New Millett Hybrid Maxed Amp - Page 3

PM "cetoole".

Tomb,

Don't take offense. I think this new Max board is great and much needed. I think the Millett is an unsung hero and deserves the new CCS and class A buffer.

Just understand that when you get to higher parts cost and designs the limitations of an LM317 become more apparent. There is nothing wrong with building this as is, but there should be an improvement with a S11. AMB said he can't measure any difference vs the LM317, but you can't measure the benefits of Blackgates either but I'll bet $'s you have used them.

As far as Steps vs S11, the question is only where does a S11 become worth the extra parts cost? There is a point where a discrete PS is worth it or people would be using LM317's on everything. We don't know till someone tries each.

No one is trashing anything. Settle down and realize that if you or I owned a $15000 tube amp, we would look to mod it. And where would I start? The power supply. That's ANY design.

Nowhere did I say one was better than the other, only that there is nothing wrong with Regal's suggestion. As for the STEPs being unworthy, I have several that I use on a daily basis. Regarding the direct comparison between the power supplies, it is you who is ultimately setting the challenge with such bold statements as "the STEPS was the best thing going for any PS, anywhere."

On another note, the MAX is a great design whose time has come. The fact that it integrates so many ideas into a singular package is brilliant. It is very well thought out, offering DIYers everything that they could want in a Millett. However, this doesn't mean there is no more room for improvement/tweaking.

You've done a great job on the MAX but you have to learn relax and not to take everything to heart.

Peace

That's not a "bold statement". It was a general consensus in the headphone DIY community. Please do not quote me out of context: you very conveniently left out the part of "up until a few weeks ago." This is referring to the fact that Amb's Sigma11 is NEW - as in didn't exist a few weeks ago.

This is not a helpful or productive line of discussion, however. We are in agreement that the Max is a special design. So, you guys buy some boards from cetoole and chop them up to your heart's content.

OK, I wasn't going to post this again after Colin's fine photo and because I also posted it in the sticky thread at top. However, maybe it will get attention back where it should be: on the Max. This is my version of the 1st prototype. The only issue was a little hum and ripple in the PS, which we solved (totally unrelated to the discussion above). I am running 27V from an 800ma supply. I also have a 1000ma supply and it runs a little cooler. It pulls about 600ma on inrush and 430ma on idle. The DB's are 2SC3422/2SA1359 biased at about 50ma (115mV) ea. Caps are Pana FC's in the PS, Pana FM's on board with Nichicon Muse ES and Wima MKP10's in boutique positions.

I have lived with this amp every day since March 11th - turning it on every night for several hours, and for most of the day and night every weekend. Listening is mostly through an M-Audio Transit. Music sources tested also include a 1st generation iPod, a Sony D-NE20, and various other PCDP's, including an old "D" model Sony. The combination of the Max with Sony D-NE20 and older D model (using Line Out) highlighted a deficiency in high bit-rate mp3's - something not easily noticed on my other amps. Ever since, I have been using FLAC files almost exclusively.

In total, it sounds better than any previous Millett - DB's or not (I've built 4). It also sounds better than some other popular DIY amps, but they will remain un-named. Besides the ever-present KSC75's, it drives an HD580 extremely well - detail and slam in abundance. The headphones also stay plugged in without worry of power up or power down. Hopefully, these remarks are free enough from unnecessary embellishment.

Tomb, your build looks very nice. Are you casing in the Hammond? Whenever you get it cased, please post some more ****! I'm pretty set on this for my next project, but we'll see if there are any more changes to the prototype. As of now, do you think there will be any changes for the final board?

regal, I appreciate your contributions to the discussion but can you honestly say that you've heard the difference in moving from a STEPS to a better power supply? Which PS? And we're still talking about a sub $200 build even using some boutique parts so I'd hardly say we're into the "higher parts cost" realm of things. The limiting factor with the Millett design is always going to be the tubes. Yes, I completely agree that moving from a wallwart to a good regulated power supply is a huge improvement but I think in this case the designers (cetoole) did a very good job of selecting a complementary topology and didn't throw money at areas just so that they might measure slightly better. Again, I'd ask that you recognize that the STEPS in commercial terms is a highly advanced power supply.

You're more than welcome to and I'd be interested to hear what you think but some parts of your earlier posts essentially said that if someone was going to bother designing a better Millett than they should have considered a better power supply. A large design consideration (as far as I could tell) with the Millett Max was to bring together known-good existing topologies so that the prototyping expenses kept to a minimum. Please understand that in large part the prototyping costs are born by folks like tomb and myself and not passed onto you, the casual builder. So you'll have to give us a bit of a break if we get a bit tweaked when we go to the trouble of donating money, time and not a little bit of effort into something only to have someone refer to the end result with a "why did you bother if you're not going to do it well" type of reply.

Again I would ask that you read what I wrote above and then re-read your previous posts. I think you'll see where tomb is coming from.

Please point me to another widely used DIY power supply design that both existed and was fully supported two months ago. The STEPS was the best DIY thing going and based on a case by case basis it very may well still be.

I think the idea here is how about letting us get done with this one before you going "tweaking it"? Some of us have been working on/with the Millett for the better part of 2 years now so it's not as if we didn't go into this project without doing a bit of research.

Way more than my 2¢ but you don't have to read it if you don't want to.

Nate

Never even thought such a thing ! The problem with typing replies in forums is people want to jump to conclusions. Heck if I wasn't seriously impressed with this great improvement to the Millett I wouldn't have even made a post.

Just understand this is the first time Head-fiers have had any exposure on this and the question is pretty benign.

For those that don't know diyforums.org is basically closed to new members.

Will there be links to a parts list, an assembly guide, estimated final cost, etc. for this particular variation of the Millett?

Second question... I just built a CMoy; how much more knowledge would it take to assemble one of these?

Nate, All very valid points, I fully appreciate where you are coming from. I can also clearly see your point of view in regards to the STEPS. As for the statement "the STEPS was the best thing going for any PS, anywhere." I took Tomb's comment to mean the holy grail of power supply design period, not in the context you described. Thus, my comment that it was such a bold statement. I think I might be getting a little bit jaded from reading all of the FOTM/fanboy threads.

Tomb, In that light, I offer you my sincerest apology.

PS. Who doesn't like chopped Millett for dinner?

Thanks, my photo is just the 1st prototype. I only posted it to get this thread back on track, though. Attention should be paid to cetoole's first post - that's the newer prototype. The back of mine was eventually cut up a little bit in trying to solve the hum/ripple noise. That was fixed, and the changes are incorporated into the new version shown in cetoole's first post.

Several of us that are building the newer 2nd prototype - the one in cetoole's photo - will be casing them up in a 1455T1601 Hammond. The board is designed to fit in the slots on that case. We will defintiely post photos when they're boxed up.

According to cetoole's comments and experience thus far with the 2nd prototype, I doubt that there will be any changes to the production boards.

The links to the current BOM, both in Excel and HTML format was posted in Post #14 of this thread. More info than you'd ever want to know is available at the link in my signature below. Total cost, including case, is less than $150. You'll need to add other things such as jacks and knobs, but that price includes the walwart, too.

Build a PIMETA next, and perhaps a TREAD to go with it, then building the Max might be possible. There are difficulties dealing with the tubes, matching transistors, the heat sinks, biasing techniques, and then there's quite a parts count. It would not be my choice for a beginner or even a successful CMoy builder, but is well within the skills of someone who has built an M3, for instance.


P.S. Pobrecito - thanks for the comments. I perhaps reacted too strongly.

If anyone would like to join that forum all they have to do is shoot me a PM here.

Yes - I think we all know who "n_maher" is, and that name has been prominently displayed as a moderator at DIYForums.org for as long as I can remember.

I am trying not to fuel this thread anymore, but - deleted -

Aw, fudge - I'm gonna go do some yardwork.

Looks awesome !!!! I love my millett hybrid, but really want to build my own from scratch, and this looks great.

Few questions though:

1. If im reading things correct then the boards are still in the prototype phase. How long will these 2nd gen boards be available for and will there be quite a limited run ?

2. Most importantly 'THE SOUND'. My millett has cerafines throughout and 2 BG standars in directly in the path, plus it's running a tread. What sort of sound improvement am i looking at with this rig:

Heavily modded marantz cd-63 (probs sounds around the same as $1500-$2000 cd-players) -----------> New MH of course -------------> Grado HF-1 (will eventually get the BEST treatment at headphile like deep woody, blackmaxx recable etc)

I believe cetoole has about 10 boards left after sending them to the prototype builders. I may be wrong, though. Easiest thing to do if you want one - PM "cetoole".

The Cerafines and BG's are great, but there's more to the new Millett Hybrid Max than good caps. You didn't say if yours has the DB's (Diamond Buffers) - that's the single biggest thing that affects the sound on a Millett, besides the tubes, of course. Even so, the Max can run the DB's at a much higher current, and sounds still better than a regular Millett with DB's. There is also more going on with cetoole's circuit around the tube and CCS. Early builders so far are reporting that every detail comes through.

There is sufficient current and voltage, along with a choice of gain by using different tubes, to satisfy most headphone types - including Grados.