# Help understanding Damping Factor and Impedance vs Frequency

Discussion in 'Sound Science' started by shrimants, Jul 17, 2018.
1. I bought the Massdrop Plus IEM. 10 ohm impedance at 1khz. Apparently it gets lower and higher than that, because frequency affects impedance. I dont understand WHY frequency affects impedance.

My source is currently an LG V20. Output impedance at 6 ohms. Does frequency affect output impedance? I'm pretty sure load affects output impedance.

And then the final major question: What is damping factor, and how does it play in to all this frequency vs impedance vs output impedance stuff?

2. If you looked at a graph of impedance vs frequency you may see how impedance changes. There are several mechanisms that can alter impedance with frequency. Basics are inductance (impedance rising with rising frequency) capacitance (impedance dropping with rising frequency, and resonances (electrical and mechanical) that are a combination of inductance, capacitance, resistance, mechanical and acoustic factors. Impedance curves are different for every headphone or IEM.
Generally, no. Most solid-state output circuits have fairly flat output impedance curves. The same is not true of valve output circuits.
No, load does not affect output impedance, but the two do interact. In the most simple terms, the source Z and load Z form a voltage divider. If the load Z is 8 to 10X the source Z at any point in the curve then the load has little effect on the applied voltage at its input terminals. However, if the load Z is lower than 8X, or in your case where the source Z is 6 ohms (how do you know that?) and the load is a nominal 10 ohms, that's enough that the load will affect the applied voltage at it's input terminals, and if the load Z changes significantly with frequency, then you'll have a frequency-dependent voltage divider (another name for an equalizer of sorts). This is not necessarily bad, though, as the IEM designer may anticipate the situation and even include the effects in his intended result.
Damping Factor is the load impedance divided by the source impedance. So a high damping factor occurs when the source is much lower Z than the load. The idea is that a transducer with a signal applied is essentially a motor driving a mass. But the inverse also occurs, there the mass has inertia and resonance, which drives the motor which acts as a generator. When the back-EMF generated is driven into a very low impedance, the system becomes "damped", with the amplifier's source Z acting as a load, which controls unwanted movement in the driver. The fallacy is that it's really impossible to get an extremely high damping factor at the driver because, in the case of speakers, there's wire and a crossover effectively raising the source Z of the driving amplifier as seen by the driver, and in headphones, the resistance of the wire becomes a factor.
There are some conditions where an undamped resonance in a transducer may be effectively damped by a high damping factor amplifier, but in practice those cases are fairly rare. It doesn't change frequency response much, other than the fact that to have a high dampling factor the amp must have a much lower source Z than the transducer load, which in turn, makes resulting voltage vs frequency variances caused by the impedance vs frequency curve of the load quite minimal. Again, that may or may not be what the designer intended.

Remember, you have to consider the entire system, the headphones/IEM, the wire, the driving amp, and any mechanical or acoustic resonances. Looking at the numbers is a starting point, and in some cases, the ending point too...just not all.

3. To scale the importance of high damping factor, notice that the amount of back-EMF that a transducer generates is a function of physical characteristics, in particular, the mass of the transducer's "motor" or diaphragm. For large drivers like woofers and subwoofers where things get quite massive, the acceleration and deceleration of that mass creates a high potential for back-EMF, whereas the masses involved in small headphone and even smaller IEM components are quite small, as is the resulting back-EFM that may require damping. This difference is scaled again against the impedance ratios of the amp and drivers, which may or may not be smaller than that of a speaker and power amp. So the obsession with high damping factors, in both speakers and headphones, is largely over rated.

4. That makes more sense. I also struggle to understand why the HD650 and bottlehead crack is touted as such a good pairing when you'd think damping factor would cause roll off. But im guessing it has to do with the specific frequency it causes a roll off in. I think for most examples i read, they always said "bass rolloff" but never took into account that the impedance is different for different frequencies of the driver.

5. Contributor
what @pinnahertz said, with a warning about the Massdrop Plus IEM which is exactly one of those gears where almost anything can have an audible impact(by that I mean the variations can reach a few dB depending on the source). not much because of the drivers being underdamped or not, but purely because of the impedance over frequency of the IEM directly affecting the signature. anything but super low impedance is going to result in upper frequencies being quieter compared to the low end. and the higher the source's impedance, the warmer the IEM(based on Marv's measurement).
in this case, the crossovers between drivers seem to be the main culprits for the massive impedance changes over frequency in that IEM.

about hd650, when you look at the impedance curve, you can see that aside from a little bass bump(around 90hz on my pair), you're not going to get much from using a high impedance amp. given that we assume the amp to have everything pretty flat and stable no matter the load.
which is probably not the case with the Bottlehead Crack? OTL amps have a lot of fans in the hd650's topic, at this point I have no idea if people run after the fame or if the fame is generated by happy customers, but here it is anyway. add the affordable price that will get a lot of people to try "because at this price I'm not really taking a risk", and I see why it can be tempting.
of course you must never forget subjective preferences. if we were to talk about objective fidelity, I doubt that this would win against some very average solid state amps. but if people like the sound, of course they should use what they like. it's the same idea with the Massdrop Plus IEM. maybe you'd like the resulting signature from a 10ohm source better than the sound from using a 0.1ohm amp? objective measurements and electrical rules can hardly answer for your preferences.

6. For the massdrop plus IEM i think i got pretty lucky. I am a huge bass head. There are very few occasions where i'll be like "this has too much bass". This is mostly for sub bass and low bass (ie stuff below 60hz or so). Mids i like normal, treble i like a bit rolled off because the cymbal "tsss" and people saying the letter S and stuff like that tend to usually make me recoil. Even on the HD650 i flinch a little sometimes, though its not that bad at all. It was really bad on the ath-m50, but the increased bass helped drown it out.

There are cans like the hd598 (original brown ones) where they had no sub bass, very little low bass, and seemed to be a high bass/low mid cannon. I would have classified those as pretty much unlistenable except for some casual youtube video browsing or maybe netflix.

That all being said, I have a fiio E11 gathering dust in my basement, so I figure i can quite easily compare the lg v20 output with the fiio E11. It wont be quite a scientific comparison since the phone's characteristics will be amplified, but i can definitely listen for changes in frequency response. Its free for me to check, so why not. Or maybe i will hardly even notice a difference. Right now i'm using RHA T10i. Everyone craps on them for being overly bassy and calls them "warm" but to me they sound fantastic (reference filter).

I'd love to try the hd650 out of an otherwise neutral source and compare it to the bottlehead crack. Im sort of done spending money on the desktop setup, but id love to see what the fuss is about.

This whole thread was mostly me trying to get a better understanding of how to conceptualize measurements. I get how to subjectively listen (obviously), i get how to read the measurements and more or less what the measurements are saying, but what do they mean? IE how do measurements correspond with specific qualities that we listen for?

At the end of the day, nothing beats simply trying stuff out but thats hardly ever an option and at least measurements can attempt to give a little bit of insight on what might be expected.

This started because i was looking at Rtings measurements, and even when they seemed to measure great they'd say "this is a mediocre sounding headphone" and not a single measurement corresponded with their opinion.

7. I will add that amplifiers do not have a flat output inpedance, unless an impedance is deliberately added. This may be the case in your V20, perhaps to protect against shortcircuits, but in this case it may be part of the impedance sensing circuitry LG phones have if they have the "quad DAC" circuit.

Amplifiers' output impedance rises with frequency, but if well designed is generally 10s of milliohms at the output, so is hopefully not an issue. The common reason for the rise is the amplifier's open loop gain reduces reduces with frequency, so there is less feedback to lower the impedance as frequency goes up. What is often not noted is the nature of the output impedance. If it is resistive buy nature it will interact linearly with the load, and be relatively inert. If it varies with amplitude or rate of change, it may introduce non-linearity not always easy to diagnose.

8. High damping factor is needed, if the transducer itself hasn't got enough of mechanical damping. In those cases electrical damping is needed and you don't have that if the damping factor isn't large enough. Some cans don't care what the damping ratio is and some do care a lot.