Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[soapfoot]

I've always been curious about something--

I know the BV8 transformer uses a U/I configuration. What are the advantages/differences of the U/I form factor over the E/I form factor in a microphone transformer?

[klaus]

Please explain for our esteemed readers what the abbreviations "U/I? and "E/I" stand for, and what they mean.

Thanks,

[Kai]

Explanation - the letters represent the shape of the iron parts of the transformer:

An U/I type transformer has two "legs" the sides of the U, that can carry individual coils.

On an E/I transformer there's only one place where coils can be applied, one wound on top of the other, on the middle line of the E.

On both types the I is mounted across the open part of the U or E after the coils are applied, to close the magnetic circuit.


If the primary and secondary coil each are distributed to both sides of the U (this means 4 coils total), externally introduced hum can be cancelled out in the U/I type.
Bit like the Hum-Bucker principle in electric guitar pickups.

This is probably the main reason Neumann uses this configuration in most mic's.
I have several Neumann brochures where they point out the fact that they use "Zweischenkel-Übertrager" ("two-leg-transformer", hence U/I type) to make the mic's insensitive to stray introduced hum, e.g. from power transformers.

Splitting the coils has another advantage: the HF resonance caused by the capacity between coil windings combined with the coil inductance can be shifted to higher frequencies.
Therefore it's easier to achieve a greater HF bandwidth.

Regards
Kai

[soapfoot]

For purposes of illustration:

EI transformer cores


EI transformer, completed


UI transformer cores


UI transformer, completed

[soapfoot]

Thanks, Kai, for that explanation, which makes a lot of sense.

[klaus]

As an example of a mic using both types of transformers (single-coil and humbucker) whose sounds I am familiar with, I'd like to bring up the early and later M49.


*The early M49 used a CT49/1 which had single primary and secondary coils, and which was originally designed for the initial tube in the mic, a Hiller MSC2, though the same transformer was used for a few more years after the switch to the AC701 tube.

*The next type, a predecessor to the final version, had a single primary but a split secondary

*The final version was the BV 11, with the same chamber and winding configurations (not specs) as the BV8 used in U47: two primaries and two secondaries- a version of the humbucker principle, though in guitar pickups the added reversal of magnet polarity helps with further noise reduction in that application.)


I have switched out transformers in the same M49 mics (which were otherwise left untouched for the test) to hear what the different building types of transformers contribute sonically.

And the difference is audible: The humbucker of the final M49 series is clearly supertior in noise and buzz suppression, and has an extra 4dB output (this is not inherent in humbuckers, but can often be found when the secondaries are connected in series, to get 200Ω output impedance). The BV11's sound is very even and balanced, with very good low-frequency extension.
However, the earlier transformers have a sex appeal of their own, despite the lower output, propensity to pick up (some) external noise sources like generators, neon lights and electric motor noise, and they do have a slight curtailment in the lows. The plus side: they are very fast and translucent, with reedy midrange processing reminiscent of (and complementary to) M7 capsules.

In the end, both types are equal in value to me as a sonic tool of high quality (the transition model is closer to the CT49/1 in sound)

[soapfoot]

exactly the type of answer I was interested in hearing. Thanks to everyone.

[Marik]

I've always been curious about something--

I know the BV8 transformer uses a U/I configuration. What are the advantages/differences of the U/I form factor over the E/I form factor in a microphone transformer?

Hello Brad,

It is hard to give a 'set in stone' answer as there are way too many variables in each of those approaches, which should be taken into account. That includes core materials, size of the core, winding techniques, ratio, etc., etc. etc.

In general, UI core is more efficient and because of the humbucking winding has better noise rejection (but the EI can also be wound with humbucking topology, or just use Mu metal shielding). They are more expensive because of the extra setup to wind a second bobbin.

Sonically, everything equal, the main difference makes the humbucking winding. Once I made an experiment, disassembled the UI transformer and then reassembled and reconnected it reversing the humbucking topology. It sounded differently--more open and transparent, with less fuzz and "thickness" in the lower mids. That's the difference I usually hear comparing the UI and EI, but again, it is hard to generalize because of many variables.

Many finest transformers ever made were done on EI cores, as well as there were many on UI, so in the end of the day only your ears can tell you what kind of sound you prefer.

Best, M 

[klaus]

Once I made an experiment, disassembled the UI transformer and then reassembled and reconnected it reversing the humbucking topology. It sounded differently--more open and transparent,

Can you explain that a bit more clearly: what exactly did you reverse?

Did you reverse the two (primary?) coils relative to each other, i.e. made the output of one the input, and the original input of the other the output, leaving both coils in series? Or did you just reverse in-and output of the series-connected primary (necessitating a reversal of the secondary as well)?

I use the polarity switching of output transformers in mics to tailor the overall timbre, as indeed reversing in- and output of the primary/secondary makes an audible difference, though no hard rules can be applied to determine which set up is better.

[Marik]

Can you explain that a bit more clearly: what exactly did you reverse?

Sorry, I should've been more clear, rather not reversed, but defeated (or eliminated).

In the original humbucking configuration the bobbins were connected as: (-) leads of Pri and Sec of the bobbin 1 to respective (-) leads of Pri and Sec of bobbin 2. Because of such configuration the induced hum is cancelled.

By reversing humbucking arrangement I meant physically flipping bobbin 2 180 degrees (that's why the lamination needs to be disassembled) and then connecting the (-) leads of Pri and Sec of the bobbin 1 to respective (+) leads of Pri and Sec of bobbin 2.

If it still unclear please let me know and I will draw it.

Best, M

[panman]

If it still unclear please let me know and I will draw it.

Hi Mark, just to be sure, that it really gets clear, would you please draw it!
Esa

[Marik]

Here we go:

[soapfoot]

Marik,

I know you make some toroidal designs, as well.  How would you characterize those with respect to both EI and UI conventional-core types?

[Marik]

Marik,

I know you make some toroidal designs, as well.  How would you characterize those with respect to both EI and UI conventional-core types?

The toroidal transformers are very different. They are the most efficient of all transformer types, and have the least of losses and distortions. As such they sound quite a bit more transparent and neutral (but by no means sterile). In comparison to EI and UI cores they are more dynamic and many customers describe the sound as fast and musical. Toroidal transformers main disadvantage is higher cost, as the production is quite a bit slower and more complicated.

Best, M

[marcus4audio]

Here we go:

Guys, I read this topic and wonder are the primary coils of the BV08 winded in same direction or opposite? Because if they are winded same way inductance of the primary will be much smaller (4times) if they are connected from one coil end to the other coil end. Correct me if I'm wrong :/

[Jim Williams]

Bill Whitlock of Jensen is the authority on this subject. Maybe he will read this and contribute.

Whenever coils are stacked there is a magnetic interaction, that can be lessened by spacing, wind geometry, etc. Whenever a coil is placed next to another coil, there is magnetic coupling. Rotating one coil 90 degrees helps minimize that. That is why speaker crossover coils are positioned at 90 degree angles, to avoid that interaction.

AKG is another mic manufacturer that aways used UI transformers with split, seperate coils. That didn't help them much, most of them don't sound very good, including those used in the 460 and 414 models.

[Oliver Archut]

Hello Guys,
The shape of the BV8/12 type transformer was specially developed by the NWDR and came in UI and LL shape (same overall size). It was first used in mic pre inputs (for 1:40 application), but was later implemented in mics too, first in the Hiller mics, then later in the M49 (1950 started the development), before Neumann started to use it in 1951 in the U47 too.

The lamination was only made by two companies, Herasus in Hanau as well as Krupp Stahl (in my hometown of Essen). Neither of those companies are still making that style of lamination, due to the fact that the production ended in the late 70s. 

Standard transformer lamination is made with little to no cut-off waste, the "I" are normally the cut outs from the two facing "E" or the "U" to be as efficient as possible, but efficiency is not always good for sound or better sound specs, so the NWDR worked closely with Krupp to develop a side ratio that yields the lowest amount of distortion and magnetic loss, but the side effect was a 40% waste. The heat-treating process was also a big secret of how that type of lamination was made.

The UI and LL transformer is not like a guitar pick-up 'Humbucker' in any form or way, but it does cancel noise differently than the pick-up does. This forum is not the best way to go into such technical details, but keeping it simple: the transformer has two legs, that are magnetically quasi out of phase. If both coils are wound the same way and then hooked up anti-paralleled, the noise cancels; this might sound similar how how the pick up cancels noise, but because no static magnet is involved, there is a bunch of additional issues that are too complex to roll out here. Aided by the incredible magnetic property, it is also quite easy to manipulate the frequency response (compared to EI types) just by arranging the windings in a different way, as well as stacking the lamination in sets of a 3: 5 ratio.
Some M49 as well as all U47 have a build in-low cut filter, a LC cut off that works with the coupling cap as primary inductance.

There is virtually no printed matter available on this subject, other than the NWDR "Hausmitteilungen" (internal communications, K.H.) and the Herasus spec sheets, both of them are in German and nearly impossible to find these days.

Neumann did several revisions of both transformers, but all of them work very similar. Trying to work on the transformer should be avoided, Neumann (they made all U47/M49 transformers in-house) used double-built silk wire that is then varnished, for lead-in wire. It crumbles away after 50 years of use in a mic, so if your transformer is still working properly, do not touch or play with it....

Best regards,

Oliver