Hi Bob,
The device in question was not originally intended for this particular function as it happens, and (as I often do!) I simply took advantage of using a proprietary device which was made for locating buried mains wires etc. in buildings.
Now that you question its workings, extremely unusually for me (!) I realise that I have never actually looked inside the enclosure, possibly because it has a plastic casing with no obvious fixings to unscrew etc. and unless it is otherwise unusable, I am reluctant to risk spoiling its construction without good reason through attempting an "unauthorised entry".
However, it has an adjustment for sensitivity which is controlled by an (unhelpful!) inset side-mounted thumbwheel with no graduations, so the degree of rotation is rather indistinct. It also has an LED of varying brightness in accordance with signals detected, and a (probably piezo) varying-level sounder within, which are very useful for this particular operation.
My best guess is that it sports a field coil within for detection purposes, together with a transistorised amplification circuit, the overall amplifier gain of which, or perhaps the input level, is controlled by the thumbwheel attenuator.
I haven't thought to even consider any possible bandwidth testing, but knowing its origins, I guess that this could be quite narrow, and centred on our UK mains frequency. i.e. 50Hz (or perhaps 100Hz)
If it ever packs up working, you can guess what will result, and I will try to remember to let you know what I find inside.
As this device is sensitive enough to detect quite well-buried wiring (IIRC under thick concrete etc.), it is ideal for this purpose and I have used it for over 20yrs now without a hitch except for an obligatory battery-change, but in reality it gets very little use of course.
Yes, I do mean simply swapping over the live & neutral connections to the primaries, and then putting the transformers under a suitable load as their emissions reduce considerably when unloaded, in my experience. In this connection, there is no doubt that this is solely due to the transformer 'reversal' as nothing other than a resistive loading is connected to the secondaries for these brief trials on my bench.
My methodology for this particular trial is to connect up and energise the transformers firstly one way around and then do the reversal. This establishes a base-line where on the lowest radiating wiring-orientation I can place the device in the required position and gradually reduce the output to virtually zero (very feint LED, and virtually no sonic output) and then do the swap-over of leads without disturbing the relative positions of anything.
As suggested, it would be more interesting if I knew how this was calibrated (i.e. if it is linear in operation, or whatever) but quite frankly I have not found the time and inclination to try to mark up the barely-visible edge of the thumbwheel and arrange a set-up of 'known' levels to carry out this task (yet!). Maybe one day, but for now it serves its purpose as there are only two choices of transformer wiring orientation, anyway.
What was initially such a surprise was the degree of difference in emissions (also when tested at alternative locations around any toroidal transformer) and, although this is a very crude method, to arrange for a similar 'reaction' from the same attenuator-setting, it usually requires the device to be moved away by tens of units in distance from the transformers in question.
Regards,
The device in question was not originally intended for this particular function as it happens, and (as I often do!) I simply took advantage of using a proprietary device which was made for locating buried mains wires etc. in buildings.
Now that you question its workings, extremely unusually for me (!) I realise that I have never actually looked inside the enclosure, possibly because it has a plastic casing with no obvious fixings to unscrew etc. and unless it is otherwise unusable, I am reluctant to risk spoiling its construction without good reason through attempting an "unauthorised entry".
However, it has an adjustment for sensitivity which is controlled by an (unhelpful!) inset side-mounted thumbwheel with no graduations, so the degree of rotation is rather indistinct. It also has an LED of varying brightness in accordance with signals detected, and a (probably piezo) varying-level sounder within, which are very useful for this particular operation.
My best guess is that it sports a field coil within for detection purposes, together with a transistorised amplification circuit, the overall amplifier gain of which, or perhaps the input level, is controlled by the thumbwheel attenuator.
I haven't thought to even consider any possible bandwidth testing, but knowing its origins, I guess that this could be quite narrow, and centred on our UK mains frequency. i.e. 50Hz (or perhaps 100Hz)
If it ever packs up working, you can guess what will result, and I will try to remember to let you know what I find inside.
As this device is sensitive enough to detect quite well-buried wiring (IIRC under thick concrete etc.), it is ideal for this purpose and I have used it for over 20yrs now without a hitch except for an obligatory battery-change, but in reality it gets very little use of course.
Yes, I do mean simply swapping over the live & neutral connections to the primaries, and then putting the transformers under a suitable load as their emissions reduce considerably when unloaded, in my experience. In this connection, there is no doubt that this is solely due to the transformer 'reversal' as nothing other than a resistive loading is connected to the secondaries for these brief trials on my bench.
My methodology for this particular trial is to connect up and energise the transformers firstly one way around and then do the reversal. This establishes a base-line where on the lowest radiating wiring-orientation I can place the device in the required position and gradually reduce the output to virtually zero (very feint LED, and virtually no sonic output) and then do the swap-over of leads without disturbing the relative positions of anything.
As suggested, it would be more interesting if I knew how this was calibrated (i.e. if it is linear in operation, or whatever) but quite frankly I have not found the time and inclination to try to mark up the barely-visible edge of the thumbwheel and arrange a set-up of 'known' levels to carry out this task (yet!). Maybe one day, but for now it serves its purpose as there are only two choices of transformer wiring orientation, anyway.
What was initially such a surprise was the degree of difference in emissions (also when tested at alternative locations around any toroidal transformer) and, although this is a very crude method, to arrange for a similar 'reaction' from the same attenuator-setting, it usually requires the device to be moved away by tens of units in distance from the transformers in question.
Regards,
Bobken, you have made an interesting revelation. Bob Crump used to do something similar, and we found one of the transformers in the HCA3500 out of polarity with the main toroid transformer, and we had it changed. This is one of the design factors in the JC-1 power amp, we test the innate polarity of the transformers, and get them aligned properly.
Of course, it is impossible, but it IS measurable! This is what happens when you make real products, and not simulations.
Of course, it is impossible, but it IS measurable! This is what happens when you make real products, and not simulations.
Probably hooking up my 'scope with an attached relay coil would provide some worthwhile results, just like I posted on this Forum several years ago when asked how 'sexing' or orientation of foil capacitors could be achieved, to determine their outer-foils. (Which is another important point to get right in HQ audio constructions).
However, in this instance, such experiments were unnecessary as I found that (for once!) there was an easier way, and I didn't need to use much imagination nor construct anything for myself.
As for Jan's very helpful and unsurprising comments, no-one suggested that this is rocket science, and indeed I went to considerable lengths to de-mystify anything here and make it quite clear that some of these trials are not very sophisticated.
What matters, though, is that my methodology described here works very well, and (for whatever reasons) it matters a lot to get this 'orientation' right if you wish to get the best sonic results. Actually, it has nothing to do with several transformers located in close proximity (which is another important, but quite separate issue) as the difference in sound is usually quite apparent when dealing with a solitary transformer.
If Jan already does what I do, he will appreciate the obvious benefits of taking this small step to ensure that his transformers are orientated in the best manner possible, but from the (usual) tone of his remarks, I am guessing that this is not so.
Regards,
However, in this instance, such experiments were unnecessary as I found that (for once!) there was an easier way, and I didn't need to use much imagination nor construct anything for myself.
As for Jan's very helpful and unsurprising comments, no-one suggested that this is rocket science, and indeed I went to considerable lengths to de-mystify anything here and make it quite clear that some of these trials are not very sophisticated.
What matters, though, is that my methodology described here works very well, and (for whatever reasons) it matters a lot to get this 'orientation' right if you wish to get the best sonic results. Actually, it has nothing to do with several transformers located in close proximity (which is another important, but quite separate issue) as the difference in sound is usually quite apparent when dealing with a solitary transformer.
If Jan already does what I do, he will appreciate the obvious benefits of taking this small step to ensure that his transformers are orientated in the best manner possible, but from the (usual) tone of his remarks, I am guessing that this is not so.
Regards,
Bob, you may have noticed that my post was not addressed at you. Rather, the (really) interesting narrative you provided was grabbed by someone else as the proverbial stick to hit simulators and a few other species, even if it was totally unrelated to the subject at hand.
I do my best to disregard such BS, but not always succesfull. Sorry.
But, coincidently, I am restoring/rebuilding an old Hiraga creation, the one-transistor power amp (Nemesis). This is a single MOSFET amp, using a 64 to 8 ohms output transformer in the drain lead. If you want to experience transformer influence in almost any physical factor of an amp, try that!
Jan Didden
I do my best to disregard such BS, but not always succesfull. Sorry.
But, coincidently, I am restoring/rebuilding an old Hiraga creation, the one-transistor power amp (Nemesis). This is a single MOSFET amp, using a 64 to 8 ohms output transformer in the drain lead. If you want to experience transformer influence in almost any physical factor of an amp, try that!
Jan Didden
It is kind of sad that when we give REAL up-to-date design info, like 'polarizing' transformers, we get hoots of derision. That is a good example of what we have been working on in the past 25 years. Of course, many here are still getting up to speed on what we tested AND PUBLISHED 30 years ago, so we usually find it difficult to fill their heads with the newer and more subtle stuff. Kind of like the difference between lower division physics and graduate level physics.
Charles has been very brave, almost to the point of being irresponsible, by revealing his more subtle design techniques in this environment. It is a 'pearls before swine' dilemma.
Still, those who want to win listening contests should take note on the more subtle techniques that we offer, for free, I might add.
Charles has been very brave, almost to the point of being irresponsible, by revealing his more subtle design techniques in this environment. It is a 'pearls before swine' dilemma.
Still, those who want to win listening contests should take note on the more subtle techniques that we offer, for free, I might add.
Charles Hansen said:
Charles,
I am not in a position to 'prove' the existence of the golden ear, but it has been demonstrated that there are at least two separate mechanisms that lead to 'super tasters.' One is that some people simply have more taste buds than others. The other is that some people produce more saliva than others (hence increasing mobility of ions, meaning more excitation of taste buds). Presumably a taster possessing both traits would be in pretty good shape if he/she wanted to perceive more 'detail' in foods or drinks.
Personally, I find it nearly inconceivable that analogous mechanisms wouldn't exist for hearing--perhaps more cilia in the inner ear, for instance.
But, of course, such a concept is unacceptable to certain members here. Everyone hears the same. No one can hear anything that is not sanctioned by the THD committee. Nothing makes any difference if they are unable to understand a mechanism that might lead to the difference.
That said, I have educated my hearing as detailed elsewhere and regard that as being indispensable. Gawd forbid that the know-nothings should do such a simple and obvious thing.
Grey
P.S.: According to my dental hygienist, I produce more saliva than any of her other patients. I have no information, anecdotal or otherwise, about the number of taste buds on my tongue. I do appear to be able to discriminate tastes that (at least some) other people cannot.
bear said:
This makes sense to me. Noise can increase the effective magnitude of a signal within a spectrum and thus make it more easily detected via resonance, autocorrelation or cross-correlation - ie components of the human auditory system. Has anybody passed such an output through a correlation spectrum analyzer?
Hi Grey,
True, there are many who are "focused on the pointer", and still others who ignore obvious problems with what can be measured and "design by ear". Over the course of many years, in excess of 25 years as it happens, I've seen proof time and time again that taking one approach or the other doesn't turn out very well. I will say that the instrument people end up designing equipment that tends to last.
If, as a designer, you don't take advantage of all the tools available to you, you are failing as a designer. I'll bet that all designers of good, well known equipment do both listen and test. Both carefully. I think that this helps the design converge to a reliable device that also sounds very good.
I feel the constant (and artificial) division between the listeners and measurer's only serves to fog the information that is presented.
Now, as for the phasing of the power transformer. I think we could all agree that "the perfect transformer" would not react any different if the phase were one way or another. This then begs the question, why? I see we are talking about toroid transformers and there may be a clue in there. Specifically, is there an unbalance in current draw due to the actual circuits being powered? This would be more pronounced at idle, as when your power demands increase, the current drawn should be symmetrical. This leads me to ask the question, are the cores air gapped? This would greatly reduce core saturation. Further, do R core or EI core transformers suffer from the same problem (both having intrinsic gaps in their cores). Could this be something as odd as rectifier diodes not matching in a bridge?
Now, the mention of an electrostatic shield has reappeared. This was a very common feature in old transformers. I think the only transformers that didn't make use of this shield were industrial control transformers (and they might) and inexpensive consumer goods. Can we just assume that such a shield does exist in all transformers we consider for use? This would seem to be a prerequisite for any transformer used in audio equipment.
I don't believe the phasing of power transformers should make any difference, assuming an ideal transformer. Also assuming there are no other transformers where magnetic coupling can occur.
-Chris
I don't think that's a fair comment to make here.
True, there are many who are "focused on the pointer", and still others who ignore obvious problems with what can be measured and "design by ear". Over the course of many years, in excess of 25 years as it happens, I've seen proof time and time again that taking one approach or the other doesn't turn out very well. I will say that the instrument people end up designing equipment that tends to last.
If, as a designer, you don't take advantage of all the tools available to you, you are failing as a designer. I'll bet that all designers of good, well known equipment do both listen and test. Both carefully. I think that this helps the design converge to a reliable device that also sounds very good.
I feel the constant (and artificial) division between the listeners and measurer's only serves to fog the information that is presented.
Now, as for the phasing of the power transformer. I think we could all agree that "the perfect transformer" would not react any different if the phase were one way or another. This then begs the question, why? I see we are talking about toroid transformers and there may be a clue in there. Specifically, is there an unbalance in current draw due to the actual circuits being powered? This would be more pronounced at idle, as when your power demands increase, the current drawn should be symmetrical. This leads me to ask the question, are the cores air gapped? This would greatly reduce core saturation. Further, do R core or EI core transformers suffer from the same problem (both having intrinsic gaps in their cores). Could this be something as odd as rectifier diodes not matching in a bridge?
Now, the mention of an electrostatic shield has reappeared. This was a very common feature in old transformers. I think the only transformers that didn't make use of this shield were industrial control transformers (and they might) and inexpensive consumer goods. Can we just assume that such a shield does exist in all transformers we consider for use? This would seem to be a prerequisite for any transformer used in audio equipment.
I don't believe the phasing of power transformers should make any difference, assuming an ideal transformer. Also assuming there are no other transformers where magnetic coupling can occur.
-Chris
anatech said:
Last? You appear to have a wildly different definition of 'last' from the one used by real people in the real world. Anyone who can say something like this with a straight face clearly hasn't compared the relative values of, say, fifty year-old McIntosh tubed pieces with solid state gear...of any vintage, no matter how good the specs.
Oh, sorry, I forgot...all those who go for tubed gear (hence keeping the demand and value up) are deluded fools blindly following evil gurus who exert mysterious mind-control powers to keep their followers in line. (Details of the workings of this putative mind control are even more scant than explanations for the operation of magic dots. For instance, while it's presumed to be good for the 'gurus,' no one ever seems to get around to explaining the benefit for their followers. [N.B.: Careful...anything you say here can just as easily be applied to followers of the meter-gurus.]) Ditto for those who unaccountably choose solid state with less than SOTA distortion specs. Poor zombies. Might as well put them out of their misery.
Perhaps you'd care to explain for us why it is that it's the ultra-low distortion pieces that seem to have the shortest shelf life...about what you'd expect from the "initially exciting, but long-term annoying" profile, in fact.
Chris, you're hopelessly disconnected from external reality.
Grey
P.S.: I'm still waiting for a nice, long list of links to posts from measurements-oriented people admitting their potential (and actual) bias, given that they already know the distortion specs of their latest magnum opus before they even listen to it.
...still waiting...
...still waiting...
...still waiting...
Chris, real designers have test equipment. As I have said many times, I can easily measure down to -120db in THD, perhaps more. I have signal averaging as well.
Right now, I am TRYING to lower the residual distortion of an AD825, donated by Scott Wurcer for a special new project. The higher order distortion products upset me. Any suggestions?
Right now, I am TRYING to lower the residual distortion of an AD825, donated by Scott Wurcer for a special new project. The higher order distortion products upset me. Any suggestions?
- Status
- Not open for further replies.