HiFiMan Deva (and the assumptions behind the HE5XX)

https://www.head-fi.org/threads/fs-hifiman-drop-he-5xx-just-arrived-today-brand-new.946418/

Edit: Impressions were added

 

Maybe this isn't the best place to post ranting about Drop, but it's been in my head for the past couple days, and prevented me from sleeping along with late night listening to my T20RP mkii. I will delete this post if it's causing too much trouble, but I just have to clear out the cache in my head for other craps at this time of the year.

My point is it's too much to ask for Drop to be the level of transparency we want them to be. The way Drop operates has already determined they aren't a bunch of tech heads that know every technical detail about audio. If they were extreme audio nerds that don't have friends and never got laid in college (cough cough me cough cough), they would make their own gears instead of collaborating with existing designs. They are more of the salesmen to serve the community, not the product developer. It seems to me that they just ask the companies to make them something that has bass and is all black. The company figures stuff out, and then Drop paraphrases what they are told to the community. It's really hard to outsmart those companies to judge whether it's BS or not, especially while keeping a job as salesmen. As a team, they would have to know enough technical details and stay posted with audio news (even across the ocean sometimes) to be able to identify the scams. For those of you who can do this, you deserve a higher salary for jobs in other industries.

Being transparent doesn't help Drop to sale. They've been pretty quiet about what improvement they did to the products, and letting the community figure out. Most of the buyers probably don't even bother to look at the reviews for more than 10 minutes. I would like to bring back the case with Airist DAC. William Tse (gosh I hate him even more after looking at his LinkedIn profile) ripped off Sosolar's design and sold the DAC through Massdrop. I still can't believe they sold 997 units of this. The buyers either don't know about the shits behind or they don't care about it. If MassDrop were to admit it was a rip off, the units wouldn't sell, and MassDrop would really hurt or even break. Will paraphrased what the company said just like this time, and William Tse's explanation was worse than the BS posters I saw at my undergrad's Senior Design Expo. For those of you who own the Airist DAC and like it, congratulations! You have the last copy of this very design! Sosolar stopped selling his PCB off Taobao immediately, and probably will never work with Drop any time soon due to Massdrop's poor response.

Spoiler: Spoiler

Your DAC only sounds half as good without Sosolar's FPGA wizardry and careful component selection.

I was around Sosolar's chat at that time, and he probably thought I was William Tse because my location was set to U.S. The chat was gone a few months after, and I couldn't join the new one. Not to blame Sosolar, the ban was really my fault. I’m kind of socialphobic and never posted anything yet still digested their knowledge. I know that's called being creepy. Later I quitted most of the chats as I acknowledge my ignorance and toxicity.

Well I miss Sosolar. His Hibiki DAC, Miska's DSC1 DAC, and a B22 built log aspired me to choose my concentration in VLSI. I learned a lot from them well before I took the corresponding lectures.

For Massdrop's fairness, it's impossible for them to identify William Tse's fraud. Anyone without technical training will be bombarded by his jargons (partly copy pasted from Sosolar's log, and the other part from the standard textbooks), let alone Will still has products from other departments to manage. Similar thing happens this time, it takes doubt and research to figure out HE5XX was actually based on DEVA. On top of that, people are losing trust in Hifiman for all the metallic timbre and unreasonable price over these years that DEVA caught few attention.

Spoiler

Just checked Will's LinkedIn page. He dropped out of Rochester Institute of Technology. This isn't to diminish Will, but instead I find the drop-outs in this country is a unique quality. They know how to work with the general public, and their presentations are naturally humble. I think this quality of Will facilitated Drop's success.

Sorry if my words are so personal. I used to be a Massdrop fanboi. K7XX would have a black and gold edition like K240M if I made the picture quicker, and there are a dozen of different Dekoni pads collecting dust in my bins. I finally learned that if you want to have the low price on Drop, you have to accept their ignorance. After all, Drop is about communities, and it wouldn't exist without communities. Be aware of the nature of its shortcomings then you won't hate it as much. It's the same as selecting audio gears. It's less about what sonic feature you want to hear, but it's more about what sonic feature you can't accept. Drop is always going to be Drop, but you can choose to be the new you.


For those of you who made it this far, I figure we are probably like minded tech head audiophiles. I'm going to share what I know about planers for you guys as a treat. Some of you probably know this already, then there should be your turns to share stories. None of this should be trade secrets though, and you should be able to verify this with google and some physics crash course. You deserve to know this as an educated buyer or enthusiastic modder. These are just the considerations you should make if you want to build a planar headphone as a basement project.

There are really two categories of planar headphones, and they really deserves different names. They react wildly different for earpads and dampening. One is the vintage style like the Yamaha and Fostex with thick Kapton tape. I only have experience with T50RPs, so this isn't my specialty, but I still want to claim they are really similar to the traditional dynamic headphones. The other kind has thin, stretched diaphragm that covers your whole ear like the Hifiman and Audeze. I owned HE400i, HE560, LCD2, AEON Flow Closed, and Verum One over the years, and had a short listen with HE1000v1(and liked non of their timbre shhhh), so I think I know enough to talk about this. Have you wondered why so many big round planar headphones that weighs a pound suddenly emerged around 2007? Maybe it's because some extreme audiophiles just suddenly decided to make their end game gear. Actually, the Orthodynamic headphones patent just expired and people can legally make and sell that design. I think the patent is DE3049803C1. Hifiman and Audeze were the early birds that take advantage of this.

There are two critical components, the magnet and the diaphragm. There's not really much to talk about the magnets. You just arrange it in the blah blah way, and running current in the blah blah way makes the diaphragm move. The good ones are available for a reasonable price on ebay, just be careful about picking the right polarity for the arrangement you want.

Now the diaphragm. The whole patent is about making the diaphragm larger than your ear trying to eliminate the effect of surroundings such as the earpad. This is the reason why they have good bass. I don't have a good analogy for this, it's like taking a bath vs. taking a shower. However, if you make it with good old Kapton tape, it's not likely to reach above 10 kHz for its weight. It might be a good thing for people hating mosquitoes, but it really upsets designers that take measurements (I think playing with measurements is as addictive as video games for releasing dopamine. It's another story to tell).

Thus you would want to make the diaphragms thinner to get more treble extension. There's the substrate and the trace. The substrate is the insulated supporting material that every company boasts with marketing BS about how thin they can achieve, but few mention the traces. The traces have to conduct current so we are stuck with metal. (Maybe some special carbon structure would suit the need here, but that's not happening till we can immigrate to Mars) Alloys are so stiff that they snap easily. The widely available metals like aluminum can only go so thin without breaking, that if you want the absolutely thinnest traces possible, you have to stick with gold. Though gold is a few times heavier, it can be made crazy thin, so the overall weight can be reduced. It's elementary material science. Regardless which metal you pick, the manufacturing process determines how thin you can make it. There are too many processes and the garage version seems to be etching. The advanced versions are probably trade secrets that I don't know about. You also need to consider how to adhere the trace to the substrate, since plastic and metal don't like each other. Adhesives could turn out to be thicker than the trace itself and void your effort to thin down the diaphragm. There are too many considerations in the diaphragms I know I don't know, and I don't know I don't know. I would love to hear more about this.

Spoiler

This part is subjective. Why do I care so much about traces?

First, the driver sounds like what they are made of. It's similar to the preference between paper cones and metal cones. The traces too thick would cause the headphone to sound metallic. Verum one has a 1:1 substrate to trace ratio, yet I still found it metallic. T50RP uses Kapton tape much thicker than the traces, and it sounds too natural even to the point of lacking aggression to me (and it makes me lack sleep by listening to albums in the night).

Second, thinner traces increase the impedance of the headphone. This is basic resistor theory I don't want to elaborate on. Fitting more traces also increases the impedance and power sensitivity. Looks like the HE5XX could use a few more. With higher impedance, headphones can pair with higher voltage higher gain amps. Amps with higher voltages are more linear, especially FET based ones. Drawing less current also keeps the output stages ideal. Those are the nature of transistors and why tube amps can achieve reasonable distortion without feedback. Higher gain also makes closed loop designs easier to stabilize. I also like to use the full travel of the volume control, because their matching sucks at both ends.

Third, it's hard to make traces thin. Being capable of making thin traces is linked to the designer's pride. Similar to making the 0.00000000x gears, it releases dopamine in the designer's brain.

After you get the thick diaphragms, you have to stretch and fix it onto the driver frame. Otherwise, it would just dangle it in the air like our favorite mod material (drum roll) toilet paper. The reason why most of the modern planars are in a round shape is to ensure even tensioning (I'm actually curious how HE1000 managed to have a teardrop shape). This process is where the ortho wall happens. The more you tension the diaphragms, the sharper the wall will be, and will also shift to higher frequency. Ideally you would shift the wall into ultrasonic, so nobody will hear it. But in this case of DEVA and HE5XX, more tension seems to shift Fs up and lower the bass. Like designing other gears, it's about finding the sweet spot. It's just hard to make a wideband transducer that makes great bass, mid, and treble at the same time.

Now you know the basics of how planar headphones are made. Oh wait, did I tell you planar headphones are not really planar headphones? The diaphragm only moves as a flat surface up to a certain frequency depending on how they were made. They also break up like your dynamics drivers do! but in a more controlled way. Headphone companies are never transparent about this, as you won't want to buy them after knowing it. It's taking the placebo bonus out of your hearing session. At some frequency, the top and bottom of the driver moves in the opposite direction as the center, and I suspect this caused the "shark fins". Then the sounds at all the frequencies above it are created by such resonances. There's really no easy way to measure the break up behavior. You could measure it with laser if you are super scientific with access to all those gears, and you could also feel the stationary spots with your finger on the driver (but for all the tensioned drivers you wouldn't be able to put it back after taking apart), or you could simulate it if you are a mechanical master (I wish to be one of them). Anyhow, acknowledging their existence rarely helps us to solve them. Sometimes life is easier to know less and take advantage of the placebo effect.


Well, that's all I want to say, and that's way too many words. I've never ever written this much in my life. I intended not to use any pictures, hyperlinks and emojis just to bore you guys out. Perhaps only the hardcore geeks reads all of it, and I hope the boredom keeps the people who don't need to see this away. I shall now be able to fall asleep without thinking about these things.

Oh crap, I meant to work on my homework after writing this. Didn't realize it would take this long. Guess I'll worry about it tomorrow.

 

You're right.

 

https://www.headphones.com/blogs/news/hifiman-he5xx-vs-hifiman-deva-do-they-sound-the-same

 

can I use Fostex T60rp balanced cable (4pin XLR to 3.5mm trrs) for Deva?