Time to DIY like it's 2009! (hobby has changed, man)

TLDR; this hobby has changed a lot in the 9 years I have been away. Many things are confusing and weird. I am particularly bummed out about the gutting of the DIY community, so I am just going to pretend like the last decade never happened and revisit all my DIY gear. It will give me some time to get back up to speed and see where I want to take this hobby.

*

It was such a simple plan. Earlier this year, COVID funk had firmly set in. So, I decided a good distraction was to re-evaluate all my headphone gear. Pull everything out of storage, re-organise to get the most out of it, and see where I might want to upgrade. Ultimately, I decided to re-case my 13-yr old DIY M3 headphone amp into something much nicer than a Hammond case and maybe upgrade my original Twisted Pear ESS 9008 DIY DAC. But maybe I should just pop into a couple of forums and see what has changed over the past few years…? Big mistake.

Let me take a step back. When I started in this hobby back in 2006, Head-Fi was the only mainstream place to get information on headphones. HD600/650 were top of the pile, Stax were spoken of in quiet whispers, and companies like Ultrasone were just starting to push the limits of pricing for high end dynamics. CD Players reigned supreme, and computer-as-source was in its infancy. Headphone amps were either rubbish, expensive, or you had to build one yourself. I both hated and loved this time. As a budding scientist, I could see that Head-Fi was largely a dumpster fire of the worst subjective audiophile bullshit, and I yearned for some objectivity. But thankfully, I fell into the awesome DIY community – my fondest memory was probably organising a local group buy for Millett Mini Max parts (Hi @TomB!) – and I found a reason to stick around.

Then, objectivity arrived in the form of NwAvGuy. It was everything I had ever hoped for, and it was horrible. People just swapped one religion for another, except with even bigger fights and even more manufactured drama. Between work, a temporarily bad money situation and general disillusionment I stopped visiting the forums in 2012, and headphone gear was progressively packed up and stored away. I did not miss any of it until this year, when I looked at and listened to all my old gear, and remembered just how awesome and intimate music sounds through headphones.

Problem is, I now see that all that drama that kept me away form this hobby is as bad as ever. The cult of measurement at ASR in one corner, and people doing things like listening to fuses in the other corner. Very disheartening. But I am also finding some things to be really excited about! The market has grown incredibly, and with people now able to enter the hobby through Religion Objective as well as Religion Subjective, there is a community that will accommodate everyone. With the larger market and mass manufacturing, the price/performance on entry level gear is astonishing. A Magni, Modi and HD6XX for under $500? That. Is. Insane. An explosion of new ortho’s that do not sound like poo? Color me intrigued. High-end manufactures like Focal that are the real deal, instead of just another Head-Fi FOTM? Shit, I need to start saving my pennies. If I can avoid the drama, I think I can enjoy this again.

But now we come to the crux of why I am posting this thread. The wide availability of great gear seems to have really killed the mid-range DIY community where I was so happy. Current DIY projects seem to be either dirt cheap junk, or ultra-high-end bespoke gear. Most new projects are also being designed for SMD, which holds zero interest from a construction point of view. The best tweaking comes not from swapping OPAMPs or tubes, but in software and transports to curate and send music collections and streaming services.

To break all these disparate issues down to their core… Without the reference point of DIY, it is just so hard for me to understand where my gear stands against what is currently available, or to know what I might need to do to take the next step 'up'. This is doubly true during COVID; meets are impossible, and even the thought of shipping gear to trial makes my butthole pucker. So, I am hoping to use this thread as a sounding board for my relearning of the hobby. I am going to go through all my old DIY gear one piece at a time, see what value it holds to me, upgrade as appropriate, and just see what sort of feedback comes through.

If you have read this far, you are surely wondering why I am posting this self-indulgent dribble on SBAF. Well, while this forum is far from drama-free, it really seems to be the only place that carefully straddles the subjectivist/objectivist divide. There is no way I am going back to Head-Fi or signing up for ASR. Head-Case would give a good assessment of the DIY, but their discussion/review of new gear is extremely limited. But most importantly, I have seen all my gear mentioned at least in-passing in old threads on SBAF, so I am optimistic this is a place where things might be mutually interesting. At the very least, I figure that keeping everything constrained within my own thread in the DIY forum makes it easier for people who are not interested to ignore me…

 

So the first gear I am going to reconsider is my Bottlehead Crack. It was my last DIY project, and the one that was the least actual DIY. But is a good starting point for this thread, because so many people are familiar with the amp, and it still seems to be close to the top of the class for its price point.

For those who are not familiar, the Crack is a SET OTL tube amp released in 2010 (pretty sure I first met @Armaegis in the Head-Fi Crack thread?). It has an unregulated CRCRC power supply, AC heaters, 12AU7 input tube, and a 6080/6AS7 output tube configured as a cathode follower. In stock form the tubes are resistor-loaded, and the headphone output is capacitor-coupled with 100 µF electrolytics. Output impedance is high at 120 ohms, so unless you just love rolled off base, current clipping, and a terrible damping factor, the amp is exclusively for high impedance phones.

In stock form, the amp was competitive with other OTL tube amps of similar vintage and was quite nice with any phones above 250 ohm impedance. But the real magic of the Crack is the beast it can become after upgrades. The ‘Speedball’ replaces the resistor loads with CCS loads for a dramatic increase in resolution and linearity. The output caps can be replaced with any number of boutique film caps, the power supply and connectors are ripe for upgrades, and tube rolling options are quite extensive.

Because I was relatively experienced at this point, I built my Crack almost fully loaded. Cardas RCA jacks, TKD 2CP2511 pot, Teflon tube sockets, Speedball mod, better specced power supply resistors, Axon 91 µF film output capacitors, and a 5998 output tube (offers more resolution and punch than 6080/6AS7 tubes, and lowers output impedance to about 80 ohms). The amp sounds glorious with the HD650 – fun, engaging, and comfortable all at the same time. But in the spirit of this thread and my journey, where does it stand now? And is there anything I could do to make it even better?

Original Crack Images:
https://i.imgur.com/9qM3rw6.jpg
https://i.imgur.com/AFkkFBB.jpg

Considering my Crack is loaded, upgrade options are limited. I did a popular power supply mod, replacing the standard rectifier diodes with zero-recovery Schottky units, to supposedly reduce power supply noise. I added Mundorf MKP 2.2 µF film bypass caps on the first and third power supply capacitors (only place they would fit). I replaced the output caps with 100 µF Mundorf MKP. And finally, I bought a boatload of new input tubes to try. Before settling on a surprisingly good JAN GE 12AU7 for daily listening, I noticed that the amp picks up a LOT of RFI, particularly with new production tubes manufactured by New Sensor Corp (Genalex, Electro Harmonix, etc.). Installing a tube shield cut the background noise to virtually nothing.

Upgraded Crack Images:
https://i.imgur.com/SueEeHs.jpg
https://i.imgur.com/jkRHIE8.jpg

So, where does this amp stand? In all my reading, this amp is still regarded as an excellent pairing with the HD650, and in my own listening I certainly do not feel like I am missing out on anything. Within the Bottlehead stable, the Crackatwoa adds a regulated power supply, but it is physically larger and adds input switching and balance controls I do not want. I have seen commercial options at twice the price that still have resistor loads, lesser pots, and lesser output caps. It would take silly money and effort to get something better. I am very confident this is a keeper.

 

Dude, you want to just borrow some stuff from me? We can do a curbside pickup or something.

 

I would certainly consider it. But the main reason I actually posted this thread is that I'm not even sure what I need or want to listen to.

I *think* I can already see how this thread is going to end....? All my DIY sources and amps are probably perfectly fine by modern standards, even if they might have terrible price/performance versus newer commercial gear. Where I am deficient, and where the hobby has advanced the most, seems to be the headphones.

The other day when those Focal Clears came up for US$1k, I almost just said "screw it i'll just buy these and then disappear again". But that's the lazy way out.....

 

A fun read so far! My own interest in audio is mainly DIY, and I stay away from commercial amps and speakers. Looking forward to seeing what else you think about your existing projects, which I suspect remain on par.

 

Glad you decided to post your "dribble". I think you have sufficient insight into your likes and the "nature of the beast" to be able to experiment as a point of enjoyment rather than some quest for the ultimate in fidelity. It is very easy to be sucked into the nervosa and I find myself walking the rope from time to time as I get distracted by new shiny things.

DIY allows a great sense of freedom because it is easy to tweak different designs and try different configurations. Before HeadWize shut down, I had been playing with a modular design. Building a single amp chassis (allowing a little extravagance) with various drop in elements to change power supply, voltage gain stage and output stage. Perhaps it is a good time to develop an ultimate SBAF tweaker box. That said, I started this approach for some PASS diy power amps and ended up with 5 chassis as I couldn't tear them down in between.

I honestly think the M3 still holds its own in today's market diy and commercial. Such a great design.

And headphones have evolved but some of the older gems still compete easily and the humble T50RP remains a potential giant killer in the right hands.

Hope you find what you are looking for and more importantly rekindle your passion for the gear and music you own/will own.

.. dB

 

OMG! It's like a DIY reunion - both dBel and Beefy (yes, I saw your post in my Walnut DAC thread, but busy, as usual).

Beefy - I share your disappointment at the DIY situation. I keep trying to nudge it forward, here and there, but it's almost like people would rather see me do it than learn how themselves. That's OK, though - maybe it'll spark a few.

If you look around at all types of "hobbies," the same thing has happened to a lot of them: Asian influence. When you can get something of the same or higher quality at a cheaper price than you can buy the parts - much less assemble it yourself - it becomes a no-brainer.

I have been deeply involved in model aircraft at various times in my life. With headphones back in the early 2000s, if you wanted an amplifier - you had to build it. Similarly, the best model aircraft were built. The manufactured ones, if they existed at all, were crap. The problem is that East Asia labor took the same kits and scratch-builts and built them for sale cheaper than the kits or parts. It took a few years, but the model aircraft RC kit business has almost disappeared altogether, when it used to be the only way you could fly something of quality.

Model railroading is similar. You used to have people building locomotives, railroad cars in kits, painting, decaling, etc. Now you have legions of hobbyists who don't even have a 4x8 layout, but own dozens of high-dollar, highly-detailed locomotives and cars, made ready to run, bursting out of their closets. Releases are pre-ordered up to a year ahead, to fund custom design and manufacturing in China/Korea/etc.

Look at AKG - they closed their factory in Austria. That was probably a Harmon business decision, but all of their manufacturing is in China, now. It's not all bad, though, as long as tight quality control is exerted by the originating companies.

You know what they say - you have to learn to play ball in the field that's available, not the one you wish you had.


P.S. Just a friendly reminder that Headwize was really in the forefront of DIY for headphones in the early-to-mid 2000s. Chu Moy did a lot for all of us, but I think it broke his heart when they hacked his site to bits. May he rest in peace ...

P.P.S. I disagree somewhat about your reference to SMD. It can be done quite successfully as DIY with the correct tools and strategy. Again, through-hole parts are also on the demise. TO-220 and TO-92 transistors are going obsolete or NRND (not recommended for new designs) every day.

 

I remember back when we had arguments on TO-99 opamp packages being superior to the DIP or SOIC models because, I don't remember... they were bigger and had better heat dissipation? I think it was mainly the LME49720HA.

My great idea that never came to fruition was recreating Jan Meier's active balanced ground, doing the "processing" side with a hacked M3, and feeding the three signals into a beta22. I don't know how his was implemented, but I had math'ed out a circuit that did some oddball things where recombining all the signals would give you a x3 gain even though everything before it was "technically" unity gain. I proofed one with a sorta-but-not-really-a-cmoy that worked. Even built up a slightly better one with some pcb's that a kind DIY'er made for me (I think it was avro_arrow? it's been so many years I don't remember). I also proof'd one design using differential receivers/drivers that used no resistors at all which was kinda neat. Then I got as far as figuring out how to hack the M3 to emulate my circuit, and then I'd only need three standard b22 boards built with whatever gain I wanted. Bought the kits for both, even had the backplane for the b22, and then... I dunno, life just stalled out and everything got shoved into the drawer and forgotten about.

I picked it up a few years later, and 4-channel balanced amps were now fully a thing. I myself had gotten into the HE-6 and all the stupid amping requirements behind it so "normal" headphone amps weren't really on my radar anymore. Casework now cost more than a finished mid-tier amp. I used to make cables, and even fully braided stuff could now be purchased at a fraction of my raw materials cost, nevermind the labour. It wasn't so much a feeling of defeat as it was... apathy?

Maybe I'll just pick up another Bottlehead kit. I don't have that same drive to go experimenting with parts anymore. My latest DIY project has been learning how to apply window film to cut down on glare and heat in the house (fantastic investment by the way, I highly recommend it. $50 of materials and a few hours of work, my west facing rooms feel so much better now)

 

Yeah, this is the balance that needs to be struck. Balancing the nervosa, the lure of shiny new things, and actually finishing up projects. Fighting OCD and hoarding instincts is so important to sanity!

Indeed! For me I think it really comes down to how easy it is to be able to create something unique. That was no problem 10+ years ago, but it has become increasingly difficult. If you have any talent at actual circuit design, and do your own P2P tube amps or something like that, the situation is much less dire. Me, having zero circuit design talent, does not have that outlet. The only way for me to create something that makes practical and economic sense is through aesthetics or intangible functional ideals (minimalism, or balls-to-the-wall crazy). I think my M3 and upcoming DAC build are good examples of that. But man, aesthetics is a tough one. I do not have a proper workshop, and FPE has become eye-poppingly expensive.

Oh, absolutely. If I had been a few years older, and started in the hobby a few years earlier, I like to think I would have found my way to Headwize. I almost certainly would have taken more engineering at university instead of biology as a result! In any case, I still owe so much to that community.

Oh, do not get me wrong, I can do SMD soldering, I am pretty good down to about 0805 for single parts, and SSOP28 has been no problem for chips. But I just don't like doing SMD. I like stuffing boards. I like bending leads at perfect 90° angles, with the part number facing upwards, with perfect spacing between part and board. SMD is, in a word, boring. But yeah, it is the future, so maybe I will just have to find a way to enjoy it.

I have always wanted to do a big nutty speaker amp, something like one of the Pass Labs First Watt clones, or one the Bottlehead 2A3 or 300B kits. Never going to happen though. My speaker rig is inherently tied up with my TV, cable box and Xbox, and there is just no wife acceptance factor. I am also way too addicted to the ease and reliability of operating a commercial integrated amp.

The bigger Bottlehead headphone amps do not hold much appeal, because they are already too good from the start. Where's the fun in a whole boatload of effort for the last 1%? That is why the Crack is so good, taking something that is just good, and making it extraordinary.

 

Well, there's always the other side of the nuttery which is "do I really want to buy all the tools and machines for a metals and woodshop to make my own casework?"

 

Heh, fair call. I reckon 90% of what I need to do could be achieved with a drill press and a scroll saw. But that last 10% does get very complicated and expensive.

 

My garage has a 240V plug at the back and a small side room that was used as a woodshop by the previous owners. I briefly considered, but then decided... no, this way lies madness

Actually it's kinda neat, when I was speaking with my neighbours when I first moved in, a lot of guys on the street a generation ago had their own little repair niches carved out to fix whatever was needed in the neighbourhood. Mine and my neighbour were the woodworking guys, the plumbing guy was two houses down, the electrical guy was across the street from him, etc. Those days are long gone though; the old guard are all in nursing homes or passed on now.

 

Have you thought about buying test equipment? An oscilloscope is really helpful -- it would keep you from having to say that a mod "supposedly" reduces power-supply noise, for example -- and various types are available for basically any budget.

And beyond that, you could buy an audio analyzer -- either purpose-built hardware or something PC-based. A decent PC-based system wouldn't be cheap, but maybe it wouldn't cost TOO much... $350 for a mic, $250 for a calibrator, $300 for an acoustic coupler, $600 for the software, $300 for a USB audio interface: That's $1800, plus a PC. Or for $300 plus PC, you can take basic more-or-less repeatable measurements with a miniDSP EARS and REW. And TIL you can buy an occluded-ear coupler with built-in mic -- allegedly meeting the appropriate specs -- on Taobao for like $100 shipped to the US. See this page for a $150 (plus iPhone) measurement rig that uses it: https://www.headphonesty.com/2020/10/how-to-measure-iems/

I would expect that last one to be useful only for -- at most -- comparisons between two measurements taken on that rig; you couldn't compare your measurements to anyone else's. But still, for $150 it would at least be fun.

Anyway, I guess my point is that if you don't already have that sort of test equipment, acquiring some will give you a different perspective on DIY/modding and/or purchasing, and it might make the hobby feel new again.

 

To catch up on some of the best gear highlighted from the past half decade.: https://www.superbestaudiofriends.org/index.php?threads/golden-schlong-awards-2020.10595/

Hottest budget finished headphone amp atm is SW51+. Amp guru @atomicbob uses it. Limited supply though.

 

Yes, great suggestion, I have absolutely considered an oscilloscope. Just to be able to look for PSU ripple and amp oscillation (concerns for my Exstata in particular) would be incredibly handy. And just looking at sine and square waves for strange behaviour would give a lot of confidence that I hadn't borked up any builds. Hell, even a better multimeter that can properly read AC on top of a DC signal would be really nice. I'll see what the budget is like after I finish this little sojourn.

Beyond that, with an audio analyser? Yeeeeeah, I can't see myself going that way. It's the sort of thing where if I couldn't compete with the big kids, I wouldn't ever be happy. And having actual data would start to feel uncomfortably like work!

Yep, read through all that going back to the start. Has been super helpful!

 

A half decent audio interface like that MOTU makes for a decent spectrum analyzer, if you know what to measure. And they are useful for acoustic measurements as well.

I too have lusted for an oscilloscope for some time, but as it turns out, the engineering dept. at my work is happy to lend me some scope time.

 

Did you notice this thread?
https://www.superbestaudiofriends.o...scilloscope-educational-interest-check.10626/

I actually went and ordered a pico in anticipation. I was unemployed at the time and figured I'd have all the time in the world to learn a new skill and maybe have something else to add to the resume. (Un)fortunately, I've since acquired some meaningful employment and sadly have not even taken the Pico out of the box since it arrived, though I'm hoping to squirrel away some time before Bob's zoom tutorials.

Either way, I'd be happy to lend you the gear if you want to test things. I don't really have much to test at the moment. Also also, I've got a Focusrite Clarett that's kinda on semi-permanent loan to sbaf people for measurements if you ever go that way. It's floating around the US at the moment (I think vtory has it) and alas getting things back into Canada is always the hard part.

 

Can't access that thread, I'm still just a lowly pleb. But from the URL it seems like it is a Pictoech 2204, 10 Mhz, 100 M samples/sec. Could be interesting!

 

What was your Head-Fi handle? to expedite friend status. Also there is contributor status which I think also unlocks club room and loaner subforum.

 

I've always been Beefy!

Fair warning, I was a huge jackass while I was on Head-Fi. I'm still a huge jackass, but I like to think I have learned to hide it better.