SUT - electrical theory and practical experience

The H 3000 already uses built in step up transformers for moving coil cartridge levels of gain. If you want to use an external step up transformer, you would need to use moving magnet cartridge inputs on the H 3000. I don’t know whether the H 3000 even has MM inputs. If it does, that’s where you could go. You cannot hook up an external step up transformer to an internal step up transformer. Personally, I think you may be chasing your tail if you are even able to try this experiment with your H 3000, because the internal step up transformers on the H 3000 are already excellent.

Dave, I’m getting my Festivus pole out of storage now. I just aired my grievances with my wife. Then we’re going out for Chinese food.

Holmz, Make no mistake, Dave (Intact Audio) knows more about this subject by far than I ever will. So if my advice to check out the Jensen website seems counter to or only partially correct in view of Dave's insight, go with what Dave says. 

Holmz, you ought to visit the Jensen Transformer website. There you will find erudite discussion with schematics and other data regarding how to load a SUT. Plus the engineers there are nice guys who might help with your specific questions, if you call them in California. 

Some of us find the science of these devices to be interesting per se.  Some others do not.  Based on his posts to start the thread and subsequently, I judge that the OP is one of those interested in the science. If he thinks the thread has drifted away from his intended question (which it has once or twice), he can say so here. If you prefer the subjective experience of "listening" as the sole means of selecting the elements of your audio system, I respect that.  I prefer a mixture of both, but anyone who is turned off by the technical language can simply stop reading.  Argumentative sidebars about whether SUTs are good or evil are beside the point of the thread; I agree. I do not nor have I ever owned a SUT nor heard one in my system, yet I do find the subject interesting, and I am open-minded as to the potential virtues, because of the testimony of so many others who do use SUTs.

Pindac, What "theorem" are you referring to?  Thanks.

antinn, In your quoted paragraph, I think they are using "Watts" as an expression of acoustic power or energy.  On that assumption, and wanting to know how acoustic watts is defined, I found this on Wiki: "Sound power or acoustic power is the rate at which sound energy is emitted, reflected, transmitted or received, per unit time.^[1] It is defined^[2] as "through a surface, the product of the sound pressure, and the component of the particle velocity, at a point on the surface in the direction normal to the surface, integrated over that surface." The SI unit of sound power is the watt (W).^[1] It relates to the power of the sound force on a surface enclosing a sound source, in air. For a sound source, unlike sound pressure, sound power is neither room-dependent nor distance-dependent. Sound pressure is a property of the field at a point in space, while sound power is a property of a sound source, equal to the total power emitted by that source in all directions."  It's not an easy concept to grasp.

"That manufacturer said that a LOMC ( with out SUT ) has a rise time between 15 to 21 microseconds and MM between 23-31 msg."

To the best of my knowledge you never previously mentioned the data for an LOMC without SUT. Anyway, I doubt any of us can hear the difference between 21 usec (high value for MC alone) and 23 usec (low value for MM).  And still one would like to know how the measurements were made, but your new information does suggest that the SUT slowed down the LOMC, without allowing for explanations based on load, reactive components, etc., because we lack the information.

Raul, I see now that it was you who posted the rise time data, which are based on your memory of something published by an electret cartridge maker who, from your memory, is said to have stated that his cartridge has a rise time of 2 microseconds, compared to 25 microseconds for an MM and 30 microseconds for an MC plus SUT.  Holmz may have the test LP and a 'scope, but so far I don't see where he posted data.  SOOOOO, why don't you post a reference to the source of the above data?  If you can find it.  Sure, there can be many reasons why the data are not a reflection of the rise time of an MC cartridge taken by itself, and I am also sure that every MC cartridge is different in that regard.  Finally, if we are only going on the basis of subjective impressions, I would say that as a whole (meaning all MI cartridges I have heard compared to all LOMC cartridges I have heard), MI cartridges sound "fastest", to me. I have no idea if actual measurements would substantiate my opinion.

Holmz, where do the data that has caused so much angst come from? (Rise time for MM at 1kHz is 25 microseconds vs for MC+SUT is 30 microseconds.) what cartridges were used? What SUT? After how many hours of use? Etc, etc. Thanks if you can reference the source of the info.

First of all, when did I say I disagree with you regarding the relative "speed" of the two types of cartridge?  Second, YOU posted the bandwidth of the Technics SUT when you were trying to claim superiority for vintage SUTs over some modern ones.  Yes, it's an actual number, I agree.  You're stuck with it if you want now to argue that the SUT used in the rise time experiment is reducing the apparent rise time of the MC used in conjunction with the SUT.  But like I said (twice at least), we lack data (frequencies, impedances, capacitance, inductance, etc), and whoever performed the measurements should certainly have re-measured the MC cartridge without the SUT in the circuit. 

Yes, I still wonder what you expect me to solve with virtually no data.  To achieve a bandwidth of 300,000Hz (at 0db), a SUT must have a rise time of at least .00000333 seconds.  Is that the solution you are looking for?  It's just a number that solves nothing.

Wish I knew what you were talking about, Raul.  Solve what?  It was you who posted the information that your vintage SUTs had such a wide bandwidth.  Now when it does not suit you to admit your own facts, you want to reduce the bandwidth so you can claim sans data that the SUT must be what limits the rise time of the MC cartridge in the scenario.  Truth is, that was a badly done experiment, just because it introduces the SUT as another variable that may or may not limit rise time.  It's really not worth further discussion.  If you think MC cartridges sound "faster" to you than MM types, so be it.  There is nothing here to deprive you of your subjective impression.

Raul, I’m thinking out loud. Just like you. You and others mentioned that a SUT is capable of an extraordinary bandwidth, out to 300K or even 500KHz. Since bandwidth is necessarily related to rise time, I’m wondering whether it’s justified to infer that a SUT is responsible for the surprising finding that an MM is faster than SUT plus MC. We’d need more data to intelligently debate the question.

If a SUT studied as a unit can exhibit a frequency response out to 300KHz, how can it be the principle culprit in determining a relatively slow rise time? Most cartridges do not respond out to 300KHz so far as I know.

If the information available from Rothwell answers your questions, that's great.  Just beware of the passages where they compare the relative virtues of active high gain stages to SUTs. Because they sell transformers. Also, if Intact Audio and Rothwell are in conflict, go with the former authority.

Keep in mind, Rothwell are selling transformers.

Raul, with regard to your question above, who is “Palmer”?  Anyway I hope I conveyed my opinion that current driven phono stages are not categorically better than voltage driven ones. They’re just a novel approach to an old problem. Again, in my opinion.

The point I was trying to make when I pointed out that transimpedance phono stages generally are just voltage driven stages with a current to voltage converter added on to the front end is that they are subject to foibles just like voltage input phono stages. As Atmasphere pointed out in another thread, a current driven stage ought theoretically to present zero input impedance, but that is impossible as there would then be no signal at the output of the converter. So they all have a finite input impedance ranging from ~2 to ~20 ohms, based only on research I’ve done on my own. That’s a fly in the ointment of perfection. Because of differences in actual input impedance ( and probably in current gain and maybe distortion of the input devices, either an op amp or transistors), different transimpedance stages seem to act differently and sound a bit different dependent upon the internal R of the cartridge. This is evident to me based on my communication with Mijostyn. He uses a Lino and I use a BMC MCCI. And each of us has experimented with a few different LOMCs. We’ve reported somewhat different results, depending on cartridge. So there’s nothing perfect about current drive or stages that use it, though they can be very competitive with the best voltage driven stage.

Also, in a voltage driven conventional stage, the input load R can be thought of as the simplest form of passive current to voltage converter.

To the notion that cartridges are “natural” current generators I would say that the general statement can only be applied to LOMC cartridges with very low internal R in relation to V output. I would also note that in a “trans impedance” stage , the first stage converts current to voltage. Thereafter and in the RIAA filter, the signal is handled as a voltage exactly as in a conventional phono stage.

Raul, to state the bandwidth of any transformer, particularly an audio SUT, without specifying the upper and lower limits of the given bandwidth in db, the impedances on the primary and secondary windings, and the other reactive components of the load, like capacitance and inductance, is fairly without meaning, I would think. But Dave said it more eloquently.  (I saw his post above when I was about two-thirds through writing this one, so what the heck; I'm hitting the "post your response" button.)

Drbond, transformer theory is a deep subject. The only person with deep knowledge who ever posts on Audiogon is Dave Slagle (Intact Audio). Otherwise you are not going to find enlightened discussion here. I suggest you do an internet search. The subject interests me as well. For example, some who dislike SUTs like to lament the concept that the signal is actually traveling along the many yards of fine wire that make up the primary winding thereby suffering degradation even befor being transformed. But I think there is some basis for treating the primary winding as a singular entity. Dave could speak more eloquently on this but also you might find such a discussion on line.

My Johnson is very quiet without a SUT, speaking for myself only.

But seriously folks, Ralph (Atma-sphere), you have often mentioned the notion that high gain phono input stages need to be designed to be insensitive to RFI.  What design elements contribute to that good behavior?  Is it simply a matter of limiting the bandwidth?

"Are there any inherent weaknesses in a SUT versus a transistor, or vice versa?"

If I didn't know you were serious, I would suspect you of being a troll based on this sentence.  The question has already been broached here in this thread.  Elsewhere on the Audiogon Analog website and on every other audio-oriented website I have ever visited, the question is hotly debated with much expenditure of wind.  In the end, you have to make up your own mind, usually by listening.

"One practical consideration is you really need a very short, low capacitance run of cable from SUT to phono stage."  Which is a good reason for having a high quality SUT built in to the phono circuit, on board.

Raul, By "passive", I only mean that it does not require or consume external power.  OK?  You can use any other definition you want.

Of course to alter AC line voltage a step down transformer of the kind you own must be inserted into the AC line. That’s precisely analogous to inserting a SUT between a cartridge and an MM phono stage in order for it to do its work. Both are passive devices. You implied that your step down is per se dependent on AC for power. I hope the difference is clear.

Elliot, in operating principle there is absolutely no difference between your step down transformer and a SUT. The functional difference is only in the respective turns ratios, and also the fact that a SUT has to handle a wide bandwidth whereas the step down always works at one fixed frequency.

Holmz, please read my long post and Larry’s too, near the top of this thread. Of course transformers transform both voltage and current in direct proportion. The product of volts X current on the primary equals the product of volts X current on the secondaries. Don’t make the subject harder or more vague than it needs to be.

Which is to say a good SUT is preferable to a mediocre SUT, on the assumption that built in SUTs are mediocre. But what about Allnic and a few others with very high quality built in SUTs? The advantages there include fewer connectors and ICs.

"euphonic coloration".  Which is to say you believe SUTs introduce a coloration.  Which is one subject of that eternal debate I mentioned.

I personally have never owned a SUT.

Like Eric says, a transformer is a device that transforms the voltage to current ratio of a power source, in either direction, where Power, which is a product of voltage times current, is a constant. So when V goes up, current (I) must go down in direct proportion to the increase in V, and vice-versa. While that happens, there is also an effect on impedance, from one side to the other of a transformer. In a SUT, the output V of a cartridge is increased in direct relation to the turns ratio of the SUT (as in all other transformers). For example, a 1:10 SUT will increase cartridge output V by 10-fold. Meantime, the current output of the cartridge is likewise reduced by 10-fold. The effect on impedance, on the other hand, is related to the square of the turns ratio. So, for a 1:10 SUT, the impedance "seen" by the cartridge on the primary side of the transformer will be reduced by 100-fold in relation to the input impedance of the phono stage. A 47K phono load resistor on a typical MM phono stage will look like 470 ohms to the MC cartridge. As to the rest of it, that is the subject of eternal debate among audiophiles, and I won’t go there.