It reminds me of this head-fi thread about paint falling from the magnets. Check it out here.
Try looking at the drivers with some back light, as shown in the pics on that page. Do you see any “specks” inside?
2 Likes
oh wow, wtf. HE6SEs with the diaphragms sticking to the magnets and Susvaras with magnets sticking together.
1 Like
Thanks to all of you for the troubleshooting 
When I tested the 2 Susvaras there was no doubt in my mind that Hifiman made a silent revision somewhere between 2020 and now (they are known for doing that). Unfortunately, I completely disregarded the other thing Hifiman are known for, inadequate QC.
So yeah, there are indeed flakes of different size and shapes inside the drivers. I guess they peeled off the magnets and got stuck to the diaphragms. But here comes the twist - It’s the 2022 unit that was bought directly from Hifiman’s website. My used 2020 unit from the EU dealer has none of these flakes.
I don’t know what to make out of this. Is there a new “bad batch” of Susvaras being sold by Hifiman, or is it just this one unit that somehow slipped through Hifiman’s otherwise immaculate QC? And what’s the idea behind painting the magnets anyway? Does neodymium oxidize? Does Audeze paint their magnets? Why is it always Hifiman who has to screw thing up?
Sorry for the rant there 
Imagine buying a $6000 headphone and finding out it has a manufacturing defect. Going through the RMA process is not going to be fun.
6 Likes
Thought I’d log in back to the forums and see if this was confirmed by others since I just got new stock pads from warranty yesterday and they do yield a different sound in my findings. I am unsure if Hifiman has confirmed new pad revisions, so I’ll use the term variants to address the two pads below for now.
To sorta preface, I’ve always wondered this because I’ve seen Crinacle’s Susvara measurements and they measure quite a bit different in the midrange and upper midrange/lower treble (from my understanding of his rig and mine on MiniDSP EARS). I just tossed it up to possibly not having hands-on experience with his rig or unit variances. My suspicions of potential variants sprung up again as I see that he’s measured 2 units that differ in the midrange to upper midrange/lower treble. Way too much to be just unit variances in my opinion, or if it is, it’s terrible QC.
Well, I got the new pads yesterday and took a listen and definitely heard a difference in the upper midrange. Here are my findings:
- The hole around 1.5 kHz is now more filled out (to more or less my preference)
- I noticed a tiny bit less energy in the area around 500 Hz - 1 kHz
- ~4 kHz is slightly pulled back to, in my opinion about neutral, in the new pads I just received. I always thought there was too much energy here with the older pads.
In more subjective findings:
- Vocals sound more natural
- Electric guitars sound less blaring and harsh
- More or less to my FR preference
- The headphones slide around less on my head due to more surface area of the earpads touching my head (more on earpad dimensions below)
Here are some measurements in addition to my subjective findings:
Pink - old variant, Blue - New variant
Note: The “old” earpad variants I received were used only a couple of times after a previous warranty replacement so they have little to no wear.
Measured on a MiniDSP EARS. HPN compensation. Left channel at 90 db. The 3-5 kHz region on the MiniDSP EARS measured on most if not all headphones is inaccurate. The MiniDSP EARS is also not accurate at measuring the treble region though it does provide a bit of insight if overlaid with another headphone’s response. I typically identify dips and peaks in the treble by ear with a tone generator - how reliable one thinks that may be is how much faith they have in my ears. …which might not be a lot understandably. ** Remember, only compare these measurements to headphones made on the same system (MiniDSP EARS) and the same compensation. In this case, there is the HEQ compensation (sorta Harman-eque IIRC), the HPN compensation (sorta diffuse field IIRC), and raw measurements (no compensation applied). Do NOT compare these measurements to headphones made on a different measuring system or you may misinterpret the comparison.
Here are my findings regarding the physical differences of the two earpads:
- There seems to be no differences in thickness on the foam used - confirmed with a caliper on all sides
- No differences in perforations on the inner linings of the earpads or material on the earpads
- The earhole opening of the older variant is noticeably taller than the newer variant by about 6mm - it may not seem like a lot but in my pad rolling and modding experience, that can and will very much change the tuning of the sound
Here are pictures of the two earpads (left - older variant, right - newer variant):
Newer variant ^
Older variant ^
Newer variant ^
Older variant ^
I guess to sort of close off my findings, I think the differences in sound are due to an earpad revision. The difference in sound is too big for me to think it’s just earpad variation. The physical differences in the two different earpads seem more intentional than not. If I may reference my impressions post on Reddit 2 years back, I mentioned that I really did not like the stock tuning of the Susvara with its huge hole around 1.5 kHz and then the rise to 4 kHz. I thought the midrange tuning was really wonky. I swapped the pads with some old Beyerdynamic(?) pads that is closer to my preference and am still using the Susvara with those pads 2+ years later. At the end of said post, I said, “Not sure if anyone from Hifiman will ever see this but I would LOVE if they could revise the pads to make them a tiny bit wider and deeper at the back (by 2-3 mm or so) and with exterior perforations to improve the tonal balance in the midrange. I think it’d be a pretty killer headphone out of the box if that happens.” I don’t think anyone from Hifiman read my post being a random nobody, but it does seem my plead for revised earpads has been granted to at least me (with much better solutions to my pad suggestion regarding perforating the exterior).
In my opinion, this new pad variant yields a better and more balanced sound.
7 Likes
The 63mm on your new variant fits with the hifiman store description. You made me curious about my own pair even though to my ears they’re perfect. Might try swapping to this if mine seem to be different.
Is it hard to swap pads on the Susvara? never tried or know how to…
They’re pretty easy to swap. There are 4 clips on the mounting piece behind the earpad that you can unhinge. I should mention though the the clips are extremely easy to break. I’m still able to get a good seal even without a clip or two but it’s still something to note.
2 Likes
Here’s a challenge to the group… I’ve ordered a Susvara and currently use a Woo Audio WA7 3rd Gen (1.5W into 4 Ohms). I’m considering upgrading - can anyone recommend a DAC/Amp combo that pairs well with the Susvara, but has a small footprint? Small desk, no room 
Ferrum Erco, ifi Pro iDSD Signature and the Burosn reference line come into mind as combos that can probably do the trick.
You didn’t mention budget or how limited you really are with size.
From the ones above I believe the ifi is your safest bet.
I’ve noticed the same 3 models.
About budget - like most I don’t want to spend money needlessly but I also want to know that the Susvara sounds as good as it can. Anywhere from $1,000 to $4,000 for a DAC/AMP combo or stack.
Size - 8 inch 20cm width / 7 inch 17.75cm depth / 4 inch 10cm height would look reasonable on my desk. I can go up to 10 inch 25cm width / 8 inch 20cm depth if necessary but it’ll take up too much space on my desk to look visually proportional.
I would go for the Ferrum Erco and Hypsos. It about fits your size constraints.
Didn’t see reviews of the Erco yet but I own the Oor and if it’s even close to it with a good DAC implementation you’ll be set.
1 Like
Get a smaller dac to go with it
Could also go TT2, though wouldn’t necessarily recommend it. Missing some performance, imo. Susvara would prefer more power.
1 Like
Really, is TT2 underpowered?
For the Susvara, yes imo. Could use it and it’ll be ok while you improve your chain. If I had the tt2 I’d look for a more powerful amp (more current) and then once I upgraded I’d use the tt2 as the dac (still has a solid amp for other headphones). But I wouldn’t use it as the primary amp for the Susvara.
TT2 from the rear balanced outs has ample power for the Susvara. I enjoyed that pairing a lot and came back to it after trying a few speaker amps. It is a very clean, black, and neutral presentation, so it’s really a matter of taste, but it’s not underpowered.
1 Like
Thanks for this thorough write up.
I sort of concluded this as well last year but didn’t do any science.
I think the newer pads are more agreeable sounding and less diffuse.
Try HE6 velours on Susvara as well. It’s astounding. HE6ifizes them quite a bit.
2 Likes
I recently received a new Susvara with 3.5mm connection. The date on the package is January 15, 2022 so it should be a new unit. I also measured the pads, the height also measures around 64mm.
1 Like
Just to confirm, you liked the IHA-6 more than the A90 for the Susvara? I currently tried the A90 and Jotunheim 2. While the Jotunheim 2 does seem to have enough power, I found it could be a bit harsh sometimes and the bass was a little looser in comparison to the A90.
Jotunheim 2 is a very good transparent enough measuring amp. I doubt you would hear those differences in a blind test.
Just focus in good current delivery as the susvara have moderetly low impedance 60ish) and extremely low sensitivity (80ish), so for heavy bass notes (high signal amplitudes; Aka high voltage), it’ll need abundant current (V=R.I) to avoid clipping if you listen in very high volumes while adding a bass shelf.
So you’re concern here would be power. Not transparency, 'cause again, both amps are very transparent.
By the way I added a bass shelf, and I too suffer from this. Soncoz are apparently getting a power amp with high level of transparency (to be confirmed) this could be the ideal candidate for the susvara. I’m following it closely.
NB: just to be clear, the issue here is the fact that with such a configuration you’ll need to use you’re amps at 99 volume or close, which is not ideal. A more powerful amp will give headroom and not stress the components to their limit.
1 Like