grounding procedures. There are many variations on the theme. But in all cases, signal and/or control connectors, intended to use shielded cable, are designed such that the shield connection by-passes the equipment enclosure, and connects the cable shield terminal directly to the 0V or signal common trace on the printed circuit board. Any noise current flowing in the shield conductor is thus injected into the circuit’s ground return conductor and the circuit will reproduce the noise. This is the dominant mechanism by which “hum” or “buzz” noise gets into an audio system – and is the reason why audio engineers adopted the practice of lifting one end of a cable shield. In other words, “hum” or ”buzz” noise, caused by loop currents flowing between two system nodes are only a problem if one or more components in a system use incorrect shield termination methods. If the cable shield can inject current into the electronic circuit, just a few microamperes will cause a loud “hum” or “buzz”. But when the cable shield bonds directly to a grounded chassis, the current flowing must be abnormally high for the noise to even reach the level of the input circuit’s self noise. This is true even for balanced input receivers that have only a modest CMRR (Common Mode Rejection Ratio).
Other problems include:
· a non-grounded enclosure;
· a non-conductive enclosure;
· no connection between the internal circuit return conductor and the enclosure;
· insulated signal and control connectors;
· external power supply with no ground conductivity between the psu and the device it serves.
By now, you realize that I am not just talking about old equipment. In fact, around 90% of currently available new audio equipment is found to be designed with these short-comings. Others will recognize that it is not just audio equipment that is designed and built with incorrect shield termination techniques.""
This is one aspect I "rail" about.
Another is star grounding. Here, in secton 4:
""4 A note on ground loops
The most common objection to a meshed earth/ground system is that it creates “ground loops” – a phrase that is now historically associated with currents flowing in earth/ground conductors, where they have been found to make serious contributions to system noise.
In a system where earth/ground potential differences exist, closing a loop between two such earth/ground points will allow current to flow in the structure. If the potential differences that exist are high, then large currents can flow in the loop. When the conductors in the loop include a segment that forms part of, or is closely coupled to, a signal or low-level power cable, substantial interference can be induced in circuits associated with that cable. When we open the loop, the current no longer flows and the interference disappears – but the potential differences remain, ready to create a similar problem when a loop is closed elsewhere in the system.
The principle of removing all ground loops and accepting the resulting high voltages between different parts of the earth/ground system is formalized in the star or single point earth/ground regime. While this type of earth/ground system is relatively easy to implement, and quite successful in simple low frequency applications, it must now be considered to be a retreat from best practice. Interference in today’s systems is usually measured in 10’s or 100’s of MHz rather than Hz. A star or single point earth/ground conductor system presents high impedance at these frequencies, and so isolates the system from earth/ground rather than couple to it. Another serious problem is that larger star systems tend to degenerate into accidentally ground-looped systems as time passes. Sections of the system get modified or updated, requiring a degree of management and control to maintain safety and equipment reliability, that is a burden on the ever-decreasing resources of modern technical departments.""
The article is dealing with the macro application. Unfortunately, nobody is really extending the understandings to applications within the same piece of equipment. For example, a star ground in a poweramp may have to contend with the output currents, 10's of amperes, and the input currents/voltages. If two wires are attached to the same star node, and one is pushing high di/dt into it, the magnetic field around that conductor will communicate to other wires connected to the node. If the node has the feedback ladder reference ground, coupling of the output di/dt can induce voltages into the feedback loop after the divider..
Again, thanks for the articles...much appreciated. The cable functionality segregation is of particular interest, as my current app has a much wider variation of signals and power levels all mixed together, 535 kcmil to microwave cables running picosecond stuff, all in the same tray structure...
Cheers, John
For harmonics generated within the equipment, one must baseline the capabilities of the equipment. JC has never done so. Massaging one cable connector until consistent (for the moment) readings are obtained (as JC mentioned earlier) is certainly not baselining the equipment.
For harmonics generated by the cable, measurements using better equipment (the AP) have confirmed there is no cable generated harmonics present at a lower level than the ST1700 is capable of.
The most difficult thing to determine, is when the cable load and configuration causes the test equipment itself to generate the harmonics. This is where JC finds himself..
When presented test results which conflict with the laws of physics as we currently understand them, a review of the situation is in order..questions:
1. What is the accuracy and reliability of the equipment?
This equipment was designed in 1979. It is not designed to drive a single ended load that is then connected directly to it's own differential input pair. As such, the test setup being used is inconsistent with the use envisioned by the equipment designers. To date, there has been NO attempt at baselining the equipment capabilities to determine if it is up to the task assigned it. In addition, on a day to day basis, even JC's results do not repeat.
2 Can the results be reproduced using modern state of the art, far higher resolution equipment designed for the purpose?
No. Several attempts have been made so far using far more accurate and test-specific equipment, and absolutely nothing has been found consistent with JC's results.
3. Are there any mechanisms available to explain the results other than a paradigm shift in the current understandings of physics?
Yes. A total lack of control of the return current through the cable shield as a result of the test design. This is a direct violation of standard engineering practices to eliminate untoward results within high accuracy test systems.
4. Are there any methods available to remove these confounding current paths?
Yes. Transformer isolation. This is a very common practice, and is what AP apparently did on their higher resolution SOTA equipment.
While we disagree at times, I cannot fathom that though. I concur that questions from a different vantage point can be fruitful.
John
In science, and debugging software and hardware, it is usually a good practice to follow the evidence. Wise (or lucky) hunches can help too, but my experience is that when the problem is eventually solved it often becomes apparent that the evidence was there all the time pointing in the right direction even if ignored for apparently sound reasons.
If fiddling with connectors causes a significant change then this suggests that the connectors may be part of the phenomenon. Assuming no gross errors (e.g. dry joints) then the main effect of fiddling with connectors will be a small change in contact resistance. This should not create or solve any problems because the signal currents should be small and the contact resistance should be a negligible part of the total resistance. We are comparing milliohms with kohms, so issues cannot be much above the -120dB level even if the contact resistance is very non-linear. An exception is if there are other currents present, which is a matter of system design.
My conclusion is that JC is measuring connectors, not cables. His measurements are probably being contaminated by something, but what it is I don't know. Fortunately, I don't need to know.
If fiddling with connectors causes a significant change then this suggests that the connectors may be part of the phenomenon. Assuming no gross errors (e.g. dry joints) then the main effect of fiddling with connectors will be a small change in contact resistance. This should not create or solve any problems because the signal currents should be small and the contact resistance should be a negligible part of the total resistance. We are comparing milliohms with kohms, so issues cannot be much above the -120dB level even if the contact resistance is very non-linear. An exception is if there are other currents present, which is a matter of system design.
My conclusion is that JC is measuring connectors, not cables. His measurements are probably being contaminated by something, but what it is I don't know. Fortunately, I don't need to know.
I would disagree. See my post a page or two back where I described what I saw when John generously allowed me to see his setup and watch some of the measurements being taken.
Let us not go there. The current code is getting away from third wire safety ground to better insulation methods for consumer electronic gear.
I don't know if the ST1700 would be included.
The IEC started the change and the UL is following it. It is simply that if you have different grounding standards and methods around the world it is difficult to build gear to meet every condition. So commercial interests have introduce the concept of looking at leakage current and using more insulation and multiple layers to meet the safety and technical requirements.
Most of the power transformers actually built with shields for audio gear that I have measured meet the requirement for leakage for medical use! I looked at that because even those small levels can cause inadequate S/N for audio systems.
The real issue is not the problems with interconnect cables or leakage current but the low level of signals that is used for consumer audio interconnections! the CD standard is a maximum of 2 volts, with an average level well below a volt. Users often then reduce this voltage to be compatible with the rest of their system!
The worst "Pro" gear I have seen could do 7 volts. That reduces the "problem" by 10+ db. Much real "Pro" gear is 15 db improvement or more. High end pro gear does at least 20 db better. "Pros" usually know not to use their power amplifiers wide open. With those simple changes most of the S/N or "interconnect" issues drop out of range.
I just got back from tuning an arena sound system. The last guy in to measure things used pink noise and mapped where the frequency response of the coverage was uneven.
I had my new guy measure the speech intelligibility using a Goldline STI-PA system. This takes about a minute per measurement. So to get a decent reading of the seating area we take several hundred readings. The first day was spent getting everything working properly. The second we started around 8 and finished after 5, just doing measurements.
Based on the readings we found a problem with curtains blocking the direct sound to some seats. Not enough direct energy to others and some loudspeakers just not working.
Now what is more important in an arena, frequency response or speech recognition? How much influence does frequency response have on speech issues? Ever use a cell phone?
BTY it takes a week to measure a stadium.
Any one care to guess what adjustments we make after taking our readings? How do the adjustments for one seating area effect other areas?
I had my new guy measure the speech intelligibility using a Goldline STI-PA system. This takes about a minute per measurement. So to get a decent reading of the seating area we take several hundred readings. The first day was spent getting everything working properly. The second we started around 8 and finished after 5, just doing measurements.
Based on the readings we found a problem with curtains blocking the direct sound to some seats. Not enough direct energy to others and some loudspeakers just not working.
Now what is more important in an arena, frequency response or speech recognition? How much influence does frequency response have on speech issues? Ever use a cell phone?
BTY it takes a week to measure a stadium.
Any one care to guess what adjustments we make after taking our readings? How do the adjustments for one seating area effect other areas?
OK, I see what you mean. Reading this thread makes me suffer from information overload! However, the fact that JC had to take great care over connectors seems to confirm that it is currents which are part of the problem. If there were no unwanted currents then connector issues would be less important.SY said:
- Status
- Not open for further replies.