Pass Labs New DIY "WHAMMY" Headphone Amp

Full kit - including a slick-looking case - now available @ $299 for pre-order; shipping by Nov 31:

https://diyaudiostore.com/collections/printed-circuit-boards/products/whammy

Serves me right for going back to the site after ordering everything separately (from a few different places - still waiting on the last of it to arrive).

Edit: more details on the various components:

https://diyaudiostore.com/collections/whammy-headphone-amplifier/products/whammy-pcb $29

https://diyaudiostore.com/collections/whammy-headphone-amplifier/products/whammy-chassis $119

https://diyaudiostore.com/collections/whammy-headphone-amplifier/products/whammy-completion-kit $189

 

That's some sweet value right there.

 

Finally got this together, after having all the parts for over a year - and apparently didn't blow it, despite having little DIY experience. It'll do desktop duty with, for now at least, Ifi Nano BL as DAC. Excuse the dust:



I used the Hammond case mainly because I'd bought it along with all the other parts before the DIYA one was announced. The Hammond (1455T2202) is smaller so saves a bit of desktop space and is quite a bit cheaper, especially if you go for the version with plastic front & rear panels (easier to cut holes in and available on their own for a few $ per pair if you screw up or want to experiment with different layouts). It makes for a tight fit for everything though, especially in front:



Pre-outs not wired yet because I hate the cheap mic cable I used for the I/O wires - the insulators are stiff yet have a low melting point. I have some Mogami coming to replace it.

For ease of construction and best aesthetics, I'd recommend going with the kit case. But if you want to save money or space by using the Hammond, here are some tips that might save you wading through all 200+ pages of the DIYA thread...

As you can see from the photos above, if you want the volume knob vertically centred you have to lift it off the board, invert it, and use wires. If you're using the plastic front & rear, ground the pot with a wire from one of the rear screws to the screw-hole at the case end (the equivalent position at the back can be used for the safety ground).

If you want the headphone socket centred both vertically and horizontally, it has to be twisted to miss the R input cap, presuming you're using the kit/BOM ones. Good luck getting anything much bigger in those positions using this case.

More space around the opamp socket can be had by using smaller caps than called for in the original BOM, considered fine in these positions because of the smooth power coming from the big-arse supply section:



These stubby little gold & black 100uF/25V Nichicons (UMW1E101MDD) have been used by a few of the DIYA people to make more room for SOIC opamps on DIP adaptors (like the OPA 1612A here), or big Bursons or other discrete things. Paradoxically though, opamp rolling then becomes most difficult with DIP-8 chips, because while there's room to push them in, it's difficult to get your fingernails under to prise them out.

Using another of these caps at position C9 (and its R-channel equivalent C28) and leaving the leads on the green C27 long enough to bend out of the way gave me additional room for the lighted front-panel switch. Power for the LED can be tapped from the front ends of resistors R10 (+ve) and R9 (-ve) whether you're using the power-supply config that uses those resistors or not (I didn't, obviously). It's recommended to use a resistor of 15k-30k Ohms in front of the power LED, but I found with the switch I used (with a white 3V LED behind a thick plastic lens), 22k looked like a dying glowworm and after some trials, 1k was appropriate. Experiment with different resistors before you solder one in.

Quick sound impressions
Woohoo, macrodynamics! This thing really kicks my inefficient modded YH-2s and T50RP3s into gear. Tonally there's a lot of top end compared to the amps I was using for these before (Hel & Vali 2) and if not increased amplitude in the lows, certainly a sense of increased power and grip there. Great imaging/layering, resolution, and blackground. The prominent top end might give some people a bit of trouble though, so it'd best paired with transducers and sources that are forgiving in that area (and free of digititus, with good USB implementation or a good DDC in front). I haven't yet tried it with my OG Modi Multibit, but I think that DACs dark tonality and multibit midrange magic would be a good complement for this amp.

The amp is dead quiet at 10 on the volume dial with my planars and HD6XX, but hiss comes in with more sensitive phones - it's definitely not something I'd use for IEMs unless you drop the gain considerably (there's discussion of this in the DIYA thread). With the not-very-sensitive 7Hertz Timeless, hiss comes in below 4 - but comfortable listening for me is around 2. I wouldn't try anything like Andromedas.

To give more meaningful comparisons vs my other amps, Vali 2 & MCTH, I'd have to spend some time listening to all in the same systems. For sure (edit: with my planars) though it's a different presentation from the Vali 2, that amp being dynamically softer, not as resolving, and relatively rolled-off in the highs. It might also enable me to compare better with my notes on Jot 2, but generalizing from what I heard of that in those systems: WHAMMY would be brighter, possibly more macrodynamic, and faster in the transients; and lacking the Jot 2's tone density in the mids and being a bit behind in note weight in the lows.

3 Opamps so far
I started with OPA 1612A and then tried LM833N and OPA2134PA, these 3 being favourites of many in the DIYA thread. I also measured the DC offsets of each.

LM833N: Not as warm or powerful in the lows as 1612A, not as extended in the highs but just as sibilant. Overall thinner sound, but fairly smooth in all areas but highs. Macrodynamics not as evident, especially bass slam.
DC: 13.68 mV R, 13.53 mV L

OPA 2134PA: Better clarity than 833N but a bit harsh/grainy in the middle-upper mids, definitely more forward there. Macrodynamics stronger than 833N. Seems to be playing at slightly higher volume than 833N? Possibly because of mids. Less warm than 1612 but good bass drum punch.
DC: 4.37 mV R, 4.45 mV L

OPA 1612A: Comes across most V-shaped of the 3. Upper bass warmth, better balanced with mids. Middle & upper mid harshness gone, mid subtleties/layering come across better. Macrodynamics as 2134, maybe a bit stronger. 'Cleaner' highs (cymbal hits) but maybe too forward. Maybe a bit less sub-bass focus on bass drum than 2134PA, or is it that there's more in the mid & upper bass with this one (better balanced)?
DC: -0.58 mV R, 1.39 mV L

1612A for me so far. I have a few more, the most interesting of which (OPA1656, OPA2156) are SOICs which I haven't yet got onto adaptors.

Anyway - fun project, satisfying outcome!

 

First off, excellent write up and review! The first compelling objective review I have seen out of this amp yet. Thanks for your sharing your work and your impressions.

And I was just going to share that I tried a 1656 in the Little Bear B4 and it was rather flat. Being the 1612 was exactly as you had described here. Very punchy and dynamic and I think it is still the best BJT op amp around. So I am not sure you will get much gains out of the 1656, but I have a couple if you need some and would like to try.

Also I think the Burson might be a good try here. I also have one of those if you would like to try that as well.

 

Neat work for a first timer

 

Hey - watch who you're calling objective there, pal

Interesting about the 1656. Thanks for the offers, but I have a few 1656 & 2156 (planning on modding the MM), just have to challenge my hand-eye co-ordination to get one of each onto DIP boards (I bought one of those EZ socket adaptors for quick trials, but the pins are too fat for the WHAMMY's socket). I have a line on the Burson V6-es too.

 

I like the Sparkos op amps in mine.

 

WHAMMY vs Vali 2 & MCTH

Yeah, I know, solid-state vs tube hybrids; but, humour me because I'll briefly compare to my notes and memory of the Jot 2, too.



I listened to these 3 amps out of the same DAC & upstream: OG Modi Multibit with v2 firmware and stock AD opamps, fed by a Pi3B-HiFiBerry Digi2 Pro wifi streamer (the black filling in the Schiit sandwich), with the headphones pictured (all modded, see here, here & here if you want the tedious details).

The opamp in the WHAMMY is OPA1612A. Tubes are a NOS Reflektor 6N23P-EV in the Vali and a Telefunken ECC88 in the MCTH. I used a Noise Nuke power filter in front of the stock MCTH wall-wart because I usually run that amp from a big LPS but that would have been hell to extract from the living room cabinet and wouldn't have left much room on the side table.

Compared to the WHAMMY:

Vali 2 (high gain)
Darker tonal balance. General character more liquid, notes rounder ('slower' transients), macrodynamics subdued (more with the planars than with HD6XX); less bass slam. Notes not as precisely rendered at the extremes, particularly highs. Rolled off in highs. Stage wider, imaging not as precise. Not as resolving (micro stuffs, both dynamics & detail). More 'romantic' (roundness and liquid character in mids combined with darker tone). Thicker tone density in mids. From the Fostexes, the Vali 2's midrange microdynamics & microdetail become significantly worse vs the WHAMMY - mids are very veiled cf. WHAMMY with these 'phones vs the difference from the HD6XX. From the Yamahas, macrodynamics are not subdued as much, despite these being quite a bit less sensitive than the Fostexes. The midrange veil of the Fostexes is also missing from the Yams (hmm, have I overdamped the T50s?). The highs (the Yams have the most treble of these 3 pairs) are not as elevated nor as well rendered as by the WHAMMY.

MCTH
More subdued in the highs than WHAMMY, not as much slam in the bass; overall tonal balance comes across as less v-shaped (do I prefer it?). Drier, squarer, more textured in the bass (I like this too), macrodynamics a bit less elsewhere too. Less liquid timbre than Vali 2, but more than WHAMMY in the mids (more rounded transients there, as opposed to in bass). Clear step up from Vali 2, technical differences from WHAMMY less marked. Maybe not quite as resolving (certainly not in treble). Clarity & blackground inferior to W. Stage maybe wider (or at least more hard-panned), imaging and instrument separation less precise. But WHAMMY can come across more 'clinical'. WHAMMY ahead in micro-stuffs, though possibly some of this perception is due to 'fake detail' from the more elevated treble. From the Fostexes, the MCTH is macrodynamically softer across the range than WHAMMY, which doesn't change a lot in that respect from the Fostexes vs the HD6XX. Bass becomes more rounded. From the Yamahas, macrodynamics are more subdued than Vali 2 (opposite of with T50s!). Otherwise good sounding; smooth but detailed treble, though not so elevated as WHAMMY (less v-shaped). Microdetail, microdynamics superior from WHAMMY.

Overall, I'm a bit conflicted between the WHAMMY and the MCTH. The WHAMMY has the better resolution, especially in the mids & highs, but is perhaps a bit hard-edged and lean there (as well as maybe having too much treble), while being nice and weighty in the lows. The MCTH has a drier, leaner, maybe more nuanced bass but more liquid mids & treble (and isn't as bright). Depending on what I was listening to, I sometimes found myself wishing I could mix the timbral character of the two amps, swapping the bass & lowest mids of one with the rest of the range of the other.

WHAMMY vs Jotunheim 2
My memory and my notes of Jot 2 tell me that the Jot has a more coherent timbre - but more on the tubey side, from comparison with the hybrids above and from reading more experienced listeners' impressions. As I said earlier, WHAMMY would be brighter, possibly more macrodynamic, and faster in the transients; and lacking the Jot 2's tone density in the mids and being a bit behind in note weight in the lows. The hint of upper mid emphasis I noted in the Jot is not there (with this opamp) in the WHAMMY, but the WHAMMY has elevated highs instead. So preferences. Of course, having balanced as well as SE I/O, and in particular a gain switch makes the Jot a lot more flexible - WHAMMY is not for IEMs.

 

The Whammy is on a whole other level using the OPA 627's. Slaughters Asgard 3.. Almost. (99.5%) as good as my tricked out M^3. Has a sense of body and tonal purity that is equal to my Sunrise 3. Absolutely great with a good RTR /NOS dac.

 

Holy crap the opa 627 is $33. Oh well, probably add one with my next mouser order.

 

They're single-channel - so you'll need 2, and a 2-to-1 adapter board!

I was too cheap for OPA2107, a dual for about the 1x627 price. A pair of 627s might be nice, but I'd be betraying my m.o.

 

They have actually gone DOWN in price. They used to be around $49 a piece. And they are single channel op amps.

1612's are about the best in bi-polar, and the 627 is about the best in JFET, but the 627's are single and you need two with an adapter.

I would also not rule out an AD827. They are also on the pricy side ($21.58 USD on Mouser), but those are dual channel. They are fast op amps for video, but still stable enough for audio.

 

That's what I'm using currently. It's my favorite out of the handful I've tried. Has really good separation - an attribute I value highly.