It would be better to get a balancing transformer. Look at Jensen Transformers.
49 responses Add your response
If you are confident you want the combination, keep your single ended cable and just have the ends going into the amp reterminated with XLRs. it obviously will NOT make this a balanced setup...but you wont have to fuss with adaptors. i have a CJ pre and a Gryphon Antileon amp (balanced only)...and i just had my Transparent Ref's reterminated at one end. i had been using Neutrik adaptors provided by Gryphon...and they seriously encouraged me not to use them long-term. After finally getting the CJ pre i wanted (and being confident i am keeping it)...i got the cable reterminated. good luck
If you are going RCA to XLR because you like the 2 components, then many people would say reterminate your RCA cable on one end...in order to get rid of the adaptor.
As for SE vs Balanced, yes many people feel there is inherent superiority of Balanced over SE technology...however, as with many things in life...it comes down to execution. A well executed SE will still sound better than a poorly executed balanced.
There are so many considerations relative to the specific gear being used that IMO "one size fits all" answers do not apply. My single-ended Tom Evans Vibe/Pulse II is far and away the quietest preamp I have owned (out of many) and running it through 2.5M rca cables into Jensen PI-2RX (rca to XLR) transformers then through very short XLR cables into balanced amps results in the best sound I have achieved in many years of doing this.
Mitch2, Rwwear, and the rest of the community: Supposing the amp in question is a nice... a VERY nice (!), little vintage EL-84 job, single-ended, of course... and you want the cream de la cream minimalist input/driver circuit for it (DIY) -- you'll need phase-splitting somewhere (at the input, or maybe even as a parafeed phase-splitting "driver" transformer, given the low voltage swing those EL-84's are looking for) -- how would the step-down versions of the Jensen transformers sound vs. the normal 1:1's? I note that 1) These are rated to handle some pretty high voltages, if need be, 2) They would present a MUCH more source [output stage] friendly load used at the input, and 3) the specs on paper far exceed those for the 1:1's. Has anyone auditioned the difference in some context?
Off hand, I don't know the input sensitivity of Podolaw's BAT, but with the tube preamp, he should have some voltage swing "to burn" with a typical Red Book spec output on the source. Then, too, advancing that volume control usually improves both channel balance tracking and the quality of sound. Does this make some sense to anyone else?
I will answer this by saying the folks at Jensen are available and IME very helpful in answering questions about the use of their transformers. They are very good at determining which of their products best suits your goals and then explaining the anticipated result of using the product. I would call them if you are serious.
Regarding channel balance tracking and sound quality improving as you advance a volume control, these may be two different things. The channel balance improvement should only be relative for volume pots (e.g., potentiometers) and should not be affected if the volume control is resistor based or a chip control. However, I suspect there could be instances where sound quality improves as the preamp is operated at higher gain (or less attenuation) because the associated componentry is operating in a more linear range resulting in a lowered noise floor and improved signal to noise ratio. This is one benefit the Rothwell people speak to when promoting their 10dB attenuators. I suggest the EE's or designers chime in on this issue since I am in the deep end now.
Yes, the pro to consumer Isomax gives you the step-down ratio (so as not to have higher output pro gear overloading a commercial consumer unit with no volume control at the input). I'm assuming they'll make this up for you with any combination of connectors you want -- maybe with your own choice of ratio as well... It wouldn't hurt to inquire. If this route sounds good to you, do compare the data pages. Folks with rolled off (up to about -3dB at 20K) NOS chip DACs, or RIAA EQ's down about one dB at 20K might well prefer not to stack another transformer in the signal path whose FR starts to roll off before 20K. I saw step-down options on site rated to 60k, and even 180K, depending on the circuits that bookend the transformer.
One more thought: If you can get custom as easily as the advertised stock items, you might give some thought to eliminating two XLR connectors by having a flying balanced lead hardwired into the Isomax unit at its output. Perhaps Jensen could accommodate you with your lead of choice if you provide 'em with the desired bits; or tackle that end of it DIY.
Another thought, though I am not the transformer answer guy, I do note that these Jensens (and most?) input transformers of this type want to be loaded by 10K up to 25K for these applications. Your preamp may not like to see this kind of load reflected at a 1:1 ratio -- and don't just go by what someone says "will work": distortions, conventionally measurable or not, tend to increase into lower impedance loads, even with [tube] preamps presenting an output impedance of only several hundred+ ohms. So check this out carefully.
Then, too, I'd guess the BAT's input impedance would need to be modified one way or another -- and IT may not like this change. An advantage of the 4:1 impedance step-down ratio shown for the pro to consumer Isomax is that it takes a 10K impedance across its secondary, and reflects 40K back to the preamp output. All that costs you is some overall available gain in your amplification chain; you probably have a surplus of that on tap as things currently stand.
No, the transformer approach is less than elementary when tube impedances are involved.
Oh, and one more thing... You COULD get a cathodyne (split load) type phase inverter pair made up as an adjunct to the inputs on the BAT. This would get you past the issues I outlined above, at the expense of more circuitry (never a free lunch...). You'd need a power supply for that circuit (not too complicated, in this case), and it would also have the downside of needing a very small value cap at its input so that the circuit can be biased. This sort of thing is also known in a somewhat different context as a bridging adapter. Something in the 6DJ8/6922/7308 range would do. Some feel even lower impedance tubes are better for this (5687, ECC99, D3a triode connected, et.al., down to the circa 800-1200 ohm Rp jobs -- there are various choices for the latter, but their characteristics tend to be all over the place, necessitating tube selection.
Astronaut Norm Thagard designed solid state versions of this in order to bridge FOUR (or was it EIGHT?!) H-K Citation II's for his monster tube system. He published a detailed write-up in 'AudioXpress' (then 'Audio Electronics', I believe), the DIY magazine. It all worked out great, reported Norm.
So, it can be done... Are you sure you want that BAT?
If you are not looking at replacing your preamp with a balanced unit anytime soon, then yes, going with the Jensen transformer would be the best route to take and provide the best results IMHO.
There are advantages of running bal. to bal. (given both circuits are true balanced)but this is not to say that the conversion would not still provide very good results. Regardless of the route you take, you should not have any impedance mismatch as the Rogue has a very low output impedance and most BAT amps have a high input impedance.
I still think the transformer route will still give you a better sound but that is only my personal opinion.
A bad idea. Balanced amps require an inverting and a non-inverting signal. I know trying to do that with an ARC amp results in half the amp output and higher distortion because you are only driving the amp with a non-inverting signal. The inverting side of the signal doesn't exist so that part of the circuit is not being driven. ARC did at one time make a powered device the took an RCA single ended input and gave you a real balanced output. You might want to consider something like that, or a good used balanced output preamp.
I apologize for hijacking the thread, but I have a question:
What if the scenario was just an opposite- balanced preamp and single-ended power amp?
Would the Jensen type converter still be the best solution?
I understand, that if I used simple XLR/RCA adapter, then I would only use "half" of the balanced preamp circuitry?
I had Rogue Magnum 99 preamp and BAT vk200 for a month or so, but soon I replaced Rogue with BAT vk30.
BAT/Rouge (with RCA-adapter) worked fine (an improvement over Linn/Rouge which I had for a couple of years), but I just could not get away with the idea that all balanced approach would be better.
I like vk30 over Rouge except the phono. The phono card from Rouge sound better for me than the vk phono card.
Later, I got an RCA-XLR cable (from AudioQuest) and it worked better than with the adapter.
Now I have vk200 and vk60 with vk30, biamping Tyler Linbrook signature systems.
"When one considers the noise cancellation properties of balanced, then you can easily see the reason for superior audio. Less noise equals better sound. "
I auditioned Joule-Electra LA-150Mk2, then I owned LA-300ME for an year and now I am blessed with LA-450ME. All these preamps have balanced amplification in the gain stage but (except 450) their inputs and outputs are single ended i.e. Joule-Electra split single ended on inout and add on output.
As much as I tried to hear any (AMY!) noise when I place my ear to the speakers I hear nothing, nada, ziltch, zero etc.
I cannot remove noise I do not hear !!!!! I am sure theere are other SE preamps without producing noise and one shoyld be careful what kind of IC he has
Most "audiophiles" love word "balanced" without realizing the immense difference between balanced input/output and balanced signal processing (amplification here) which cancel not so much noise as DISTORTIONS !!!!!!!! and this make immense difference in distortion-free music presentation.
08-22-11: Bob_reynoldsBob, that's an excellent question.
I'm not entirely sure of the answer, but I believe it relates to differences in inter-winding stray capacitance (i.e., between primary and secondary) of output transformers vs. input transformers. I've noticed that the schematic representations of the Jensen input transformers that I've looked at all show Faraday shields, which would greatly reduce that stray capacitance, while none of their output transformers that I've seen include those shields. I have no particular knowledge of why that may be; presumably it has some relation to the fact that output transformers have to be able to drive much longer cable lengths.
Lack of a Faraday shield in the output transformers results in very substantial inter-winding capacitance. For instance, the this output transformer is spec'd as having 22,000 pf of capacitance between windings, as measured at 1kHz.
A result of that capacitance will be that the input stage of the amp would see, "looking back" at the output of a preamp + output transformer combination, an impedance which is very unbalanced between the two legs at high frequencies (since one side of the transformer primary is connected to preamp ground, while the other side is not). That imbalance will significantly degrade the cmrr of the amp's balanced input stage, at the high frequencies for which that capacitance becomes significant.
Hopefully someone more knowledgeable about transformers than I am will comment further, but I suspect that the explanation is along those lines.
Yes, Al, I think you have it right. Bob, the up to 60K bandwidth I mentioned above was for the Jensen output transformer, and the out to 180K was for the step-down input version -- with Faraday shield. For more on the benefits of input transformers, see Stuart Yaniger's DIY buffer preamp article at DIYaudio.com. Stuart [sp?] graphically shows how the Jensen input tranny not only lowers input noise, but also how it lowers his simple circuit's distortion. Interesting... almost like magic. This treatment is relatively in depth, from an audiophile's perspective; but not overlong.
No one's mentioned the other advantage of deriving your balanced (or quasi-balanced, if you prefer) signal at the BAT's input -- the simple practical advantage of using the far more common unbalanced
interconnect from the preamp... more common for Audiogon shoppers and experimenters, anyway.
Hi guys, I just had a chance to check in and see Bob's question about the input transformers. I believe these minimized any impedance differences, as long as you keep the balanced IC's from the transformers to the amps very short. Jensen says 1M or less, but shorter is better. Of course the ideal approach would be to mount the transformers internally with RCA inputs to the amps.
Between talking with Jensen and going over their literature, I remember that the input transformers were clearly the better approach for the rca/XLR interface. Virtually no sonic penalty with regards to measurements (I believe some very small input loss - like 1.5 dB) which I cannot detect even when comparing to balanced preamps. They work with loads as low as 10K ohm and I believe the impedence issues are somewhat related to the load, and to the transformer to amp cable length. My highish 100K ohm input amps result in an optimal impedance situation. The transformers also provide noise rejection benefits. Here is a link to read,[url]http://www.jensen-transformers.com/datashts/pi2xx.pdf[/url]
The guys at Jensen and the associated documentation do a better job of explaining than I can. Compared to giving up equipment you like the price of trying the transformers make them a stupid easy decision. I remember initially trying the output transformers based on someone's suggestion but they were noisy and didn't sound good, I called Jensen and they told me "you want the input transformers." They swapped me for the input transformers and I later purchased their mono version to keep cables very short - mine are like 2 feet long from the transformers to the amps. Call Jensen, they are very friendly and knowledgeable.
Once again you raise excellent questions.
In the absence of a transformer, the cmrr of an active balanced input stage will generally be degraded significantly when driven by an unbalanced source. Good cmrr requires a close match between, among other things, the input impedances of the two legs. Since the output impedances of whatever is driving those two legs are in series with those input impedances, and since an unbalanced source has an output impedance of essentially zero on one of the legs but not the other, an impedance imbalance will result.
A transformer will not be subject to that effect, as I understand it. It will provide an output that (for signals whose characteristics fall within the limits of what it is designed to handle) is simply proportional to the instantaneous voltage difference between its two primary terminals, regardless of source impedance differences.
What may be important, however, if an input transformer is used in conjunction with an unbalanced source, is assuring that pickup of noise, rfi/emi, etc., is as common mode as possible (i.e., as equal as possible between the two legs). That means that even though the source is unbalanced, a coaxial or other non-symmetrical rca cable preferably should not be used. Preferably a shielded twisted pair or other cable that has symmetrical signal and return conductors should be used, terminated with rca connectors, with the shield grounded at the source end.
The degree to which the cable configuration may be important, though, would be dependent on whether the primary contributors to noise are ground loop effects, or rfi/emi pickup. Several of the Jensen writeups suggest that more often than not the former is the more significant problem in home audio systems that have unbalanced interfaces. To the extent that that is true in any given system, it seems to me that input transformers and output transformers, even if connected with non-symmetrical rca cables, may be comparably effective, since either one will break a ground loop. Although in the case of an output transformer, the high inter-winding capacitance I mentioned in my previous post may lessen its effectiveness with respect to ground loop-related noise that is at high frequencies, since that noise may to some extent simply couple through the capacitance.
Fine job again, Al. Bill Whitlock and team have documented transformerless input receiver schemes that improve significantly on the all too common poor designs out there in the pro world (with respect to the issues Al has raised); but these would not normally be considered too pretty from the audiophile standpoint. I vote as well for a close reading of the Jensen literature, even if you are not a prospective transformer user (unless you're already a world class expert on grounding issues in sensitive systems -- this stuff is beyond elementary). Check the K & K Audio website for links to Dave Davenport's detailed discussion of similar issues: a second, somewhat different perspective on the same problems can throw some very helpful extra illumination on the important concepts. And again, Mr. Yaniger's piece is on point here.
As Al discusses, perhaps the cables also make a difference. My rca IC's are a balanced (they call them that when both legs are the same) twisted pair type shielded cable from Furutech made from PCOCC copper. There is literally no noticable noise from the speakers at virtually any typical volume level, unless music is playing.
The reason you use balanced lines is lower noise and lower distortion, plus the opportunity to eliminate the sonic role that the cable has in the system. BTW, length (or shortness) of the cables has nothing to do with it!
Now if you are using a single-ended source with a balanced amplifier, you will not be able to realize this advantage. Thus comes the idea of a transformer. You would use the transformer directly at the output of the source, keeping the cables between the source and the transformer as short as possible. However the transformer (including the Jensens, which are excellent) will have a sonic artifact which is easily heard, even if properly loaded. If the transformer is kept as close to the source with the cables to it as short as possible, CMRR degradation issues will be minimized.
Now, if the main interconnect cable can be short, you are better off just making the connection with a single-ended cable. The amp, BTW, is not going to give a hoot if it has differential inputs, and the BAT does.
But if the cable is to be very long, over 20 feet, then there will be an advantage to using the transformer solution (assuming an excellent transformer), despite the artifact it introduces, because that artifact will be *less* than that of a single-ended cable being run long distances like that.
Mari555 asks about the other way 'round, where the preamp is balanced and the amp single-ended. There are several ways to do this. The most elegant is to modify the amp with a balanced input, and even though the amp is single-ended, it is possible to set it up so that the amp uses both the non-inverting and the inverting connections of the XLR.
The second method is to use an input transformer that does the conversion.
Finally, an adapter can be used to good effect, and while this technique is cheap and easy- you get what you pay for- it does not take advantage of the balanced output of the preamp and so the resulting sound will not be as good.
Just FWIW, we built the world's first balanced line preamps. We've been answering these questions for a long time.
For Atmasphere: Your knowledge, helpfulness, and long experience is much appreciated here. If you follow the set-up in your second paragraph above to the letter, Mr. K.,(and I highlight your spec of short cables in this case), are you better off using one of the Jensen input transformers, for some of the reasons cited earlier in the thread... or should you employ the line output type? I allow that the answer may be different, depending on the preamp in question. Thanks.
Now if you are using a single-ended source with a balanced amplifier, you will not be able to realize this advantage. Thus comes the idea of a transformer. You would use the transformer directly at the output of the source, keeping the cables between the source and the transformer as short as possible.Ralph, with all due respect, this solution did not work for me using the Tom Evans Vibe/Pulse 2 single-ended preamp into Clayton M300 (balanced input only) monoblocks. It is also contrary to what I was told by Jensen, who recommended using the input transformers next to the amps with very short balanced cables to the amps, which is the solution that has worked very well.
However the transformer (including the Jensens, which are excellent) will have a sonic artifact which is easily heard, even if properly loaded.In my case, where my preamp output impedance is very low (12 ohms) and my amp input impedance is 100K ohms, "easily heard" would be an overstatement, better replaced with "barely heard" or perhaps "negligible." Also, the acceptance of this solution can be dependent on the gear. In my case, based on my desire to stay with SS devices, the TEAD preamp with transformers beat every balanced (SS) preamp (without transformers) alternative I have tried to-date.
Al, you're exactly on the right path regarding the performance of input vs. output transformers. Put another way, output transformers require tighter primary-to-secondary power coupling to maintain low output impedance, and the resulting low distortion and low noise figure. A side-effect of this is the necessity of a larger core, and higher leakage reactances between the primary and secondary. Input transformers on the other hand need to transfer very little power, and so can make effective use of Faraday shields and have lower leakage reactances, at the cost of the requirement of needing carefully controlled secondary impedances for good performance. But in both cases, the distribution of these reactances across the two windings can be controlled in the design of the transformer, and frequently an input transformer will work best with its secondary grounded on one side, or the primary with an output transformer.
A huge complicating factor is the fact that the design and performance of "balanced" inputs varies wildly in high-end audio . . . I would divide them into two "worlds", depending on whether the circuit after the balanced input is balanced differential, or conventional unbalanced. Both have myriad potential design issues.
The main issue with balanced-line-to-differential-circuit input stages is that most of them really offer no common-mode rejection at all, that is, a common-mode voltage on the input translates into a common-mode voltage on the output . . . hopefully (but not always) the common-mode voltage gain is less than the differential mode. The result is that any tiny gain or impedance imbalance within the equipment or cables (and in the following equipment, if it's of similar design) will result in the common-mode (noise) voltage becoming differential-mode (signal) voltage. It's also frequently more suceptible to RF interference than an unbalanced input (there are two input paths), and under no circumstances will the circuit work correctly if fed from an unbalanced source. An input transformer can improve things tremendously on all fronts.
The problem with balanced-line-to-unbalanced-circuit input stages is usually that many of the simpler designs have an impedance balance that's maintained by the open-loop gain of the input circuit, and the critical matching of resistors and circuit trace capacitances . . . and since this is never perfect, the CMRR is poor and usually falls as frequency increases. This can be improved by the buffered "instrumentation opamp" topology, but still all of these approaches almost always result in increased noise over an unbalanced input, as a result of the Johnson noise in the resistors forming the differential subtraction. Here again, a high-quality input transformer almost always performs better, especially because input RFI networks aren't required.
When interfacing with source imbalances or an unbalanced output, CMRR is determined by the ratio of the differential input impedance to the common-mode impedance. In the overwhelming majority of both types of input stages, the common-mode impedance is one-quarter that of the differential-mode impedance, making the impedance balance VERY critical, with very little noise rejection from an unbalanced source. There are two ways of dealing with this . . . raise the common-mode impedance, or lower the differential-mode (signal) impedance. Atmasphere advocates the latter with a 600 ohm terminating resistor . . . the obvious disadvantage is that the overwhelming majority of equipment on the market will perform more poorly into the lower impedance load.
Transformers do the opposite, they raise the common-mode impedance . . . which is why they still work well from an unbalanced source. Input transformers will generally have a higher common-mode impedance than output transformers as a result of the lower leakage reactances mentioned above.
Seakayaker, the issue with the use of a transformer is one of loading. If the transformer is not loaded, it may well be that it will express the inter-capacitive qualities of its windings rather than the turns ratio of the windings.
IOW the frequency response could go to hell. Loading of the transformer is paramount if you want to get the most out of it.
This type of application would normally use a line out transformer.
Mitch2, I appreciate that with some equipment the artifacts of the transformer will be less audible. The area I hear the most problems is in the bass- my speakers go down to 20Hz so if the bass is drying up due to low frequency phase shift its pretty audible. That is why I prefer to direct-couple.
Kirkus: just to set the record straight, the reason I advocate 600 ohms is not to improve CMRR (we have pretty good numbers there already as our gear is fully differential with very effective constant current sources) but to swamp out the inductive, resistive and capacitive effects of the interconnect cable, rendering it inaudible while the signal remains intact. I do agree and it is one of my frequent beefs that most high end audio products do not support the balanced standard and so fall right flat on their respective faces when trying to drive 600 ohms.
Just a couple of quick points here.
First - the notion that only one-half of the internal circuit will be used when supplying the single-ended signal is total nonsense. While it indeed might be true in case of some (I presume rare) products, all BAT preamps have differential inputs and internal circuits.
Two - converting the single-ended signal into balanced with a transformer most likely will be counterproductive, as the preamp itself will do that conversion just fine - see item #1.
Three... balanced circuits have other advantages besides the interface noise immunity. A very important one is the nature of interaction between the gain stages and the power supply - it is many times easier to build a **good** power supply for the balanced circuit than it is for the single-ended one.
The adapters will work great, your loss will not come from them, but from the fact that you will not be using your preamp to its fullest potential.
Sorry Atmasphere, I didn't mean to imply that CMRR was the raison d'etre for your inclusion of the termination resistor. It usually only shows a handful of dB improvement for the CMRR in practice, sometimes a little more at higher frequencies. A certain REL subwoofer input circuit comes to mind, and this is a single-opamp differntial input stage.
I do think that it's highly inaccurate to characterize line output transformers, as a group, as requiring a substantial or specific load on the secondary to acheive proper transient response. While there have certainly been vast quantities of poor-quality transformers produced over the decades, tolerance of a wide range of secondary load impedances is one of the fundamental differences between output and input transformers. On a modern line output transformer, it is in fact the tight coupling between the windings and their higher inter-winding capacitances that allow the primary to provide damping for the secondary. This of course requires that the primary be driven from a very low impedance source, as indicated on transformer datasheets.
Sonically, I feel that the best input transformers are virtually transparent, with the main audible artifacts a result of huge, measureable improvements in bandwidth control and noise rejection. For output transformers, I think I usually hear do have a very slight discernable signature, but it tends to be mainly at low frequencies, and frequently a rather welcome coloration. But one example of the astounding potential of output transformers can be found in any of the Audio Precision analog-based generators . . . maybe as the patents for this famous circuit near expiration we might see something similar in an audio product that we actually listen to.
I do think that it's highly inaccurate to characterize line output transformers, as a group, as requiring a substantial or specific load on the secondary to acheive proper transient response.
All of the line output transformers I have ever seen need some kind of load, but its not so much for transient response as for flat frequency response.
All of the line output transformers I have ever seen need some kind of load, but its not so much for transient response as for flat frequency response.Proper transient response and flat frequency response of course go hand in hand, especially in the HF/ultrasonic region, where a poor quality transformer tends to resonate.
But for a bit of perspective, I pulled out a John Hardy M-2 mic preamp and ran a couple of sweeps. This uses a Jensen JT-11-BMCF output transformer, and its response deviation between 150, 600, and 100K loads were all within +/- 0.1 dB of each other from 20Hz to 50KHz. Measuring the AP System One itself (Bruce Hofer's patented transformer output circuit) delivers virtually indistinguishable performance into all three loads (less than +/- 0.01 dB of each other) from 20Hz to 200KHz.
Pretty good for a bit of iron, eh?
Talk2me, several references to your amp that I've seen indicate that it has rca as well as xlr inputs. Is that not the case with your particular amp?
In any event, the reason you didn't find an explicit answer as to the best way of interfacing rca outputs to xlr inputs is that there is no answer that is universally applicable and universally agreed upon. Relevant variables include cable length, component impedances, susceptibility to ground loop effects, rfi/emi environment, deep bass extension of the speakers, etc.
In your case, I found a post somewhere indicating that the balanced input impedance of your amp is an unusually low 2,000 ohms. If so, that probably rules out the Jensen input transformer approach, and limits you to a choice between a Jensen output transformer, such as their model DM2-2RX (the "RX" version is the same as the "XX" version referred to in the data sheet, except that it has rca input connectors), or the adapters you are presently using, or adapter cables. As Mitch2 suggested, you should speak with a Jensen rep before finalizing a choice of one of their transformers.
See section 2 of this paper for further information. Note that for all three of the methods that are illustrated the cable has symmetrical signal and return conductors (i.e., it is not a coaxial unbalanced cable).
It is indeed! Did you get a chance to check the bandwidth at 10Hz and 5 Hz also?I just ran an existing scripted procedure on the M-2, and it started at 20Hz. The Gen-Mon results I did manually, but I neglected to change the LF end of the sweep. And the System One only goes down to 10Hz . . . for subsonic results I have to use a function generator, then measure with a 'scope . . . big pain. THD results are even tougher; I need to use a DSO sync'd to the generator to acquire the waveform, average 64x to reduce noise, perform an FFT with a rectangular window sized to the exact generator frequency, then calculate the THD from the sum of the harmonics. Ugh.
Jensen's datasheets are probably more accurate at subsonic frequencies than my measurements . . . Deane Jensen was actually a pioneer in many of the techniques used to acheive accurate results in this region.
Hi Talk2me . . . Al managed to find much more information about your amp than I did, and his advice is spot-on as always. If the Burmester amp indeed only has a balanced input with a 2K input impedance, then there's a chance it's only a 1K impedance with their adapter, depending on the design of the input stage. It'd be a good idea to get Burmester's recommendation, and verify with the preamp manufacturer that it's comfortable driving a 1K-2K load.
Another transformer manufacturer that deserves recommendation is Lundahl, and the owner Per Lundahl has been very responsive to technical inquires in years past. They manufacture some input transformers that can be configured in many different ways . . . some configurations could work well for a small step-down ratio to a lower-impedance input such as the one on your amp, and present a less demanding load to the preamp. I think his offerings in pre-packaged plug-in solutions may be a bit more limited, but definately worth checking into.
Kirkus, yes that does sound like a pain.
I've always been a fan of bandwidth, well below 20Hz and well above 20KHz to maintain proper phase reproduction in the audio passband. Both areas can be hard for transformers (which is why I prefer to direct-couple) and is why I usually look into the loading issue if transformers are to by used; its all about out-of-band extension.