Video

Valid questions. A frequency sweep shows all of the transformers to be completely flat, however this is deceiving, as you can clearly tell with your ears. Not only is a line output (which would drive such test) much lower impedance than an electromagnetic pickup, the output impedance is also flat while that of a pickup is curved. There is also the issue of transient response. I could attempt to model it all, but in the end I believe the ears are a better test. Transformers are one of those things that seem simple but get more and more complicated the deeper you look at them.

Good question. That is exactly what I expected and I had hoped to perhaps use any sound difference to advantage, but what I found in practice is what ribbon makers have known for a long time. As long as the ribbon is uniform (not piston corrugation), then the response is dominated by the mass of the ribbon and the resonant frequency. These can be equal between standard corrugation and herringbone, and they do not sound different, so long as the ribbons are tensioned to the same resonant frequency.

@@alderaudio that is interesting!!

The one place I know to point you to is retrogearshop. They've got a lot of desirable vintage transformers and maybe comparing to their photos can help with identification? They have a lot but are priced accordingly.

Hi Dirk, thanks for the feedback. I've struggled to manage the technical depth of these transformers while still making the content accessible. There is a lot to it! But I'll try to let you in on just a bit more in what I can fit here in a comment. The biggest huge difference in this case is that my transformer is toroidal, which comes with a whole host of changes. I've highlighted a few, but really the whole design just ends up different because you have different inductance per wrap, wire gauge, capacitance, just kind of everything. So my intention was never to clone, I just took the high inductance and low leakage away from the vintage model as guiding principles. As far as saturation we are WAY far away from that using just a pickup. I have intentionally pushed these both to saturation and it takes a decent bit of juice. As far as phase, I know Jensen likes to talk about that, which I assume is where you are coming from? Their point is that phase shift can occur in the low end as a result of the natural high pass filter of every transformer. What's the key to pushing that shift lower and lower out of audio range? Higher and higher inductance. So both of these perform really well in that regard. The TL-20 also measures lower in THD at 30hz.

@@alderaudio interesting, thank you for sharing!!

Big questions here, hard to answer quickly. First, unless you are running miles of cable the capacitance is going to be the biggest thing that that has an effect on the input. There's been a lot written about this and it's interaction with pickups that you can find and read online if you like. And as far as input stages, manufactures don't get to just set it wherever they want, the input impedance that is appropriate is linked to the design choices of the first stage. So if you have an op amp vs a FET vs a tube, things change. The input impedance options are limited by the overall design.

shots fired :) what's wrong with A? Won't that glowing midrange around 1k translate best within a full mix?

Well, you won't have impedance matching problems, so that part is sort of fine, but you will have level problems and possibly ground problems. Because both devices are referenced separately to ground you may get a hum. The input level will also be really loud. But you can turn down the level in your daw and give it a try if you like. It may work. If you have hum, you can probably solve it with a reamp box, which will also bring down the level.

@@alderaudio no ground problems here, but about the level, would the pad button on the preamp help, or is that not enough or at a different point in the circuit?

@@alvincornistamusic8754 the pad could help and is probably a good idea but really if you have no hum it's just a gain staging thing. All you need is to not distort the input on your preamp, so whether you want to use the pad, turn down the output on your interface, or turn down the source right inside the DAW, you can get it to the right level using any combination.

If you watch the video, his explanation say that impedance change by frequency. He didn't want it to be super technical so he didn't convey inductive and capacitive reactance.

I have realized over time that my statement in the video isn't clear enough to some people. I'm talking about phase in this specific context of electrical impedance, as in the phase angle difference between the voltage and the current waves. I am not at all talking about all the effects of phase cancelation that can happen with sound waves. That is very important.

Well, if you have the exact same input with switchable impedance, than in general I'd say you are not going to perceive any difference as long as the input impedance is at least 5x the output impedance of the device you are using. So really just equal above that point, not better. However, sometimes the high impedance input is designed with a different circuit which is not as well suited. For example, it will sound worse rather than better if you attempt to use an adaptor to plug an xlr mic into a high impedance 1/4" input. In your case, if you were choosing between DIs, I'd guess the other parts of the DI (the tranny or circuit) is going to dominate any noticeable difference in sound.

It looks like the VTB2281 is a quad coil 600:600 ohm transformer, sometimes called a repeating coil. The Western Electric 111C is a sought after vintage model wired similarly. You can wire them 1:1, 2:1, or 1:2. However this configuration is very different from the A9J, which is a ratio of 1:12. In this DI context we are actually using it backwards as 12:1. The need for the big ratio is to adapt the high impedance bass guitar signal to a lower impedance XLR input. So very different applications and not interchangeable, just on the ratio alone. All the other specs like the inductance and DC resistance would also be very different between the two based on their intended use.

Thank you for taking the time to explain that to me, very helpful.

Thanks for the question, I can understand where your points come from. Perhaps in the future I will make a longer video with less hand-wavy explanations of some technical details. Both ribbon mics and dynamic mics have a large impedance peak in the low end that is caused by the resonant frequency of the element. The same thing applies to a loudspeaker, which is very much like a dynamic mic. (Google "loudspeaker impedance" and you will see the graphs with the hump in the low end, often accompanied by poor explanations) Because that peak is often somewhere from 20-100Hz, I generalized that impedance goes up as frequency goes down. In reality, it goes up, reaches a max at resonance, and then goes back down again below the range of human hearing. Here is a stack exchange that has a good discussion on the subject: electronics.stackexchange.com/questions/341595/understanding-loudspeaker-impedance Condensers are a different matter, as they have a powered circuit, they were not my focus for this video, which I could have pointed out. I build ribbons and my experience is mainly rooted there.

I kind of want to, but as I am a maker and seller of products, including now a boutique DI transformer, I try to treat my competition the way I'd like to be treated. So I don't publish names of comparisons. However, all three do have a nominal ratio of 12:1. Also, IMO Impedance stats on transformers are misleading. They are not actually a measurement of the transformer, but rather a statement of what impedance circuit the transformer is designed to be used with. Inductance, which hardly ever published, is a far more useful stat, and transformers listed as having the same impedance can have very different inductance.

@@alderaudio I wish you luck in your business. I mean that. Thanks again, it was a very useful video. Yes, the inductance (permeability of the metal) is a great specification to look at and most TX makers don’t publish this. But so is density, and they don’t publish that either. I asked, cause you published the name of Triad, which still makes the their famous transformers. Why not publish Jensen and Edcor or Cinemag and Neutrik or Hammond and Chinese off eBay, too? Oh well. When people do a comparison, company or not, they usually tell you who the competition is. The biggest difference seemed like a comparison between a transformer with higher nickel content vs higher steel content, but since you didn’t publish the names and models of all three, we can’t know for sure. Nickel is clean, Iron is colored and steel is crunchy/dirty. All are useful for different purposes in the audio world. Honestly, the Triad A series was a more modern transformer design, probably higher nickel content, right? It still rivals the clarity of current offerings, like Jensen and Cinemag. You should compare an older design and maybe USA vs UK, German, Russian, or Japanese.

You have a good point in that Triad does produce a remake. I had thought of it as a vintage thing and therefore not active competition. If you really want to know what they are I'm willing to tell you over email. Shoot me a message at tyler@alderaudio.com if you are interested. As far as inductance and cores, I don't tend to think there is as much in the idea of different metals sounding different as some believe. The inductance doesn't come from the core alone but rather a combination of the magnetic permeability of the core and the number of wraps. The number of wraps will change everything. So if you try to compare an iron core to one with some nickel content, you will need a ridiculously different number of wraps to get the same inductance between the two. Those wraps come with different resistance and distributed capacitance and whatnot, so I tend to think the difference in the sound is more dominated by those side effects of the material that change the overall design rather than something about the sound of the material itself. Saturation is also a big factor that is also changed by both material and wrap count. And yes, anything small enough to fit in a small box that can function as a good DI transformer is going to need very high inductance and thus a high permeability core such as 80% nickel. @@dickdixon6409

@@alderaudio those are all very good points, I see why you think of the difference in tone and saturation in all of the changes you have to make to a design to accommodate a new Metal with a different permeability, density, etc…. Thanks again, it is a very fascinating subject. I will email you.

Most measurements were done with an LCR meter at 100hz. Selectable measurement frequency at 100hz is important because higher frequencies get you into the resonance region of the transformer and give erroneous results. Primary impedance, Q, and DC resistance are measured on the leads of the primary with the secondary open. Leakage is measured the same with the secondary shorted. Mutual capacitance is measured by shorting both sides of the transformer, connecting guard to ground, and connecting meter clips one to each set of windings. Mutual capacitance is requires an oscilloscope, signal generator, and a calculation based on resonant frequency. I also used an oscilloscope resonant method to verify primary inductance.

It's not. Impedance isn't ANYTHING like explained here...